WO2012083274A2 - Methods for diagnosis, prognosis and methods of treatment - Google Patents

Abstract

The present invention provides an approach for the determination of the activation states of a plurality of proteins in single cells. This approach permits the rapid detection of heterogeneity in a complex cell population based on activation states, expression markers and other criteria, and the identification of cellular subsets that exhibit correlated changes in activation within the cell population. Moreover, this approach allows the correlation of cellular activities or properties. In addition, the use of modulators of cellular activation allows for characterization of pathways and cell populations. The process is effective at characterizing MDS.

Description

METHODS FOR DIAGNOSIS, PROGNOSIS AND METHODS OF TREATMENT

CROSS-REFERENCE

[0001] This application claims priority from U.S. Provisional Application Serial Nos. 61/565,935, filed December 1, 2011 and 61/423,918, filed December 16, 2010. This application also relates to U.S. Patent Application Serial No. 12/910,769, filed October 22, 2010, U.S. Patent Application Serial No. 13/083,156, filed April 8, 2011, and U.S. Patent Application Serial No. 12/713,165, filed February 25, 2010. Each of these applications is hereby expressly incorporated by reference in their entirety.

BACKGROUND OF THE INVENTION

[0002] Many conditions are characterized by disruptions in cellular pathways that lead, for example, to aberrant control of cellular processes, with uncontrolled growth and increased cell survival. These disruptions are often caused by changes in the activity of molecules participating in cellular pathways. For example, alterations in specific signaling pathways have been described for many cancers. Despite the increasing evidence that disruption in cellular pathways mediate the detrimental transformation, the precise molecular events underlying these transformations in diseases remain unclear. As a result, therapeutics may not be effective in treating conditions involving cellular pathways that are not well understood. Thus, the successful diagnosis of a condition and use of therapies will require knowledge of the cellular events that are responsible for the condition pathology.

[0003] Acute myeloid leukemia (AML), myelodysplasia syndrome (MDS), and myeloproliferative neoplasms (MPN) are examples of disorders that arise from defects of hematopoietic cells of myeloid origin. These hematopoietic disorders are recognized as clonal diseases, which are initiated by somatic and/or inherited mutations that cause dysregulated signaling in a progenitor cell. The wide range of possible mutations and accompanying signaling defects accounts for the diversity of disease phenotypes and response to therapy observed within this group of disorders. For example, some leukemia patients respond well to treatment and survive for prolonged periods, while others die rapidly despite aggressive treatment. Some patients with

myelodysplastic syndrome suffer only from anemia while others transform to an acute myeloid leukemia that is difficult to treat. Despite the emergence of new therapies to treat these disorders the percentage of patients who do not benefit from current treatment is still high. Patients that are resistant to therapy experience significant toxicity and have very short survival times. While various staging systems have been developed to address this clinical heterogeneity, they cannot accurately predict at diagnosis the prognosis or predict response to a given therapy or the clinical course that a given patient will follow.

[0004] Accordingly, there is a need for a biologically based clinically relevant re-classification of these disorders that can inform on disease management at the individual level. This classification, based upon the biologic commonalities of the disorders above, will aid clinicians in both prognosis and therapeutic selection at the individual patient level thus improving patient outcomes e.g. survival and quality of life.

[0005] There are also needs for a biologically based clinically relevant re-classification of these disorders to aid in new drug target identification and drug screening for agents that may be active against myeloid malignancies.

SUMMARY OF THE INVENTION

[0006] One embodiment of the present invention is a method of determining the progression of myelodysplasia syndrome (MDS) to acute myeloid leukemia (AML) in an individual, comprising: contacting one or more MDS cells (bone marrow or peripheral blood cells from an individual with a diagnosis or suspected diagnosis of MDS) with at least one modulator from the group comprising hematopoietic stem cell growth factors; contacting one or more MDS cells from the individual with at least one modulator from the group comprising lineage-specific hematopoietic factors comprising TPO, EPO, G-CSF and M-CSF; determining an activation level of one or more activatable elements from the following pathways, comprising the group consisting essentially of PBKinase, MAPK, or JAK/STAT; determining an activation level of one or more activatable elements from a cell cycle pathway; and determining the progression of MDS to AML.

[0007] Another embodiment of the present invention is a method of determining the progression of MDS to AML in an individual, comprising: contacting one or more MDS cells from an individual with at least one modulator from the group comprising FLT3-L, SCF, G-CSF, GM- CSF, and IL-6; contacting one or more MDS cells with at least one modulator from the group comprising G-CSF, M-CSF, TPO, and EPO; determining an activation level of one or more activatable elements from the group comprising p-Erk, p-S6, p-AKT, p-STATl, p-STAT3, p- STAT5, or p-H2AX; determining an activation level of one or more activatable elements from the group comprising p-H3, Ki-67, cyclin A2 or cyclin Bl ; and determining the progression of MDS to AML.

[0008] A further embodiment is a method wherein the determining the progression of MDS to AML step comprises comparing the results of the determining of the activatable element steps to a profile that indicates progression to AML. The method further comprises using age-matched controls. The activation state can be selected from the group consisting of extracellular protease exposure, novel hetero-oligomer formation, glycosylation state, phosphorylation state, acetylation state, methylation state, biotinylation state, glutamylation state, glycylation state, hydroxylation state, isomerization state, prenylation state, myristoylation state, lipoylation state,

phosphopantetheinylation state, sulfation state, ISGylation state, nitrosylation state,

palmitoylation state, SUMOylation state, ubiquitination state, neddylation state, citrullination state, deamidation state, disulfide bond formation state, proteolytic cleavage state, translocation state, changes in protein turnover, multi-protein complex state, oxidation state, multi-lipid complex, and biochemical changes in cell membrane. The activation state can be a

phosphorylation state and the activatable element can be selected from the group consisting of proteins, carbohydrates, lipids, nucleic acids and metabolites. The activatable element can be a protein capable of being phosphorylated and/or dephosphorylated. The method can further comprise determining the presence or absence of one or more cell surface markers, intracellular markers, or combination thereof and the intracellular markers can be independently selected from the group consisting of proteins, carbohydrates, lipids, nucleic acids and metabolites.

[0009] One embodiment of the invention is a method wherein said activation level is determined by a process comprising the binding of a binding element which is specific to a particular activation state of the particular activatable element and in one embodiment the binding element can comprise an antibody. The method can also include the use of an instrument for flow cytometry, immunofluorescence, confocal microscopy, immunohistochemistry,

immunoelectronmicroscopy, nucleic acid amplification, gene array, protein array, mass spectrometry, patch clamp, 2-dimensional gel electrophoresis, differential display gel electrophoresis, microsphere-based multiplex protein assays, ELISA, and label- free cellular assays to determine the activation level of one or more intracellular activatable element in single cells.

[0010] Another embodiment of the invention provides a method of choosing a treatment for an individual with a diagnosis or suspected diagnosis of MDS, comprising contacting one or more bone marrow or peripheral blood cells from the individual with a therapeutic agent; determining a level of DNMT 1 , DNMT 3a, or DNMT 3b in one or more cells from the individual; determining an activation level of one or more activatable elements in the cells which indicates cytostasis; determining an activation level of one or more activatable elements in the cells which indicates cytotoxicity; determining a level of CD34+ cells; wherein the determining steps comprise a first result; repeating the determining steps at a later time point to create a second result; comparing the first and second results; and making a decision regarding a therapy based on the results of the comparison. Another embodiment of the invention provides a method of choosing a treatment for an individual with a diagnosis or suspected diagnosis of MDS, comprising contacting one or more bone marrow or peripheral blood cells from the individual with at least one modulator from the group Lenalidomide, Azacitidine, Decitabine or Vorinostat; determining a level of DNMT 1, DNMT 3a, or DNMT 3b in the cells; determining an activation level of one or more activatable elements in the cells from the group comprising cyclin Bl , cyclin A2, Ki-67, cPARP, Amine Aqua or p-H2AX; determining a level of CD34+ cells; wherein the determining steps comprise a first result; repeating the determining steps at a later time point to create a second result;

comparing the first and second results; and making a decision regarding a therapy based on the results of the comparison. [0011] Another embodiment of the invention provides a method of screening a drug that is in development as a candidate therapeutic to treat an MDS patient, comprising contacting one or more bone marrow or peripheral blood cells from an individual with a diagnosis or suspected diagnosis of MDS with a candidate therapeutic either in vivo or in vitro; determining a level of DNMT 1, DNMT 3a, or DNMT 3b in the cells; determining an activation level of one or more activatable elements in the cells from the group comprising cyclin Bl, cyclin A2, Ki-67, cPA P, Amine Aqua or p-H2AX; determining a level of CD34+ cells; wherein the determining steps comprise a first result; repeating the determining steps at a later time point to create a second result; comparing the first and second results; and determining if the candidate therapeutic is effective as a treatment for MDS.

[0012] The first and second determining times can be between 0 and 36 hours for the first step and between 10 and 80 hours for the second step or increments of 15 minutes therein.

[0013] Other aspects described above for the AML progression embodiment can apply to the therapy and screening embodiments.

INCORPORATION BY REFERENCE

[0014] All publications, patents, and patent applications mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent, or patent application was specifically and individually indicated to be incorporated by reference.

BRIEF DESCRIPTION OF THE DRAWINGS

[0015] The novel features of the invention are set forth with particularity in the appended claims. A better understanding of the features and advantages of the present invention will be obtained by reference to the following detailed description that sets forth illustrative embodiments, in which the principles of the invention are utilized, and the accompanying drawings of which:

[0016] Figure 1 shows CD34+ frequency in healthy, Low Risk (LR) and High Risk (HR) MDS

patients.

[0017] Figure 2 shows Cleaved-PARP Levels (ERF) Measured at t=0.

[0018] Figure 3 shows Dacogen, Vidaza, and Zolinza Exhibit Differential Effects on DNMT (DNA Methyl-Transferase) Protein Levels: Healthy vs. High Risk MDS Samples.

[0019] Figure 4 shows Vidaza and Zolinza Disrupt Cell Cycle by Arresting CD34+ Cells in M Phase:

Increased p-H3 Levels.

[0020] Figure 5 shows Lenalidomide induces p-Stat3 signaling in Low Risk but not Healthy nRBC Subpopulations.

[0021] Figure 6 shows Distinct Vidaza (Azacitidine) & Dacogen (Decitabine ) Responses Observed in AML Samples.

[0022] Figure 7 shows ONI 03 Frequency of CD34+ Cells - Whole Bone Marrow for patient 103. [0023] Figure 8 shows ONI 03 Frequency of CD34+ Cells - Whole Bone Marrow for patient 104.

[0024] Figure 9 shows FLT3L Responsiveness - Fold.

[0025] Figure 10 shows SCF Responsiveness - Fold.

[0026] Figure 11 shows G-CSF Responsiveness - Fold.

[0027] Figure 12 shows GM-CSF Responsiveness - Fold.

[0028] Figure 13 shows an early block in RBC development in LR MDS.

[0029] Figure 14 shows no difference in CD34+ cell frequency or signaling with age.

[0030] Figure 15 shows increased nRBC frequency, but decreased EPO responsiveness with age.

[0031] Figure 16 shows a table with the functional characterization of LR MDS.

[0032] Figure 17 shows RAEB with high percent nRBC vs. healthy have hyper response to EPO.

[0033] Figure 18 shows RARS with high percent nRBC vs. healthy have low to normal EPO

response.

[0034] Figure 19 shows LR MDS with high percent myeloid vs. healthy have hyper response to GCSF while those with low percent myeloid v healthy have poor response.

[0035] Figure 20 shows a comparison of healthy and low risk MDS in modulated and unmodulated states.

[0036] Figure 21 shows a visual relationship of MDS cell types using SPADE.

DETAILED DESCRIPTION OF THE INVENTION

[0037] The present invention incorporates information disclosed in other applications and texts. The following patent and other publications are hereby incorporated by reference in their entireties: Haskell et al, Cancer Treatment, 5^th Ed., W.B. Saunders and Co., 2001; Alberts et al., The Cell, 4^th Ed., Garland Science, 2002; Vogelstein and Kinzler, The Genetic Basis of Human Cancer, 2d Ed., McGraw Hill, 2002; Michael, Biochemical Pathways, John Wiley and Sons, 1999;

Weinberg, The Biology of Cancer, 2007; Immunobiology, Janeway et al. 7^th Ed., Garland, and Leroith and Bondy, Growth Factors and Cytokines in Health and Disease, A Multi Volume Treatise, Volumes 1A and IB, Growth Factors, 1996. Other conventional techniques and descriptions can be found in standard laboratory manuals such as Genome Analysis: A

Laboratory Manual Series (Vols. I-IV), Using Antibodies: A Laboratory Manual, Cells: A Laboratory Manual, PCR Primer: A Laboratory Manual, and Molecular Cloning: A Laboratory Manual (all from Cold Spring Harbor Laboratory Press), Stryer, L. (1995) Biochemistry (4th Ed.) Freeman, New York, Gait, "Oligonucleotide Synthesis: A Practical Approach" 1984, IRL Press, London, Nelson and Cox (2000), Lehninger, Principles of Biochemistry 3rd Ed., W. H. Freeman Pub., New York, N.Y. and Berg et al. (2002) Biochemistry, 5th Ed., W. H. Freeman Pub., New York, N.Y.; and Sambrook, Fritsche and Maniatis. "Molecular Cloning A laboratory Manual" 3rd Ed. Cold Spring Harbor Press (2001), all of which are herein incorporated in their entirety by reference for all purposes. [0038] Patents and applications that are also incorporated by reference in their entirety include U.S. Patent Nos. 7,381,535, 7,393,656, 7,695,924 and 7,695,926 and U.S. Patent Application Nos. 10/193,462; 1 1/655,785; 1 1/655,789; 1 1/655,821 ; 1 1/338,957, 12/877,998; 12/784,478;

12/730, 170; 12/703,741 ; 12/687,873; 12/617,438; 12/606,869; 12/713, 165; 12/293,081 ;

12/581 ,536; 12/776,349; 12/538,643; 12/501 ,274; 61/079,537; 12/501,295; 12/688,851 ;

12/471 , 158; 12/910,769; 12/460,029; 12/432,239; 12/432,720; 12/229,476; 12/877,998;

13/083, 156; 61/469812; 61/436,534; 61/423,918; 61/557,831 ; 61/542,910; 61/499, 127;

61/317, 187; and 61/353, 155; 61/515,660; PCT No. PCT/US201 1/029845; and PCT No.

PCT/US201 1/48332.

[0039] Some commercial reagents, protocols, software and instruments that are useful in some

embodiments of the present invention are available at the Becton Dickinson Website

http://www.bdbiosciences.com/features/products/, and the Beckman Coulter website, http://www.beckmancoulter.com/Default.asp ?bhfv=7. Relevant articles include High-content single-cell drug screening with phosphospecific flow cytometry, Krutzik et al., Nature Chemical Biology, 23 December 2007; Irish et al., FLt3 ligand Y591 duplication and Bcl-2 over expression are detected in acute myeloid leukemia cells with high levels of phosphorylated wild-type p53, Neoplasia, 2007, Irish et al. Mapping normal and cancer cell signaling networks: towards single- cell proteomics, Nature, Vol. 6 146-155, 2006; and Irish et al., Single cell profiling of potentiated phospho-protein networks in cancer cells, Cell, Vol. 1 18, 1 -20 July 23, 2004; Schulz, K. R., et al., Single-cell phospho-protein analysis by flow cytometry, Curr Protoc Immunol, 2007, 78:8 8.17.1 -20; Krutzik, P.O., et al., Coordinate analysis of murine immune cell surface markers and intracellular phosphoproteins by flow cytometry, J Immunol. 2005 Aug 15; 175(4):2357-65; Krutzik, P.O., et al., Characterization of the murine immunological signaling network with phosphospecific flow cytometry, J Immunol. 2005 Aug 15; 175(4):2366-73; Shulz et al., Current Protocols in Immunology 2007, 78:8.17.1 -20; Stelzer et al. Use of Multiparameter Flow

Cytometry and Immunophenotyping for the Diagnosis and Classification of Acute Myeloid Leukemia, Immunophenotyping, Wiley, 2000; and Krutzik, P.O. and Nolan, G. P., Intracellular phospho-protein staining techniques for flow cytometry: monitoring single cell signaling events, Cytometry A. 2003 Oct;55(2):61 -70; Hanahan D. , Weinberg, The Hallmarks of Cancer, CELL, 2000 Jan 7; 100(1) 57-70; Krutzik et al, High content single cell drug screening with

phophospecific flow cytometry, Nat Chem Biol. 2008 Feb;4(2): 132-42. Experimental and process protocols and other helpful information can be found at http:/proteomices. stanford.edu. The articles and other references cited below are also incorporated by reference in their entireties for all purposes.

[0040] One embodiment of the present invention involves the classification, diagnosis, selection of a therapeutic to treat MDS, prognosis of disease or outcome after administering a therapeutic to treat MDS. Another embodiment of the invention involves monitoring and predicting outcome of disease. Another embodiment is the progression of MDS to AML. Another embodiment is drug screening using some of the methods of the invention, to determine which drugs may be useful in particular diseases. In other embodiments, the invention involves the identification of new draggable targets, that can be used alone or in combination with other treatments. The invention allows the selection of patients for specific target therapies. The invention allows for delineation of subpopulations of cells associated with a disease that are differentially susceptible to drugs or drug combinations. In another embodiment, the invention allows to demarcate subpopulations of cells associated with a disease that have different genetic subclone origins. In another embodiment, the invention provides for the identification of a cell type, that in combination other cell type(s), provide ratiometric or metrics that singly or coordinately allow for surrogate identification of subpopulations of cells associated with a disease, diagnosis, prognosis, disease stage of the individual from which the cells were derived, response to treatment, monitoring and predicting outcome of disease. Another embodiment involves the analysis of apoptosis, DNA methyl transferase (DNMT) proteins, drug transport and/or drug metabolism. In performing these processes, one preferred analysis method also involves looking at cell signals and/ or expression markers. One embodiment of cell signal analysis involves the analysis of phosphorylated proteins by the use of flow cytometers in that analysis. In one embodiment, a signal transduction-based classification of MDS can be performed using clustering of phospho- protein patterns or biosignatures.

[0041] In some embodiments, the present invention provides methods for classification, diagnosis, prognosis of disease and outcome after administering a therapeutic to treat the disease by characterizing a plurality of pathways in a population of cells, especially cells that show dysplasia or are dysmorphic or cells that show other characteristics of the diagnosis of MDS. In some embodiments, a treatment is chosen based on the characterization of plurality of pathways in single cells. In some embodiments, characterizing a plurality of pathways in single cells comprises determining whether apoptosis pathways, cell cycle pathways, DNMT proteins, signaling pathways, or DNA damage pathways are functional in an individual based on the activation levels of activatable elements within the pathways, where a pathway is functional if it is permissive for a response to a treatment. For example, when the apoptosis, cell cycle, signaling, and DNA damage pathways are functional the individual can respond to treatment, and when at least one of the pathways is not functional the individual does not have disease that respond to treatment. In some embodiments, if the apoptosis and DNA damage pathways are functional the individual has disease that is responsive to treatment.

[0042] In some embodiments, the characterization of pathways in conditions such as MDS shows disruptions in cellular pathways that are reflective of increased proliferation, increased survival, evasion of apoptosis, insensitivity to anti-growth signals and other mechanisms. In some embodiments, the disruption in these pathways can be revealed by exposing a cell to one or more modulators that mimic one or more environmental cues. Biology determines response to therapy. For example, without intending to be limited to any theory, responsive cells treated with a drug will undergo cell death through activation of DNA damage and apoptosis pathways. However, a non-responsive cell might escape apoptosis through disruption in one or more pathways that allows the cell to survive. For instance, a non-responsive cell might have increased concentration of a drug transporter (e.g., MPR-1), which causes the drug to be removed from the cells. A non- responsive cell might also have disruptions in one or more pathways involve in proliferation, cell cycle progression and cell survival that allows the cell to survive. A non-responsive cell may have a DNA damage response pathway that fails to communicate with apoptosis pathways. A non-responsive cell might also have disruptions in one or more pathways involve in proliferation, cell cycle progression and cell survival that allows the cell to survive. The disruptions in these pathways can be revealed, for example, by exposing the cell to a growth factor such as FLT3L or G-CSF. In addition, the revealed disruptions in these pathways can allow for identification of target therapies that will be more effective in a particular patient and can allow the identification of new draggable targets, which therapies can be used alone or in combination with other treatments. Expression levels of proteins, such as drug transporters and receptors, may not be as informative by themselves for disease management as analysis of activatable elements, such as phosphorylated proteins. However, expression information may be useful in combination with the analysis of activatable elements, such as phosphorylated proteins.

[0043] In some embodiments of the invention, cell health is considered in analyzing the results. See for example PCT /US201 1/48332 and U.S. Patent App. Ser. No. 61/436,534. Also, quality control methods are part of the present invention. See U.S. Provisional App. Ser. No.

61/557,831.

[0044] A significant fraction of cells with high cleaved PARP levels or low MCL-1 levels, before or without treatment with, e.g., a modulator, can indicate that some cells are undergoing apoptosis before treatment with a modulator. For some experiments, the activation state or activation level of an activatable element in an untreated sample of cells may be attributable to cells undergoing apoptosis due to one or more reasons related to sample processing (e.g., shipment conditions, cryogenic storage, thawing of cryogenically stored cells, etc.). If the apoptotic cells are not physically removed from the analysis, or data from apoptotic cells is not removed from an analysis of cell signaling data, apoptotic cells (which can be cleaved PARP positive or MCL-1 negative) can negatively impact the measurement of treatment (e.g., with a modulator) induced activation of an activatable element, e.g., phosphorylation of a phosphorylation site, and cause a misleading view of the signaling potential for the specific cell population being studied.

[0045] The discussion below describes some of the preferred embodiments with respect to particular diseases. However, it should be appreciated that the principles may be useful for the analysis of many other diseases as well. [0046] Introduction

[0047] Hematopoietic cells are blood-forming cells in the body. Hematopoiesis (development of blood cells) begins in the bone marrow and depending on the cell type, further maturation occurs either in the periphery or in secondary lymphoid organs such as the spleen or lymph nodes. Hematopoietic disorders are recognized as clonal diseases, which are initiated by somatic and/or inherited mutations that cause dysregulated signaling in a progenitor cell. The wide range of possible mutations and accompanying signaling defects accounts for the diversity of disease phenotypes observed within this group of disorders. Hematopoietic disorders fall into three major categories: Myelodysplastic syndromes, myeloproliferative disorders, and acute leukemias. Examples of hematopoietic disorders include non-B lineage derived, such as acute myeloid leukemia (AML), Chronic Myeloid Leukemia (CML), non-B cell acute lymphocytic leukemia (ALL), myelodysplastic disorders, myeloproliferative disorders, polycythemias, thrombocythemias, or non-B atypical immune lymphoproliferations. Examples of B-Cell or B cell lineage derived disorder include Chronic Lymphocytic Leukemia (CLL), B lymphocyte lineage leukemia, Multiple Myeloma, acute lymphoblastic leukemia (ALL), B-cell pro- lymphocytic leukemia, precursor B lymphoblastic leukemia, hairy cell leukemia or plasma cell disorders, e.g., amyloidosis or Waldenstrom's macroglobulinemia.

[0048] Acute myeloid leukemia (AML), myelodysplastic syndrome (MDS), and myeloproliferative neoplasms (MPN) are examples of distinct myeloid hematopoietic disorders. However, it is recognized that these disorders share clinical overlap in that 30% of patients with MDS and 5- 10% of patients with MPN will go on to develop AML.

[0049] Myelodysplastic Syndromes (MDS)

[0050] Myelodysplastic syndromes (MDS) constitute a heterogeneous group of hematologic disorders characterized by ineffective hematopoiesis and dysplasia with varying risks of transformation to acute myeloid leukemia (AML). In addition, evidence of a cellular immunologic response has been implicated in the pathogenesis of a subset of MDS patients (Melchert, et al., Current Opinion in Haematology 2007 Vol. 14, p 123-129.).

[0051] MDS is predominantly a disease of the elderly. Median age of diagnosis MDS is 68 years.

MDS has an overall age-adjusted annual incidence of 3.3 per 100,000, and the rate increases with age to 10 per 100,000 among those aged 70 years or older. Approximately 55%> of patients die within 3 years of diagnosis. (Rollison et al. Epidemiology of myelodysplastic syndromes and chronic myeloproliferative neoplasms in the United States, 2001-2004, using data from the NAACCR and SEER programs.) Blood (2008) vol. 112 (1) pp. 45-52) Patients with high-risk MDS generally survive for approximately one year. Morbidity and mortality are a result of complications of cytopenias or transformation to acute myeloid leukemia. One of the major morbidities of MDS found in the vast majority of ( -60-80%)) patients is symptomatic anemia, with associated fatigue. Other cytopenias include neutropenia (-50-60%) and thrombocytopenia ( -40-60%). Dysfunctional neutrophils cause an increased risk of infection. Decreased platelets are associated with bleeding. (PETER L. GREENBERG, et. al. Myelodysplasia Syndromes. The American Society of Hematology. 2002, p.136-61.)

[0052] Causes

[0053] The initiating event for MDS is DNA injury in a hematopoietic progenitor cell. The

disruption of genes that control the balance of growth and differentiation results in the clonal proliferation of defective progeny, which are eliminated by apoptosis before they fully mature. The excessive apoptosis contributes to the peripheral cytopenias characteristic of the MDS phenotype. Accumulated genetic damage, particularly anti-apoptotic mutations, may result in neoplastic transformation to acute leukemia. (AUI C, et. al. Pathogenesis, etiology and epidemiology of myelodysplasia syndromes. Haematologica. 1998, vol. 83, p.71 -86;

HELLSTROM-LINDBERG E, et. al. Achievements in understanding and treatment of myelodysplasia syndromes. Hematology (American Society of Hematology Education Program), 2000, p.110-132; BARRCTT J, et al. Myelodysplasia syndrome and aplastic anemia: diagnostic and conceptual uncertainties. Leukemia Research, 2000, vol.24, p.595-596.)

[0054] Similar to AML, MDS may develop in individuals who have been exposed to environmental or occupational toxins that increase the likelihood of somatic mutations, including, but not limited to: Cancer chemotherapy, e.g., alkylating agents and topoisomerase II inhibitors, excess ionizing radiation, e.g., atomic bombs and radiotherapy for malignant diseases, and industrial chemicals, e.g., benzene, pesticides, fertilizers, herbicides, heavy metals, stone and cereal dusts, nitro- organic explosives, petroleum and diesel derivatives, and organic solvents (benzene, toluene, xylene, and chloramphenicol).

[0055] Symptoms

[0056] MDS is characterized by cytopenias (anemia, neutropenia, thrombocytopenia) of any or all of the three hematopoietic lineages (red blood cells, white blood cells and platelets) with varying degrees of severity. The common symptoms include fatigue, bruising, and/or bleeding, pallor, ecchymosis, epistaxis, gingival bleeding, and bacterial infections. Patients may be asymptomatic at diagnosis. Bleeding (due to lack of platelets) and infection (due to lack of WBCs) are the two most serious complications in MDS patients. MDS is sometimes underdiagnosed, since patients suffering from mild to moderate anemia are attributed to a chronic disease or a mild renal insufficiency.

[0057] Diagnosis [0058] A combination of bone marrow cellular morphology (to detect multilineage dysplasia in the bone) and cytogenetics (to detect characteristic clonal abnormalities) is used for the diagnosis of MDS. Basic diagnostic criteria involve microscopic morphological examination of bone marrow using a variety of histological stains. Dysplasia, particularly of megakaryocytes, evidence of disruption of the normal marrow architecture, such as abnormal localization of immature precursors (ALIP), and an estimate of the blast percentage are important diagnostic findings in bone marrow examinations. Bone marrows are also examined for dysgranulopoiesis,

dysmegakaryocytopoiesis, and dyserythropoiesis. Dysgranulopoiesis include abnormalities in primary granules such as decreased or absent secondary granules, large granules or decreased staining, and nuclear abnormalities or increased blasts. Examples of dysmegakaryocytopoiesis include micromegakaryocytes, large mononuclear or binuclear forms, multiple small nuclei, and reduced numbers. Dyserythropoiesis is characterized by more than 15 percent ringed siderob lasts, nuclear fragments, multiple nuclei, nuclear lobation, internuclear bridges, megaloblastic erythropoiesis, macronormoblastic erythropoiesis, irregular cytoplasmic staining, or less than 5 percent erythroid cells. Such morphologic dysplasias are however not specific for MDS. Mild megaloblastic changes without dyspoiesis in other cell lines are not considered sufficient for a diagnosis ofMDS.

[0059] In addition to a bone marrow aspirate with biopsy, and a CBC with differential, one usually orders a reticulocyte count, serum EPO,ferritin, B12, and folate to differentiate other causes and to optimize treatment of the anemia. Other helpful tests in MDS include HLA typing (if platelet support and/or potential marrow transplant), HLA-DR 15 typing (for possible administration of immunosuppressive therapies), FLAER test (to differentiate MDS from a PNH clone), and a JAK2 mutation if the patient has thrombocytosis (to differentiate essential thrombocythemia).

[0060] Deletions or amplifications of large chromosomal regions are more commonly observed in MDS, compared with the balanced translocations commonly observed in de novo AML.

Cytogenetic data help stratify patients in terms of diagnosis and evaluating prognosis for survival and risk of transformation to AML (HOFMAN WK, et al. Myelodysplasia syndrome. Annual Review of Medicine. 2005, vol.56, p.1 -16). Characteristic chromosomal deletions involve chromosome 5 [del(5q),-5], chromosome 1 1 [del(l l q)], chromosome 12 [del(12q)], chromosome 20 [del(20q)], chromosome 7 [del(7q),-7], chromosome 17 [del(17p)], and chromosome 13

[del(13q)]. Other frequent structural and/or numerical chromosomal aberrations include trisomy 8, trisomy 21 , and inversion 3(q21 q26). Rare reciprocal translocations include t(l;7)(ql 0;pl 0), t(l; 3)(p36;q21), t(3;3)(q21 ;q26), t(6;9)(p23;q34), and t(5; 12)-fusion between PDGFR-β and TEL(ETV-6), (q33;pl 3); t(5;7)(q33; l 1.2).

[0061] Deletion of chromosomal region 5q31 (5q-) is the most frequent genetic lesion in MDS and is present in more than 20 percent of MDS patients, garnering its own WHO classification. The pathogenic event associated with this genetic lesion has been traced to the hemizygous deletion of RPS14, which encodes a ribosomal subunit protein, and is also implicated in Diamond-Blackfan anemia. (Ebert BL, et al. Identification of RPS 14 as a 5q- syndrome gene by RNA interference screen. Nature, 2008, Vol. 451 , No. 17, pp 335-340)

[0062] A chromosomal abnormality commonly implicated in the progression of MDS is monosomy 7q. (STEPHENSON J, et al. Possible co-existence of RAS activation and monosomy 7 in the leukemic transformation of myelodysplastic syndromes. Leukemia Research, 1995, vol.19, p.741-8). While 5q- is associated with favorable prognosis, uniparental disomy in 7q confers substantially lower prognosis (3 months vs. 39 months survival). (Itzykson R, et al. Meeting report: myelodysplastic syndromes at ASH 2007. Leukemia. 2008, Vol. 22, pp 893-897).

[0063] Low risk (LR) MDS is characterized by cytopenias that arise through inefficient

hematopoiesis. In the majority of cases anemia is an early and prominent clinical finding despite the presence of normal to elevated levels of serum EPO and EPOR (Jacobs BJH73 : 1989, Backx Leuk 10: 1996). However, when bone marrow (BM) cells from patients with MDS are cultured in the presence of EPO, erythroid colony formation is reduced compared to healthy controls (Baines Leuk Res 14: 1990, Backx Leukemia 75: 1993). Furthermore, phosphorylation of STAT5 measured in nuclear extracts by EMSA is absent or greatly reduced in patients with MDS in response to in- vitro stimulation with EPO (Hoefsloot Blood 89: 1997). Despite the presence of cytopenias, MDS BM cells tend to be hypercellular. Increased proliferation of myeloid precursors in patients with MDS, with the potential for the acquisition of mutations, puts these patients at an increased risk for the evolution of the disease to acute myeloid leukemia.

[0064] A substantial fraction of MDS patients appear cytogenetically normal because they harbor submicroscopic chromosomal lesions. Recently, SNP array-based methods have been used to detect cryptic genetic lesions in this class of patients, although this is not yet standard in clinical practice. (Itzykson R, et al. Meeting report: myelodysplastic syndromes at ASH 2007.

Leukemia. 2008, Vol. 22, pp 893-897). Furthermore, molecular genotyping assays are now being used experimentally to screen for known pathogenic mutations to help stratify MDS patients.

[0065] In the context of MDS, multiparameter flow cytometry is used to measure abnormal light scatter properties of dysplastic cells, abnormal antigen density, loss of antigens, and asynchronous expression of antigens which are normally co-expressed during myeloid maturation, and these parameters may correlate to the grade of the disease. (STETLER STEVENSON M, et al.

Diagnostic utility of flow cytometric immunophenotyping in myelodysplastic syndromes. Blood. 2001 , vol.98, p.979-987.)

[0066] One embodiment of the invention combines one or more of these existing tests with the

analysis of signalling mediated by receptors to diagnose disease especially MDS. All tests may be performed in one location and provided as a single service to physicians or other caregivers. [0067] Cell-signaling pathways and differentiating factors involved

[0068] Regulation of hematopoiesis in MDS is complex and multiple factors are involved. Genetic alterations in signaling molecules have been extensively studied in MDS. These molecules include transcription factors, receptors for growth factors, RAS signaling molecules, and cell cycle regulators.

[0069] In the early stages of MDS, there is an increased frequency of apoptosis resulting in

intramedullary apoptotic bodies. Advanced MDS, which may transform to AML, is characterized by increased proliferation and antiapoptotic factors, such as mutations in p53, RAS, C-MPL or FMS. (Aul et al. Evaluating the prognosis of patients with myelodysplasia syndromes, Ann Hematol (2002) vol. 81 (9) pp. 485-97)

[0070] Genetic alterations in the RAS signaling pathway are frequently seen in MDS. The RAS

signaling pathway normally promotes cellular proliferation and differentiation. By contrast, pathogenic RAS pathway mutations generally cause continuous kinase activity and signal transduction. The cell surface receptor for macrophage colony stimulating factor (M-CSF), encoded by the FMS gene, normally promotes cellular proliferation and differentiation of monocyte and macrophages, and is upstream of RAS signaling. Activating mutations in this gene are found in 10% of MDS cases, and are associated with poor survival and increased risk of transformation to AML. (PADUA RA, et al. RAS, FMS and p53 mutations and poor clinical outcome in myelodysplasias: a 10-year follow-up. Leukemia, 1998, vol.12, p.887-892; TOBAL K, et al. Mutation of the human FMS gene (M-CSF receptor) in myelodysplasia syndromes and acute myeloid leukemia. Leukemia, 1990, vol.4, p.486-489.)

[0071] Activating mutations in FLT3, a receptor-type tyrosine kinase also upstream of RAS

signaling, have been reported in 3-5% of MDS cases. (Georgiou et al. Serial determination of FLT3 mutations in myelodysplasia syndrome patients at diagnosis, follow up or acute myeloid leukemia transformation: incidence and their prognostic significance. Br J Haematol (2006) vol. 134 (3) pp. 302-6) Inactivation of the neurofibromatosis type 1 (NF1) gene, normally a negative regulator of RAS signaling, has also been implicated in the progression of MDS. (Stephenson J, et al. Possible co-existence of RAS activation and monosomy 7 in the leukemic transformation of myelodysplasia syndromes. Leukemia Res 1995;19:741-8). Gain-of-function mutations have also been reported in PTPN11 in patients with MDS. (NEUBAUER A, et al. Mutations in the ras proto-oncogenes in patients with myelodysplasia syndromes. Leukemia. 1994, vol.8, p.638-641). Among the RAS genes themselves, mutations of the N-RAS (herein also "N-ras") gene are the most frequent and are detected in 20 to 30 percent of human leukemias and approximately 16 percent of MDS cases. K-RAS mutations are found at approximately half that frequency. The majority of studies suggest that RAS mutations in MDS are associated with poor survival and increased probability of developing AML. (YUNIS JJ, et al. Mechanisms of ras mutation in myelodysplasia syndrome. Oncogene. 1989, vol.4, p.609-614; Aul et al. Evaluating the prognosis of patients with myelodysplasia syndromes. Ann Hematol (2002) vol. 81 (9) pp. 485- 97).

[0072] Although less frequently, AML1 , C/EBPa, TEL (ETV6) and p53 genes are also a target of mutations in MDS. AML1 -binding sites exist upstream of several genes encoding factors and receptors that determine the lineage specificity of hematopoietic cells. (OKUDA T, et al. AML1 , the target of multiple chromosomal translocations in human leukemia, is essential for normal fetal liver hematopoiesis. Cell. 1996, vol.84, p.321-30.) C/EBPa is an important mediator of granulocyte differentiation and regulates the expression of multiple granulocyte-specific genes including the granulocyte colony-stimulating factor (G-CSF) receptor, neutrophil elastase and myeloperoxidase. C/EBPa knockout mice display a profound block in granulocyte differentiation (COLLINS SJ, et al. Multipotent hematopoietic cell lines derived from C/EBPa (-/-) knockout mice display granulocyte macrophage-colony-stimulating factor, granulocyte-colony-stimulating factor and retinoic acid- induced granulocytic differentiation. Blood. 2001, vol.98, p.2382-8). This suggests that any mutation in C/EBPa will result in defective hematopoiesis. TEL function is essential for the establishment of hematopoiesis of all lineages in the bone marrow, suggesting a critical role for TEL in the normal transition of the hematopoietic activity from fetal liver to bone marrow. Experiments conducted on the role of TEL genes indicate an ineffective hematopoiesis in the case of an alteration in these genes. (WANG LC, et al. The TEL/ETV6 gene is required specifically for hematopoiesis in the bone marrow. Genes and Development. 1998, vol.12, p.2392-402). Mutations or deletions causing inactivation of the p53 gene in both the alleles have been shown to predispose the cells to neoplastic transformation. Inactivation is detected in 5 to 10 percent of cases of clinically advanced MDS, indicating that p53 mutations may play a role in leukemic progression of MDS. (SUGIMOTO K, et al. Mutations of the p53 gene in MDS and MDS-derived leukemia. Blood. 1993, vol.81 , p.3022-6.)

[0073] Apoptotic genes (increased bcl-2 expression) (KUROTAKI H, et al. Apoptosis, bcl-2

expression and p53 accumulation in MDS, MDS derived acute myeloid leukemia and de novo acute myeloid leukemia. Acta Haematologica, 2000, vol.102, p. l 15-123.) And mutations in genes including CHK2, p53, MLL have been implicated in the pathogenesis of MDS (HOFMANN WK, et al. Mutation analysis of the DNA-damage checkpoint gene CHK2 in myelodysplasia syndromes and acute myeloid leukemias. Leukemia Research, 2001, vol.25, p.333-338;

KIKUKAWA M, et al. Study of p53 in elderly patients with myelodysplasia syndromes by immunohistochemistry and DNA analysis. American Journal of Pathology. 1999, vol.155, p.71 7-721 ; POPPE B, et al. Expression analyses identify MEL as a prominent target of 1 lq23 amplification and support an etiologic role for MLL gain of function in myeloid malignancies. Blood. 2004, vol.103, p.229-235.)

[0074] Dysregulation of genes that encode angiogenic factors involved in the growth of

hematopoietic cells may play important role in pathogenesis of MDS. (PRUNERI G, et al. Angiogenesis in myelodysplasia syndromes. British Journal of Cancer, 1999, vol.81 , p.1398- 1401.) The immunomodulatory cytokine, TNF-a has been shown to express strong inhibitory activity in hematopoiesis. (BROXMEYER HE, et al. The suppressive influences of human tumor necrosis factors on bone marrow hematopoietic progenitor cells from normal donors and patients with leukemia: synergism of tumor necrosis factor and interferon-gamma. Journal of

Immunology. 1986, vol.36, p.4487-4495.) Other cytokines reportedly involved in the processes leading to ineffective hematopoiesis in MDSs include TGF-β, IL-Ι β, and TNF -related signaling molecules TRADD/FADD, RIP, and TNF-related apoptosis inducing ligand (TRAIL)

(SAWANOBORI M, et al. Expression of TNF receptors and related signaling molecules in the bone marrow from patients with myelodysplasia syndromes. Leukemia Research, 2003, vol.27, p.583-591 ; PLASILOVA M, et al. TRAIL (Apo2L) suppresses growth of primary human leukemia and myelodysplasia progenitors. Leukemia, 2002, vol.16, p.67-73.)

[0075] One embodiment of the invention will look at any of the cell signaling pathways described above in classifying diseases, such as MDS. Modulators can be designed to investigate these pathways and any relevant parallel pathways.

[0076] In some embodiments, the invention provides a method for diagnosing, prognosing,

determining progression, predicting response to treatment or choosing a treatment for MDS or rationale combinations of drugs, or identification of new potentially draggable targets the method, the method comprising the steps of (a) subjecting a cell population from the individual to a plurality of distinct modulators, optionally, in separate cultures, (b) characterizing a plurality of pathways in one or more cells from the separate cultures comprising determining an activation level of at least one activatable element in at least three pathways, where the pathways are selected from the group consisting of apoptosis, cell cycle, signaling, or DNA damage pathways, and (c) correlating the characterization with diagnosing, prognosing, determining progression, predicting response to treatment or choosing a treatment for MDS, in an individual, where the pathways characterization is indicative of the diagnosing, prognosing, determining progression, response to treatment or the appropriate treatment for MDS. In some embodiments, the individual has a predefined clinical parameter and the characterization of multiple pathways in combination with the clinical parameter is indicative of the diagnosis, prognosis, determining progression, predicting response to treatment or choosing a treatment for MDS, in an individual. Examples of predetermined clinical parameters include, but are not limited to,

biochemical/molecular markers.

[0077] In some embodiments, the activatable elements can demarcate MDS cell subpopulations that have different genetic subclone origins. In some embodiments, the activatable elements can demarcate MDS subpopulations that, in combination with additional surface molecules, can allow for surrogate identification of MDS cell subpopulations. In some embodiments, the activatable elements can demarcate MDS subpopulations that can be used to determine other protein, epitope-based, RNA, mRNA, siRNA, or metabolic markers that singly or coordinately allow for surrogate identification of MDS cell subpopulations, disease stage of the individual from which the cells were derived, diagnosis, prognosis, response to treatment, or new draggable targets. In some embodiments, the pathways characterization allows for the delineation of MDS cell subpopulations that are differentially susceptible to drugs or drug combinations. In other embodiments, the cell types or activatable elements from a given cell type will, in combination with activatable elements in other cell types, provide ratiometric or metrics that singly or coordinately allow for surrogate identification of MDS cell subpopulations, disease stage of the individual from which the cells were derived, diagnosis, prognosis, response to treatment, or new draggable targets.

[0078] Therapeutic agents effective against the disease

[0079] There are many treatments for MDS. The treatment option typically includes choice of therapy on the basis of risk factors such as patient's age, MDS subtype, and prognostic score. The most commonly used prognostic score for MDS, the International Prognostic Scoring System (IPSS), is calculated based on bone marrow blast percentage, cytogenetics, and the number of cytopenias. Depending on the IPSS score and the patient's symptoms, different treatment paths are pursued.

[0080] Supportive care is important in the treatment of all patients with MDS. One aspect of

supportive care is transfusion therapy which involves blood transfusion (red blood cells or platelets). Red blood cell transfusions are generally performed when the patient has symptoms of fatigue in combination with low red cell numbers or low red cell numbers and an inability to make new red blood cells. Platelet transfusions are generally performed when the patient is bleeding, has a low platelet count and is not producing adequate platelets to prevent bleeding, or having a procedure that may cause bleeding. Patients who receive frequent red blood cell transfusions may suffer from tissue and organ damage due to accumulation of iron. Reactive oxygen species generated by labile plasma iron are a principal cause of cellular injury and organ dysfunction in patients with iron overload which affects survival and increases the risk of leukemia. Iron chelation therapy is recommended to the patient in these cases. This therapy uses drugs such as deferasirox, which can chelate extra iron and remove it from the body through the urinary passage.

[0081] Low-risk MDS patients are generally empirically treated with growth factor therapy.

Erythropoietin (EPO) therapy is most effective in patients with serum EPO <200 IU/L, low-int-1 IPSS, and an absence of transfusion requirement. A recent study of 403 patients with MDS s/p EPO +/- GCSF showed a 50% overall response rate to this therapy (Park et al., Blood 2008 111 :574-582).

[0082] EPO is thought to overcome reduced sensitivity of erythroid precursors to EPO at the initial level of signal transduction. (Hoefsloot LH, et al. Erythropoietin-induced activation of STAT5 is impaired in the myelodysplasia syndrome. Blood. 1997, vol.89, p.1690- 1700). Reports show a comparable erythroid response rate when using EPO alone or EPO plus filgrastim (G-CSF) (response rate 49 percent versus 51 percent), whereas higher EPO dose schedules were found to have higher response rate than standard EPO dose schedules. (Moyo VM et al. Treating the anemia of MDS with erythropoietin: Impact of higher dose compared to combination with G/GM-CSF. Proceeding from the American Society of Clinical Oncology Conference. Chicago, IL. 2007. Abstract 7082.)

[0083] Hematide, a novel synthetic pegylated peptidic compound, acts as an erythropoiesis

stimulating agent that binds to and activates the erythropoietin receptor. It could restore hemoglobin to the target range and eliminate the need for red blood cell transfusions, though hematide is immunologically distinct from EPO

(http://www.takeda.com/press/article_28646.html). Another growth factor thrombopoeitin (TPO), the ligand for the c-mpl receptor, is a major regulator of platelet production in vivo. It has been indicated in several studies that TPO increases platelet counts, platelet size, and increases isotope incorporation into platelets of recipient animals. Platelet count begins to increase after 3 to 5 days. TPO is thought to affect megakaryocytopoiesis in several ways: (1) it increases the size and number of megakaryocytes; (2) it produces an increase in DNA content, in the form of polyploidy, in megakaryocytes; (3) it increases megakaryocyte endomitosis; (4) it produces increased maturation of megakaryocytes; and (5) it produces an increase in the percentage of precursor cells, in the form of small acetylcholinesterase-positive cells, in the bone marrow. Romiplostim, a recombinant Fc-peptide fusion protein, is a thrombopoietin receptor agonist which can be used for identification of treatments effective in improving thrombocytopenia. It has recently been used in Phase II trials for MDS. However, its use is complicated by side effects such as disturbances of the gastrointestinal system, and arthralgia.

[0084] Immunosuppressive therapy (1ST) has emerged as an effective therapy for a subset of MDS patients with clonal amplification of T lymphocytes. T cell clones have been identified in 50% of MDS patients and have been implicated in suppression of hematopoiesis through CD8 cytotoxic T lymphocytes. Immunosuppressive agents like anti-thymocyte globulin, alone or in combination with cyclosporine, inhibit the effects of T-cell clones. Patients enriched for response to this therapy include the younger age group (<60 years), those requiring little to no red blood cell transfusion, those with marrow hypocellularity, those with the presence of paroxysmal nocturnal hemoglobinuria clone, and those with human leukocyte antigen (HLA)-DR15 phenotype. Using this enrichment criteria, recent data show a 30%> response rate with improved overall survival and a decrease in transformation to AML (Sloand et al JCO 2008 26:2505-251 1).

[0085] The immunomodulatory drugs are agents that target both the MDS clone and the bone marrow microenvironment and have notable erythropoietic activity in patients with low-risk MDS.

Lenalidomide, an amino- derivative of thalidomide with greater potency and minimal neurotoxicity, has erythropoietic and cytogenetic remitting activity. The efficacy of lenalidomide is greatest in patients with deletions of chromosome 5q. In this subset, lenalidomide produces and maintains red cell transfusion independence in the majority of low-risk patients for about two years. In a study of 148 patients with MDS RBC dependent anemia and 5q-, 67% of patients achieved a major erythroid response defined as RBC transfusion independence and an absence of any RBC transfusion during any consecutive 56 days (8 weeks) and Hgb increase of at least lg/dL during the treatment period (List A, et al. Lenalidomide in the myelodysplasia syndrome with chromosome 5q deletion. New England Journal of Medicine. 2006, vol.355, p.1456-1465.) Among the vast majority of MDS patients (over 80 percent) without the 5q- chromosomal defect, only about 26% respond to lenalidomide. (Blood 2008 1 1 1 :86-93 (Raza et al)

[0086] One study of gene expression profiling identified a cohesive set of erythroid-specific genes used as erythroid gene expression signature to predict the response of lenalidomide. The reduced expression of the erythroid gene signature in responders suggested a defect in erythroid differentiation. This suggests that it might be possible to use the response signature to develop a test that can predict the patients with MDS who will benefit from treatment with lenalidomide. (Benjamin L. Ebert et al. An Erythroid Differentiation Signature Predicts Response to

Lenalidomide in Myelodysplasia Syndrome. PLoS Medicine. Feb 2008. Vol.5, no.2, p.312-322).

[0087] Hypomethylating drugs such as Azacytidine (Vidaza) and Decitabine (Dacogen) have been approved for all IPSS scores of MDS. This class of drugs is thought to induce differentiation in the affected cells by preventing DNA methylation. Azacytidine is the first FDA-approved drug for the treatment of MDS. It is a pyrimidine analog that inhibits DNA methyl transferase. A CALGB study indicates that treatment with azacytidine produced higher response rate, improved quality of life, reduced risk of transformation to AML and extended life expectancy. (Silverman LR et al. Randomized controlled trial of azacytidine in patients with the myelodysplasia syndrome: a study of the cancer and leukaemia group B. Journal of Clinical Oncology. 2002. vol.20, p.2429-2440). Median survival was significantly prolonged to 24.4 months as compared to 15 months with conventional care, with greatest improvement observed in patients with chromosome 7 abnormalities, including monosomy 7. (Lim ZY et al. Outcomes of MDS patients with chromosome abnormalities treated with 5-azacytidine. Program and abstracts of 49^th Annual Meeting of the American Society of haematology. Dec.2007. Atlanta, Georgia. Abstract 1449). Further azacytidine treatment delays the progress of MDS to AML to 13 months as compared to 7.6 months in patients given only conventional care. A relatively higher number of patients treated with azacytidine achieved complete remission (CR) and hematologic improvements as compared to best supportive care. Decitabine (DNA methyl transferase inhibitor) is the second FDA-approved drug for treatment of patients with MDS. 170 patients were studied with an overall response rate of 17%> (9%>CR and 8%>PR) with a median duration of response of 10.3 months and a tend toward increased time to AML transformation (Cancer 2006 106: 1794-80 (Kantarjian et al)

[0088] Combination of hy omethylating agents with histone deacetylase (HDAC) inhibitors (MGCD- 0103) is under trial and preliminary data suggests major responses including CR, partial remission or marrow CR in 35% of patients with refractory MDS and 50% of previously untreated patients. (Itzykson et al. Meeting report: myelodysplasia syndromes at ASH 2007. Leukemia (2008) vol. 22 (5) pp. 893-7).

[0089] Chemotherapy with stem cell transplants is a method for giving high dose chemotherapy and replacing blood-forming cells, which have been destroyed by the cancer treatment. The stem cells of healthy donors are used for infusion in patients who have undergone chemotherapy. These reinfused stem cells grow into (and restore) the blood cells in the body. Although transplant can be curative in MDS, it is often limited by the patient's performance status and the availability of donors. Transplantation appears to be most beneficial for children with refractory cytopenias and adults with chemotherapy-related MDS, which represent only a small fraction of the MDS population. (Itzykson et al. Meeting report: myelodysplasia syndromes at ASH 2007. Leukemia (2008) vol. 22 (5) pp. 893-7).

[0090] A large number of treatment approaches are under the process of development. Agents under investigation include Arsenic trioxide (apoptosis inducer), Sorafenib (tyrosine kinase inhibitor), Vorinostat and valproic acid (histone deacetylase inhibitors), tipifarnib and lonafarnib (farnesyl transferase), bevacizumab (anti-VEGF monoclonal antibody that inhibits angiogenesis), FG-2216 (hypoxia-inducible factor stabilizer), ezatiostat (glutathione S I transferase inhibitor), clofarabine (nucleoside analog). (ALAN F. LIST, et al. Insights into the pathogenesis, Classification, and treatment of Myelodysplasia Syndromes, Semin. Hematol. 2008 Jan.; 45(1) 31 -8).

Pharmacologic differentiators, such as TLK199, (liposomal glutathione derivative) mediate proliferation and differentiation of myeloid precursors and production of GM-CSF. A TLK-199 trial on MDS patients showed hematologic improvement in all three hematopoietic lineages - erythrocytes, neutrophils, platelets. Toxicities were limited to infusion reactions, nausea, chills and bone pain. The thrombopoiesis-stimulating agent, IL-1 1 , is an indirect thrombopoietic cytokine that helps to combat platelet dysfunction and thrombocytopenia in MDS. The major side effects of this drug include fever, fluid retention, peripheral edema, pleural effusions and atrial arrhythmias. Pegylated, recombinant human megakaryocyte growth and development factor (PEG-rHuMGDF) stimulates megakaryocyte and platelet production by binding to c-Mpl receptors.

[0091] One embodiment of the invention involves the use of multiparameter flow cytometry to

examine the biology and signalling pathways in myelodysplasia syndrome to classify MDS identification of possible draggable targets, and inform on likelihood of response to agents such as growth factors (e.g. EPO), immunosuppressive agents (e.g. anti-thymocyte globulin (ATG)+/- cyclosporinA (CsA)), epigenetic modulators (e.g. hypomethylators Azacytidine and Decitabine and HDAC inhibitors), immune-modulators (e.g. Lenalidomide), or a rationale combination of the above.

[0092] One embodiment of the invention involves the use of multiparametric flow cytometry to examine the biology and signalling pathways in myelodysplasia syndrome to determine likelihood of progression to AML.

[0093] One embodiment of the invention involves the use of multiparametric flow cytometry to examine the biology and signalling pathways in myelodysplasia syndrome to determine likelihood of response to a candidate therapeutic in development for the treatment of MDS.

[0094] One embodiment of the invention involves the use of multiparametric flow cytometry to examine the biology and signalling pathways of myeloid disorders to aid in classification and therapeutic selection and identification of new potentially draggable targets.

[0095] In some embodiments, the invention provides a method for predicting a response to a

treatment or choosing a treatment for MDS or designing rationale combinations of drugs, in an individual. In some embodiments, the method further comprises determining whether the apoptosis, cell cycle, signaling, or DNA damage pathways are functional in the individual based on the activation levels of the activatable elements, wherein a pathway is functional if it is permissive for a response to a treatment, where if the apoptosis, cell cycle, signaling, and DNA damage pathways are functional the individual can respond to treatment, and where if at least one of the pathways is not functional the individual cannot respond to treatment.

[0096] General Methods

[0097] Embodiments of the invention may be used to diagnose, predict or to provide therapeutic decisions for disease treatment. In some embodiments, the invention may be used to determine the progression of MDS to AML, identify new draggable targets and to design drug

combinations. The following will discuss instruments, reagents, kits, and the biology involved with these and other diseases. One aspect of the invention involves contacting a hematopoietic cell with a modulator; determining the activation states of a plurality of activatable elements in the cell; and classifying the cell based on said activation state.

[0098] In some embodiments, this invention is directed to methods and compositions, and kits for analysis, drug screening, diagnosis, prognosis, for methods of disease treatment and prediction. In some embodiments, the present invention involves methods of analyzing experimental data. In some embodiments, the physiological status of cells present in a sample (e.g. clinical sample) is used, e.g., in diagnosis or prognosis of a condition, patient selection for therapy using some of the agents identified above, to monitor treatment, modify therapeutic regimens, and to further optimize the selection of therapeutic agents which may be administered as one or a combination of agents. Hence, therapeutic regimens can be individualized and tailored according to the data obtained prior to, and at different times over the course of treatment, thereby providing a regimen that is individually appropriate. In some embodiments, a compound is contacted with cells to analyze the response to the compound.

[0099] In some embodiments, the present invention is directed to methods for classifying a sample derived from an individual having or suspected of having a condition, e.g., a neoplastic or a hematopoietic condition, such as MDS. The invention allows for identification of prognostically and therapeutically relevant subgroups of conditions and prediction of the clinical course of an individual. The methods of the invention provide tools useful in the treatment of an individual afflicted with a condition, such as MDS, including but not limited to methods for assigning a risk group, methods of predicting an increased risk of relapse, methods of predicting an increased risk of developing secondary complications, methods of choosing a therapy for an individual, methods of predicting progression of MDS to AML, methods of predicting duration of response, response to a therapy for an individual, methods of determining the efficacy of a therapy in an individual, and methods of determining the prognosis for an individual. The present invention provides methods that can serve as a prognostic indicator to predict the course of a condition, e.g. whether the course of a neoplastic or a hematopoietic condition in an individual will be aggressive or indolent, thereby aiding the clinician in managing the patient and evaluating the modality of treatment to be used. In another embodiment, the present invention provides information to a physician to aid in the clinical management of a patient so that the information may be translated into action, including treatment, prognosis or prediction.

[00100] In some embodiments, the invention is directed to methods of characterizing a plurality of pathways in single cells. Exemplary pathways include apoptosis, cell cycle, signaling, or DNA damage pathways. In some embodiments, the characterization of the pathways is correlated with diagnosing, prognosing or determining condition progression in an individual. In some embodiments, the characterization of the pathways is correlated with predicting response to treatment or choosing a treatment in an individual. In some embodiments, the characterization of the pathways is correlated with finding a new draggable target. In some embodiments, the pathways' characterization in combination with a predetermined clinical parameter is indicative of the diagnosis, prognosis or progression of the condition. In some embodiments, the pathways' characterization in combination with a predetermined clinical parameter is indicative of a response to treatment or of the appropriate treatment for an individual. In some embodiments, the characterization of the pathways in combination with a predetermined clinical parameter is indicative a new draggable target.

[00101] In some embodiments, the invention is directed to methods for determining the activation level of one or more activatable elements in a cell upon treatment with one or more modulators. The activation of an activatable element in the cell upon treatment with one or more modulators can reveal operative pathways in a condition that can then be used, e.g., as an indicator to predict course of the condition, to identify risk group, to predict an increased risk of developing secondary complications, to choose a therapy for an individual, to predict response to a therapy for an individual, to determine the efficacy of a therapy in an individual, and to determine the prognosis for an individual. In some embodiments, the operative pathways can reveal whether apoptosis, cell cycle, DNMT, signaling, or DNA damage pathways are functional in an individual, where a pathway is functional if it is permissive for a response to a treatment. In some embodiments, when apoptosis, cell cycle, signaling, and DNA damage pathways are functional the individual can respond to treatment, and if at least one of the pathways is not functional the individual cannot respond to treatment. In some embodiments, when the apoptosis, DNMT, and DNA damage pathways are functional the individual can respond to treatment. In some embodiments, the operative pathways can reveal new draggable targets.

[00102] In some embodiments, the invention is directed to methods of determining a phenotypic

profile of a population of cells by exposing the population of cells to a plurality of modulators in separate cultures, determining the presence or absence of an increase in activation level of an activatable element in the cell population from each of the separate culture and classifying the cell population based on the presence or absence of the increase in the activation of the activatable element from each of the separate culture. In some embodiments at least one of the modulators is an inhibitor. In some embodiments, the presence or absence of an increase in activation level of a plurality of activatable elements is determined. In some embodiments, each of the activatable elements belongs to a particular pathway and the activation level of the activatable elements is used to characterize each of the particular pathways. In some

embodiments, a plurality of pathways are characterized by exposing a population of cells to one or a plurality of modulators in separate cultures, determining the presence or absence of an increase in activation levels of a plurality of activatable elements in the cell population from each of the separate culture, wherein the activatable elements are within the pathways being characterized and classifying the cell population based on the characterizations of said multiple pathways. In some embodiments, the activatable elements and modulators are selected from the activatable elements and modulators listed herein or in Tables 1 , 2, 3 or 5 of U.S. Patent App. Serial No. 13/083, 156. In some embodiments, the activatable elements and modulators are selected from the activatable elements and modulators listed in Table 12 of U.S. Patent App. Serial No. 13/083, 156 and are used to predict response duration in an individual after treatment. U.S. Patent App. Serial No. 13/083, 156 is incorporated by reference above.

[00103] In some embodiments, the invention is directed to methods for classifying a cell by

determining the presence or absence of an increase in activation level of an activatable element or protein level in combination with additional expression markers. In some embodiments, expression markers or drug transporters, such as CD34, CD45, CD235a, or CD 71 , or activatable elements or proteins such as DNMT 1 , 3a, or 3b, p-H2AX, Ki-67, cyclin A2 cyclin Bl , pSTAT 1 , 3, or 5, AKT, or p-Erk, pS6 and others noted below, can also be used for stratifying responders and non-responders. Surface markers may also be used such as those identified in U.S.

Provisional App. Serial No. 61/557,831 including surface markers include CD3, CD4, CD5, CD7, CD8, CDl lb, CDl lc, CD14, CD15, CD16, CD19, CD20, CD22, CD25, CD27, CD28, CD33, CD34, CD38, CD40, CD45, CD56, CD69, CD71, CD80, CD1 17, CD138, CD161, CD235a, CD235b, Terl 19, GP-130, IgM, IgD, IgE, IgG, IgA, CCR5, CCR3, TLR2,and TLR4.

04] The expression markers may be detected using many different techniques, for example using nodes from flow cytometry data (see the articles and patent applications referred to above). Other common techniques employ expression arrays (commercially available from Affymetrix, Santa Clara CA or Illumina, San Diego, CA), taqman (commercially available from ABI, Foster City CA), SAGE (commercially available from Genzyme, Cambridge MA), sequencing techniques (see the commercial products from Helicos, 454, US Genomics, and ABI) and other commonly known assays. See Golub et al., Science 286: 531 -537 (1999). Expression markers are measured in unstimulated cells to know whether they have an impact on functional apoptosis. This provides implications for treatment and prognosis for the disease. Under this hypothesis, the amount of drug transporters correlates with the response of the patient and non-responders may have more levels of drug transporters (to move a drug out of a cell) as compared to responders. In some embodiments, the invention is directed to methods of classifying a cell population by contacting the cell population with at least one modulator that affects signaling mediated by receptors selected from the group comprising of growth factors, mitogens and cytokines. In some embodiments, the invention is directed to methods of classifying a cell population by contacting the cell population with at least one modulator that affects signaling mediated by receptors;

determining the activation states of a plurality of activatable elements in the cell comprising; and classifying the cell based on said activation states and expression levels. In some embodiments, the cell population is also exposed in a separate culture to at least one modulator that affects, slows or stops the growth of cells and/or induces apoptosis of cells. In some embodiments, the modulator is selected from the group consisting of EPO, G-CSF, the combination of EPO and G- CSF, Lenalidomide, and the combination of Lenalidomide and EPO, GM-CSF, IL-6, IL-27, TPO, Vidaza, Dacogen, or Zolinza (SAHA). In some embodiments, the cell population is also exposed in a separate culture to at least one modulator that is an inhibitor. In some embodiments, the cell population in a hematopoietic cell population. In some embodiments, the invention is directed to methods of correlating and/or classifying an activation state of an MDS cell with a clinical outcome in an individual by subjecting the MDS cell from the individual to a modulator, determining the activation levels of a plurality of activatable elements, and identifying a pattern of the activation levels of the plurality of activatable elements to determine the presence or absence of an alteration in signaling, where the presence of the alteration is indicative of a clinical outcome. In some embodiments, the activatable elements can demarcate MDS cell subpopulations that have different genetic subclone origins. In some embodiments, the activatable elements can demarcate MDS subpopulations that can be used to determine other protein, epitope-based, RNA, mRNA, siRNA, or metabolomic markers that singly or coordinately allow for surrogate identification of MDS cell subpopulations, disease stage of the individual from which the cells were derived, diagnosis, prognosis, response to treatment, or new draggable targets. In some embodiments, the pathways characterization allows for the delineation of MDS cell subpopulations that are differentially susceptible to drugs or drug combinations. In other embodiments, the cell types or activatable elements from a given cell type will, in combination with activatable elements in other cell types, provide ratiometric or metrics that singly or coordinately allow for surrogate identification of MDS cell subpopulations, disease stage of the individual from which the cells were derived, diagnosis, prognosis, response to treatment, or new draggable targets.

[00105] In some embodiments, the invention provides methods to determine dosing and scheduling of drugs. Drug selection, dosing, and dosing schedules can be guided by the effect of the drug on activatable elements in patient cells. In some embodiments, the invention may identify whether a patient responds to a drug, and therefore may be used to identify effective drugs for treating that patient. In some embodiments, the invention may be used to select drugs for combination therapies based on how a primary drug affects cell signaling or cell cycle progression in cell lines or patient samples: the invention may identify side effects, or biological processes that decrease efficacy of the drug. Based on these observations, combination treatments may be selected based on their ability to reduce side effects or enhance the efficacy of the primary drug. For example, the DNA methyltransferase inhibitors Vidaza^® cytidine analog (5-Azacytidine) and Dacogen^® cytidine analog (5-Aza-2'-deoxycytidine) are used to treat Acute Myeloid Leukemia (AML), a disease characterized by the overproliferation of undifferentiated cells. See U.S. Provisional App. No. 61/120,320, hereby incorporated by reference, for a more detailed description of AML, other hematologic malignancies, and current therapies and their mechanisms of action.

Overexpression of DNA methyltransferases DNMTl , DMNT3a, and DMNT3b is associated with higher MDS disease risk. Additionally, DNMT3A is commonly mutated in AML and is associated with poor prognosis. See Hopfer O. et al., Aberrant promoter methylation in MDS hematopoietic cells during in vitro lineage-specific differentiation is differently associated with DNMT isoforms (2009), Leukemia Research 33 pp. 434-442; Langer, F. et al. (2005), Up- regulation of DNA methyltransferases DNMTl , 3A, and 3B in myelodysplasia syndrome, Leukemia Research 29, pp. 325-329, Hou et al. (201 1) DNMT3A mutations in acute myeloid leukemia-stability during disease evolution and clinical implication, Blood 201 1 , Novl O which are hereby incorporated by reference.

[00106] Vidaza^® cytidine analog and Dacogen^® cytidine analog are both pyrimidine analogs that inhibit DNA methyltransferase activity by incorporating into nucleic acids. By promoting DNA demethylation, Vidaza^® cytidine analog and Dacogen^® cytidine analog affect regulation of cells, such as cells affected by AML. Other drugs for the treatment of cancers, such as AML, include: Arsenic trioxide (apoptosis inducer), sorafenib (tyrosine kinase inhibitor), gemtuzumab ozogamicin (Mylotarg), vorinostat and valproic acid (histone deacetylase inhibitors), tipifarnib and lonafarnib (farnesyl transferase and RAF/RAS/ERK inhibitor), bevacizumab and ranibizumab (anti-EDGF monoclonal antibody that inhibits angiogenesis), ezatiostat (glutathione SI transferase inhibitor), and clofarabine (nucleoside analog). A combination of

hypomethylating agents with histone deacetylase (HDAC) inhibitors (MGCD-0103) is under trial for MDS and preliminary data suggests major responses (Itzykson et al., Meeting report:

myelodysplasia syndromes at ASH 2007, Leukemia (2008) vol. 22 (5) pp. 893-7. See also Griffiths, E.A., and Gore, S.D., DNA Methyltransferase and Histone Deacetylase Inhibitors in the Treatment of Myelodysplastic Syndromes, Semin. Hematol. (2008) January 45(1) pp. 23-30. As one embodiment of the invention demonstrates, Vidaza^® cytidine analog and Dacogen^® cytidine analog treatments elicit different responses as measured by different responses within different phases of the cell cycle, such as can be seen with Dacogen^® cytidine analog inducing arrest at S phase, and Vidaza^® cytidine analog inducing cell death.

[00107]The subject invention also provides kits for use in determining the physiological status of cells in a sample, the kit comprising one or more modulators, inhibitors, specific binding elements for signaling molecules, reagents, instructions, and may additionally comprise one or more therapeutic agents. The above reagents for the kit are all recited and listed in the present application. The kit may further comprise a software package for data analysis of the cellular state and its physiological status, which may include reference profiles for comparison with the test profile and comparisons to other analyses as referred to above. The kit may also include instructions for use for any of the above applications.

[00108] In some embodiments, the invention provides methods, including methods to determine the physiological status of a cell, e.g., by determining the activation level of an activatable element upon contact with one or more modulators. In some embodiments, the invention provides methods, including methods to classify a cell according to the status of an activatable element in a cellular pathway. In some embodiments, the cells are classified by analyzing the response to particular modulators and by comparison of different cell states, with or without modulators. The information can be used in prognosis and diagnosis, including susceptibility to disease(s), status of a diseased state and response to changes, in the environment, such as the passage of time, treatment with drugs or other modalities. The physiological status of the cells provided in a sample (e.g. clinical sample) may be classified according to the activation of cellular pathways of interest. The cells can also be classified as to their ability to respond to therapeutic agents and treatments. The physiological status of the cells can provide new draggable targets for the development of treatments. These treatments can be used alone or in combination with other treatments. The physiological status of the cells can be used to design combination treatments.

[00109] In some embodiments, the present invention can determine an appropriate therapeutic

compound to administer to an individual having MDS. One embodiment of the invention is a method of choosing a treatment for an individual having MDS, comprising: contacting one or more MDS cells with a therapeutic agent; determining a level of DNMT 1 , DNMT 3a, or DNMT 3b in one or more cells from the individual; determining an activation level of one or more activatable elements in one or more cells from the individual which indicates cytostasis;

determining an activation level one or more activatable elements in one or more cells from the individual which indicates cytotoxicity; determining the level of CD34+ cells; repeating the four determining steps at a later time point and comparing the results; making a decision regarding a therapy based on the results of the comparison. Another embodiment of the invention is a method of screening a drug that is in development as a candidate therapeutic to treat an MDS patient, comprising: contacting one or more MDS cells with a candidate therapeutic either in vivo or in vitro; determining a level of DNMT 1, DNMT 3a, or DNMT 3b in one or more cells from the individual; determining an activation level one or more activatable elements in one or more cells from said individual from the group Ki-67, cyclin A2, cyclin Bl ; determining an activation level one or more activatable elements in one or more cells from said individual from the group cPARP, Amine Aqua or p-H2AX; determining the level of CD34+ cells; repeating the four determining steps at a later time point and comparing the results; determining if the candidate therapeutic is effective as a treatment for MDS.

[00110] One or more cells or cell types, or samples containing one or more cells or cell types, can be isolated from body samples. The cells can be separated from body samples by centrifugation, elutriation, density gradient separation, apheresis, affinity selection, panning, FACS,

centrifugation with Hypaque, solid supports (magnetic beads, beads in columns, or other surfaces) with attached antibodies, etc. By using antibodies specific for markers identified with particular cell types, a relatively homogeneous population of cells may be obtained.

Alternatively, a heterogeneous cell population can be used. Cells can also be separated by using filters. For example, whole blood can also be applied to filters that are engineered to contain pore sizes that select for the desired cell type or class. Rare pathogenic cells can be filtered out of diluted, whole blood following the lysis of red blood cells by using filters with pore sizes between 5 to 10 μηι, as disclosed in U.S. Patent Application No. 09/790,673. Once a sample is obtained, it can be used directly, frozen, or maintained in appropriate culture medium for short periods of time. Methods to isolate one or more cells for use according to the methods of this invention are performed according to standard techniques and protocols well-established in the art. See also U.S. Patent App. Ser. Nos. 12/432,720, 12/229,476, and 12/432,239. See also, the commercial products from companies such as BD and BCI as identified above. [00111] See also U.S. Patent Nos. 7,381,535 and 7,393,656. All of the above patents and applications are incorporated by reference as stated above.

[00112]In some embodiments, the cells are cultured post collection in a media suitable for revealing the activation level of an activatable element (e.g. RPMI, DMEM) in the presence, or absence, of serum such as fetal bovine serum, bovine serum, human serum, porcine serum, horse serum, or goat serum. When serum is present in the media it could be present at a level ranging from

0.0001 % to 30%.

[00113]In some embodiments, the cells are hematopoietic cells. Examples of hematopoietic cells include but are not limited to pluripotent hematopoietic stem cells, B-lymphocyte lineage progenitor or derived cells, T-lymphocyte lineage progenitor or derived cells, NK cell lineage progenitor or derived cells, granulocyte lineage progenitor or derived cells, monocyte lineage progenitor or derived cells, megakaryocyte lineage progenitor or derived cells and erythroid lineage progenitor or derived cells.

[00114]The term "patient" or "individual" as used herein includes humans as well as other mammals.

The methods generally involve determining the status of an activatable element. The methods also involve determining the status of a plurality of activatable elements.

[00115]In some embodiments, the invention provides a method of classifying a cell by determining the presence or absence of an increase in activation level of an activatable element in the cell upon treatment with one or more modulators, and classifying the cell based on the presence or absence of the increase in the activation of the activatable element. In some embodiments of the invention, the activation level of the activatable element is determined by contacting the cell with a binding element that is specific for an activation state of the activatable element, such as a phosphorylation state. In some embodiments, a cell is classified according to the activation level of a plurality of activatable elements after the cell have been subjected to a modulator. In some embodiments of the invention, the activation levels of a plurality of activatable elements are determined by contacting a cell with a plurality of binding elements, where each binding element is specific for an activation state of an activatable element.

[00116]The classification of a cell according to the status of an activatable element can comprise classifying the cell as a cell that is correlated with a clinical outcome. In some embodiments, the clinical outcome is the prognosis and/or diagnosis of a condition. In some embodiments, the clinical outcome is the presence or absence of a neoplastic or a hematopoietic condition as such myelodysplasia syndrome (MDS). In some embodiments, the clinical outcome is the staging or grading of a neoplastic or hematopoietic condition. Examples of staging include, but are not limited to, aggressive, indolent, benign, refractory, Roman Numeral staging, TNM Staging, Rai staging, Binet staging, WHO classification, FAB classification, IPSS score, WPSS score, limited stage, extensive stage, staging according to cellular markers, occult, including information that may inform on time to progression, progression free survival, overall survival, or event- free survival.

[00117]The classification of a cell according to the status of an activatable element can comprise classifying a cell as a cell that is correlated to a patient response to a treatment. In some embodiments, the patient response is selected from the group consisting of complete response, partial response, nodular partial response, no response, progressive disease, stable disease and adverse reaction.

[00118]The classification of a rare cell according to the status of an activatable element can comprise classifying the cell as a cell that can be correlated with minimal residual disease or emerging resistance. See U.S. Patent App. Serial No. 12/432,720 which is incorporated by reference.

[00119]The classification of a cell according to the status of an activatable element can comprise selecting a method of treatment. Example methods of treatments include, but are not limited to chemotherapy, biological therapy, radiation therapy, bone marrow transplantation, Peripheral stem cell transplantation, umbilical cord blood transplantation, autologous stem cell

transplantation, allogeneic stem cell transplantation, syngeneic stem cell transplantation, surgery, induction therapy, maintenance therapy, watchful waiting, and other therapy.

[00120] A modulator can be an activator, an inhibitor or a compound capable of impacting cellular signaling networks. Modulators can take the form of a wide variety of environmental cues and inputs. Examples of modulators include but are not limited to growth factors, mitogens, cytokines, adhesion molecules, drugs, hormones, small molecules, polynucleotides, antibodies, natural compounds, lactones, chemotherapeutic agents, immune modulators, carbohydrates, proteases, ions, reactive oxygen species, radiation, physical parameters such as heat, cold, UV radiation, peptides, and protein fragments, either alone or in the context of cells, cells themselves, viruses, and biological and non-biological complexes (e.g. beads, plates, viral envelopes, antigen presentation molecules such as major histocompatibility complex).

[00121] In some embodiments, the modulator is an activator. In some embodiments the modulator is an inhibitor. In some embodiments, the invention provides methods for classifying a cell by contacting the cell with an inhibitor, determining the presence or absence of an increase in activation level of an activatable element in the cell, and classifying the cell based on the presence or absence of the increase in the activation of the activatable element. In some embodiments, a cell is classified according to the activation level of a plurality of activatable elements after the cells have been subjected to an inhibitor. In some embodiments, the inhibitor is an inhibitor of a cellular factor or a plurality of factors that participates in a signaling cascade in the cell. In some embodiments, the inhibitor is a phosphatase inhibitor. Examples of phosphatase inhibitors include, but are not limited to H2O2, siRNA, miRNA, Cantharidin, (-)-p- Bromotetramisole, Microcystin LR, Sodium Orthovanadate, Sodium Pervanadate, Vanadyl sulfate, Sodium oxodiperoxo(l , 10-phenanthroline)vanadate, bis(maltolato)oxovanadium(IV), Sodium Molybdate, Sodium Perm olybdate, Sodium Tartrate, Imidazole, Sodium Fluoride, β- Glycerophosphate, Sodium Pyrophosphate Decahydrate, Calyculin A, Discodermia calyx, bpV(phen), mpV(pic), DMHV, Cypermethrin, Dephostatin, Okadaic Acid, NIPP-1 , N-(9, 10- Dioxo-9, 10-dihydro-phenanthren-2-yl)-2,2-dimethyl-propionamide, a-Bromo-4- hydroxyacetophenone, 4-Hydroxyphenacyl Br, a-Bromo-4-methoxyacetophenone, 4- Methoxyphenacyl Br, a-Bromo-4-(carboxymethoxy)acetophenone, 4-(Carboxymethoxy)phenacyl Br, and bis(4-Trifluoromethylsulfonamidophenyl)-l ,4-diisopropylbenzene, phenylarsine oxide, Pyrrolidine Dithiocarbamate, and Aluminum fluoride. In some embodiments, the phosphatase inhibitor is H2O2.

[00122] In some embodiments, the methods of the invention provide methods for classifying a cell population or determining the presence or absence of a condition in an individual by subjecting a cell from the individual to a modulator and an inhibitor, determining the activation level of an activatable element in the cell, and determining the presence or absence of a condition based on the activation level. In some embodiments, the activation level of a plurality of activatable elements in the cell is determined. The inhibitor can be an inhibitor as described herein. In some embodiments, the inhibitor is a phosphatase inhibitor. In some embodiments, the inhibitor is H2O2. The modulator can be any modulator described herein. In some embodiments, the methods of the invention provides for methods for classifying a cell population by exposing the cell population to a plurality of modulators in separate cultures and determining the status of an activatable element in the cell population. In some embodiments, the status of a plurality of activatable elements in the cell population is determined. In some embodiments, at least one of the modulators of the plurality of modulators is an inhibitor. The modulator can be at least one of the modulators described herein. In some embodiments of the invention, the status of an activatable element is determined by contacting the cell population with a binding element that is specific for an activation state of the activatable element. In some embodiments, the status of a plurality of activatable elements is determined by contacting the cell population with a plurality of binding elements, where each binding element is specific for an activation state of an activatable element.

[00123] In some embodiments, the methods of the invention provide methods for determining a

phenotypic profile of a population of cells by exposing the population of cells to a plurality of modulators in the same or in separate cultures, determining the presence or absence of an increase in activation level of an activatable element in the cell population and classifying the cell population based on the presence or absence of the increase in the activation of the activatable element from each of the separate culture. In some embodiments, the phenotypic profile is used to characterize multiple pathways in the population of cells.

[00124] Patterns and profiles of one or more activatable elements are detected using the methods

known in the art including those described herein. In some embodiments, patterns and profiles of activatable elements that are cellular components of a cellular pathway or a signaling pathway are detected using the methods described herein. For example, patterns and profiles of one or more phosphorylated polypeptides are detected using methods known in art including those described herein.

[00125] In some embodiments, cells (e.g. normal cells) other than the cells associated with a condition (e.g. cancer cells) or a combination of cells are used, e.g., in assigning a risk group, predicting an increased risk of relapse, predicting an increased risk of developing secondary complications, choosing a therapy for an individual, predicting response to a therapy for an individual, determining the efficacy of a therapy in an individual, and/or determining the prognosis for an individual. Cells other than cells associated with a condition (e.g. cancer cells) are in fact reflective of the condition process. For instance, in the case of cancer, infiltrating immune cells might determine the outcome of the disease. Alternatively, a combination of information from the cancer cell plus the immune cells in the blood that are responding to the disease, or reacting to the disease can be used for diagnosis or prognosis of the cancer.

[00126] In some embodiments, the invention provides methods to carry out multiparameter flow

cytometry for monitoring phospho-protein responses to various factors in MDS at the single cell level. Phospho-protein members of signaling cascades and the kinases and phosphatases that interact with them are required to initiate and regulate proliferative signals in cells. Apart from the basal level of protein phosphorylation alone, the effect of potential drug molecules on these network pathways was studied to discern unique cancer network profiles, which correlate with the genetics and disease outcome. Single cell measurements of phospho-protein responses reveal shifts in the signaling potential of a phospho-protein network, enabling categorization of cell network phenotypes by multidimensional molecular profiles of signaling. See U.S. Patent No. 7,393,656. See also Irish et. al., Single cell profiling of potentiated phospho-protein networks in cancer cells. Cell. 2004, vol. 118, p.1-20.

[00127] Flow cytometry is useful in a clinical setting, since relatively small sample sizes, as few as 10,000 cells, can produce a considerable amount of statistically tractable multidimensional signaling data and reveal key cell subsets that are responsible for a phenotype. See U.S. Patent App. Serial No. 12/432,720.

[00128] Cytokine response panels have been studied to survey altered signal transduction of cancer cells by using a multidimensional flow cytometry file which contained at least 30,000 cell events. In one embodiment, this panel is expanded and the effect of growth factors and cytokines on primary AML samples studied. See U.S. Patent Nos. 7,381,535 and 7,393,656. See also Irish et. al., CELL Jul 23;118(2):217-28. In some embodiments, the analysis involves working at multiple characteristics of the cell in parallel after contact with the compound. For example, the analysis can examine DNMT proteins, drug transporter function; drug transporter expression; drug metabolism; drug activation; cellular redox potential; signaling pathways; DNA damage repair; and apoptosis.

[00129] In some embodiments, the MDS or other panel of modulators is further expanded to examine the process of DNA damage, apoptosis, drug transport, drug metabolism, DNMT proteins, and the use of peroxide to evaluate phosphatase activity. Analysis can assess the ability of the cell to undergo the process of apoptosis after exposure to the experimental drug in an in vitro assay as well as how quickly the drug is exported out of the cell or metabolized. The drug response panel can include but is not limited to detection of phosphorylated Chk2, Cleaved Caspase 3, Caspase 8, PARP and mitochondria-released Cytoplasmic Cytochrome C. Analysis can assess phosphatase activity after exposure of cells to phosphatase inhibitors including but not limited to hydrogen peroxide (H₂O₂), H₂O₂ + SCF and H₂O₂ + IFNa. The response panel to evaluate phosphatase activity can include but is not limited to the detection of the phosphorylated activatable element recited above. Later, the samples may be analyzed for the expression of drug transporters such as MDR1/PGP, MRP1 and BCRP/ABCG2. Samples may also be examined for XIAP, Survivin, Bcl-2, MCL-1, Bim, Ki-67, Cyclin Dl , ID1 and Myc.

[00130]Another method of the present invention is a method for determining the prognosis and

therapeutic selection for an individual with MDS. Using the signaling nodes and methodology described herein, multiparametric flow could separate a patient into "responsive" or "non- responsive". Furthermore, for those patients unlikely to benefit from therapy, the individual's blood or marrow sample could reveal signaling biology that corresponds to either in-vivo or in- vitro sensitivity to a class of drugs including but not limited to direct drug resistance modulators, anti-Bcl-2 or pro-apoptotic drugs, proteosome inhibitors, DNA methyl transferase inhibitors, histone deacetylase inhibitors, anti-angiogenic drugs, farnesyl transferase inhibitors, FLt3 ligand (herein also FLT3-L,Flt-3 ligand, and FLT3L) inhibitors, or ribonucleotide reductase inhibitors. An individual with MDS with a complete response to induction therapy could further benefit from the present invention. The individual's blood or marrow sample could reveal signaling biology that corresponds to likelihood of benefit from further chemotherapy versus myeloablative therapy followed by and stem cell transplant versus reduced intensity therapy followed by stem cell transplantation.

[00131] In some embodiments, the invention provides a method for diagnosing, prognosing,

determining progression, predicting response to treatment or choosing a treatment for MDS in an individual where the individual has a predefined clinical parameter.

[00132] In some embodiments, the invention provides a method for predicting a response to a

treatment or choosing a treatment for MDS in an individual, the method comprising the steps: (a) subjecting a cell population from the individual to one or more distinct modulators; (b) determining an activation level of at least one activatable element; and (c) predicting a response to a treatment or choosing a therapeutfic for MDS in the individual based on the activation level of said activatable elements. In some embodiments, the method further comprises determining whether the apoptosis, cell cycle, signaling, or DNA damage pathways are functional in the individual based on the activation levels of the activatable elements, wherein a pathway is functional if it is permissive for a response to a treatment, where if the apoptosis, cell cycle, signaling, and DNA damage pathways are functional the individual can respond to treatment, and where if at least one of the pathways is not functional the individual cannot respond to treatment. In some embodiments, the method further comprises determining whether the apoptosis, cell cycle, signaling, or DNA damage pathways are functional in the individual based on the activation levels of the activatable elements, wherein a pathway is functional if it is permissive for a response to a treatment, where if the apoptosis and DNA damage pathways are functional the individual can respond to treatment. In some embodiments, the method further comprises determining whether the apoptosis, cell cycle, signaling, or DNA damage pathways are functional in the individual based on the activation levels of the activatable elements, wherein a pathway is functional if it is permissive for a response to a treatment, where a therapeutic is chosen depending of the functional pathways in the individual.

[00133] In some embodiments, if the activation levels of an activatable element within any pathway are higher than a threshold level in response to a cytokine it may be indicative that an individual cannot respond to treatment. In some embodiments, a treatment is chosen based on the ability of the cells to respond to treatment.

[00134] In some embodiments, a diagnosis, prognosis, a prediction of outcome such as response to treatment or relapse is performed by analyzing the two or more phosphorylation levels of two or more proteins each in response to one or more modulators. The phosphorylation levels of the independent proteins can be measured in response to the same or different modulators. Grouping of data points increases predictive value.

[00135] In some embodiments, the invention provides methods for predicting response to a treatment for MDS wherein the positive predictive value (PPV) is higher than 60, 70, 80, 90, 95, or 99.9 %. In some embodiments, the invention provides methods for predicting response to a treatment for MDS wherein the PPV is equal or higher than 95%. In some embodiments, the invention provides methods for predicting response to a treatment for MDS wherein the negative predictive value (NPV) is higher than 60, 70, 80, 90, 95, or 99.9 %. In some embodiments, the invention provides methods for predicting response to a treatment for MDS wherein the NPV is higher than 85 %.

[00136] In some embodiments, the invention provides methods for predicting risk of relapse at 2 years, wherein the PPV is higher than 60, 70, 80, 90, 95, or 99.9 %. In some embodiments, the invention provides methods for predicting risk of relapse at 2 years, wherein the PPV is equal or higher than 95%. In some embodiments, the invention provides methods for predicting risk of relapse at 2 years, wherein the NPV is higher than 60, 70, 80, 90, 95, or 99.9 %. In some embodiments, the invention provides methods for predicting risk of relapse at 2 years, wherein the NPV is higher than 80 %. In some embodiments, the invention provides methods for predicting risk of relapse at 5 years, wherein the PPV is higher than 60, 70, 80, 90, 95, or 99.9 %. In some embodiments, the invention provides methods for predicting risk of relapse at 5 years, wherein the PPV is equal or higher than 95%. In some embodiments, the invention provides methods for predicting risk of relapse at 5 years, wherein the NPV is higher than 60, 70, 80, 90, 95, or 99.9 %. In some embodiments, the invention provides methods for predicting risk of relapse at 5 years, wherein the NPV is higher than 80 %. In some embodiments, the invention provides methods for predicting risk of relapse at 10 years, wherein the PPV is higher than 60, 70, 80, 90, 95, or 99.9 %. In some embodiments, the invention provides methods for predicting risk of relapse at 10 years, wherein the PPV is equal or higher than 95%. In some embodiments, the invention provides methods for predicting risk of relapse at 10 years, wherein the NPV is higher than 60, 70, 80, 90, 95, or 99.9 %>. In some embodiments, the invention provides methods for predicting risk of relapse at 10 years, wherein the NPV is higher than 80 %>.

[00137] In some embodiments, the p value in the analysis of the methods described herein is below 0.05, 04, 0.03, 0.02, 0.01, 0.009, 0.005, or 0.001. In some embodiments, the p value is below 0.001. Thus in some embodiments, the invention provides methods for diagnosing, prognosing, determining progression or predicting response for treatment of MDS wherein the p value is below 0.05, 04, 0.03, 0.02, 0.01, 0.009, 0.005, or 0.001. In some embodiments, the p value is below 0.001. In some embodiments, the invention provides methods for diagnosing, prognosing, determining progression or predicting response for treatment of MDS wherein the AUC value is higher than 0.5, 0.6, 07, 0.8 or 0.9. In some embodiments, the invention provides methods for diagnosing, prognosing, determining progression or predicting response for treatment of MDS wherein the AUC value is higher than 0.7. In some embodiments, the invention provides methods for diagnosing, prognosing, determining progression or predicting response for treatment of MDS wherein the AUC value is higher than 0.8. In some embodiments, the invention provides methods for diagnosing, prognosing, determining progression or predicting response for treatment of MDS wherein the AUC value is higher than 0.9.

[00138]Another method of the present invention is a method for determining the prognosis and

therapeutic selection for an individual with myelodysplasia or MDS. Using the signaling nodes and methodology described herein, multiparametric flow cytometry could separate a patient into one of five groups consisting of: "AML-like", where a patient displays signaling biology that is similar to that seen in acute myelogenous leukemia (AML) requiring intensive therapy, "EPO- Responsive", where a patient's bone marrow or potentially peripheral blood, shows signaling biology that corresponds to either in-vivo or in-vitro sensitivity to erythropoietin, "Lenalidomide responsive", where a patient's bone marrow or potentially peripheral blood, shows signaling biology that corresponds to either in-vivo or in-vitro sensitivity to Lenalidomide, "Auto-immune", where a patient's bone marrow or potentially peripheral blood, shows signaling biology that corresponds to sensitivity to cyclosporine A(CSA) and anti-thymocyte globulin(ATG).

[00139]In those cases where an individual is classified as "AML-like", the individual's blood or marrow sample could reveal signaling biology that corresponds to either in-vivo or in-vitro sensitivity to cytarabine or to a class of drugs including but not limited to direct drug resistance modulators, anti-Bcl-2 or pro-apoptotic drugs, proteosome inhibitors, DNA methyl transferase inhibitors, histone deacetylase inhibitors, anti-angiogenic drugs, farnesyl transferase inhibitors, FLt3 ligand inhibitors, or ribonucleotide reductase inhibitors.

[00140] In some embodiments of the invention, different gating strategies can be used in order to analyze only blasts in the sample of mixed population after treatment with the modulator. These gating strategies can be based on the presence of one or more specific surface marker expressed on each cell type. In some embodiments, the first gate eliminates cell doublets so that the user can focus on singlets. The following gate can differentiate between dead cells and live cells and subsequent gating of live cells classifies them into blasts, monocytes and lymphocytes. A clear comparison can be carried out to study the effect of potential modulators, such as G-SCF on activatable elements in: ungated samples, blasts, monocytes, granulocytes and lymphocytes by using two-dimensional contour plot representations of Stat5 and Stat3 phosphorylation (x and Y axis) of patient samples. The level of basal phosphorylation and the change in phosphorylation in both Stat3 and Stat5 phosphorylation in response to G-CSF can be compared. G-CSF increases both STAT3 and STAT5 phosphorylation and this dual signaling can occur concurrently

(subpopulations with increases in both pSTAT 3 and pSTAT5) or individually (subpopulations with either an increase in phospho pSTAT 3 or pSTAT5 alone). The advantage of gating is to get a clearer picture and more precise results of the effect of various activatable elements on blasts. For an example gating strategy see Hoefsloot LH, Lowenberg B, et al Blood. 1997

Marl ;89(5): 1690-700 which is hereby incorporated by reference in its entirety.

[00141] In some embodiments, a gate is established after learning from a responsive subpopulation.

That is, a gate is developed from one data set. This gate can then be applied retrospectively or prospectively to other data sets. The cells in this gate can be used for the diagnosis or prognosis of a condition. The cells in this gate can also be used to predict response to a treatment or for treatment selection. The mere presence of cells in this gate may be indicative of a diagnosis, prognosis, or a response to treatment. In some embodiments, the presence of cells in this gate at a number higher than a threshold number may be indicative of a diagnosis, prognosis, or a response to treatment.

[00142] Some methods of analysis, also called metrics are: 1) measuring the difference in the log of the median fluorescence value between an unstimulated fluorochrome-antibody stained sample and a sample that has not been treated with a stimulant or stained (log (MFI_Unstimui_ated stained) - log (MFIoated unstained)), 2) measuring the difference in the log of the median fluorescence value between a stimulated fluorochrome-antibody stained sample and a sample that has not been treated with a stimulant or stained (log (MFI_Stimui_{ated Stained}) - log(MFI_Gated unstained)), 3) Measuring the change between the stimulated fluorochrome-antibody stained sample and the unstimulated fluorochrome-antibody stained sample log (MFI_Stimui_{ated Stained}) - log (MFLjnstimuiated stained), also called "fold change in median fluorescence intensity", 4) Measuring the percentage of cells in a Quadrant Gate of a contour plot which measures multiple populations in one or more dimension 5) measuring MFI of phosphor positive population to obtain percentage positivity above the background; and 6) use of multimodality and spread metrics for large sample population and for subpopulation analysis. Other metrics used to analyze data are population frequency metrics measuring the frequency of cells with a described property such as cells positive for cleaved PARP (% PARP+) , or cells positive for p-S6 and p-Akt. Similarly, measurements examining the changes in the frequencies of cells may be applied such as the Change in % PARP + which would measure the % PARP+stimulated Stained " % PARP+Unstimulated Stained- The AUCunstim metric also measures changes in population frequencies measuring the frequency of cells to become positive compared to an unstimulated condition. The metrics can be used to measure apoptosis. For example, these metrics can be applied to cleaved Caspase-3 and Caspase-8, e.g., Change in % Cleaved Caspase-3 or Cleaved Caspase-8.

[00143] Other possible metrics include third-color analysis (3D plots); percentage positive and relative expression of various markers; clinical analysis on an individual patient basis for various parameters, including, but not limited to age, race, cytogenetics, mutational status, blast percentage, CD34+ percentage, time of relapse, survival, etc. In alternative embodiments, there are other ways of analyzing data, such as third color analysis (3D plots), which can be similar to Cytobank 2D, plus third D in color. See the patent applications incorporated by reference above, including U.S. Provisional Application Serial No. 61/515,660.

[00144] Other metrics shown below include fold change, U_u, and Total Phospho. Fold change is the measure of the shift in the median value of the population of cells that display modulation of signaling relative to the basal state. It is calculated as log₂Fold. U_u is the measure of proportion of cells that display induction of signaling relative to basal activity. It is a rank based metric based on Mann- Whitney U statistic. Total Phospho is the measure of the shift in the median value of the population of cells that display modulation of signaling relative to the

autofluorescence state. It is calculated as log₂Total Phospho.

[00145] Disease Conditions

[00146] The methods of the invention are applicable to any condition in an individual involving, indicated by, and/or arising from, in whole or in part, altered physiological status in a cell. The term "physiological status" includes mechanical, physical, and biochemical functions in a cell. In some embodiments, the physiological status of a cell is determined by measuring characteristics of cellular components of a cellular pathway. Cellular pathways are well known in the art. In some embodiments the cellular pathway is a signaling pathway. Signaling pathways are also well known in the art (see, e.g., Hunter T., Cell 100(1): 113-27 (2000); Cell Signaling Technology, Inc., 2002 Catalogue, Pathway Diagrams pgs. 232-253 and U.S. Patent Application Serial No. 12/460,029). A condition involving or characterized by altered physiological status may be readily identified, for example, by determining the state in a cell of one or more activatable elements, as taught herein.

[00147] In some embodiments, the present invention is directed to methods for classifying one or more cells in a sample derived from an individual having or suspected of having a condition, such as MDS. In some embodiments, the invention allows for identification of prognostically and therapeutically relevant subgroups of the conditions and prediction of the clinical course of an individual. In some embodiments, the invention provides methods of classifying a cell according to the activation levels of one or more activatable elements in a cell from an individual having or suspected of having a condition. In some embodiments, the classification includes classifying the cell as a cell that is correlated with a clinical outcome. The clinical outcome can be the prognosis and/or diagnosis of a condition, and/or staging or grading of a condition. In some embodiments, the classifying of the cell includes classifying the cell as a cell that is correlated with a patient response to a treatment. In some embodiments, the classifying of the cell includes classifying the cell as a cell that is correlated with minimal residual disease or emerging resistance.

[00148]Activatable elements

[00149] The methods and compositions of the invention may be employed to examine and profile the status of any activatable element in a cellular pathway, or collections of such activatable elements. Single or multiple distinct pathways may be profiled (sequentially or simultaneously), or subsets of activatable elements within a single pathway or across multiple pathways may be examined (again, sequentially or simultaneously). In some embodiments, apoptosis, signaling, cell cycle and/or DNA damage pathways are characterized in order to classify one or more cells in an individual. The characterization of multiple pathways can reveal operative pathways in a condition that can then be used to classify one or more cells in an individual. In some embodiments, the classification includes classifying the cell as a cell that is correlated with a clinical outcome. The clinical outcome can be the prognosis and/or diagnosis of a condition, and/or staging or grading of a condition. In some embodiments, the classifying of the cell includes classifying the cell as a cell that is correlated with a patient response to a treatment. In some embodiments, the classifying of the cell includes classifying the cell as a cell that is correlated with minimal residual disease or emerging resistance.

[00150] As will be appreciated by those in the art, a wide variety of activation events can find use in the present invention. In general, the basic requirement is that the activation results in a change in the activatable protein that is detectable by some indication (termed an "activation state indicator"), preferably by altered binding of a labeled binding element or by changes in detectable biological activities (e.g., the activated state has an enzymatic activity which can be measured and compared to a lack of activity in the non-activated state). What is important is to differentiate, using detectable events or moieties, between two or more activation states (e.g. "off and "on").

[00151] The activation state of an individual activatable element is either in the on or off state. As an illustrative example, and without intending to be limited to any theory, an individual

phosphorylatable site on a protein can activate or deactivate the protein. Additionally, phosphorylation of an adapter protein may promote its interaction with other components/proteins of distinct cellular signaling pathways. The terms "on" and "off," when applied to an activatable element that is a part of a cellular constituent, are used here to describe the state of the activatable element, and not the overall state of the cellular constituent of which it is a part. Typically, a cell possesses a plurality of a particular protein or other constituent with a particular activatable element and this plurality of proteins or constituents usually has some proteins or constituents whose individual activatable element is in the on state and other proteins or constituents whose individual activatable element is in the off state. Since the activation state of each activatable element is measured through the use of a binding element that recognizes a specific activation state, only those activatable elements in the specific activation state recognized by the binding element, representing some fraction of the total number of activatable elements, will be bound by the binding element to generate a measurable signal. The measurable signal corresponding to the summation of individual activatable elements of a particular type that are activated in a single cell is the "activation level" for that activatable element in that cell. The measurable signal can be produced by the binding element and/or the activatable element. The measurable signal can be produced by the activatable element after the activatable element has been dissociated from the binding element.

[00152] Activation levels for a particular activatable element may vary among individual cells so that when a plurality of cells is analyzed, the activation levels follow a distribution. The distribution may be a normal distribution, also known as a Gaussian distribution, or it may be of another type. Different populations of cells may have different distributions of activation levels that can then serve to distinguish between the populations.

[00153] In some embodiments, the basis for classifying cells is that the distribution of activation levels for one or more specific activatable elements will differ among different phenotypes. A certain activation level, or more typically a range of activation levels for one or more activatable elements seen in a cell or a population of cells, is indicative that that cell or population of cells belongs to a distinctive phenotype. Other measurements, such as cellular levels (e.g., expression levels) of biomolecules that may not contain activatable elements, may also be used to classify cells in addition to activation levels of activatable elements; it will be appreciated that these levels also will follow a distribution, similar to activatable elements. Thus, the activation level or levels of one or more activatable elements, optionally in conjunction with levels of one or more levels of biomolecules that may or may not contain activatable elements, of cell or a population of cells may be used to classify a cell or a population of cells into a class. Once the activation level of intracellular activatable elements of individual single cells is known they can be placed into one or more classes, e.g., a class that corresponds to a phenotype. A class encompasses a class of cells wherein every cell has the same or substantially the same known activation level, or range of activation levels, of one or more intracellular activatable elements. For example, if the activation levels of five intracellular activatable elements are analyzed, predefined classes of cells that encompass one or more of the intracellular activatable elements can be constructed based on the activation level, or ranges of the activation levels, of each of these five elements. It is understood that activation levels can exist as a distribution and that an activation level of a particular element used to classify a cell may be a particular point on the distribution but more typically may be a portion of the distribution.

[00154] In some embodiments, the basis for classifying cells may use the position of a cell in a contour or density plot. The contour or density plot represents the number of cells that share a characteristic such as the activation level of activatable proteins in response to a modulator. For example, when referring to activation levels of activatable elements in response to one or more modulators, normal individuals and patients with a condition might show populations with increased activation levels in response to the one or more modulators. However, the number of cells that have a specific activation level (e.g. specific amount of an activatable element) might be different between normal individuals and patients with a condition. Thus, a cell can be classified according to its location within a given region in the contour or density plot. In other embodiments, the basis for classifying cells may use a series of population clusters whose centers, centroids, boundaries, relative positions describe the state of a cell, the diagnosis or prognosis of a patient, selection of treatment, or predicting response to treatment or to a combination of treatments, or long term outcome.

[00155] In some embodiments, the basis for classifying cells may use an N-dimensional Eigen map that describe the state of a cell, the diagnosis or prognosis of a patient, selection of treatment, or predicting response to treatment or to a combination of treatments, or long term outcome.

[00156] In other embodiments, the basis for classifying cells may use a Bayesian inference network of activatable elements interaction capabilities that together, or in part, describe the state of a cell, the diagnosis or prognosis of a patient, selection of treatment, or predicting response to treatment or to a combination of treatments, or long term outcome. See U.S. Patent Application Serial No. 1 1/338,957 and U.S. Patent Publication No. 2007/0009923 entitled Use of Bayesian Networks for Modeling Signaling Systems, incorporated herein by reference on its entirety. [00157] In addition to activation levels of intracellular activatable elements, levels of intracellular or extracellular biomolecules, e.g., proteins, may be used alone or in combination with activation states of activatable elements to classify cells. Further, additional cellular elements, e.g., biomolecules or molecular complexes such as RNA, DNA, carbohydrates, metabolites, and the like, may be used in conjunction with activatable states or expression levels in the classification of cells encompassed here.

[00158] In some embodiments, cellular redox signaling nodes are analyzed for a change in activation level. Reactive oxygen species (ROS) are involved in a variety of different cellular processes ranging from apoptosis and necrosis to cell proliferation and carcinogenesis. ROS can modify many intracellular signaling pathways including protein phosphatases, protein kinases, and transcription factors. This activity may indicate that the majority of the effects of ROS are through their actions on signaling pathways rather than via non-specific damage of

macromolecules. The exact mechanisms by which redox status induces cells to proliferate or to die, and how oxidative stress can lead to processes evoking tumor formation are still under investigation. See Mates, JM et al., Arch Toxicol. 2008 May:82(5):271 -2; Galaris D., et al., Cancer Lett. 2008 Jul 18;266(l)21 -9.

[00159] Reactive oxygen species can be measured. One example technique is by flow cytometry. See Chang et al., Lymphocyte proliferation modulated by glutamine: involved in the endogenous redox reaction; Clin Exp Immunol. 1999 September; 1 17(3): 482-488. Redox potential can be evaluated by means of an ROS indicator, one example being 2',7'-dichlorofluorescein-diacetate (DCFH-DA) which is added to the cells at an exemplary time and temperature, such as 37°C for 15 minutes. DCF peroxidation can be measured using flow cytometry. See Yang KD, Shaio MF. Hydroxyl radicals as an early signal involved in phorbol ester- induced monocyte differentiation of HL60 cells. Biochem Biophys Res Commun. 1994;200: 1650-7 and Wang JF, Jerrells TR, Spitzer JJ. Decreased production of reactive oxygen intermediates is an early event during in vitro apoptosis of rat thymocytes. Free Radic Biol Med. 1996;20:533-42.

[00160] In some embodiments, other characteristics that affect the status of a cellular constituent may also be used to classify a cell. Examples include the translocation of biomolecules or changes in their turnover rates and the formation and disassociation of complexes of biomolecule. Such complexes can include multi-protein complexes, multi-lipid complexes, homo- or hetero-dimers or oligomers, and combinations thereof. Other characteristics include proteolytic cleavage, e.g. from exposure of a cell to an extracellular protease or from the intracellular proteolytic cleavage of a biomolecule.

[00161] In some embodiments, cellular pH is analyzed. See June, CH and Moore, and JS, Curr Protoc Immunol, 2004 Dec;Chapter 5:Unit 5.5; Leyval, D et al., Flow cytometry for the intracellular pH measurement of glutamate producing Corynebacterium glutamicum, Journal of Microbiological Methods, Volume 29, Issue 2, 1 May 1997, Pages 121 -127; Weider, ED, et al., Measurement of intracellular H using flow cytometry with carboxy-SNARF-1. Cytometry, 1993 Nov; 14(8):916- 21 ; and Valli, M, et al., Intracellular pH Distribution in Saccharomyces cerevisiae Cell

Populations, Analyzed by Flow Cytometry, Applied and Environmental Microbiology, March 2005, p. 1515-1521, Vol. 71 , No. 3.

[00162] In some embodiments, the activatable element is the phosphorylation of immunoreceptor tyrosine-based inhibitory motif (ITIM). An immunoreceptor tyrosine-based inhibition motif (ITIM), is a conserved sequence of amino acids (S/I/V/LxYxxI/V/L) that is found in the cytoplasmic tails of many inhibitory receptors of the immune system. After ITIM-possessing inhibitory receptors interact with their ligand, their ITIM motif becomes phosphorylated by enzymes of the Src family of kinases, allowing them to recruit other enzymes such as the phosphotyrosine phosphatases SHP-1 and SHP-2, or the inositol-phosphatase called SHIP. These phosphatases decrease the activation of molecules involved in cell signaling. See Barrow A, Trowsdale J (2006). "You say IT AM and I say ITIM, let's call the whole thing off: the ambiguity of immunoreceptor signalling". Eur J Immunol 36 (7): 1646-53. When phosphorylated, these phospho-tyrosine residues provide docking sites for the Shps which may result in transmission of inhibitory signals and effect the signaling of neighboring membrane receptor complexes (Paul et al., Blood (2000 96:483).

[00163] ITIMs can be analyzed by flow cytometry.

[00164] Additional elements may also be used to classify a cell, such as the expression level of

extracellular or intracellular markers, nuclear antigens, enzymatic activity, protein expression and localization, cell cycle analysis, chromosomal analysis, cell volume, and morphological characteristics like granularity and size of nucleus or other distinguishing characteristics. For example, B cells can be further subdivided based on the expression of cell surface markers such as CD 19, CD20, CD22 or CD23.

[00165] Alternatively, predefined classes of cells can be aggregated or grouped based upon shared characteristics that may include inclusion in one or more additional predefined class or the presence of extracellular or intracellular markers, similar gene expression profile, nuclear antigens, enzymatic activity, protein expression and localization, cell cycle analysis,

chromosomal analysis, cell volume, and morphological characteristics like granularity and size of nucleus or other distinguishing cellular characteristics.

[00166] In some embodiments, the physiological status of one or more cells is determined by

examining and profiling the activation level of one or more activatable elements in a cellular pathway. In some embodiments, a cell is classified according to the activation level of a plurality of activatable elements. In some embodiments, a hematopoietic cell is classified according to the activation levels of a plurality of activatable elements. In some embodiments, 1 , 2, 3, 4, 5, 6, 7, 8, 9, 10 or more activatable elements may be analyzed in a cell signaling pathway. In some embodiments, the activation levels of one or more activatable elements of a hematopoietic cell are correlated with a condition. In some embodiments, the activation levels of one or more activatable elements of a hematopoietic cell are correlated with a neoplastic or hematopoietic condition as described herein. Examples of hematopoietic cells include, but are not limited to, MDS cells.

[00167] In some embodiments, the activation level of one or more activatable elements in single cells in the sample is determined. Cellular constituents that may include activatable elements include without limitation proteins, carbohydrates, lipids, nucleic acids and metabolites. The activatable element may be a portion of the cellular constituent, for example, an amino acid residue in a protein that may undergo phosphorylation, or it may be the cellular constituent itself, for example, a protein that is activated by translocation, change in conformation (due to, e.g., change in pH or ion concentration), by proteolytic cleavage, degradation through ubiquitination and the like. Upon activation, a change occurs to the activatable element, such as covalent modification of the activatable element (e.g., binding of a molecule or group to the activatable element, such as phosphorylation) or a conformational change. Such changes generally contribute to changes in particular biological, biochemical, or physical properties of the cellular constituent that contains the activatable element. The state of the cellular constituent that contains the activatable element is determined to some degree, though not necessarily completely, by the state of a particular activatable element of the cellular constituent. For example, a protein may have multiple activatable elements, and the particular activation states of these elements may overall determine the activation state of the protein; the state of a single activatable element is not necessarily determinative. Additional factors, such as the binding of other proteins, pH, ion concentration, interaction with other cellular constituents, and the like, can also affect the state of the cellular constituent.

[00168] In some embodiments, the activation levels of a plurality of intracellular activatable elements in single cells are determined. In some embodiments, at least about 2, 3, 4, 5, 6, 7, 8, 9, 10 or more than 10 intracellular activatable elements are determined.

[00169] Activation states of activatable elements may result from chemical additions or modifications of biomolecules and include biochemical processes such as glycosylation, phosphorylation, acetylation, methylation, biotinylation, glutamylation, glycylation, hydroxy lation, isomerization, prenylation, myristoylation, lipoylation, phosphopantetheinylation, sulfation, ISGylation, nitrosylation, palmitoylation, SUMOylation, ubiquitination, neddylation, citrullination, amidation, and disulfide bond formation, disulfide bond reduction. Other possible chemical additions or modifications of biomolecules include the formation of protein carbonyls, direct modifications of protein side chains, such as o-tyrosine, chloro-, nitrotyrosine, and dityrosine, and protein adducts derived from reactions with carbohydrate and lipid derivatives. Other modifications may be non-covalent, such as binding of a ligand or binding of an allosteric modulator. [00170] One example of a covalent modification is the substitution of a phosphate group for a hydroxyl group in the side chain of an amino acid (phosphorylation). A wide variety of proteins are known that recognize specific protein substrates and catalyze the phosphorylation of serine, threonine, or tyrosine residues on their protein substrates. Such proteins are generally termed "kinases." Substrate proteins that are capable of being phosphorylated are often referred to as phosphoproteins (after phosphorylation). Once phosphorylated, a substrate phosphoprotein may have its phosphorylated residue converted back to a hydroxyl one by the action of a protein phosphatase that specifically recognizes the substrate protein. Protein phosphatases catalyze the replacement of phosphate groups by hydroxyl groups on serine, threonine, or tyrosine residues. Through the action of kinases and phosphatases a protein may be reversibly phosphorylated on a multiplicity of residues and its activity may be regulated thereby. Thus, the presence or absence of one or more phosphate groups in an activatable protein is a preferred readout in the present invention.

[00171]Another example of a covalent modification of an activatable protein is the acetylation of histones. Through the activity of various acetylases and deacetlylases the DNA binding function of histone proteins is tightly regulated. Furthermore, histone acetylation and histone deactelyation have been linked with malignant progression. See Nature, 2004 May 27; 429(6990): 457-63.

[00172]Another form of activation involves cleavage of the activatable element. For example, one form of protein regulation involves proteolytic cleavage of a peptide bond. While random or misdirected proteolytic cleavage may be detrimental to the activity of a protein, many proteins are activated by the action of proteases that recognize and cleave specific peptide bonds. Many proteins derive from precursor proteins, or pro-proteins, which give rise to a mature isoform of the protein following proteolytic cleavage of specific peptide bonds. Many growth factors are synthesized and processed in this manner, with a mature isoform of the protein typically possessing a biological activity not exhibited by the precursor form. Many enzymes are also synthesized and processed in this manner, with a mature isoform of the protein typically being enzymatically active, and the precursor form of the protein being enzymatically inactive. This type of regulation is generally not reversible. Accordingly, to inhibit the activity of a

proteolytically activated protein, mechanisms other than "reattachment" must be used. For example, many proteolytically activated proteins are relatively short-lived proteins, and their turnover effectively results in deactivation of the signal. Inhibitors may also be used. Among the enzymes that are proteolytically activated are serine and cysteine proteases, including cathepsins and caspases respectively.

[00173] In one embodiment, the activatable enzyme is a caspase. The caspases are an important class of proteases that mediate programmed cell death (referred to in the art as "apoptosis"). Caspases are constitutively present in most cells, residing in the cytosol as a single chain proenzyme. These are activated to fully functional proteases by a first proteolytic cleavage to divide the chain into large and small caspase subunits and a second cleavage to remove the N-terminal domain. The subunits assemble into a tetramer with two active sites (Green, Cell 94:695-698, 1998). Many other proteolytically activated enzymes, known in the art as "zymogens," also find use in the instant invention as activatable elements.

[00174] In an alternative embodiment the activation of the activatable element involves prenylation of the element. By "prenylation", and grammatical equivalents used herein, is meant the addition of any lipid group to the element. Common examples of prenylation include the addition of farnesyl groups, geranyl groups, myristoylation and palmitoylation. In general these groups are attached via thioether linkages to the activatable element, although other attachments may be used.

[00175] In alternative embodiment, activation of the activatable element is detected as intermolecular clustering of the activatable element. By "clustering" or "multimerization", and grammatical equivalents used herein, is meant any reversible or irreversible association of one or more signal transduction elements. Clusters can be made up of 2, 3, 4, etc., elements. Clusters of two elements are termed dimers. Clusters of 3 or more elements are generally termed oligomers, with individual numbers of clusters having their own designation; for example, a cluster of 3 elements is a trimer, a cluster of 4 elements is a tetramer, etc.

[00176] Clusters can be made up of identical elements or different elements. Clusters of identical elements are termed "homo" dimers, while clusters of different elements are termed "hetero" clusters. Accordingly, a cluster can be a homodimer, as is the case for the 2-adrenergic receptor.

[00177] Alternatively, a cluster can be a heterodimer, as is the case for GABA _B-R. In other

embodiments, the cluster is a homotrimer, as in the case of TNFa, or a heterotrimer such the one formed by membrane-bound and soluble CD95 to modulate apoptosis. In further embodiments the cluster is a homo-oligomer, as in the case of Thyrotropin releasing hormone receptor, or a hetero-oligomer, as in the case of TGF i .

[00178] In another embodiment, the activation or signaling potential of elements is mediated by

clustering, irrespective of the actual mechanism by which the element's clustering is induced. For example, elements can be activated to cluster a) as membrane bound receptors by binding to ligands (ligands including both naturally occurring and synthetic ligands), b) as membrane bound receptors by binding to other surface molecules, or c) as intracellular (non-membrane bound) receptors binding to ligands.

[00179] In some embodiments, the activatable element is a protein. Examples of proteins that may include activatable elements include, but are not limited to kinases, phosphatases, lipid signaling molecules, adaptor/scaffo Id proteins, cytokines, cytokine regulators, ubiquitination enzymes, adhesion molecules, cytoskeletal/contractile proteins, heterotrimeric G proteins, small molecular weight GTPases, guanine nucleotide exchange factors, GTPase activating proteins, caspases, proteins involved in apoptosis, cell cycle regulators, molecular chaperones, metabolic enzymes, vesicular transport proteins, hydroxylases, isomerases, deacetylases, methylases, demethylases, tumor suppressor genes, proteases, ion channels, molecular transporters, transcription

factors/DNA binding factors, regulators of transcription, and regulators of translation. Examples of activatable elements, activation states and methods of determining the activation level of activatable elements are described in US Publication Number 20060073474 entitled "Methods and compositions for detecting the activation state of multiple proteins in single cells" and US Publication Number 20050112700 entitled "Methods and compositions for risk stratification" the content of which are incorporate here by reference. See also U.S. Provisional App. Nos.

61/048,886 and 61/048,920; and Shulzet al., Current Protocols in Immunology 2007, 78:8.17.1 - 20.

80] In some embodiments, the activatable protein (or other measurable protein) is selected from the group consisting of DNMT 1 , 3a, 3b, pH2AX, pH3, cyclin B, STATs 1 , 3, and 5. Other embodiments include HER receptors, PDGF receptors, Kit receptor, FGF receptors, Eph receptors, Trk receptors, IGF receptors, Insulin receptor, Met receptor, Ret, VEGF receptors, TIE1 , TIE2, FAK, Jakl , Jak2, Jak3, Tyk2, Src, Lyn, Fyn, Lck, Fgr, Yes, Csk, Abl, Btk, ZAP70, Syk, IRAKs, cRaf, ARaf, BRAF, Mos, Lim kinase, ILK, Tpl, ALK, TGFP receptors, BMP receptors, MEKKs, ASK, MLKs, DLK, PAKs, Mek 1 , Mek 2, MKK3/6, MKK4/7, ASKl,Cot, NIK, Bub, Myt 1 , Weel , Casein kinases, PDK1 , SGK1, SGK2, SGK3, Aktl , Akt2, Akt3, p90Rsks, p70S6 Kinase, Prks, PKCs, PKAs, ROCK 1, ROCK 2, Auroras, CaMKs, MNKs, AMPKs, MELK, MARKs, Chkl , Chk2, LKB-1 , MAPKAPKs, Piml , Pim2, Pim3, IKKs, Cdks, Jnks, Erks, IKKs, GSK3a, GSK3P, Cdks, CLKs, PKR, PI3-Kinase class 1 , class 2, class 3, mTor, SAPK/JNK1 ,2,3, p38s, PKR, DNA-PK, ATM, ATR, Receptor protein tyrosine phosphatases (RPTPs), LAR phosphatase, CD45, Non receptor tyrosine phosphatases (NPRTPs), SHPs, MAP kinase phosphatases (MKPs), Dual Specificity phosphatases (DUSPs), CDC25 phosphatases, Low molecular weight tyrosine phosphatase, Eyes absent (EYA) tyrosine phosphatases, Slingshot phosphatases (SSH), serine phosphatases, PP2A, PP2B, PP2C, PP1 , PP5, inositol phosphatases, PTEN, SHIPs, myotubularins, phosphoinositide kinases, phopsholipases, prostaglandin synthases, 5-lipoxygenase, sphingosine kinases, sphingomyelinases,

adaptor/scaffold proteins, She, Grb2, BLNK, LAT, B cell adaptor for PI3-kinase (BCAP), SLAP, Dok, KSR, MyD88, Crk, CrkL, GAD, Nek, Grb2 associated binder (GAB), Fas associated death domain (FADD), TRADD, TRAF2, RIP, T-Cell leukemia family, IL-2, IL-4, IL-8, IL-6, interferon γ, interferon a, suppressors of cytokine signaling (SOCs), Cbl, SCF ubiquitination ligase complex, APC/C, adhesion molecules, integrins, Immunoglobulin- like adhesion molecules, selectins, cadherins, catenins, focal adhesion kinase, pl30CAS, fodrin, actin, paxillin, myosin, myosin binding proteins, tubulin, eg5/KSP, CENPs, β-adrenergic receptors, muscarinic receptors, adenylyl cyclase receptors, small molecular weight GTPases, H-Ras, K-Ras, N-Ras, Ran, Rac, Rho, Cdc42, Arfs, RABs, RHEB, Vav, Tiam, Sos, Dbl, PRK, TSC1 ,2, Ras-GAP, Arf-GAPs, Rho-GAPs, caspases, Caspase 2, Caspase 3, Caspase 6, Caspase 7, Caspase 8, Caspase 9, Bcl-2, Mcl-1 , Bcl-XL, Bcl-w, Bcl-B, Al , Bax, Bak, Bok, Bik, Bad, Bid, Bim, Bmf, Hrk, Noxa, Puma, IAPs, XIAP, Smac, Cdk4, Cdk 6, Cdk 2, Cdkl , Cdk 7, Cyclin D, Cyclin E, Cyclin A, Cyclin B, Rb, i 6, pl4Arf, p27KIP, p21CIP, molecular chaperones, Hsp90s, Hsp70, Hsp27, metabolic enzymes, Acetyl-CoAa Carboxylase, ATP citrate lyase, nitric oxide synthase, caveolins, endosomal sorting complex required for transport (ESCRT) proteins, vesicular protein sorting (Vsps), hydroxylases, prolyl-hydroxylases PHD-1 , 2 and 3, asparagine hydroxylase FIH transferases, Pinl prolyl isomerase, topoisomerases, deacetylases, Histone deacetylases, sirtuins, histone acetylases, CBP/P300 family, MYST family, ATF2, DNA methyl transferases, Histone H3K4 demethylases, H3K27, JHDM2A, UTX, VHL, WT-1 , p53, Hdm, PTEN, ubiquitin proteases, urokinase-type plasminogen activator (uPA) and uPA receptor (uPAR) system, cathepsins, metalloproteinases, esterases, hydrolases, separase, potassium channels, sodium channels, , multi-drug resistance proteins, P-Gycoprotein, nucleoside transporters, Ets, Elk, SMADs, Rel-A (p65-NFKB), CREB, NFAT, ATF-2, AFT, Myc, Fos, Spl, Egr-1, T-bet, β- catenin, HIFs, FOXOs, E2Fs, SRFs, TCFs, Egr-1, β-catenin, FOXO, STAT1, STAT 3, STAT 4, STAT 5, STAT 6, p53, WT-1, HMGA, pS6, 4EPB-1 , eIF4E-binding protein, RNA polymerase, initiation factors, elongation factors.

[00181] In another embodiment the activatable element is a nucleic acid. Activation and deactivation of nucleic acids can occur in numerous ways including, but not limited to, cleavage of an inactivating leader sequence as well as covalent or non-covalent modifications that induce structural or functional changes. For example, many catalytic RNAs, e.g. hammerhead ribozymes, can be designed to have an inactivating leader sequence that deactivates the catalytic activity of the ribozyme until cleavage occurs. An example of a covalent modification is methylation of DNA. Deactivation by methylation has been shown to be a factor in the silencing of certain genes, e.g. STAT regulating SOCS genes in lymphomas. See Leukemia, February 2004; 18(2): 356-8. SOCS 1 and SHP1 hypermethylation in mantle cell lymphoma and follicular lymphoma: implications for epigenetic activation of the Jak/STAT pathway. Chim C S, Wong K Y, Loong F, Srivastava G.

[00182] In another embodiment the activatable element is a small molecule, carbohydrate, lipid or other naturally occurring or synthetic compound capable of having an activated isoform. In addition, as pointed out above, activation of these elements need not include switching from one form to another, but can be detected as the presence or absence of the compound. For example, activation of cAMP (cyclic adenosine mono-phosphate) can be detected as the presence of cAMP rather than the conversion from non-cyclic AMP to cyclic AMP.

[00183] In some embodiments of the invention, the methods described herein are employed to

determine the activation level of an activatable element, e.g., in a cellular pathway. Methods and compositions are provided for the classification of a cell according to the activation level of an activatable element in a cellular pathway. The cell can be a hematopoietic cell. Examples of hematopoietic cells include but are not limited to pluripotent hematopoietic stem cells, granulocyte lineage progenitor or derived cells, monocyte lineage progenitor or derived cells, macrophage lineage progenitor or derived cells, megakaryocyte lineage progenitor or derived cells and erythroid lineage progenitor or derived cells.

[00184] In some embodiments, the cell is classified according to the activation level of an activatable element, e.g., in a cellular pathway comprises classifying the cell as a cell that is correlated with a clinical outcome. In some embodiments, the clinical outcome is the prognosis and/or diagnosis of a condition. In some embodiments, the clinical outcome is the presence or absence of a neoplastic or a hematopoietic condition. In some embodiments, the clinical outcome is the staging or grading of a neoplastic or hematopoietic condition. Examples of staging include, but are not limited to, aggressive, indolent, benign, refractory, Roman Numeral staging, TNM Staging, Rai staging, Binet staging, WHO classification, FAB classification, IPSS score, WPSS score, limited stage, extensive stage, staging according to cellular markers such as ZAP70 and CD38, occult, including information that may inform on time to progression, progression free survival, overall survival, or event-free survival.

[00185] In some embodiments, methods and compositions are provided for the classification of a cell according to the activation level of an activatable element, e.g., in a cellular pathway wherein the classification comprises classifying a cell as a cell that is correlated to a patient response to a treatment. In some embodiments, the patient response is selected from the group consisting of complete response, partial response, nodular partial response, no response, progressive disease, stable disease and adverse reaction.

[00186] In some embodiments, methods and compositions are provided for the classification of a cell according to the activation level of an activatable element, e.g., in a cellular pathway wherein the classification comprises classifying the cell as a cell that is correlated with minimal residual disease or emerging resistance.

[00187] In some embodiments, methods and compositions are provided for the classification of a cell according to the activation level of an activatable element, e.g., in a cellular pathway wherein the classification comprises selecting a method of treatment. Example of methods of treatments include, but are not limited to, chemotherapy, biological therapy, radiation therapy, bone marrow transplantation, Peripheral stem cell transplantation, umbilical cord blood transplantation, autologous stem cell transplantation, allogeneic stem cell transplantation, syngeneic stem cell transplantation, surgery, induction therapy, maintenance therapy, and watchful waiting.

[00188] Generally, the methods of the invention involve determining the activation levels of an

activatable element in a plurality of single cells in a sample

[00189] Measurements of cytoxicity can include with apoptosis, necrosis, and/or autophagy, including but not limited to caspase cleavage products such as dye substrates, cleaved PARP, cleaved cytokeratin 18, total or cleaved caspases (i.e. caspase-2, caspase-3, caspase-6, caspase-7, caspase-8, caspase-9, caspase-10, or other caspase family members) cytochrome C, apoptosis inducing factor (AIF), Inhibitor of Apoptosis (IAP) family members, survivin, as well as other molecules such as Bcl-2 family members including anti-apoptotic proteins (MCL-1, BCL-2, BCL-XL), BH3-only apoptotic sensitizers (PUMA, NOXA, Bim, Bad), and pro-apoptotic proteins (Bad, Bax), p53, c-myc proto-oncogene, APO-l/Fas/CD95, TRADD, FADD, FasL, growth stimulating genes, or tumor suppressor genes, mitochondrial membrane dyes, Annexin-V, 7-AAD, Amine Aqua, trypan blue, propidium iodide or other viability dyes. In some

embodiments the compounds are cPARP, Amine Aqua, or pH2AX among other compounds.

[00190] Measurements of cytostasis can include , cyclin or cyclin dependent kinase (cdk) proteins, such as cyclin A, cyclin A2, cyclin B, eye line Bl, cyclin D, cyclin E, KI-67, CDK1, CDK2, CDK3, CDK4, CDK5, CDK6, CDK7, CDK8, CDK9, CDK10, CDK11, CDK12, CDK13;

regulators of cyclin-cdk complexes, such as Wee, CDK-activating kinase (CAK), Cdc20 and Cdc25; retinoblastoma susceptibility protein (Rb); cell cycle inhibitor proteins, such as cip/kip family proteins, such as p21, p27, p57; p53; Tumor Growth Factor beta (TGF ); INK4a/ARF family proteins such as pl6INK4a and pl4ARF. Other cell cycle pathway activatable elements include, but are not limited to, Plks such as Plkl, BrdU, p-mpm2, p-aurora, Histone H3, and components of the DNA Damage checkpoint response such as Chkl, pChkl, Chk2, pChk2, pH2AX. In some embodiments cyclin Bl is used as the measurement for cytostasis.

[00191] Signaling Pathways

[00192] In some embodiments, the methods of the invention are employed to determine the status of an activatable element in a signaling pathway. In some embodiments, a cell is classified, as described herein, according to the activation level of one or more activatable elements in one or more signaling pathways. Signaling pathways and their members have been described. See (Hunter T. Cell Jan. 7, 2000;100(1): 13-27). Exemplary signaling pathways include the following pathways and their members: The MAP kinase pathway including Ras, Raf, MEK, ERK and elk; the PI3K/Akt pathway including PI-3-kinase, PDK1, Akt and Bad; the NF-κΒ pathway including IKKs, IkB and the Wnt pathway including frizzled receptors, beta-catenin, APC and other co- factors and TCF (see Cell Signaling Technology, Inc. 2002 Catalog pages 231-279 and Hunter T., supra.). In some embodiments of the invention, the correlated activatable elements being assayed (or the signaling proteins being examined) are members of the MAP kinase, Akt, NFkB, WNT, RAS/RAF/MEK/ERK, JNK/SAPK, p38 MAPK, Src Family Kinases, JAK/STAT and/or PKC signaling pathways. See U.S. 12/460,029, including that application's Figure 1.

[00193] In some embodiments, the status of an activatable element within the PI3K/AKT, or MAPK pathways in response to a growth factor or mitogen is determined. In some embodiments, the activatable element within the PI3K/AKT or MAPK pathway is selected from the group consisting of Akt, p-Erk, p38. [00194] In some embodiments, the status of an activatable element within JAk/STAT pathways in response to a cytokine is determined. In some embodiments, the activatable element within the JAK/STAT pathway is selected from the group consisting of p-Stat3, p-Stat5, p-Statl, and p- Stat6.

[00195] In some embodiments, the status of an activatable element within the phospholipase C

pathway in response to an inhibitor is determined. In some embodiments, the activatable element within the phospholipase C pathway is selected from the group consisting of p-Slp-76, and Plcg2 and the inhibitor is H202.

[00196] In some embodiments, the status of a phosphatase in response to an inhibitor is determined. In some embodiments, the inhibitor is H2O2.

[00197] In some embodiments, the methods of the invention are employed to determine the status of a signaling protein in a signaling pathway known in the art including those described herein.

Exemplary types of signaling proteins within the scope of the present invention include, but are not limited to kinases, kinase substrates (i.e. phosphorylated substrates), phosphatases, phosphatase substrates, binding proteins (such as 14-3-3), receptor ligands and receptors (cell surface receptor tyrosine kinases and nuclear receptors)). Kinases and protein binding domains, for example, have been well described (see, e.g., Cell Signaling Technology, Inc., 2002

Catalogue "The Human Protein Kinases" and "Protein Interaction Domains" pgs. 254-279).

[00198] See U.S. Patent App. Ser. No. 12/910,769 for more information on signaling pathways.

[00199] Modulators

[00200] In some embodiments, the methods and composition utilize a modulator. A modulator can be an activator, a therapeutic compound, an inhibitor or a compound capable of impacting a cellular pathway. Modulators can also take the form of environmental cues and inputs.

[00201] Modulation can be performed in a variety of environments. In some embodiments, cells are exposed to a modulator immediately after collection. In some embodiments where there is a mixed population of cells, purification of cells is performed after modulation. In some embodiments, whole blood is collected to which a modulator is added. In some embodiments, cells are modulated after processing for single cells or purified fractions of single cells. As an illustrative example, whole blood can be collected and processed for an enriched fraction of lymphocytes that is then exposed to a modulator. Modulation can include exposing cells to more than one modulator. For instance, in some embodiments, cells are exposed to at least 2, 3, 4, 5, 6, 7, 8, 9, or 10 modulators. See U.S. Patent Application 12/460,029 which is incorporated by reference.

[00202] In some embodiments, cells are cultured post collection in a suitable media before exposure to a modulator. In some embodiments, the media is a growth media. In some embodiments, the growth media is a complex media that may include serum. In some embodiments, the growth media comprises serum. In some embodiments, the serum is selected from the group consisting of fetal bovine serum, bovine serum, human serum, porcine serum, horse serum, and goat serum. In some embodiments, the serum level ranges from 0.0001% to 30 %. In some embodiments, the growth media is a chemically defined minimal media and is without serum. In some

embodiments, cells are cultured in a differentiating media.

[00203] Modulators include chemical and biological entities, and physical or environmental stimuli.

Modulators can act extracellularly or intracellularly. Chemical and biological modulators include growth factors, mitogens, cytokines, drugs, immune modulators, ions, neurotransmitters, adhesion molecules, hormones, small molecules, inorganic compounds, polynucleotides, antibodies, natural compounds, lectins, lactones, chemotherapeutic agents, biological response modifiers, carbohydrate, proteases and free radicals. Modulators include complex and undefined biologic compositions that may comprise cellular or botanical extracts, cellular or glandular secretions, physiologic fluids such as serum, amniotic fluid, or venom. Physical and

environmental stimuli include electromagnetic, ultraviolet, infrared or particulate radiation, redox potential and pH, the presence or absences of nutrients, changes in temperature, changes in oxygen partial pressure, changes in ion concentrations and the application of oxidative stress. Modulators can be endogenous or exogenous and may produce different effects depending on the concentration and duration of exposure to the single cells or whether they are used in combination or sequentially with other modulators. Modulators can act directly on the activatable elements or indirectly through the interaction with one or more intermediary biomolecule. Indirect modulation includes alterations of gene expression wherein the expressed gene product is the activatable element or is a modulator of the activatable element.

[00204] In some embodiments the modulator is selected from the group consisting of growth factors, mitogens, cytokines, adhesion molecules, drugs, hormones, small molecules, polynucleotides, antibodies, natural compounds, lactones, chemotherapeutic agents, immune modulators, carbohydrates, proteases, ions, reactive oxygen species, peptides, and protein fragments, either alone or in the context of cells, cells themselves, viruses, and biological and non-biological complexes (e.g. beads, plates, viral envelopes, antigen presentation molecules such as major histocompatibility complex). In some embodiments, the modulator is a physical stimulus such as heat, cold, UV radiation, and radiation. Examples of modulators, include but are not limited to EPO, G-CSF, Lenalidomide, and the combination of Lenalidomide and EPO, GM-CSF, IL-6, IL- 22, TPO, Vidaza, Dacogen, or Zolinza (SAHA).

[00205] In one embodiment the modulators are stem cell growth factors. They include Activins, BMPs (Bone Morphogenetic Proteins), Common beta Chain Receptor Family, Common gamma Chain Receptor Family, EGF Family, FGF Family, Growth/Differentiation Factors (GDFs), Hedgehog Family, IGF Family, IL-6 Family, SCF, Flt-3 Ligand & M-CSF, VEGF & PDGF Families, and Wnt Pathways. [00206] Activins, members of the TGF-beta superfamily, are disulfide- linked dimeric proteins originally purified from gonadal fluids as proteins that stimulated pituitary follicle stimulating hormone (FSH) release. Activin proteins have a wide range of biological activities, including mesoderm induction, neural cell differentiation, bone remodeling, hematopoiesis and roles in reproductive physiology. Activin isoforms and other members of the TGF-beta superfamily exert their biological effects by binding to heteromeric complexes of a type I and a type II serine- threonine kinase receptor, both of which are essential for signal transduction.

[00207] Activins are homodimers or heterodimers of the various beta subunit isoforms, while inhibins are heterodimers of a unique alpha subunit and one of the various beta subunits. Five beta subunits (mammalian beta A, beta B, beta C, beta E and Xenopus beta D) have been cloned to date. The activin/inhibin nomenclature reflects the subunit composition of the proteins: Activin A (beta A - beta A), Activin B (beta B - beta B), Activin AB (beta A - beta B), Inhibin A (alpha - beta A) and Inhibin B (alpha - beta B).

[00208] Bone Morphogenetic Proteins (BMPS) are secreted signaling molecules that comprise a

subfamily of the TGF-beta superfamily and were originally identified as regulators of cartilage and bone formation. There are at least 20 structurally and functionally related BMPs, most of which play roles in embryogenesis and morphogenesis of various tissues and organs. Biologically active BMPs are usually homodimers containing a characteristic cysteine knot structure.

Heterodimers, BMP-2/BMP-7 and BMP-4/BMP-7 have also been suggested to exist and function in vivo. They are more potent inducers of bone formation than their respective homodimers. In addition, heterodimers, but not homodimers, are ventral mesoderm inducers. Heterodimer activity may be mediated by a different or additional receptor subtype.

[00209] Decapentaplegic (Dpp) is one of at least five TGF-beta superfamily ligands identified in the Drosophila genome. Dpp, a functional ortholog of mammalian BMP-2 and BMP-4, is a morphogen and plays an essential role in Drosophila development. Dpp regulates embryonic dorsal- ventral polarity and is required for gut morphogenesis and outgrowth and patterning of imaginal disks.

[00210] Common beta Chain Receptor Family . The receptors for human granulocyte macrophage- colony stimulating factor (GM-CSF), IL-3 and IL-5 are comprised of a cytokine-specific alpha chain, and a common beta chain. The common beta chain on GM-CSF, IL-3, and IL-5 receptors interacts with all three ligands, promoting some degree of overlap between their regulation of hematopoietic cell signaling.

[00211] GM-CSF is produced by a number of different cell types, including activated T cells, B cells, macrophages, mast cells, endothelial cells and fibroblasts, in response to cytokine or immune and inflammatory stimuli. IL-3, also known as mast cell growth factor, is a pleiotropic factor produced primarily by activated T cells. IL-3 can stimulate the proliferation and differentiation of pluripotent hematopoietic stem cells, as well as various lineage committed progenitors. IL-5 is a T cell-derived factor that promotes the proliferation, differentiation and activation of eosinophils. In mice, IL-5 is also a growth and differentiation factor for B cells.

[00212] Common gamma Chain Receptor Family. When IL-2 receptor alpha is associated with the IL- 2 receptor beta and gamma chains, a high affinity heterotrimeric receptor complex that transduces IL-2 signals is formed. The gamma chain is also the common signaling subunit of the high affinity receptor complex for IL-2, IL-4, IL-7, IL-9, IL-15 and IL-21. Interleukin 2 (IL-2) was initially identified as a T cell growth factor that is produced by T cells following activation by mitogens or antigens. Since then, it has also been shown to stimulate the growth and

differentiation of B cells, natural killer (NK) cells, lymphocyte activated killer (LAK) cells, monocytes/macrophages, and oligodendrocytes.

[00213] There is some degree of overlap between the biological effects of IL-4, IL-7, IL-9, and IL-15 on hematopoietic and non-hematopoietic cells, including B and T cells, monocytes, macrophages, mast cells, myeloid and erythroid progenitors, fibroblasts, and endothelial cells. In contrast, IL-21 is a pleiotropic cytokine, produced by T, NKT, and dendritic cells, which modulates lymphoid and myeloid cell functions.

[00214] EGF Family The members of the EGF family are best known for their ability to stimulate cell growth and proliferation and are important for many developmental processes including promoting mitogenesis and differentiation of mesenchymal and epithelial cells. EGF family members have at least one common structural motif, the EGF domain, which consists of six conserved cysteine residues forming three disulfide bonds. Most are synthesized in membrane- associated pro forms before liberation by proteolytic cleavage. Family members include EGF, Neuregulins, Amphiregulin, Betacellulin, and others. The activity of EGF family members is mediated by the EGF R/ErbB receptor tyrosine kinases. When unregulated, members of this family and their receptors are known to be involved in tumor formation.

[00215] EGF Ligands include Amphiregulin, Betacellulin/BTC, EGF, EGF-L6, Epigen, Epiregulin, HB-EGF, LRIG1 and 3, Neuregulin-l/NRGl , Neuregulin-1 alpha/NRGl alpha, Neuregulin-1 beta 1/NRGl beta 1 , Neuregulin-1 Isoform GGF2, Neuregulin-1 Isoform SMDF, Neuregulin- 3/NRG3, TGF-alpha, TMEFFl/Tomoregulin-1 , TMEFF2/Tomoregulin-2. See

http://www.rndsystems.com/resources/images/6639.pdf

[00216] The Fibroblast Growth Factors (FGFs) constitute a large family of proteins involved in many aspects of development including cell proliferation, growth, and differentiation. They act on several cell types to regulate diverse physiologic functions including angiogenesis, cell growth, pattern formation, embryonic development, metabolic regulation, cell migration, neurotrophic effects, and tissue repair. FGF family activities are mediated by receptor tyrosine kinases and are facilitated by heparan sulfate. Family members have been implicated in several disorders of bone growth, as well as in tumor formation and progression. The family includes FGF acidic, basic, FGF 3, 4, 5, 6, 7, 8, 9, 10, 1 1, 17, 19, 20, 21, 22, and 23. [00217] Growth/differentiation factors (GDF-1 to GDF-15) are members of the BMP family of TGF-beta superfamily proteins. They are produced as inactive preproproteins which are then cleaved and assembled into active secreted homodimers. GDF dimers are disulfide- linked with the exception of GDF-3 and -9. GDF proteins are important during embryonic development, particularly in the skeletal, nervous, and muscular systems.

[00218] Hedgehog Family The vertebrate hedgehog family is represented by at least three members:

Desert Hedgehog (Dhh), Indian Hedgehog (Ihh), and Sonic Hedgehog (Shh). Hedgehog proteins undergo autocatalytic processing and modification that is critical for signaling activity. The precursor protein is cleaved to yield an N-terminal domain and a C-terminal domain. Auto- processing of Hedgehog also causes the covalent attachment of cholesterol to the C-terminal side of the N-terminal domain. The N-terminal domain retains all known signaling capabilities, while the C-terminal domain is responsible for the intramolecular precursor processing, acting as a cholesterol transferase. Hedgehog signaling occurs through two transmembrane proteins: Patched (Ptc), a twelve-pass protein binds Hedgehog ligand; while Smoothened (Smo), a seven-pass protein is a signal transducer. Hedgehog signaling is involved in diverse areas of development. Among these are neurogenesis, hematopoiesis, bone formation, and gonad development.

[00219] Hedgehog Related Molecules & Regulators include BOC, CDO, DISP1 , Gasl , GLI-1 , 2, 3, Glypican 3, GSK-3 alpha/beta, GSK-3 alpha, GSK-3 beta, Hip, LIN-41 , MED 12, Patched 1/PTCH, and Patched 2/PTCH2.

[00220] Insulin- like growth factor (IGF)-I (also known as somatomedin C and somatomedin A) and IGF-II (multiplication stimulating activity or MSA) belong to the family of insulin- like growth factors that are structurally homologous to proinsulin. Mature IGF-I and IGF-II share approximately 70% sequence identity. Both IGF-I and IGF-II are expressed in many tissues and cell types and may have autocrine, paracrine and endocrine functions. Mature IGF-I and IGF-II are highly conserved between the human, bovine and porcine proteins (100% identity), and exhibit cross-species activity.

[00221] IGF-I receptor is a disulfide-linked heterotetrameric transmembrane protein consisting of two alpha and two beta subunits. Both the alpha and beta subunits are encoded within a single receptor precursor cDNA. The proreceptor polypeptide is proteolytically cleaved and disulfide- linked to yield the mature heterotetrameric receptor. The alpha subunit of IGF-I receptor is extracellular while the beta subunit has an extracellular domain, a transmembrane domain and a cytoplasmic tyrosine kinase domain. The IGF-I receptor is highly expressed in all cell types and tissues.

[00222] The superfamily of insulin- like growth factor (IGF) binding proteins include the six high- affinity IGF binding proteins (IGFBP) and at least four additional low-affinity binding proteins referred to as IGFBP related proteins (IGFBP-rP). All IGFBP superfamily members are cysteine- rich proteins with conserved cysteine residues, which are clustered in the amino- and carboxy- terminal thirds of the molecule. IGFBPs modulate the biological activities of IGF proteins. Some IGFBPs may also have intrinsic bioactivity that is independent of their ability to bind IGF proteins. Post-translational modifications of IGFBP, including glycosylation, phosphorylation and proteolysis, have been shown to modify the affinities of the binding proteins to IGF. ALS (Acid Labile Subunit) is a liver-derived protein that exists in a ternary complex with Insulin-like Growth Factor (IGF)-binding Protein-3 (IGFBP-3) or IGFBP-5, and either IGF-I or IGF-II. ALS increases the half- life of IGF/IGFBP complexes in circulation. IGF Ligands include IGF-I, IGF- II, andIGFL-3.

[00223] IL-6 Family includes IL-6, IL-1 1 , leukemia inhibitory factor (LIF), oncostatin M (OSM), cardiotrophin-1 (CT-1), ciliary neurotrophic factor (CNTF), and cardiotrophin-like cytokine (CLC) belong to the IL-6 family of cytokines. Inclusion in the IL-6 family is based on a helical cytokine structure and receptor subunit makeup. Members of the IL-6 family of cytokines activate the signal transducing receptor protein, glycoprotein 130 (gpl 30). As they share a common signal transducer, IL-6 family cytokines display both unique and overlapping biologic activities on multiple hematopoietic lineages. In addition, the IL-6 family also includes IL-31. The receptor heterodimer for IL-31 consists of a unique gpl 30-like receptor chain IL-31RA, and the receptor subunit oncastatin M receptor (OSMR) beta that is shared with OSM. IL-6 family binding to the gpl 30 alpha subunit induces homodimerization and subsequent activation of Janus kinases (JAK), followed by activation of signal transducers and activators of transcription (STAT1 and STAT3). The 11-6 family includes Cardiotrophin-1 /CT-1 , CLC, CLC/CNTF R alpha Chimera, CLF-1, CLF-1/CLC Complex, CNTF, CNTF R alpha, G-CSF, G-CSF R/CD1 14, gpl 30, IL-6, IL-6 R alpha, IL-1 1 , IL-1 1 R alpha, IL-31 , IL-31 RA, Jakl , 2, 3, Leptin/OB, Leptin R, LIF, and LIF R alpha, Neuropoietin/NP, Oncostatin M/OSM, and OSM R beta.

[00224] SCF, Flt-3 Ligand and M-CSF contain a 4-helix bundle structure in the extracellular domain and 4 conserved cysteines. Their receptors are tyrosine kinases.

[00225] VEGF & PDGF Families VEGF and PDGF family members are potent mitogenic and

angiogenic factors with critical roles in tumor formation as well as embryonic development and wound healing. Relatedness of these growth factors is based on both sequence and structural similarity. Each member contains an 80-90 amino acid VEGF/PDGF homology domain with several conserved cysteine residues, which are important for the formation of the characteristic cysteine knot structure. The family includes Neuropilin-1 and 2, PIGF, P1GF-2, VEGF,

VEGF/PIGF Heterodimer, Neuropilins 1 and 2, VEGF Family Ligands including PIGF, VEGF-B, P1GF-2, VEGF-C, VEGF, VEGF-D, and VEGF/PIGF. Heterodimer

PDGF Family Ligands PDGF, PDGF-BB, PDGF-A, PDGF-C, PDGF-AA, PDGF-CC, PDGF- AB, PDGF-D, PDGF-B, and PDGF-DD.

[00226] Wnt Pathways The molecular name Wnt is derived from Wingless, the Drosophila

melanogaster segment-polarity gene, and Integrase-1, the vertebrate homologue. The Wnt signaling pathway is a highly conserved signal transduction cascade that has a central role in embryonic development, tissue regeneration, and a host of other biological processes. There are three established Wnt signaling pathways: 1) the canonical pathway, involving beta-Catenin, 2) the planar cell polarity (PCP) pathway, and 3) the Wnt-Ca²⁺ pathway.

[00227] Target cell populations respond to secreted Wnt morphogens in a concentration dependent manner, such that the gradient of Wnt concentration determines the resulting gene expression and cellular differentiation. These critical actions make Wnt molecules central to the signal transduction pathways that underlie cell proliferation, survival and differentiation. The importance of Wnt signaling is underlined by the fact that deregulation of the Wnt pathway results in cancer and other disease conditions. In addition, recent studies have shown that Wnt molecules also play a role in the immune system. Wnt signaling has been shown to regulate T- cell development and activation, and dendritic cell maturation.

[00228] Stem cell growth factors are commercially available from R and D Systems, Minneapolis MN, Sigma Aldrich or Merck Chemical.

[00229] In some embodiments the modulator can be a lineage-specific growth factor, such as EPO, TPO, G-CSF or M-CSF. These growth factors work on committed precursor cells and typically as single agents. See K. Kaushansky, N Engl J Med 2006;354:2034-45.

[00230] In some embodiments, the modulator is an activator. In some embodiments the modulator is an inhibitor. In some embodiments, cells are exposed to one or more modulator. In some embodiments, cells are exposed to at least 2, 3, 4, 5, 6, 7, 8, 9, or 10 modulators. In some embodiments, cells are exposed to at least two modulators, wherein one modulator is an activator and one modulator is an inhibitor. In some embodiments, cells are exposed to at least 2, 3, 4, 5, 6, 7, 8, 9, or 10 modulators, where at least one of the modulators is an inhibitor.

[00231] In some embodiments, the cross-linker is a molecular binding entity. In some embodiments, the molecular binding entity is a monovalent, bivalent, or multivalent is made more multivalent by attachment to a solid surface or tethered on a nanoparticle surface to increase the local valency of the epitope binding domain.

[00232] In some embodiments, the inhibitor is an inhibitor of a cellular factor or a plurality of factors that participates in a cellular pathway (e.g. signaling cascade) in the cell. In some embodiments, the inhibitor is a phosphatase inhibitor. Examples of phosphatase inhibitors include, but are not limited to H2O2, siRNA, miRNA, Cantharidin, (-)-p-Bromotetramisole, Microcystin LR, Sodium Orthovanadate, Sodium Pervanadate, Vanadyl sulfate, Sodium oxodiperoxo(l , 10- phenanthroline)vanadate, bis(maltolato)oxovanadium(IV), Sodium Molybdate, Sodium Perm olybdate, Sodium Tartrate, Imidazole, Sodium Fluoride, β-Glycerophosphate, Sodium

Pyrophosphate Decahydrate, Calyculin A, Discodermia calyx, bpV(phen), mpV(pic), DMHV, Cypermethrin, Dephostatin, Okadaic Acid, NIPP-1 , N-(9, 10-Dioxo-9, 10-dihydro-phenanthren-2- yl)-2,2-dimethyl-propionamide, a-Bromo-4-hydroxyacetophenone, 4-Hydroxyphenacyl Br, a- Bromo-4-methoxyacetophenone, 4-Methoxyphenacyl Br, a-Bromo-4- (carboxymethoxy)acetophenone, 4-(Carboxymethoxy)phenacyl Br, and bis(4- Trifluoromethylsulfonamidophenyl)-l ,4-diisopropylbenzene, phenylarsine oxide, Pyrrolidine Dithiocarbamate, and Aluminum fluoride. In some embodiments, the phosphatase inhibitor is H₂0₂.

[00233] In some embodiments, the activation level of an activatable element in a cell is determined by contacting the cell with at least 2, 3, 4, 5, 6, 7, 8, 9, or 10 modulators. In some embodiments, the activation level of an activatable element in a cell is determined by contacting the cell with at least 2, 3, 4, 5, 6, 7, 8, 9, or 10 modulators where at least one of the modulators is an inhibitor. In some embodiments, the activation level of an activatable element in a cell is determined by contacting the cell with an inhibitor and a modulator, where the modulator can be an inhibitor or an activator. In some embodiments, the activation level of an activatable element in a cell is determined by contacting the cell with an inhibitor and an activator. In some embodiments, the activation level of an activatable element in a cell is determined by contacting the cell with two or more modulators.

[00234] In some embodiments, a phenotypic profile of a population of cells is determined by

measuring the activation level of an activatable element when the population of cells is exposed to a plurality of modulators in separate cultures.

[00235] Gating

[00236] In another embodiment, a user may analyze the signaling in subpopulations based on surface markers. For example, the user could look at CD34, CD45, CD235a, or CD71. In another alternative embodiment, a user may analyze the data based on intracellular markers, such as transcription factors or other intracellular proteins; based on a functional assay (i.e. dye negative "side population" aka drug transporter + cells, or fluorescent glucose uptake, or based on other fluorescent markers). In some embodiments, a gate is established after learning from a responsive subpopulation. That is, a gate is developed from one data set after finding a population that correlates with a clinical outcome. This gate can then be applied retrospectively or prospectively to other data sets.

[00237] In some embodiments where flow cytometry is used, prior to analyzing of data the populations of interest and the method for characterizing these populations are determined. For instance, there are at least two general ways of identifying populations for data analysis: (i) "Outside-in" comparison of Parameter sets for individual samples or subset (e.g., patients in a trial). In this more common case, cell populations are homogenous or lineage gated in such a way as to create distinct sets considered to be homogenous for targets of interest. An example of sample-level comparison would be the identification of signaling profiles in tumor cells of a patient and correlation of these profiles with non-random distribution of clinical responses. This is considered an outside-in approach because the population of interest is pre-defined prior to the mapping and comparison of its profile to other populations, (ii) "Inside-out" comparison of Parameters at the level of individual cells in a heterogeneous population. An example of this would be the signal transduction state mapping of mixed hematopoietic cells under certain conditions and subsequent comparison of computationally identified cell clusters with lineage- specific markers. This could be considered an inside-out approach to single cell studies as it does not presume the existence of specific populations prior to classification. A major drawback of this approach is that it creates populations which, at least initially, require multiple transient markers to enumerate and may never be accessible with a single cell surface epitope. As a result, the biological significance of such populations can be difficult to determine. The main advantage of this unconventional approach is the unbiased tracking of cell populations without drawing potentially arbitrary distinctions between lineages or cell types.

[00238] Each of these techniques capitalizes on the ability of flow cytometry to deliver large amounts of multiparameter data at the single cell level. For cells associated with a condition (e.g.

neoplastic or hematopoietic condition), a third "meta-level" of data exists because cells associated with a condition (e.g. cancer cells) are generally treated as a single entity and classified according to historical techniques. These techniques have included organ or tissue of origin, degree of differentiation, proliferation index, metastatic spread, and genetic or metabolic data regarding the patient.

[00239] In some embodiments, the present invention uses variance mapping techniques for mapping condition signaling space. These methods represent a significant advance in the study of condition biology because it enables comparison of conditions independent of a putative normal control. Traditional differential state analysis methods (e.g., DNA microarrays, subtractive Northern blotting) generally rely on the comparison of cells associated with a condition from each patient sample with a normal control, generally adjacent and theoretically untransformed tissue. Alternatively, they rely on multiple clusterings and reclusterings to group and then further stratify patient samples according to phenotype. In contrast, variance mapping of condition states compares condition samples first with themselves and then against the parent condition population. As a result, activation states with the most diversity among conditions provide the core parameters in the differential state analysis. Given a pool of diverse conditions, this technique allows a researcher to identify the molecular events that underlie differential condition pathology (e.g., cancer responses to chemotherapy), as opposed to differences between conditions and a proposed normal control.

[00240] In some embodiments, when variance mapping is used to profile the signaling space of patient samples, conditions whose signaling response to modulators is similar are grouped together, regardless of tissue or cell type of origin. Similarly, two conditions (e.g. two tumors) that are thought to be relatively alike based on lineage markers or tissue of origin could have vastly different abilities to interpret environmental stimuli and would be profiled in two different groups.

[00241] When groups of signaling profiles have been identified it is frequently useful to determine whether other factors, such as clinical responses, presence of gene mutations, and protein expression levels, are non-randomly distributed within the groups. If experiments or literature suggest such a hypothesis in an arrayed flow cytometry experiment, it can be judged with simple statistical tests, such as the Student's t-test and the X² test. Similarly, if two variable factors within the experiment are thought to be related, the Pearson, and/or Spearman are used to measure the degree of this relationship.

[00242] Examples of analysis for activatable elements are described in US publication number

20060073474 entitled "Methods and compositions for detecting the activation state of multiple proteins in single cells" and US publication number 200501 12700 entitled "Methods and compositions for risk stratification" the content of both are incorporate here by reference.

[00243] Detection

[00244] In practicing the methods of this invention, the detection of the status of the one or more activatable elements can be carried out by a person, such as a technician in the laboratory.

Alternatively, the detection of the status of the one or more activatable elements can be carried out using automated systems (see U.S. Patent App. Serial No. 12/606,869). In either case, the detection of the status of the one or more activatable elements for use according to the methods of this invention is performed according to standard techniques and protocols well-established in the art.

[00245] One or more activatable elements can be detected and/or quantified by any method that detect and/or quantitates the presence of the activatable element of interest. Such methods may include radioimmunoassay (RIA) or enzyme linked immunoabsorbance assay (ELISA),

immunohistochemistry, immunofluorescent histochemistry with or without confocal microscopy, reversed phase assays, homogeneous enzyme immunoassays, and related non-enzymatic techniques, Western blots, whole cell staining, immunoelectronmicroscopy, nucleic acid amplification, gene array, protein array, mass spectrometry, patch clamp, 2-dimensional gel electrophoresis, differential display gel electrophoresis, microsphere-based multiplex protein assays, label- free cellular assays and flow cytometry, etc. U.S. Pat. No. 4,568,649 describes ligand detection systems, which employ scintillation counting. These techniques are particularly useful for modified protein parameters. Cell readouts for proteins and other cell determinants can be obtained using fluorescent or otherwise tagged reporter molecules. Flow cytometry methods are useful for measuring intracellular parameters.

[00246] In some embodiments, the present invention provides methods for determining an activatable element's activation profile for a single cell. The methods may comprise analyzing cells by flow cytometry on the basis of the activation level of at least two activatable elements. Binding elements (e.g. activation state-specific antibodies) are used to analyze cells on the basis of activatable element activation level, and can be detected as described below. Alternatively, non- binding elements systems as described above can be used in any system described herein.

[00247] Detection of cell signaling states may be accomplished using binding elements and labels.

Cell signaling states may be detected by a variety of methods known in the art. They generally involve a binding element, such as an antibody, and a label, such as a fluorchrome to form a detection element. Detection elements do not need to have both of the above agents, but can be one unit that possesses both qualities. These and other methods are well described in Patents and applications that are also incorporated by reference in their entirety include U.S. Patent Nos. 7,381,535, 7,393,656, 7,695,924 and 7,695,926 and U.S. Patent Application Nos. 10/193,462; 1 1/655,785; 1 1/655,789; 1 1/655,821 ; 1 1/338,957, 12/877,998; 12/784,478;

12/730, 170; 12/703,741 ; 12/687,873; 12/617,438; 12/606,869; 12/713, 165; 12/293,081 ;

12/581 ,536; 12/776,349; 12/538,643; 12/501 ,274; 61/079,537; 12/501,295; 12/688,851 ;

12/471 , 158; 12/910,769; 12/460,029; 12/432,239; 12/432,720; 12/229,476; 12/877,998;

13/083, 156; U.S. Provisional Patent Application Nos. 61/469,812; 61/436,534; 61/423,918; 61/557,831 ; 61/542,910; 61/499, 127; 61/317, 187; and 61/353, 155; 61/515,660; PCT No.

PCT/US201 1/029845; and PCT No. PCT/US201 1/48332 which are all incorporated by reference in their entireties.

[00248] In one embodiment of the invention, it is advantageous to increase the signal to noise ratio by contacting the cells with the antibody and label for a time greater than 5, 6, 7, 8, 9, 10, 1 1 , 12, 13, 14, 15, 16, 17, 18, 19, 20, 24 or up to 48 or more hours.

[00249] When using fluorescent labeled components in the methods and compositions of the present invention, it will recognized that different types of fluorescent monitoring systems, e.g., cytometric measurement device systems, can be used to practice the invention. In some embodiments, flow cytometric systems are used or systems dedicated to high throughput screening, e.g. 96 well or greater microtiter plates. Methods of performing assays on fluorescent materials are well known in the art and are described in, e.g., Lakowicz, J. R., Principles of Fluorescence Spectroscopy, New York: Plenum Press (1983); Herman, B., Resonance energy transfer microscopy, in: Fluorescence Microscopy of Living Cells in Culture, Part B, Methods in Cell Biology, vol. 30, ed. Taylor, D. L. & Wang, Y.-L., San Diego: Academic Press (1989), pp. 219-243; Turro, N. J., Modern Molecular Photochemistry, Menlo Park: Benjamin/Cummings Publishing Col, Inc. (1978), pp. 296-361.

[00250] Fluorescence in a sample can be measured using a fluorimeter. In general, excitation

radiation, from an excitation source having a first wavelength, passes through excitation optics. The excitation optics cause the excitation radiation to excite the sample. In response, fluorescent proteins in the sample emit radiation that has a wavelength that is different from the excitation wavelength. Collection optics then collect the emission from the sample. The device can include a temperature controller to maintain the sample at a specific temperature while it is being scanned. According to one embodiment, a multi-axis translation stage moves a microtiter plate holding a plurality of samples in order to position different wells to be exposed. The multi-axis translation stage, temperature controller, auto-focusing feature, and electronics associated with imaging and data collection can be managed by an appropriately programmed digital computer. The computer also can transform the data collected during the assay into another format for presentation. In general, known robotic systems and components can be used.

[00251] Other methods of detecting fluorescence may also be used, e.g., Quantum dot methods (see, e.g., Goldman et al., J. Am. Chem. Soc. (2002) 124:6378-82; Pathak et al. J. Am. Chem Soc. (2001) 123 :4103-4; and Remade et al., Proc. Natl. Sci. USA (2000) 18:553-8, each expressly incorporated herein by reference) as well as confocal microscopy. In general, flow cytometry involves the passage of individual cells through the path of a laser beam. The scattering the beam and excitation of any fluorescent molecules attached to, or found within, the cell is detected by photomultiplier tubes to create a readable output, e.g. size, granularity, or fluorescent intensity.

[00252] In some embodiments, the activation level of an activatable element is measured using

Inductively Coupled Plasma Mass Spectrometer (ICP-MS). A binding element that has been labeled with a specific element binds to the activatable. When the cell is introduced into the ICP, it is atomized and ionized. The elemental composition of the cell, including the labeled binding element that is bound to the activatable element, is measured. The presence and intensity of the signals corresponding to the labels on the binding element indicates the level of the activatable element on that cell (Tanner et al. Spectrochimica Acta Part B: Atomic Spectroscopy, 2007 Mar;62(3): 188-195.). See U.S. Patent App. Serial No. 12/521 ,272.

[00253] The detecting, sorting, or isolating step of the methods of the present invention can entail fluorescence-activated cell sorting (FACS) techniques, where FACS is used to select cells from the population containing a particular surface marker, or the selection step can entail the use of magnetically responsive particles as retrievable supports for target cell capture and/or background removal. A variety of flow cytometers or FACS systems are known in the art and can be used in the methods of the invention (see e.g., W099/54494, filed Apr. 16, 1999; U.S. Pat. Application Pub. No. 20010006787, filed Jul. 5, 2001 , each expressly incorporated herein by reference). See also U.S. Patent Nos. 7,939,278, 7,563,584, and 7,393,656.

[00254] In some embodiments, a FACS cell sorter (e.g. a FACSVantage™ Cell Sorter, Becton

Dickinson Immunocytometry Systems, San Jose, Calif.) is used to sort and collect cells based on their activation profile (positive cells) in the presence or absence of an increase in activation level in an activatable element in response to a modulator. Other flow cytometers that are commercially available include the LSR II and the Canto II both available from Becton Dickinson. See Shapiro, Howard M., Practical Flow Cytometry, 4th Ed., John Wiley & Sons, Inc., 2003 for additional information on flow cytometers.

[00255] In some embodiments, the cells are first contacted with fluorescent- labeled activation state- specific binding elements (e.g. antibodies) directed against specific activation state of specific activatable elements. In such an embodiment, the amount of bound binding element on each cell can be measured by passing droplets containing the cells through the cell sorter. By imparting an electromagnetic charge to droplets containing the positive cells, the cells can be separated from other cells. The positively selected cells can then be harvested in sterile collection vessels. These cell-sorting procedures are described in detail, for example, in the FACSVantage™. Training Manual, with particular reference to sections 3-1 1 to 3-28 and 10-1 to 10-17, which is hereby incorporated by reference in its entirety. See the patents, applications and articles referred to, and incorporated above for detection systems.

[00256] In another embodiment, positive cells can be sorted using magnetic separation of cells based on the presence of an isoform of an activatable element. In such separation techniques, cells to be positively selected are first contacted with specific binding element (e.g., an antibody or reagent that binds an isoform of an activatable element). The cells are then contacted with retrievable particles (e.g., magnetically responsive particles) that are coupled with a reagent that binds the specific element. The cell-binding element-particle complex can then be physically separated from non-positive or non-labeled cells, for example, using a magnetic field. When using magnetically responsive particles, the positive or labeled cells can be retained in a container using a magnetic field while the negative cells are removed. These and similar separation procedures are described, for example, in the Baxter Immunotherapy Isolex training manual which is hereby incorporated in its entirety.

[00257] In some embodiments, methods for the determination of a receptor element activation state profile for a single cell are provided. The methods comprise providing a population of cells and analyze the population of cells by flow cytometry. Preferably, cells are analyzed on the basis of the activation level of at least two activatable elements. In some embodiments, a multiplicity of activatable element activation-state antibodies is used to simultaneously determine the activation level of a multiplicity of elements.

[00258] In some embodiments, cell analysis by flow cytometry on the basis of the activation level of at least two elements is combined with a determination of other flow cytometry readable outputs, such as the presence of surface markers, granularity and cell size to provide a correlation between the activation level of a multiplicity of elements and other cell qualities measurable by flow cytometry for single cells.

[00259]As will be appreciated, the present invention also provides for the ordering of element

clustering events in signal transduction. Particularly, the present invention allows the artisan to construct an element clustering and activation hierarchy based on the correlation of levels of clustering and activation of a multiplicity of elements within single cells. Ordering can be accomplished by comparing the activation level of a cell or cell population with a control at a single time point, or by comparing cells at multiple time points to observe subpopulations arising out of the others.

[00260] The present invention provides a valuable method of determining the presence of cellular subsets within cellular populations. Ideally, signal transduction pathways are evaluated in homogeneous cell populations to ensure that variances in signaling between cells do not qualitatively nor quantitatively mask signal transduction events and alterations therein. As the ultimate homogeneous system is the single cell, the present invention allows the individual evaluation of cells to allow true differences to be identified in a significant way.

[00261] Thus, the invention provides methods of distinguishing cellular subsets within a larger cellular population. As outlined herein, these cellular subsets often exhibit altered biological

characteristics (e.g. activation levels, altered response to modulators) as compared to other subsets within the population. For example, as outlined herein, the methods of the invention allow the identification of subsets of cells from a population such as primary cell populations, e.g. peripheral blood mononuclear cells that exhibit altered responses (e.g. response associated with presence of a condition) as compared to other subsets. In addition, this type of evaluation distinguishes between different activation states, altered responses to modulators, cell lineages, cell differentiation states, etc.

[00262]As will be appreciated, these methods provide for the identification of distinct signaling

cascades for both artificial and stimulatory conditions in complex cell populations, such a peripheral blood mononuclear cells, or naive and memory lymphocytes.

[00263] When necessary cells are dispersed into a single cell suspension, e.g. by enzymatic digestion with a suitable protease, e.g. collagenase, dispase, etc; and the like. An appropriate solution is used for dispersion or suspension. Such solution will generally be a balanced salt solution, e.g. normal saline, PBS, Hanks balanced salt solution, etc., conveniently supplemented with fetal calf serum or other naturally occurring factors, in conjunction with an acceptable buffer at low concentration, generally from 5-25 mM. Convenient buffers include HEPES 1 phosphate buffers, lactate buffers, etc. The cells may be fixed, e.g. with 3% paraformaldehyde, and are usually permeabilized, e.g. with ice cold methanol; HEPES-buffered PBS containing 0.1% saponin, 3% BSA; covering for 2 min in acetone at -200C; and the like as known in the art and according to the methods described herein.

[00264] In some embodiments, one or more cells are contained in a well of a 96 well plate or other commercially available multiwell plate. In an alternate embodiment, the reaction mixture or cells are in a cytometric measurement device. Other multiwell plates useful in the present invention include, but are not limited to 384 well plates and 1536 well plates. Still other vessels for containing the reaction mixture or cells and useful in the present invention will be apparent to the skilled artisan.

[00265] The addition of the components of the assay for detecting the activation level or activity of an activatable element, or modulation of such activation level or activity, may be sequential or in a predetermined order or grouping under conditions appropriate for the activity that is assayed for. Such conditions are described here and known in the art. Moreover, further guidance is provided below (see, e.g., in the Examples).

[00266] In some embodiments, the activation level of an activatable element is measured using

Inductively Coupled Plasma Mass Spectrometer (ICP-MS). A binding element that has been labeled with a specific element binds to the activatable. When the cell is introduced into the ICP, it is atomized and ionized. The elemental composition of the cell, including the labeled binding element that is bound to the activatable element, is measured. The presence and intensity of the signals corresponding to the labels on the binding element indicates the level of the activatable element on that cell (Tanner et al. Spectrochimica Acta Part B: Atomic Spectroscopy, 2007 Mar;62(3): 188-195.).

[00267] As will be appreciated by one of skill in the art, the instant methods and compositions find use in a variety of other assay formats in addition to flow cytometry analysis. For example, DNA microarrays are commercially available through a variety of sources (Affymetrix, Santa Clara CA) or they can be custom made in the lab using arrayers which are also know (Perkin Elmer). In addition, protein chips and methods for synthesis are known. These methods and materials may be adapted for the purpose of affixing activation state binding elements to a chip in a prefigured array. In some embodiments, such a chip comprises a multiplicity of element activation state binding elements, and is used to determine an element activation state profile for elements present on the surface of a cell.

[00268]In some embodiments, a chip comprises a multiplicity of the "second set binding elements," in this case generally unlabeled. Such a chip is contacted with sample, preferably cell extract, and a second multiplicity of binding elements comprising element activation state specific binding elements is used in the sandwich assay to simultaneously determine the presence of a multiplicity of activated elements in sample. Preferably, each of the multiplicity of activation state-specific binding elements is uniquely labeled to facilitate detection.

[00269] In some embodiments, confocal microscopy can be used to detect activation profiles for

individual cells. Confocal microscopy relies on the serial collection of light from spatially filtered individual specimen points, which is then electronically processed to render a magnified image of the specimen. The signal processing involved confocal microscopy has the additional capability of detecting labeled binding elements within single cells, accordingly in this embodiment the cells can be labeled with one or more binding elements. In some embodiments the binding elements used in connection with confocal microscopy are antibodies conjugated to fluorescent labels, however other binding elements, such as other proteins or nucleic acids are also possible.

[00270] In some embodiments, the methods and compositions of the instant invention can be used in conjunction with an "In-Cell Western Assay." In such an assay, cells are initially grown in standard tissue culture flasks using standard tissue culture techniques. Once grown to optimum confluency, the growth media is removed and cells are washed and trypsinized. The cells can then be counted and volumes sufficient to transfer the appropriate number of cells are aliquoted into microwell plates (e.g., Nunc™ 96 Microwell™ plates). The individual wells are then grown to optimum confluency in complete media whereupon the media is replaced with serum-free media. At this point controls are untouched, but experimental wells are incubated with a modulator, e.g. EGF. After incubation with the modulator cells are fixed and stained with labeled antibodies to the activation elements being investigated. Once the cells are labeled, the plates can be scanned using an imager such as the Odyssey Imager (LiCor, Lincoln Nebr.) using techniques described in the Odyssey Operator's Manual v 1.2, which is hereby incorporated in its entirety. Data obtained by scanning of the multiwell plate can be analyzed and activation profiles determined as described below.

[00271] In some embodiments, the detecting is by high pressure liquid chromatography (HPLC), for example, reverse phase HPLC, and in a further aspect, the detecting is by mass spectrometry.

[00272] These instruments can fit in a sterile laminar flow or fume hood, or are enclosed, self- contained systems, for cell culture growth and transformation in multi-well plates or tubes and for hazardous operations. The living cells may be grown under controlled growth conditions, with controls for temperature, humidity, and gas for time series of the live cell assays. Automated transformation of cells and automated colony pickers may facilitate rapid screening of desired cells.

[00273] Flow cytometry or capillary electrophoresis formats can be used for individual capture of magnetic and other beads, particles, cells, and organisms.

[00274] Flexible hardware and software allow instrument adaptability for multiple applications. The software program modules allow creation, modification, and running of methods. The system diagnostic modules allow instrument alignment, correct connections, and motor operations.

Customized tools, labware, and liquid, particle, cell and organism transfer patterns allow different applications to be performed. Databases allow method and parameter storage. Robotic and computer interfaces allow communication between instruments.

[00275] In some embodiments, the methods of the invention include the use of liquid handling

components. The liquid handling systems can include robotic systems comprising any number of components. In addition, any or all of the steps outlined herein may be automated; thus, for example, the systems may be completely or partially automated. See U.S. Patent App. Serial No. 12/432,239. [00276] As will be appreciated by those in the art, there are a wide variety of components which can be used, including, but not limited to, one or more robotic arms; plate handlers for the positioning of microplates; automated lid or cap handlers to remove and replace lids for wells on non-cross contamination plates; tip assemblies for sample distribution with disposable tips; washable tip assemblies for sample distribution; 96 well loading blocks; cooled reagent racks; microtiter plate pipette positions (optionally cooled); stacking towers for plates and tips; and computer systems.

[00277] Fully robotic or microfluidic systems include automated liquid-, particle-, cell- and organism- handling including high throughput pipetting to perform all steps of screening applications. This includes liquid, particle, cell, and organism manipulations such as aspiration, dispensing, mixing, diluting, washing, accurate volumetric transfers; retrieving, and discarding of pipet tips; and repetitive pipetting of identical volumes for multiple deliveries from a single sample aspiration. These manipulations are cross-contamination- free liquid, particle, cell, and organism transfers. This instrument performs automated replication of microplate samples to filters, membranes, and/or daughter plates, high-density transfers, full-plate serial dilutions, and high capacity operation.

[00278] In some embodiments, chemically derivatized particles, plates, cartridges, tubes, magnetic particles, or other solid phase matrix with specificity to the assay components are used. The binding surfaces of microplates, tubes or any solid phase matrices include non-polar surfaces, highly polar surfaces, modified dextran coating to promote covalent binding, antibody coating, affinity media to bind fusion proteins or peptides, surface- fixed proteins such as recombinant protein A or G, nucleotide resins or coatings, and other affinity matrix are useful in this invention.

[00279] In some embodiments, platforms for multi-well plates, multi-tubes, holders, cartridges,

minitubes, deep-well plates, microfuge tubes, cryovials, square well plates, filters, chips, optic fibers, beads, and other solid-phase matrices or platform with various volumes are accommodated on an upgradable modular platform for additional capacity. This modular platform includes a variable speed orbital shaker, and multi-position work decks for source samples, sample and reagent dilution, assay plates, sample and reagent reservoirs, pipette tips, and an active wash station. In some embodiments, the methods of the invention include the use of a plate reader.

[00280] In some embodiments, thermocycler and thermoregulating systems are used for stabilizing the temperature of heat exchangers such as controlled blocks or platforms to provide accurate temperature control of incubating samples from 0° C to 100° C.

[00281] In some embodiments, interchangeable pipet heads (single or multi-channel) with single or multiple magnetic probes, affinity probes, or pipetters robotically manipulate the liquid, particles, cells, and organisms. Multi-well or multi-tube magnetic separators or platforms manipulate liquid, particles, cells, and organisms in single or multiple sample formats.

[00282] In some embodiments, the instrumentation will include a detector, which can be a wide variety of different detectors, depending on the labels and assay. In some embodiments, useful detectors include a microscope(s) with multiple channels of fluorescence; plate readers to provide fluorescent, ultraviolet and visible spectrophotometric detection with single and dual wavelength endpoint and kinetics capability, fluorescence resonance energy transfer (FRET), luminescence, quenching, two-photon excitation, and intensity redistribution; CCD cameras to capture and transform data and images into quantifiable formats; and a computer workstation.

[00283] In some embodiments, the robotic apparatus includes a central processing unit which

communicates with a memory and a set of input/output devices (e.g., keyboard, mouse, monitor, printer, etc.) through a bus. Again, as outlined below, this may be in addition to or in place of the CPU for the multiplexing devices of the invention. The general interaction between a central processing unit, a memory, input/output devices, and a bus is known in the art. Thus, a variety of different procedures, depending on the experiments to be run, are stored in the CPU memory.

[00284] These robotic fluid handling systems can utilize any number of different reagents, including buffers, reagents, samples, washes, assay components such as label probes, etc.

[00285] Any of the steps above can be performed by a computer program product that comprises a computer executable logic that is recorded on a computer readable medium. For example, the computer program can execute some or all of the following functions: (i) exposing reference population of cells to one or more modulators, (ii) exposing reference population of cells to one or more binding elements, (iii) detecting the activation levels of one or more activatable elements, (iv) characterizing one or more cellular pathways and/or (v) classifying one or more cells into one or more classes based on the activation level.

[00286] The computer executable logic can work in any computer that may be any of a variety of types of general-purpose computers such as a personal computer, network server, workstation, or other computer platform now or later developed. In some embodiments, a computer program product is described comprising a computer usable medium having the computer executable logic (computer software program, including program code) stored therein. The computer executable logic can be executed by a processor, causing the processor to perform functions described herein. In other embodiments, some functions are implemented primarily in hardware using, for example, a hardware state machine. Implementation of the hardware state machine so as to perform the functions described herein will be apparent to those skilled in the relevant arts.

[00287] The program can provide a method of determining the status of an individual by accessing data that reflects the activation level of one or more activatable elements in the reference population of cells.

[00288] Analysis

[00289] Advances in flow cytometry have enabled the individual cell enumeration of up to thirteen simultaneous parameters (De Rosa et al., 2001) and are moving towards the study of genomic and proteomic data subsets (Krutzik and Nolan, 2003; Perez and Nolan, 2002). Likewise, advances in other techniques (e.g. microarrays) allow for the identification of multiple activatable elements. As the number of parameters, epitopes, and samples have increased, the complexity of experiments and the challenges of data analysis have grown rapidly. An additional layer of data complexity has been added by the development of stimulation panels which enable the study of activatable elements under a growing set of experimental conditions. See Krutzik et al, Nature Chemical Biology Feb. 2008. Methods for the analysis of multiple parameters are well known in the art. See U.S. Patent App. Serial. No. 12/501,295 for gating analysis. See U.S. Provisional App. Serial No. 61/515,660 for analysis metrics.

[00290] In some embodiments where flow cytometry is used, flow cytometry experiments are

performed and the results are expressed as fold changes using graphical tools and analyses, including, but not limited to a heat map or a histogram to facilitate evaluation. One common way of comparing changes in a set of flow cytometry samples is to overlay histograms of one parameter on the same plot. Flow cytometry experiments ideally include a reference sample against which experimental samples are compared. Reference samples can include normal and/or cells associated with a condition (e.g. tumor cells). See also U.S. Patent App. Serial. No.

12/501 ,295 for visualization tools.

[00291] The patients are stratified based on nodes that inform the clinical question using a variety of metrics. To stratify the patients between those patients with No Response (NR) versus a Complete Response (CR), a prioritization of the nodes can be made according to statistical significance (such as p-value from a t-test or Wilcoxon test or area under the receiver operator characteristic (ROC) curve ) or their biological relevance. See the methods described herein for methods for analyzing the cell signaling pathway data. For example, there are at least four methods to analyze data, such as from MDS patients. Other characteristics such as expression markers may also be used. For example the fold over isotype can be used (e.g., log2(MFIstain)- Log2(MFIisotype)) or % positive above Isotype.

[00292] The present application illustrates the use of four metrics used to analyze data from cells that may be subject to a disease. For example, the "basal" metric is calculated by measuring the autofluorescence of a cell that has not been stimulated with a modulator or stained with a labeled antibody. The "total phospho" metric is calculated by measuring the autofluorescence of a cell that has been stimulated with a modulator and stained with a labeled antibody. The "fold change" metric is the measurement of the total phospho metric divided by the basal metric. The quadrant frequency metric is the frequency of cells in each quadrant of the contour plot

[00293] A user may also analyze multimodal distributions to separate cell populations. In some

embodiments, metrics can be used for analyzing bimodal and spread distribution. In some embodiments, a Mann- Whitney U Metric is used.

[00294] In some embodiments, metrics that calculate the percent of positive above unstained and metrics that calculate MFI of positive over untreated stained can be used. [00295] A user can create other metrics for measuring the negative signal. For example, a user may analyze a "gated unstained" or ungated unstained autofluorescence population as the negative signal for calculations such as "basal" and "total". This is a population that has been stained with surface markers such as CD33 and CD45 to gate the desired population, but is unstained for the fluorescent parameters to be quantitatively evaluated for node determination. However, every antibody has some degree of nonspecific association or "stickyness" which is not taken into account by just comparing fluorescent antibody binding to the autofluorescence. To obtain a more accurate "negative signal", the user may stain cells with isotype-matched control antibodies. In addition to the normal fluorescent antibodies, in one embodiment, (phospho) or non phosphopeptides which the antibodies should recognize will take away the antibody's epitope specific signal by blocking its antigen binding site allowing this "bound" antibody to be used for evaluation of non-specific binding. In another embodiment, a user may block with unlabeled antibodies. This method uses the same antibody clones of interest, but uses a version that lacks the conjugated fluorophore. The goal is to use an excess of unlabeled antibody with the labeled version. In another embodiment, a user may block other high protein concentration solutions including, but not limited to fetal bovine serum, and normal serum of the species in which the antibodies were made, i.e. using normal mouse serum in a stain with mouse antibodies. (It is preferred to work with primary conjugated antibodies and not with stains requiring secondary antibodies because the secondary antibody will recognize the blocking serum). In another embodiment, a user may treat fixed cells with phosphatases to enzymatically remove phosphates, then stain.

[00296] In alternative embodiments, there are other ways of analyzing data, such as third color

analysis (3D plots), which can be similar to Cytobank 2D, plus third D in color.

[00297] One embodiment of the present invention is software to examine the correlations among

phosphorylation or expression levels of pairs of proteins in response to stimulus or modulation. The software examines all pairs of proteins for which phosphorylation and/or expression was measured in an experiment. The Total phoshometric (sometimes called "FoldAF") is used to represent the phosphorylation or expression data for each protein; this data is used either on linear scale or log2 scale.

[00298] For each protein pair under each experimental condition (unstimulated, stimulated, or treated with drug/modulator), the Pearson correlation coefficient and linear regression line fit are computed. The Pearson correlation coefficients for samples representing responding and non- responding patients are calculated separately for each group and compared to the unperturbed (unstimulated) data. The following additional metrics are derived:

(1) Delta CRNR unstim: the difference between Pearson correlation coefficients for each protein pair for the responding patients and for the non-responding patients in the basal or unstimulated state. (2) Delta CRNR stim: the difference between Pearson correlation coefficients for each protein pair for the responding patients and for the non-responding patients in the stimulated or treated state.

(3) DeltaDelta CRNR: the difference between Delta CRNRstim and Delta

CRNRunstim.

[00299] The correlation coefficients, line fit parameters (R, p-value, and slope), and the three derived parameters described above are computed for each protein-protein pair. Protein-protein pairs are identified for closer analysis by the following criteria:

(1) Large shifts in correlations within patient classes as denoted by large positive or negative values (top and bottom quartile or 10^th and 90^th percentile) of the DeltaDelta CRNR parameter.

(2) Large positive or negative (top and bottom quartile or 10^th and 90^th percentile) Pearson correlation for at least one patient group in either unstimulated or stimulated/treated condition.

(3) Significant line fit (p-value <= 0.05 for linear regression) for at least one

patient group in either unstimulated or stimulated/treated condition.

[00300]A11 pair data is plotted as a scatter plot with axes representing phosphorylation or expression level of a protein. Data for each sample (or patient) is plotted with color indicating whether the sample represents a responder (generally blue) or non-responder (generally red). Further line fits for responders, non-responders and all data are also represented on this graph, with significant line fits (p-value <= 0.05 in linear regression) represented by solid lines and other fits represented by dashed line, enabling rapid visual identification of significant fits. Each graph is annotated with the Pearson correlation coefficient and linear regression parameters for the individual classes and for the data as a whole. The resulting plots are saved in PNG format to a single directory for browsing using Picasa. Other visualization software can also be used.

[00301] Each protein pair can be further annotated by whether the proteins comprising the pair are connected in a "canonical" pathway. In the current implementation canonical pathways are defined as the pathways curated by the NCI and Nature Publishing Group. This distinction is important; however, it is likely not an exclusive way to delineate which protein pairs to examine. High correlation among proteins in a canonical pathway in a sample may indicate the pathway in that sample is "intact" or consistent with the known literature. One embodiment of the present invention identifies protein pairs that are not part of a canonical pathway with high correlation in a sample as these may indicate the non-normal or pathological signaling. This method will be used to identify stimulator/modulator-stain-stain combinations that distinguish classes of patients.

[00302] In some embodiments, nodes and/or nodes/metric combinations can be analyzed and

compared across sample for their ability to distinguish among different groups (e.g., CR vs. NR patients) using classification algorithms. Any suitable classification algorithm known in the art can be used. Examples of classification algorithms that can be used include, but are not limited to, multivariate classification algorithms such as decision tree techniques: bagging, boosting, random forest, additive techniques: regression, lasso, bblrs, stepwise regression, nearest neighbors or other methods such as support vector machines.

[00303] In some embodiments, nodes and/or nodes/metric combinations can be analyzed and

compared across sample for their ability to distinguish among different groups (e.g., CR vs. NR patients) using random forest algorithm. Random forest (or random forests) is an ensemble classifier that consists of many decision trees and outputs the class that is the mode of the class's output by individual trees. The algorithm for inducing a random forest was developed by Leo Breiman (Breiman, Leo (2001). "Random Forests". Machine Learning 45 (1): 5-32.

doi: 10.1023/A: 1010933404324) and Adele Cutler. The term came from random decision forests that was first proposed by Tin Kam Ho of Bell Labs in 1995. The method combines Breiman's "bagging" idea and the random selection of features, introduced independently by Ho (Ho, Tin (1995). "Random Decision Forest". 3rd Int'l Conf. on Document Analysis and Recognition, pp. 278-282; Ho, Tina (1998). "The Random Subspace Method for Constructing Decision Forests". IEEE Transactions on Pattern Analysis and Machine Intelligence 20 (8): 832-844.

doi: 10.1 109/34.709601) and Amit and Geman (Amit, Y.; Geman, D. (1997). "Shape quantization and recognition with randomized trees". Neural Computation 9 (7): 1545-1588.

doi: 10.1 162/neco. l 997.9.7.1545) in order to construct a collection of decision trees with controlled variation.

[00304] In some embodiments, nodes and/or nodes/metric combinations can be analyzed and

compared across sample for their ability to distinguish among different groups (e.g., CR vs. NR patients) using lasso algorithm. The method of least squares is a standard approach to the approximate solution of overdetermined systems, i.e. sets of equations in which there are more equations than unknowns. "Least squares" means that the overall solution minimizes the sum of the squares of the errors made in solving every single equation. The best fit in the least-squares sense minimizes the sum of squared residuals, a residual being the difference between an observed value and the fitted value provided by a model.

[00305] In some embodiments, nodes and/or nodes/metric combinations can be analyzed and

compared across sample for their ability to distinguish among different groups (e.g., CR vs. NR patients) using BBLRS model building methodology.

\ 2> tfADescription of the BBLRS model building methodology

[00307]Production of bootstrap samples:_A large number of bootstrap samples are first generated with stratification by outcome status to insure that all bootstrap samples have a representative proportion of outcomes of each type. This is particularly important when the number of observations is small and the proportion of outcomes of each type is unbalanced. Stratification under such a scenario is especially critical to the composition of the out of bag (OOB) samples, since only about one-third of observations from the original sample will be included in each OOB sample.

[00308]i?est subsets selection of main effects:_Best subsets selection is used to identify the

combination of predictors that yields the largest score statistic among models of a given size in each bootstrap sample. Models having from 1 to 2xN/10 are typically entertained at this stage, where N is the number of observations. This is much larger than the number of predictors generally recommended when building a generalized linear prediction model (Harrell, 2001) but subsequent model building rules are applied to reduce the likelihood of over- fitting. At the conclusion of this step, there will be a "best" main effects model of each size for each bootstrap sample, though the number of unique models of each size may be considerably fewer.

[00309]Determination of the optimal model size (for main effects) : Each of the unique "best" models of each size, identified in the previous step, are fit to each of a subset of the bootstrap samples, where the number of bootstrap samples in the subset is under the control of the user (i.e. a tuning parameter) so that the processing time required at this step can be controlled. For each of the bootstrap samples in the subset, the median SBC of the "best" models of the same size is calculated and the model size yielding the lowest median SBC in that bootstrap sample is identified. The optimal model size is then determined as the size for which the median SBC is smallest most often over the subset of bootstrap samples.

[00310]Identification of the top models of the best size:_At this stage, all previously identified "best" models of the optimal size are fit to every bootstrap sample. A number of top models are then selected as those with the highest values of the margin statistic (a measure from the logistic model of the difference in the predicted probabilities of CR, between NR patients with the highest predicted probabilities and CR patients with the lowest predicted probabilities). In order to limit the processing time required in subsequent steps, the number of top models selected is under the control of the user.

[00311]Identification of important two-way interactions: Vox each of the top main effects models identified in the previous step, models are constructed on every bootstrap sample, with main effects forced into the model and with stepwise selection used to identify important two-way interactions among the set of all possible pair-wise combinations of the main effects. The nominal significance level for entry and removal of interaction terms is under the control of the user. Significance levels greater than 0.05 are often used for entry because of the low power many studies have to detect interactions and because safeguards against over-fitting are applied subsequently.

[00312] At this stage, collections of full models (main effects and possibly some two-way interactions among them) have been constructed (on the set of all bootstrap samples) for each unique set of main effects identified in the previous step. The top full models in each collection are then chosen as those constructed most frequently over all bootstrap samples, where winners are decided among tied models by the lowest mean SBC and then the highest mean AUROC. The number of full models in each collection that are advanced to the next step is under the control of the user.

[00313]Selection of the effects in the final model: Eac full model advanced to this step is fit to every bootstrap sample and the median margin statistic for each model over the bootstrap samples is calculated. The model with the highest median margin statistic is selected as the final model. If there are ties, the model with the lowest mean SBC is selected.

[00314] Technically, the procedure described here results in the selection of the effects (main effects and possibly two-way interactions) to be included in the final model, but not specification of the model itself. The latter includes the effects and the specific regression coefficients associated with the intercept and each of the model effects.

[00315] Specification of the final model: T e effects in the final model are then fit to the complete dataset using Firth's method to apply shrinkage to the regression coefficient estimates. The model effects and their estimated regression coefficients (plus the estimate of the intercept) comprise the final model.

[00316] Another method of the present invention relates to display of information using scatter plots.

Scatter plots are known in the art and are used to visually convey data for visual analysis of correlations. See U.S. Patent No. 6,520, 108. The scatter plots illustrating protein pair correlations can be annotated to convey additional information, such as one, two, or more additional parameters of data visually on a scatter plot.

[00317] Previously, scatter plots used equal size plots to denote all events. However, using the

methods described herein two additional parameters can be visualized as follows. First, the diameter of the circles representing the phosphorylation or expression levels of the pair of proteins may be scaled according to another parameter. For example they may be scaled according to expression level of one or more other proteins such as transporters (if more than one protein, scaling is additive, concentric rings may be used to show individual contributions to diameter).

[00318] Second, additional shapes may be used to indicate subclasses of patients. For example they could be used to denote patients who responded to a second drug regimen or where CRp status. Another example is to show how samples or patients are stratified by another parameter (such as a different stim-stain-stain combination). Many other shapes, sizes, colors, outlines, or other distinguishing glyphs may be used to convey visual information in the scatter plot.

[00319] In this example the size of the dots is relative to the measured expression and the box around a dot indicates a NRCR patient that is a patient that became CR (Responsive) after more aggressive treatment but was initially NR (Non-Responsive). Patients without the box indicate a NR patient that stayed NR. [00320] Applying the methods of the present invention, the Total Phospho metric for p-Akt and p- Statl are correlated in response to hydrogen peroxide ("HOOH") treatment. On log2 scale the Pearson correlation coefficient for p-Akt and p-Statl in response to H2O2 for samples from patients who responded to first treatment is 0.89 and the p-value for linear regression line fit is 0.0075. In contrast there appeared to be no correlation observed for p-Akt and p-Statl in HOOH treated samples from patients annotated as "NR" (non-responder) or "NRCR" (initial non- responder, who responded to later more intensive treatment). Further there are no significant correlations observed for these proteins in any patient class for untreated samples.

[00321] The Total phospho metric for p-Erk and p-CREB also appeared to be correlated in response to IL-3, IL-6, and IL-27 treatment in samples from non-responding patients (NR and NR-CR). When considering all data in log2 scale the Pearson correlation coefficients for p-Erk and p- CREB in response to IL-3, IL-6, and IL-27 for samples from patients who did not respond to first treatment are 0.74, 0.76, 0.81, respectively, and the respective p-values for linear regression line fits are <0.0001, <0.0001, and <0.0001. In contrast there appeared to be no correlation observed for p-Erk and p-Creb in IL-3, IL-6, and IL-27 experiments for patients annotated as "CR".

[00322] In some embodiments, analyses are performed on healthy cells. In some embodiments, the health of the cells is determined by using cell markers that indicate cell health. In some embodiments, cells that are dead or undergoing apoptosis will be removed from the analysis. In some embodiments, cells are stained with apoptosis and/or cell death markers such as PARP or Aqua dyes. Cells undergoing apoptosis and/or cells that are dead can be gated out of the analysis. In other embodiments, apoptosis is monitored over time before and after treatment. For example, in some embodiments, the percentage of healthy cells can be measured at time zero and then at later time points and conditions such as: 24h with no modulator, and 24h with a modulator. In some embodiments, the measurements of activatable elements are adjusted by measurements of sample quality for the individual sample, such as the percent of healthy cells present. See U.S. Provisional App. Serial No. 61/436,534.

[00323] In some embodiments, a regression equation will be used to adjust raw node readout scores for the percentage of healthy cells at 24 hours post-thaw. In some embodiments, means and standard deviations will be used to standardize the adjusted node readout scores.

[00324] Before applying the SCNP classifier, raw node-metric signal readouts (measurements) for samples will be adjusted for the percentage of healthy cells and then standardized. The adjustment for the percentage of healthy cells and the subsequent standardization of adjusted measurements is applied separately for each of the node-metrics in the SCNP classifier.

[00325] The following formula can be used to calculate the adjusted, normalized node-metric

measurement (z) for each of the node-metrics of each sample:

z = ((x - (b₀ + b_l x pcthealthy)) - residual _ mean) I residual _ sd , [00326] where is the raw node-metric signal readout, b₀ and b are the coefficients from the regression equation used to adjust for the percentage of healthy cells ( pcthealthy ), and residual _ mean and residual _ sd are the mean and standard deviation, respectively, for the adjusted signal readouts in the training set data. The values of

b₀ , b_! , residual _ mean, and residual _ sd for each node-metric are included in the embedded object below, with values of the latter two parameters stored in variables by the same name. The values of the b₀ and b_x parameters are contained on separate records in the variable named

"estimate". The value for b₀ is contained on the record where the variable "parameter" is equal to "Intercept" and the value for b_x is contained on the record where the variable "parameter" is equal to "percenthealthy24Hrs". The value of pcthealthy will be obtained for each sample as part of the standard assay output. The SCNP classifier will be applied to the z values for the node- metrics to calculate the continuous SCNP classifier score and the binary induction response assignment (pNR or pCR) for each sample.

[00327] In some embodiments, the measurements of activatable elements are adjusted by

measurements of sample quality for the individual cell populations or individual cells, based on markers of cell health in the cell populations or individual cells. Examples of analysis of healthy cells can be found in See U.S. Provisional App. Serial No. 61/436,534., the content of which is incorporated herein by reference in its entirety for all purposes.

[00328] In some embodiments, the invention provides methods of diagnosing, prognosing,

determining progression, predicting a response to a treatment or choosing a treatment for acute leukemia, myelodysplasia syndrome or myeloproliferative neoplasms in an individual, the method comprising: (1) classifying one or more hematopoietic cells associated with acute leukemia, myelodysplasia syndrome or myeloproliferative neoplasms in said individual by a method comprising: a) subjecting a cell population comprising said one or more hematopoietic cells from said individual to modulator conditions , b) determining an activation level of activatable elements in one or more cells from said individual, and c) classifying said one or more hematopoietic cells based on said activation levels in response to modulator conditions using multivariate classification algorithms such as decision tree techniques: bagging, boosting, random forest, additive techniques: regression, lasso, bblrs, stepwise regression, nearest neighbors or other methods such as support vector machines (2) making a decision regarding a diagnosis, prognosis, progression, response to a treatment or a selection of treatment for acute leukemia, myelodysplasia syndrome or myeloproliferative neoplasms in said individual based on said classification of said one or more hematopoietic cells. In some embodiments, classifying further comprises identifying a difference in kinetics of said activation level. In some embodiments, the measurements of activatable elements are made only in healthy cells as determined using markers of cell health. In some embodiments, the measurements of activatable elements are adjusted by measurements of sample quality for the individual sample, such as the percent of healthy cells present. [00329] Other methods of analysis are important to discretize and visualize data. Spanning-tree progression analysis of density-normalized events (SPADE) is a recently developed

computational approach to facilitate analysis of heterogeneous cytometry data. The program is designed to arrange cell events into clusters based on antibody staining intensity, which has two advantages over traditional gating methods. First, since cells with distinct levels of antibody staining are clustered separately, SPADE can enable fine resolution of cell subsets with slightly differing antigen expression levels. Second, as every phenotypic marker in a given experiment is simultaneously used to generate a SPADE tree, seemingly homogeneous cell populations may be further discretized based on varying staining intensity of secondary markers that a user may not have appreciated with traditional gating strategies. The workflow of the SPADE algorithm is as follows: 1) Cytometry data is subjected to density dependent down sampling to increase the frequency of rare cell types. 2) Cells are clustered by marker expression levels into a user- defined number of clusters. 3) A minimum spanning-tree is constructed, linking each of the cell clusters. 4) Cell events that were discarded during the down-sampling step are then up-sampled into the appropriate cluster. Thus, a SPADE tree is built that gives information on which cell clusters are closely related by marker expression, as well as the median and variance of each marker in a cluster, and the percentage of cells in the sample that fall into each cluster. See Extracting a cellular hierarchy from high-dimensional cytometry data with SPADE. Qiu P, Simonds EF, Bendall SC, Gibbs KD Jr, Bruggner RV, Linderman MD, Sachs K, Nolan GP, Plevritis SK. Nat Biotechnol. 201 1 Oct 2;29(10):886-91 and Bendall SC, Simonds EF, Qiu P, Amir ED, Krutzik PO, Finck R, Bruggner RV, Melamed R, Trejo A, Ornatsky OI, Balderas RS, Plevritis SK, Sachs K, Pe'er D, Tanner SD, Nolan GP. Single-cell mass cytometry of differential immune and drug responses across a human hematopoietic continuum. Science. 201 1 May 6; 332(6030):687-96.

[00330] Drug Screening

[00331]Another embodiment of the present invention is a method for screening drugs that are in development and indicated for patients that have been diagnosed with, myelodysplasia (MDS).

[00332] Using the signaling nodes and methodology described herein, multiparametric flow cytometry could be used in-vitro to predict both on and off-target cell signaling effects. Using an embodiment of the present invention, the bone marrow or peripheral blood obtained from a patient diagnosed with MDS could be divided and part of the sample subjected to a therapeutic. Modulators could then be added to the untreated and treated specimens. Activatable elements, including the proposed target of the therapeutic, or those that may be affected by the therapeutic (off-target) can then be assessed for an activation state. This activation state can be used to predict the therapeutics' potential for on and off target effects prior to first in human studies. [00333] Using the signaling nodes and methodology described herein, one embodiment of the present invention, such as multiparametric flow cytometry, could be used after in-vivo exposure to a therapeutic in development for patients that have been diagnosed with MDS to determine both on and off-target effects. Using an embodiment of the present invention, the bone marrow or peripheral blood (fresh, frozen, ficoll purified, etc.) obtained from a patient diagnosed with MDS at time points before and after exposure to a given therapeutic may be subjected to a modulator as above. Activatable elements including the proposed target of the therapeutic, or those that may be affected by the therapeutic (off-target) can then be assessed for an activation state. This activation state can then be used to determine the on and off target signaling effects on the bone marrow or blast cells.

[00334] The apoptosis and peroxide panel study may reveal new biological classes of stratifying nodes for drug screening. Some of the important nodes could include changes in response to peroxide alone or in combination with growth factors or cytokines. These important nodes are induced Cleaved Caspase 3 and Cleaved Caspase 8, and etoposide induced p-Chk2, peroxide (H₂O₂) induced p-SLP-76, peroxide (H₂O₂) induced p-PLCy2 and peroxide (H₂O₂) induced P-Lck. The apoptosis panel may include but is not limited to, detection of changes in phosphorylation of Chk2, changes in amounts of cleaved caspase 3, cleaved caspase 8, cleaved poly (ACP ribose) polymerase PA P, cytochrome C released from the mitochondria these apoptotic nodes are measured in response to agents that included but are not limited to DNA damaging agents such as Etoposide, Mylotarg, AraC and daunorubicin either alone or in combination as well as to the global kinase inhibitor staurosporine.

[00335] Using the signaling nodes and methodology described herein, multiparametric flow cytometry could be used to find new target for treatment (e.g. new draggable targets). Using an

embodiment of the present invention, the bone marrow or peripheral blood obtained from a patient diagnosed with MDS could be divided and part of the sample subjected to one or more modulators (e.g. GM-CSF or PMA). Activatable elements (e.g. JAKs/STATs/AKT) can then be assessed for an activation state. This activation state can be used to predict find new target molecule for new existing therapeutics. These therapeutics can be used alone or in combination with other treatments for the treatment of MDS.

[00336] Kits

[00337] In some embodiments the invention provides kits. Kits provided by the invention may

comprise one or more of the state-specific binding elements described herein, such as phospho- specific antibodies. A kit may also include other reagents that are useful in the invention, such as modulators, fixatives, containers, plates, buffers, therapeutic agents, instructions, and the like.

[00338] In some embodiments, the kit comprises one or more of the phospho-specific antibodies specific for the proteins selected from the group consisting of PI3-Kinase (p85, pi 10a, pi 10b, pl l Od), Jakl , Jak2, SOCs, Rac, Rho, Cdc42, Ras-GAP, Vav, Tiam, Sos, Dbl, Nek, Gab, PRK, SHP1, and SHP2, SHIP1, SHIP2, sSHIP, PTEN, She, Grb2, PDK1, SGK, Aktl , Akt2, Akt3, TSC1,2, Rheb, mTor, 4EBP-1, p70S6Kinase, S6, LKB-1 , AMPK, PFK, Acetyl-CoAa

Carboxylase, DokS, Rafs, Mos, Tpl2, MEK1/2, MLK3, TAK, DLK, MKK3/6, MEKK1,4, MLK3, ASK1 , MKK4/7, SAPK/JNK1 ,2,3, p38s, Erkl/2, Syk, Btk, BLNK, LAT, ZAP70, Lck, Cbl, SLP-76, PLCyl , PLCy 2, STAT1 , STAT 3, STAT 4, STAT 5, STAT 6, FAK, pl 30CAS, PAKs, LIMK1/2, Hsp90, Hsp70, Hsp27, SMADs, Rel-A (p65-NFKB), CREB, Histone H2B, HATs, HDACs, PKR, Rb, Cyclin D, Cyclin E, Cyclin A, Cyclin B, PI 6, pl4Arf, p27KIP, p21 CIP, Cdk4, Cdk6, Cdk7, Cdkl , Cdk2, Cdk9, Cdc25,A/B/C, Abl, E2F, FADD, TRADD, TRAF2, RIP, Myd88, BAD, Bcl-2, Mcl-1 , Bcl-XL, Caspase 2, Caspase 3, Caspase 6, Caspase 7, Caspase 8, Caspase 9, IAPs, Smac, Fodrin, Actin, Src, Lyn, Fyn, Lck, NIK, ΙκΒ, p65(RelA), IKKa, PKA, PKCa, PKCP, PKC9, PKC8, CAMK, Elk, AFT, Myc, Egr-1 , NFAT, ATF-2, Mdm2, p53, DNA-PK, Chkl , Chk2, ATM, ATR, Pcatenin, CrkL, GSK3a, GSK3 , and FOXO. In some embodiments, the kit comprises one or more of the phospho-specific antibodies specific for the proteins selected from the group consisting of Erk, Syk, Zap70, Lck, Btk, BLNK, Cbl, PLCy2, Akt, RelA, p38, S6. In some embodiments, the kit comprises one or more of the phospho-specific antibodies specific for the proteins selected from the group consisting of Aktl , Akt2, Akt3, SAPK/JNK1,2,3, p38s, Erkl/2, Syk, ZAP70, Btk, BLNK, Lck, PLCy, PLCy 2, STAT1 , STAT 3, STAT 4, STAT 5, STAT 6, CREB, Lyn, p-S6, Cbl, NF-κΒ, GSK3 ,

CARMA/Bcll O and Tcl-1.

[00339] Kits provided by the invention may comprise one or more of the modulators described herein.

In some embodiments, the kit comprises one or more modulators selected from the group consisting of H₂0₂, PMA, BAFF, April, SDFl a, CD40L, IGF-1, Imiquimod, polyCpG, IL-7, IL- 6, IL-10, IL-27, IL-4, IL-2, IL-3, thapsigardin and a combination thereof.

[00340] The state-specific binding element of the invention can be conjugated to a solid support and to detectable groups directly or indirectly. The reagents may also include ancillary agents such as buffering agents and stabilizing agents, e.g., polysaccharides and the like. The kit may further include, where necessary, other members of the signal-producing system of which system the detectable group is a member (e.g., enzyme substrates), agents for reducing background interference in a test, control reagents, apparatus for conducting a test, and the like. The kit may be packaged in any suitable manner, typically with all elements in a single container along with a sheet of printed instructions for carrying out the test.

[00341] Such kits enable the detection of activatable elements by sensitive cellular assay methods, such as IHC and flow cytometry, which are suitable for the clinical detection, prognosis, and screening of cells and tissue from patients, such as leukemia patients, having a disease involving altered pathway signaling. [00342] Such kits may additionally comprise one or more therapeutic agents. The kit may further comprise a software package for data analysis of the physiological status, which may include reference profiles for comparison with the test profile.

[00343] Such kits may also include information, such as scientific literature references, package insert materials, clinical trial results, and/or summaries of these and the like, which indicate or establish the activities and/or advantages of the composition, and/or which describe dosing, administration, side effects, drug interactions, or other information useful to the health care provider. Such information may be based on the results of various studies, for example, studies using experimental animals involving in vivo models and studies based on human clinical trials. Kits described herein can be provided, marketed and/or promoted to health providers, including physicians, nurses, pharmacists, formulary officials, and the like. Kits may also, in some embodiments, be marketed directly to the consumer.

[00344] In some embodiments, the invention provides a kit comprising: (a) at least two modulators selected from the group consisting of Staurosporine, Etoposide, Mylotarg, Daunorubicin, AraC, G-CSF, IFNg, IFNa, IL-27, IL-3, IL-6, IL-10, FLT3L, SCF, G-CSF, SCF, G-CSF, SDFl a, LPS, PMA, Thapsigargin and H202; b) at least three binding elements specific to a particular activation state of the activatable element selected from the group consisting of p-Slp-76, p- Plcg2, p-Stat3, p-Stat5, p-Statl , p-Stat6, P-Creb, Parp+, Chk2, Rel-A (p65-NFKB), p-AKT, p-S6, p-ERK, Cleaved Caspase 8, Cytoplasmic Cytochrome C, and p38; and (c) instructions for diagnosis, prognosis, determining acute myeloid leukemia progression and/or predicting response to a treatment for acute myeloid leukemia in an individual. In some embodiments, the kit further comprises a binding element specific for a cytokine receptor or drug transporter are selected from the group consisting of MDR1 , ABCG2, MRP, P-Glycoprotein, CXCR4, FLT3, and c-kit. In some embodiments, the binding element is an antibody.

[00345] The following examples serve to more fully describe the manner of using the above-described invention, as well as to set forth the best modes contemplated for carrying out various aspects of the invention. It is understood that these examples in no way serve to limit the true scope of this invention, but rather are presented for illustrative purposes. All references cited herein are expressly incorporated by reference in their entireties.

[00346] EXAMPLES

[00347] Example 1

[003481 Materials and Methods

[00349] The present illustrative example represents how to analyze cells in one embodiment of the present invention. There are several steps in the process, such as the stimulation step, the staining step and the flow cytometry step. The stimulation step of the phospho-flow procedure can start with vials of frozen cells and end with cells fixed and permeabilized in methanol. Then the cells can be stained with an antibody directed to a particular protein of interest and then analyzed using a flow cytometer.

[00350] The materials used in this invention include thawing medium which comprises PBS-CMF + 10% FBS + 2mM EDTA; 70 urn Cell Strainer (BD); anti-CD45 antibody conjugated to Alexa 700 (Invitrogen) used at 1 ul per sample; propidium iodide (PI) solution (Sigma 10ml, 1 mg/ml) used at 1 ug/ml; RPMI + 1% FBS medium; media A comprising RPMI + 1% FBS + IX

Penn/Strep; Live/Dead Reagent, Amine Aqua (Invitrogen); 2 ml, 96-Deep Well, U-bottom polypropylene plates (Nunc); 300ul 96-Channel Extended-Length D.A.R.T. tips for Hydra (Matrix); Phosphate Buffered Saline (PBS) (MediaTech); 16%> Paraformaldehyde (Electron Microscopy Sciences); 100% Methanol (EMD) stored at -20C; Transtar 96 dispensing apparatus (Costar); Transtar 96 Disposable Cartridges (Costar, Polystyrene, Sterile); Transtar reservoir (Costar); and foil plate sealers.

[00351] Thawing cell and live/dead staining

[00352] Frozen cells are thawed in a 37°C water bath and gently resuspended in the vial and

transferred to the 15 mL conical tube. The 15 mL tube is centrifuged at 930 RPM (200xg) for 8 minutes at room temperature. The supernatant is aspirated and the pellet is gently resuspended in 1 mL media A. The cell suspension is filtered through a 70 um cell strainer into a new 15 mL tube. The cell strainer is rinsed with 1 mL media A and another 12ml of media A into the 15 mL tube. The cells are mixed into an even suspension. A 20 μΕ aliquot is immediately removed into a 96-well plate containing 180 μΕ PBS + 4%> FBS + CD45 Alexa 700 + PI to determine cell count and viability post spin. After the determination, the 15 mL tubes are centrifuged at 930 RPM (200xg) for 8 minutes at room temperature. The supernatant is aspirated and the cell pellet is gently resuspended in 4 mL PBS + 4 μΕ Amine Aqua and incubated for 15 min in a 37°C incubator. 10 mL RPMI + 1%> FBS is added to the cell suspension and the tube is inverted to mix the cells. The 15 mL tubes are centrifuged at 930 RPM (200xg) for 8 minutes at room

temperature. The cells are resuspended in Media A at the desired cell concentration (1.25x 10⁶/mL). For samples with low numbers of cells (<18.5 x 10⁶), the cells are resuspended in up to 15 mL media. For samples with high numbers of cells (>18.5 x 10⁶), the volume is raised to 10 mL with media A and the desired volume is transferred to a new 15 mL tube, and the cell concentration is adjusted to 1.25 x 10⁶ cells/ml. 1.6mL of the above cell suspension

(concentration at 1.25 x 10⁶ cells/ml) is transferred into wells of a multi-well plate. From this plate, 80ul is dispensed into each well of a subsequent plate. The plates are covered with a lid (Nunc) and placed in a 37°C incubator for 2 hours to rest.

[003531 Cell Stimulation [00354] A concentration for each stimulant that is five folds more (5X) than the final concentration is prepared using Media A as diluent. 5X stimuli are arrayed into wells of a standard 96 well v- bottom plate that correspond to the wells on the plate with cells to be stimulated.

[00355] Preparation of fixative: Stock vial contains 16% paraformaldehyde which is diluted with PBS to a concentration that is 1.5X. The stock vial is placed in a 37°C water bath.

[00356] Adding the stimulant: The cell plate(s) are taken out of the incubator and placed in a 37°C water bath next to the pipette apparatus. The cell plate is taken from the water bath and gently swirled to resuspend any settled cells. With pipettor, the stimulant is dispensed into the cell plate and vortexed at "7" for 5 seconds. The deep well plate is put back into the water bath.

[00357] Adding Fixative: 200μ1 of the fixative solution (final concentration at 1.6%) is dispensed into wells and then mixed on the titer plate shaker on high for 5 seconds. The plate is covered with foil sealer and incubated in a 37°C water bath for 10 minutes. The plate is spun for 6 minutes at 2000 rpm at room temperature. The cells are aspirated using a 96 well plate aspirator (VP Scientific). The plate is vortexed to resuspend cell pellets in the residual volume. The pellet is ensured to be dispersed before the Methanol step (see cell permeabilization) or clumping will occur.

[00358] Cell Permeabilization: Permeability agent, for example methanol, is added slowly and while the plate is vortexing. To do this, the cell plate is placed on titer plate shaker and made sure it is secure. The plate is set to shake using the highest setting. A pipetter is used to add 0.6 mis of 100%) methanol to the plate wells. The plate(s) are put on ice until this step has been completed for all plates. Plates are covered with a foil seal using the plate roller to achieve a tight fit. At this stage the plates may be stored at -80°C.

[003591 Staining Protocol

[00360] Reagents for staining include FACS/Stain Buffer-PBS + 0.1% Bovine serum albumen (BSA) + 0.05%) Sodium Azide; Diluted Bead Mix-lmL FACS buffer + 1 drop anti-mouse Ig Beads + 1 drop negative control beads. The general protocol for staining cells is as follows, although numerous variations on the protocol may be used for staining cells:

[00361] Cells are thawed if frozen. Cells are pelleted at 2000 rpm 5 minutes. Supernatant is aspirated with vacuum aspirator. Plate is vortexed on a "plate vortex" for 5-10 seconds. Cells are washed with lmL FACS buffer. Repeat the spin, aspirate and vortex steps as above. 50μΕ of FACS/stain buffer with the desired, previously optimized, antibody cocktail is added to two rows of cells at a time and agitate the plate. The plate is covered and incubated in a shaker for 30 minutes at room temperature (RT). During this incubation, the compensation plate is prepared. For the compensation plate, in a standard 96 well V-bottom plate, 20 μΐ. of "diluted bead mix" is added per well. Each well gets 5μΕ of 1 fluorophore-conjugated control IgG (examples: Alexa488, PE, Pac Blue, Aqua, Alexa647, Alexa700). For the Aqua well, add 200uL of Aqua-/+ cells. Incubate the plate for 10 minutes at RT. Wash by adding 200 μΕ FACS/stain buffer, centrifuge at 2000 rpm for 5 minutes, and remove supernatant. Repeat the washing step and resuspend the cells/beads in 200μL· FACS/stain buffer and transfer to a U-bottom 96 well plate. After 30min, lmL FACS/stain buffer is added and the plate is incubated on a plate shaker for 5 minutes at room temperature. Centrifuge, aspirate and vortex cells as described above. lmL FACS/stain buffer is added to the plate and the plate is covered and incubated on a plate shaker for 5 minutes at room temperature. Repeat the above two steps and resuspend the cells in 75 μΐ FACS/stain buffer. The cells are analyzed using a flow cytometer, such as a LSRII (Becton Dickinson). All wells are selected and Loader Settings are described below: Flow Rate: 2 uL/sec; Sample Volume: 40 uL; Mix volume: 40 uL; Mixing Speed: 250uL/sec; # Mixes: 5; Wash Volume: 800uL; STANDARD MODE. When a plate has completed, a Batch analysis is performed to ensure no clogging.

[003621 Gating protocol

[00363]Data acquired from the flow cytometer are analyzed with Flowjo software (Treestar, Inc). The Flow cytometry data is first gated on single cells (to exclude doublets) using Forward Scatter Characteristics Area and Height (FSC-A, FSC-H). Single cells are gated on live cells by excluding dead cells that stain positive with an amine reactive viability dye (Aqua-Invitrogen). Live, single cells are then gated for subpopulations using antibodies that recognize surface markers as follows: CD45++, CD33- for lymphocytes, CD45++, CD33++ for monocytes + granulocytes and CD45+, CD33+ for leukemic blasts. Signaling, determined by the antibodies that interact with intracellular signaling molecules, in these subpopulation gates that select for "lymphs", "monos+grans", and "blasts" is analyzed.

[003641 Gating of flow cytometry data to identify live cells and the lymphoid and myeloid

subpopulations

[00365] Flow cytometry data can be analyzed using several commercially available software programs including FACSDiva™, FlowJo, and Winlist™. The initial gate is set on a two-parameter plot of forward light scatter (FSC) versus side light scatter (SSC) to gate on "all cells" and eliminate debris and some dead cells from the analysis. A second gate is set on the "live cells" using a two- parameter plot of Amine Aqua (a dye that brightly stains dead cells, commercially available from Invitrogen) versus SSC to exclude dead cells from the analysis. Subsequent gates are set using antibodies that recognize cell surface markers and in so doing define cell sub-sets within the entire population. A third gate is set to separate lymphocytes from all myeloid cells (acute myeloid leukemia cells reside in the myeloid gate). This is done using a two-parameter plot of CD45 (a cell surface antigen found on all white blood cells) versus SSC. The lymphocytes are identified by their characteristic high CD45 expression and low SSC. The myeloid population typically has lower CD45 expression and a higher SSC signal allowing these different populations to be discriminated. The gated region containing the entire myeloid population is also referred to as the PI gate. See Hoefsloot LH, Lowenberg B, et al Blood. 1997

Marl ;89(5): 1690-700 which is hereby incorporated by reference in its entirety.

[00366] The data can then be analyzed using various metrics, such as basal level of a protein or the basal level of phosphorylation in the absence of a stimulant, total phosphorylated protein, or fold change (by comparing the change in phosphorylation in the absence of a stimulant to the level of phosphorylation seen after treatment with a stimulant), on each of the cell populations that are defined by the gates in one or more dimensions. These metrics are then organized in a database tagged by: the Donor ID, plate identification (ID), well ID, gated population, stain, and modulator. These metrics tabulated from the database are then combined with the clinical data to identify nodes that are correlated with a pre-specified clinical variable (for example; response or non-response to therapy) of interest.

[00367]The present application utilizes all of the disclosure of U.S. Patent App. Serial No. 12/910,769 and the example below should be read with that disclosure. Additionally, U.S. Provisional App. Serial No. 61/381 ,067 and PCT/US201 1/01565 are likewise relevant to the present disclosure, especially, example 1 and its corresponding figures.

[00368] Example 2

[00369] The objectives of this example were to simultaneously compare the functional effects of a panel of modulators on different signaling pathways (such as the PI3K and the Janus Kinases (Jak) signal transducers and activators of transcription (Stat) pathway) to identify specific proteomic profiles associated with the biological activity of and response to ON 01910.Na in MDS pts.

[00370] MDS bone marrow samples from two patients (103 and 104) were collected at baseline and after treatment cycles 1 , 3, 5, & 7. Bone marrow mononuclear cells (BMMCs) were isolated and cryopreserved for longitudinal analysis. Activation of signaling pathways was measured with fluorochrome-conjugated antibodies that recognize p-Erkl/1 (T202/Y204), p-Akt (S473), p-S6 (S235/236), p-Statl (Y701), p-Stat3 (Y705), and p-Stat5 (Y694). BMMCs were modulated with FMS-like tyrosine kinase 3 ligand (FLTL3), stem cell factor (SCF), granulocyte colony stimulating factor (G-CSF), or granulocyte-monocyte colony stimulating factor (GM-CSF) for 15 minutes. Cells were processed for SCNP by fixation, permeabilization, and incubation with fluorochrome-conjugated antibodies.

[00371] SCNP analyses in Pt ONI 03 (progressed to AML after completion of trial) showed that frequency of CD34+ cells increased during the course of the clinical trial; when modulated with either FLT3L or SCF, compared to baseline findings, CD34+ cells exhibited increased p-S6 and p-Akt responsiveness with treatment; and interestingly, CD34+ cell responsiveness to G-CSF decreased (p-Statl and p-Stat5) while no signaling was observed in response to GM-C SF. In contrast, SCNP analyses in Pt ONI 04 (stable disease), showed that frequency of CD34+ cells was maintained throughout treatment; when modulated with either FLT3L or SCF, CD34+ cells exhibited decreased p-S6, p-Akt, p-Erk (slight); and while CD34+ cell responsiveness to G-CSF decreased (p-Statl and p-Stat5), a robust p-Stat5 response was induced by GM-CSF which increased during the course of the clinical trial.

[00372] The data here suggest that measurement of intracellular signaling responses using SCNP is feasible using BMMCs from patients with MDS. Interrogating key signaling pathways thought to be involved in the action of ON 01910.Na provides a tool for developing functional proteomic signaling profiles associated with biological activity of this novel cell cycle active agent.

[00373] Example 3

[00374] The present example was designed: to describe cell viability in MDS samples by measuring levels of c-PA P at time zero, before addition of modulator; to identify proteomic profiles in cryopreserved bone marrow mononuclear cells (BMMCs) or peripheral blood mononuclear cells (PBMCs) from patients diagnosed with MDS by MDS risk category (low risk, high risk); to identify proteomic profiles in cryopreserved BMMCs from healthy donors using both the Low Risk MDS and High Risk MDS panels of signaling nodes; and to compare proteomic profiles between low risk and high risk MDS samples using a common panel of signaling nodes for low or high risk MDS samples with cells remaining after assaying the low or high risk panel of signaling nodes, respectively.

[00375]Forty (40) samples were divided into three categories; healthy BMMCs (15 samples); MDS high risk (1 1 samples); and MDS low risk (14 samples). (Many of these samples were also used in Example 1 of U.S. Patent App. Ser. No 12/713, 165). Nodes were selected for testing in each category. For example, the MDS low risk group was tested with the low risk panel, and common panel. The MDS high risk group was tested with the high risk and common panel. The healthy BMMCs were tested with the low risk, high risk, and common panel. The low risk panel contained the following modulators: EPO, G-CSF, EPO + G-CSF, Lenalidomide, and

Lenalidomide and EPO and the following activatable elements: p-STATl, p-STAT3, p-STAT5, p-AKT, p-S6 and p-Erk. The common panel contained modulators GM-CSF, IL-6, IL-27 and TPO; the activatable elements were p-STATl , p-STAT3, p-STAT5; and the surface markers were for lymphoid , stem, and CD45 isoforms. The high risk panel contained the modulators: Vidaza, Dacogen, and Zolinza (SAHA); the activatable elements were DNMT1 , DNMT3a, DNMT3b (where total protein is measured), phospho histone H3 (p-H3), phospho histone H2AX (p- H2AX), and Cyclin Bl . (See U.S. Ser. No 12/713,165.) Surface markers and phenotypic markers were used to identify specific subpopulations. For example, the surface markers were CD 45, CD34, CD235a, and CD71.

[00376] Side Scatter (SSC) and CD45 were used to identify lymphoid cells, myeloid cells, and

nucleated red blood cells (nRBC). Within myeloid cells, CD34 and CD45 were used to identify CD34+ stem cells. The nRBC subsets were divided into four groups according to their differentiation state and maturity. Within nRBCs, CD71 (Transferrin receptor) and CD235ab (Glycophorin A) were used to identify nRBC developmental stages: 1) ml CD71 - CD235ab- Immature nRBCs (nRBCl A), 2) m2 CD71+ CD235ab- Early Erythroblasts (nRBC IB), 3) m3 CD71+ CD235ab+ Normoblasts (nRBCl C), 4) m4 CD71 - CD235ab+ more Mature RBCs (nRBC ID). See Hoefsloot J.H., Lowenberg B., et al Blood 1997 Mar 1 ;89(5): 1690-700.

[00377] SCNP analysis was conducted in a manner similar to that shown in example 1 above and example 1 of U.S. Patent App. Serial No. 12/910,769, example 1 of U.S. Patent App. Ser. No. 12/713, 165 or in example 1 of U.S. Provisional App. Serial No. 61/381 ,067. Upon analysis, increased CD34+ cells identified many of the high risk MDS patients as AML. For example, the frequency of the CD34+ cells in the healthy group was below 10% and was from 0 to 70% in the high risk group. See Figure 1. Cleaved PARP levels were measured at time 0 (baseline measurement with no modulators) which showed that the high and low risk groups were generally higher than healthy group. Lower risk samples appeared to have a higher cleaved PARP level, which indicates that they were less healthy and higher risk cells had a higher cleaved PARP level than healthy samples. See Figure 2. Subsequent analysis can use cleaved PARP as a gating metric. (See U.S. Provisional App. Ser. No. 61/436,534).

[00378] High risk cells from MDS patients were modulated with: Hypomethylating agents (Dacogen and Vidaza), HDAC inhibitor (Zolinza, also known as SAHA or Vorinostat) and Lenalidomide. There are differences between Dacogen and Vidaza as shown in Figure 3. Cell populations are plotted from left to right with population 1 on the left and population 4 on the right. Population #1 is Lymph, #2 is nRBCl , #3 is Myeloid, and #4 is CD34+. The results show that Dacogen and Vidaza exhibit differential effects on DNMT protein levels and cell cycle progression. Dacogen decreases DNMT protein levels in Healthy BMMCs and increases DNMT protein levels in High Risk BMMCs as is apparent in lymph, myeloid, and CD34+ cells. See the general position of the data points below the Uu line in Figure 3. Also compare with Figures 1 and 2 from U.S. Patent App. Ser. No 12/713, 165. Vidaza decreases DNMT protein levels in Healthy BMMCs and increases DNMT1 and DNMT3b protein levels in the lymphoid cells from High Risk BMMCs, and decreases DNMT3a protein levels in lymphs, while decreasing DNMT3a in myeloid and CD34+ cells within High Risk BMMCs. We can also see a modulation of DNMT 3a and 3b levels in response to Zolinza in healthy and high risk patients with different effects seen in specific cell populations (decreasing levels in Myeloid cells).

[00379] Figure 4 shows that Vidaza disrupts cell cycle by arresting CD34+ cells in M phase as

evidenced by increased p-H3 levels. Cell populations are plotted from left to right with population 1 on the left and population 4 on the right. Population #1 is Lymph-MDS, #2 is nRBCl , #3 is PI , and #4 is STEM. Zolinza causes DNA damage through double stranded breaks as evidenced by p-H2AX and an increase in p-H3 to produce an M phase block to cell division (increased Cyclin Bl and p-H3). Vidaza arrests CD34+ cells in the M phase of cell cycle. Dacogen does not appear to alter cell cycle progression. There was a limited quantity of high risk samples which did not all for an assay to be conducted.

[00380] Figure 6 shows the effects of Vidaza and Dacogen on apoptosis readouts Cleaved PARP and AmineAqua in AML samples. Both Vidaza and Dacogen are capable of inducing apoptosis in AML samples, in a patient specific manner. This demonstrates that individual samples have distinct responses to either Vidaza or Dacogen and highlights the ability to identify samples sensitive to either Vidaza or Dacogen using apoptosis readouts.

[00381] Figure 5 shows that the effects of Erythropoietin and Lenalidomide in healthy and low risk MDS patients. Erythropoietin (EPO) responsiveness delineates healthy nRBC development by signaling and phenotype. Low Risk samples lack p-Stat3 induction and exhibit altered p-Stat5 response. Lenalidomide treated healthy samples exhibit no response. p-Stat3 and p-Stat5 are induced in Low risk nRBC subpopulations. Lenalidomide responsive and nonresponsive patients observed.

[00382] Short term modulation with Lenalidomide induces p-Statl, p-Stat3, and p-Stat5 signals in Low Risk but not Healthy BMMCs. Data for nRBC subsets are plotted in Figure 5 which shows differential signaling responses to lenalidomide. Signaling was induced by lenalidomide in lymphoid, nRBC, myeloid, and CD34+ cells. No synergy was observed in dual

erythropoietin/lenalidomide (short term) modulation. Low Risk samples exhibit a developmental block in nRBC development. For example, they lack mature nRBC sub-population (m4); and early erythrob lasts (m2) fail to induce p-Stat3 signaling. Cell populations are plotted from left to right with population 1 on the left and population 5 on the right. Population #1 is nRBClD, #2 is nRBCI C, #3 is nRBC IB, #4 is nRBC 1 A, and #5 is nRBCl . Short term modulation with EPO showed that healthy samples all responded to EPO while a subset of MDS patients responded.

[00383] These results show that determination of healthy and placement of cells into low and high risk groups can be accomplished by SCNP. The results also indicate that progression of MDS to AML may be determined using SCNP. Also, the individual patient differences can be seen for the various drugs above which suggests SCNP usefulness for measuring/predicting drug responses for these agents. The process can enable a user to see functional differences between some structurally similar drugs, like Dacogen and Vidaza which differ by a hydroxyl group.

[00384] Example 4

[00385] Single-cell Network Profiling (SCNP) to Evaluate the Proteomic Profiles Associated with ON

01910.Na Treatment of MDS Patients

[00386] Single Cell Network Profiling (SCNP) is used to simultaneously measure the effects of

multiple modulators (including drugs) on intracellular signaling cascades at the single cell level.

ON 01910.Na has been reported to inhibit polo-like kinase 1 , PI3-kinase and Akt pathways.

[00387]The objectives of the study were to simultaneously compare the functional effects of a panel of modulators on different signaling pathways (such as the PI3K and the Janus Kinases (Jak) signal transducers and activators of transcription (Stat) pathway) to identify specific proteomic profiles associated with the biological activity of and response to ON 01910.Na in MDS patients.

[00388] MDS patient bone marrow samples were collected at baseline and after treatment cycles 1 , 3, 5, & 7. Bone marrow mononuclear cells (BMMCs) were isolated and cryopreserved for longitudinal analysis. Activation of signaling pathways was measured with fluorochrome- conjugated antibodies that recognize p-Erkl/1 (T202/Y204), p-Akt (S473), p-S6 (S235/236), p- Statl (Y701), p-Stat3 (S727), and p-Stat5 (Y694). BMMCs were modulated with FMS-like tyrosine kinase 3 ligand (FLTL3), stem cell factor (SCF), granulocyte colony stimulating factor (G-CSF), or granulocyte-monocyte colony stimulating factor (GM-CSF) for 15 minutes. Cells were processed for SCNP by fixation, permeabilization, and incubation with fluorochrome- conjugated antibodies.

[00389] SCNP analyses in patient ONI 03 (progressed to AML after completion of trial) showed that frequency of CD34+ cells increased during the course of the clinical trial. Compare patient 103 as shown in Figure 7 with patient 104 as shown in Figure 8. When modulated with either FLT3L or SCF, compared to baseline findings, CD34+ cells exhibited increased p-S6 and p-Akt responsiveness with treatment, See Figures 9 and 10. Interestingly, CD34+ cell responsiveness to G-CSF decreased (p-Statl and p-Stat5) while no signaling was observed in response to GM-CSF. See Figures 1 1 and 12. In contrast, SCNP analyses in Pt ONI 04 (stable disease), showed that frequency of CD34+ cells was maintained throughout treatment; when modulated with either FLT3L or SCF, CD34+ cells exhibited decreased p-S6, p-Akt, p-Erk (slight). While CD34+ cell responsiveness to G-CSF decreased (p-Statl and p-Stat5), a robust p-Stat5 response was induced by GM-CSF which increased during the course of the clinical trial.

[00390] The data here suggest that measurement of intracellular signaling responses using SCNP is feasible using BMMCs from patients with MDS. Interrogating key signaling pathways thought to be involved in the action of ON 01910.Na provides a tool for developing functional proteomic signaling profiles associated with biological activity of this novel cell cycle active agent.

[00391] Example 5

[00392] Two sets of bone marrow mononuclear cell (BMMC) samples obtained from healthy donors were analyzed. One healthy bone marrow sample set consisted of nine cryopreserved BMMCs derived from patients undergoing planned orthopedic procedures at Williamson Medical Center in Tennessee. Another healthy sample set consisted of six cryopreserved BMMCs that were purchased from a commercial source (All-Cells).

[00393]A total of nine MDS cryopreserved BMMC collected from previously untreated patients with a diagnosis of low risk MDS were analyzed. MDS samples were acquired from patients treated at MD Anderson Cancer Center (MDACC) between May 1999 and September 2008. One additional MDS sample was purchased from a commercial source (Conversant). All patients consented, in accordance with the Declaration of Helsinki, for the collection and use of their samples for institutional review board (IRB)-approved research purposes. Clinical data were de-identified in compliance with Health Insurance Portability and Accountability Act (HIPAA) regulations. Sample inclusion criteria included diagnosis of low risk MDS, post thaw cell viability and signaling (samples must have >50% healthy cells in the zero-hour unmodulated condition, and must signal to modulators of use in the gated populations), collection prior to the initiation of treatment, and availability of clinical annotations.

[003941 SCNP Assays

[00395] Cocktails of fluorochrome-conjugated antibodies were used to measure phosphorylated intracellular signaling molecules and cell lineage markers in healthy and MDS cells. Measurements were performed at basal state and after extracellular modulation with growth factors or therapeutic agents.

[00396] SCNP assays were done similar to that described above in example 1. Cryopreserved samples were thawed at 37°C, washed and centrifuged in RPMI cell culture media and 60% fetal bovine serum (FBS). The cells were re-suspended, filtered to remove debris and washed in RPMI/1%> FBS, before staining with Live/Dead Fixable Amine Aqua Viability Dye to distinguish nonviable cells. The cells were then re-suspended in RPMI/10%> FBS, aliquoted to 150,000 cells per condition and rested for 2-2.5 hours at 37°C before undergoing cell-based functional assays. Cells were incubated with modulators, such as 3 IU/ml final concentration of EPO and 50 ng/ml final concentration of G-CSF, at 37°C for 15 minutes, fixed with 1.6% paraformaldehyde (final concentration) for 10 minutes at 37°C, pelleted and permeabilized with 100%) ice-cold methanol and stored at -80°C. Subsequently, cells were washed with fluorescence-activated cell sorting buffer (PBS/0.5%) bovine serum albumin/0.05%) NaN3), pelleted, and stained with cocktails of fluorochrome-conjugated antibodies These cocktails included antibodies against four phenotypic markers for gating cell populations (e.g., CD45, CD34, CD 71 and CD235a), up to three antibodies against intracellular signaling molecules (e.g., p-Statl, 3, 5, p-Akt, P-Erk 1/2, and p- S6) or cPARP for an eight-color flow cytometry assay. Isotype controls or phosphopeptide blocking experiments were performed to characterize each phospho-antibody. The modulators were EPO (nRBCs) and G-CSF (myeloid and CD34+); the intracellular readouts are p-STATl , 3, 5, for both EPO and G-CSF and p-Akt, p-S6, and p-Erk were added for G-CSF. Surface markers were CD45 and CD34 for myeloid cells and Cd45 CD235a and CD71 for erythroid cells.

[00397] /ow Cytometry Data Acquisition and Analysis

[00398] Flow cytometry data were acquired on a LSRII flow cytometer using FACSDiva software (BD Biosciences). All flow cytometry data were analyzed with FlowJo (TreeStar Software) or Winlist (Verity House Software). Dead cells and debris were excluded by forward scatter (FSC), side scatter (SSC), and Amine Aqua Viability Dye measurement. Myeloid cells were identified as cells that express the SSC versus CD45 characteristics of myeloid blasts and monocytoide cells. Stem cells were identified as cells that express CD45^MiCD34⁺. Nucleated red blood cells (nRBCs) were identified as cells that express the SSC versus CD45^Locharacteristics. Additional lineage markers such as CD71 and CD235a were used for further identification of nRBC sub-populations (e.g., CD71⁺CD235a^~: early erythrob lasts; CD71⁺CD235a⁺: normoblasts; CD71^"CD235a⁺: more mature RBCs). See Hoefsloot J.H., Lowenberg B., et al Blood 1997 Mar 1 ;89(5): 1690-700.

[00399] A minimum of 100 viable cells for nRBCs and a minimum of 200 viable cells for each of the other gated populations were acquired for analysis.

[00400] Statistical Analysis and Stratifying Node Selection

[00401]Metrics

[00402] Median fluorescence intensity (MFI) was computed for each node from the fluorescence intensity levels for the cells. Equivalent Number of Reference Fluorophores (ERF)(Shults, Cytometry B Clin Cytom 2006, Purvis Cytometry 1998, Wang L Cytometry Part A 2008) is a transformed value of the MFI values. ERF was computed using a calibration line determined by fitting observations of a standardized set of 8-peak rainbow beads for all fluorescent channels (Spherotech Libertyville, IL; Cat. No. RFP-30-5A) to standard values assigned by the manufacturer. The ERF values were then used to compute a variety of metrics to measure the biology of functional signaling proteins. To measure basal levels of signaling in the resting, unmodulated state, the "Basal" metric was designed. With modulation, the "Fold" metric identifies the inducibility or responsiveness of a protein or pathway. The "Total" metric was developed to assess the magnitude of total activated protein. To demonstrate how many cells show signaling/functional (comparing ranking of the cells in the modulated vs. unmodulated states), the "Uu" metric was applied.

[00403]The following metrics were included in the analysis: Equivalent Number of Reference Fluorophores (ERF): calibrated median fluorescence intensity. Basal ERF ("Basal"): defined as log₂(ERFu„_modui_ated / ERFAutofiuoresence). Fold Change ERF ("Fold"): defined as log₂(ERF_Modui_ated / ERF_Umnoduiated). Total Phospho ERF ("TotalPhospho"): defined as log2(ERF_Moduiated /

ERFAutofluorescence)

[00404] Uu is the Mann- Whitney U statistic comparing the ERF values of the modulated and unmodulated wells that has been scaled to the unit interval (0, 1) for a given donor and specimen type (e.g., BMMCs). For surface markers, the Percent Positive ("PercentPos") was used to quantify the frequency of cells positive for a surface marker relative to a control antibody.

[00405] Statistical Analysis

a. Statistical Analysis methods for the objectives in this study follow:

[00406] Cell viability after thaw and ficoll gradient

b. Method: The frequency of cells in a two-dimensional flow plot quadrant region [i.e., defined by low levels of Amine Aqua Viability Dye and low levels of caspase product cleaved poly-(ADP-ribose) polymerase (PARP); Amine Aqua^", cleaved PARP^" quadrant] was used to identify viable cells. c. Sample viability was determined by the percent of healthy cells in the time-zero unmodulated condition: Percentage of Healthy cells = [Amine Aqua^" Cleaved PARP^" Cells/ Scatter Gated Cells] x 100%

[00407] Effect of age on bone marrow samples from healthy donors

d. Method: Data graphs, R2 and P values were computed by Tableau (Tableau Software) to assess association between age and induced signaling (e.g., EPO modulated p-Stat5 activity in nucleated RBCs) or frequency of cell subsets (i.e., myeloid cells, stem cells or nRBC sub-populations) in each healthy donor.

[00408] Comparison of proteomic profiles between low risk MDS samples and age-matched bone marrow samples from healthy donors

[00409] Method: A heat map was made to visually inspect the range of induced signaling based on node/metrics and cell populations of interest (i.e., EPO or G-CSF modulated JAK/STAT pathway activity in myeloid cells, stem cells or nRBC sub-populations) across all donors. Data graphs were constructed by Tableau (Tableau Software) to display association between clinical diagnosis (e.g., RAEB, RARS) and signaling responsiveness or frequency of cell subsets in each donor.

[00411]Patient and Sample Characteristics

[00412] Modulated SCNP was evaluated on two sets of healthy and one set of low risk MDS samples.

[00413]Effect of Donor Age on Bone Marrow Signaling Function

[00414] The effects of donor age on the frequency and signaling function of CD34 expressing cells (CD34+), myeloid cells and nRBC cells was evaluated. The frequency of CD34+ cells was found to be independent of donor age (R²=0.006; p=0.783) suggesting any observed signaling differences were likely due to donor age rather than bone marrow source (Figure 14). Function of CD34+ cells was tested by examining GCSF induced phosphorylation of the STAT and PI3K pathways in both younger and older patient samples. Examination of signaling nodes using all 3 metrics (Uu, Fold change, and Total phospho-see materials and methods) revealed no significant correlations with age for p-STAT3, p-Akt, p-Erk or p-S6 between older and younger populations. However, a weak correlation with age for GCSF induced p-STAT5 was noted when used with the Uu metric (R²=0.352, p=0.025) (Figure 14). A weak-moderate correlation with older age for decreased frequency of myeloid and increased frequency of nRBC populations were observed (R²=0.326, p=0.026; and R²=0.494, p=0.003 respectively). However, there were no significant correlations between age and GCSF induced p-STAT3 or p-STAT5 for all 3 metrics indicating no difference in GCSF mediated myeloid signaling between older and younger bone marrow sources.

[00415] Conversely, a striking correlation between older age and decreased p-STAT5 in nRBC was identified. Further supporting this difference is that the correlation was noted in all 3 metrics examined (Table). This correlation held true when early erythroblasts and normoblasts, the subpopulations of nRBC in bone marrow that are known to be EPO responsive, were examined (Figure 15).

[00416] These data underscore the importance of using age-matched comparisons when evaluating differences between healthy and pathologic signaling in hematologic disorders (e.g. AML or MDS for example) and suggest that some age-related differences in hematopoiesis may be due to differences in signaling pathway activation. Therefore, age-matched control BMMb samples were used for evaluating pathologic signaling in low risk MDS samples.

[004171LOW Risk MDS BMMC Versus Age-matched Healthy BMMC

[00418]Block in Erythroid Differentiation in MDS

[00419] Consistent with prior findings of ineffective erythropoiesis in MDS BM, a block in erythroid differentiation was observed in LR MDS BMMb compared to healthy age match control (Figure 13). In healthy control BMMb samples early erythroid RBC (CD71+ CD235a-) comprised a minority while normoblasts (CD71+ CD235a+) and more mature RBC marrow (CD71 - CD235a+) comprised the majority percentage of the nRBC compartment. Conversely, fewer more mature RBC (CD71 - CD235a+) and increased early erythroid RBC (CD71+ CD235a-) were observed in a subset of marrow from LR MDS (Figure 13).

[00420]&4£ can be divided by nRBC and Myeloid Frequency and Function

[00421] Of the five BM from patients with a diagnosis of RAEB two (Pt 3 and Pt 8) had an increased percentage of erythroid elements versus healthy age-matched control. In response to EPO both RAEB samples showed an increased phosphorylation of STAT5 (Uu = 0.77 and 0.68 versus a median of 0.57 for healthy control) compared to healthy marrow (Figures 16 and 17). Conversely, those with low nRBC frequency for age displayed normal EPO mediated STAT5 phosphorylation response. These data held true when nRBC subsets (early erythroid and normoblasts) were examined.

[00422] RAEB BM with low nRBC frequency versus age-matched control had an increased percentage of the myeloid BM compartment (Figure 16). When myeloid and CD34 cells were examined for signaling function they displayed a hyper-response of STAT3 and STAT5 phosphorylation to GCSF versus control (Figures 16, 17, and 19).

[00423]R^RS with High Frequency of nRBC Demonstrate poor EPO signaling

[00424] Bone marrow samples from two patients with RARS were examined for erythroid and myeloid cell frequency and signaling function. Both BMMb samples displayed an increased percentage of nRBC and a low to normal percentage of myeloid and CD34+ elements. Unlike RAEB samples with similar nRBC frequency (Pt 3 and Pt 8) that displayed a robust hyper- response in STAT5 phosphorylation versus control, the RARS nRBC, showed low to normal EPO signaling versus age match control. Furthermore both RARS samples showed low to normal GCSF STAT3 and STAT5 phosphorylation (See Figures 16, 17, and 19).

[00425]MDS Samples with Increase Myeloid Frequency have increased STAT Activation [00426] Of the nine LR MDS samples four (Pts 4,6, 1 1 , and 13) showed increased and two (Pts 5 and 14) showed decreased percentage of myeloid BM elements versus age-matched healthy control. When examined for signaling response to GCSF phosphorylation of STAT3 and STAT5 were increased in the former and decreased in the latter versus age match control.

[00427]No synergy in vitro between EPO +GCSF

[00428] EPO and GCSF have been used together in the clinic to treat anemia associated with a subset of patients with LR MDS. We hypothesized that the addition of EPO may increase GCSF- mediated STAT3 and STAT5 phosphorylation in myeloid and CD34+ cells and that GCSF may synergize with EPO to increase STAT5 phosphorylation in nRBC. Neither STAT5 in nRBC or STAT3 or STAT5 in myeloid or CD34+ BMMb showed increased phosphorylation in response to EPO and GCSF versus EPO or GCSF alone respectively.

[00429] SCNP identified signaling differences related to physiologic (e.g. aging) and pathologic (e.g. MDS) conditions in bone marrow cell subpopulations. Specifically: aging was associated with a functional impairment in nRBC EPO response, while myeloid cells appeared to be unaffected in their signaling response to GCSF (STATs or PI3K pathway). Also, cell population numbers and signaling profiles distinguished LR MDS patients from healthy age-matched controls. For example, EPO RESPONSE (in nRBC): RAEB with High % RBC precursors (vs healthy) with increased EPO→p-STAT5 response and RARS with High % RBC precursors (vs healthy) with low /normal EPO→p-STAT5 response. Also, GCSF RESPONSE (in

Myeloid/CD34+): LR MDS with High % myeloid cells (vs healthy) with robust GCSF→p- STAT3 & p-STAT5 responses and LR MDS with Low % myeloid cells (vs healthy) with poor GCSF→p-STAT3 & p-STAT5 responses.

[00430] Signaling profiles classified RAEB patients into 2 categories based on differences in EPO- and GCSF-induced signaling. For example, compared to age-matched healthy "older" controls, one subset was characterized by a high percentage of RBC precursors (CD451o nRBC) and increased p-STAT5 levels in response to EPO and B: The other subset was characterized by a high percentage of myeloid cells with robust GCSF-induced p-STAT3 & p-STAT5 responses in both total myeloid and CD34+ cells

[00431] These data demonstrate the importance of using healthy age-matched controls to define disease-associated signaling (e.g. for MDS), and highlight the ability of SCNP to enhance disease classification based upon functional biology.

[00432] Example 6

[00433] We used SPADE to identify cell types and signaling potential in healthy and MDS cells.

[00434] Data was obtained from 14 MDS patients and 3 healthy donors from Example 5. See

Figure 20 which shows the difference between healthy and MDS cells. For example, two dimensional contour plots were prepared in which three healthy donor cells were assayed after being unmodulated, modulated with G-CSF, and modulated with EPO. CD45 was plotted versus pSTAT5. These plots show that the healthy responses were similar. In contrast, six MDS patient cells were plotted under similar conditions which showed a wide variation in responses.

[00435] We used SPADE to cluster cell events based on intensity of parameters a) light scatter, b) phenotypic markers (such as CD34, CD45, CD71 and CD235a) and c) functional markers, such as activatable elements including protein phosphorylation and DNA damage elements. A single SPADE tree was built from samples across multiple MDS patients and healthy donors. Each sample was activated by a modulator relevant for a given activatable element (for example, G-CSF stimulation of phospho-STAT5). This lead to the identification of cell clusters that exist in healthy donors but are absent in some MDS patients, or, alternatively, unexpected cell types that expressed phenotypic and/or functional markers that exist in some MDS patients but not in healthy donors. In addition, the percentage of cells assigned to each cell cluster is used to determine the frequency of cell types within a sample.

[00436] SPADE was used to produce Figure 21 which revealed cell types that only exist in some MDS patients. A single SPADE tree was built from G-CSF modulated bone marrow from the 14 MDS patients and 3 healthy donors. Cells were clustered based on the intensity of side scatter, CD34, CD45, CD71, CD235a, phospho-STAT5. Two branches of cell clusters were absent in the trees from healthy donors, meaning that cell types existed in MDS patient sample set that do not exist in healthy. See Figure 21 in which the question mark indicates two heavy lined ovals in which cell types are missing in healthy patients. These cell types were identified as CD34- phospho-STAT5+ (MDS 003) and CD34+ CD45- phospho-STAT5+ (MDS 004). The SPADE trees in Figure 21 cont. 1 shows that the cell types that are missing from the healthy diagram are present in the MDS patients. Percent total can also be viewed in SPADE.

[00437] We also used SPADE to identify cell signaling potential in bone marrow MDS patients.

SPADE was used to cluster cell events based on intensity of parameters a) light scatter and b) phenotypic markers (such as CD34, CD45, CD71 and CD235a). The SPADE tree was built from samples in the basal state and multiple activated from a single donor. Then, the change in intensity of each activatable element in each modulated condition was determined for each cell cluster. This method enabled the identification of responsiveness for multiple cell types. In preliminary studies, we used this technique to identify cell clusters that responded to G-CSF stimulation with STAT5 phosphorylation, then assessed if these same cell clusters were competent to respond to GM-CSF with activation of STAT5. Patients that progressed to AML contained cells that were G-CSF responsive but deficient in GM-CSF mediated signaling (ex: patient ON 122). However, in healthy donors and patients with stable disease (ex: patient ON 108), all cell clusters that showed increased phospho-STAT5 in the G-CSF condition also responded to GM-CSF. In a separate analysis, healthy donors contained cell clusters that activated STAT5 in response to EPO and in these same clusters AKT phosphorylation was induced by SCF. Cell clusters that responded to both EPO and AKT were also identified in one MDS patient that exhibited elevated nucleated red blood cell (nRBC) frequency in the bone marrow following treatment with On 01910.Na. In contrast, patients with depressed nRBC levels lacked cell clusters that responded to both EPO and SCF.

38] While preferred embodiments of the present invention have been shown and described herein, it will be obvious to those skilled in the art that such embodiments are provided by way of example only. Numerous variations, changes, and substitutions will now occur to those skilled in the art without departing from the invention. It should be understood that various alternatives to the embodiments of the invention described herein may be employed in practicing the invention. It is intended that the following claims define the scope of the invention and that methods and structures within the scope of these claims and their equivalents be covered thereby.

Claims

1. A method of determining a progression of myelodysplasia syndrome (MDS) to acute myeloid leukemia (AML) in an individual with a diagnosis or suspected diagnosis of MDS, comprising: contacting one or more bone marrow or peripheral blood cells from the individual with at least one modulator from the group comprising hematopoietic stem cell growth factors;

contacting one or more bone marrow or peripheral blood cells from the individual with at least one modulator from the group comprising lineage-specific hematopoietic factors comprising TPO, EPO, G-CSF and M-CSF;

determining an activation level of one or more activatable elements of the cells from the following pathways, comprising the group consisting essentially of cell cycle, PBKinase, MAPK, or JAK/STAT; and

determining the progression of MDS to AML.

A method in accordance with claim 1, wherein the lineage-specific hematopoietic factor is EPO or G-CSF.

A method in accordance with claim 1 , wherein the hematopoietic stem cell growth factors are selected from the group consisting essentially of FLT3-L, SCF, G-CSF, GM-CSF, and IL-6.

4. A method in accordance with claim 1, wherein the activatable elements from the PBKinase, MAPK, or JAK/STAT pathways are p-Erk, p-S6, p-AKT, p-STATl, p-STAT3, or p-STAT5.

5. A method in accordance with claim 1, wherein the activatable elements from the cell cycle

pathway are p-FB, cyclin A2, Ki-67, or cyclin Bl .

6. A method in accordance with claim 1, wherein the determining the progression of MDS to AML step comprises comparing the results of the determining of the activatable element steps to a profile that indicates progression to AML.

7. The method in accordance with claim 1, further comprising using age-matched controls.

8. The method in accordance with claim 1, wherein said activation level is based on the activation state selected from the group consisting of extracellular protease exposure, novel hetero- oligomer formation, glycosylation state, phosphorylation state, acetylation state, methylation state, biotinylation state, glutamylation state, glycylation state, hydroxylation state, isomerization state, prenylation state, myristoylation state, lipoylation state, phosphopantetheinylation state, sulfation state, ISGylation state, nitrosylation state, palmitoylation state, SUMOylation state, ubiquitination state, neddylation state, citrullination state, deamidation state, disulfide bond formation state, proteolytic cleavage state, translocation state, changes in protein turnover, multi- protein complex state, oxidation state, multi-lipid complex, and biochemical changes in cell membrane.

9. The method in accordance with claim 8, wherein said activation state is a phosphorylation state.

10. The method in accordance with claim 1, wherein said activatable element is selected from the group consisting of proteins, carbohydrates, lipids, nucleic acids and metabolites.

11. The method in accordance with claim 10, wherein said activatable element is a protein capable of being phosphorylated and/or dephosphorylated.

12. The method in accordance with claim 1, further comprising determining the presence or absence of one or more cell surface markers, intracellular markers, or combination thereof.

13. The method in accordance with claim 12, wherein said cell surface markers and said intracellular markers are independently selected from the group consisting of proteins, carbohydrates, lipids, nucleic acids and metabolites.

14. The method in accordance with claim 1, wherein said activation level is determined by a process comprising the binding of a binding element which is specific to a particular activation state of the particular activatable element.

15. The method in accordance with claim 14, wherein said binding element comprises an antibody.

16. The method in accordance with claim 1, wherein the step of determining the activation level comprises the use of an instrument for flow cytometry, immunofluorescence, confocal microscopy, immunohistochemistry, immunoelectronmicroscopy, nucleic acid amplification, gene array, protein array, mass spectrometry, patch clamp, 2-dimensional gel electrophoresis, differential display gel electrophoresis, microsphere-based multiplex protein assays, ELISA, and label- free cellular assays to determine the activation level of one or more intracellular activatable element in single cells.

17. A method of determining a progression of myelodysplasia syndrome (MDS) to acute myeloid leukemia (AML) in an individual with a diagnosis or suspected diagnosis of MDS, comprising: contacting one or more bone marrow or peripheral blood cells from the individual with at least one modulator from the group comprising FLT3-L, SCF, G-CSF, GM-CSF, and IL-6; contacting one or more bone marrow or peripheral blood cells from the individual with at least one modulator from the group comprising G-CSF, M-CSF, TPO, and EPO;

determining an activation level of one or more activatable elements from the group comprising p-

Erk, p-S6, p-AKT, p-STATl, p-STAT3, p-STAT5, p-H2AX, p-H3, cyclin A2, Ki-67, or cyclin

Bl ;

and

determining the progression of MDS to AML.

18. A method in accordance with claim 17, wherein the determining the progression of MDS to AML step comprises comparing the results of the determining of the activatable element steps to a profile that indicates progression to AML.

19. The method in accordance with claim 17, further comprising using age-matched controls.

20. The method in accordance with claim 17, wherein said activation level is based on the activation state selected from the group consisting of extracellular protease exposure, novel hetero- oligomer formation, glycosylation state, phosphorylation state, acetylation state, methylation state, biotinylation state, glutamylation state, glycylation state, hydroxylation state, isomerization state, prenylation state, myristoylation state, lipoylation state, phosphopantetheinylation state, sulfation state, ISGylation state, nitrosylation state, palmitoylation state, SUMOylation state, ubiquitination state, neddylation state, citrullination state, deamidation state, disulfide bond formation state, proteolytic cleavage state, translocation state, changes in protein turnover, multi- protein complex state, oxidation state, multi-lipid complex, and biochemical changes in cell membrane.

The method in accordance with claim 20, wherein said activation state is a phosphorylation state.

22. The method in accordance with claim 17, wherein said activatable element is selected from the group consisting of proteins, carbohydrates, lipids, nucleic acids and metabolites.

23. The method in accordance with claim 22, wherein said activatable element is a protein capable of being phosphorylated and/or dephosphorylated.

24. The method in accordance with claim 17, wherein said method further comprises determining the presence or absence of one or more cell surface markers, intracellular markers, or combination thereof.

25. The method in accordance with claim 24, wherein said cell surface markers and said intracellular markers are independently selected from the group consisting of proteins, carbohydrates, lipids, nucleic acids and metabolites.

26. The method in accordance with claim 17, wherein said activation level is determined by a process comprising the binding of a binding element which is specific to a particular activation state of the particular activatable element.

27. The method in accordance with claim 26, wherein said binding element comprises an antibody.

28. The method in accordance with claim 17, wherein the step of determining the activation level comprises the use of an instrument for flow cytometry, immunofluorescence, confocal microscopy, immunohistochemistry, immunoelectronmicroscopy, nucleic acid amplification, gene array, protein array, mass spectrometry, patch clamp, 2-dimensional gel electrophoresis, differential display gel electrophoresis, microsphere-based multiplex protein assays, ELISA, and label- free cellular assays to determine the activation level of one or more intracellular activatable element in single cells.

29. A method of choosing a treatment for an individual with a diagnosis or suspected diagnosis of MDS, comprising:

contacting one or more bone marrow or peripheral blood cells from the individual with a therapeutic agent;

determining a level of DNMT 1, DNMT 3 a, or DNMT 3b in one or more cells from the individual;

determining an activation level of one or more activatable elements in the cells which indicates cytostasis;

determining an activation level of one or more activatable elements in the cells which indicates cytotoxicity;

determining a level of CD34+ cells;

wherein the determining steps comprise a first result;

repeating the determining steps at a later time point to create a second result;

comparing the first and second results; and

making a decision regarding a therapy based on the results of the comparison.

30. The method in accordance with claim 29, wherein the therapeutic agent is selected from the group consisting essentially of Lenalidomide, Azacitidine, Decitabine or Vorinostat.

31. The method in accordance with claim 29, wherein the activatable element which indicates

cytostasis is cyclin Bl .

32. The method in accordance with claim 29, wherein the activatable element which indicates

cytotoxicity is cPARP, Amine Aqua or p-H2AX.

33. The method in accordance with claim 29, further comprising using age-matched controls.

34. The method in accordance with claim 29, wherein said activation level is based on the activation state selected from the group consisting of extracellular protease exposure, novel hetero- oligomer formation, glycosylation state, phosphorylation state, acetylation state, methylation state, biotinylation state, glutamylation state, glycylation state, hydroxylation state, isomerization state, prenylation state, myristoylation state, lipoylation state, phosphopantetheinylation state, sulfation state, ISGylation state, nitrosylation state, palmitoylation state, SUMOylation state, ubiquitination state, neddylation state, citrullination state, deamidation state, disulfide bond formation state, proteolytic cleavage state, translocation state, changes in protein turnover, multi- protein complex state, oxidation state, multi-lipid complex, and biochemical changes in cell membrane.

35. The method in accordance with claim 34, wherein said activation state is a phosphorylation state.

36. The method in accordance with claim 29, wherein said activatable element is selected from the group consisting of proteins, carbohydrates, lipids, nucleic acids and metabolites.

37. The method in accordance with claim 36, wherein said activatable element is a protein capable of being phosphorylated and/or dephosphorylated.

38. The method in accordance with claim 29, further comprising determining the presence or absence of one or more cell surface markers, intracellular markers, or combination thereof.

39. The method in accordance with claim 38, wherein the cell surface markers comprise CD 45, CD 235ab, or CD 71.

40. The method in accordance with claim 38, wherein said cell surface markers and said intracellular markers are independently selected from the group consisting of proteins, carbohydrates, lipids, nucleic acids and metabolites.

41. The method in accordance with claim 29, wherein said activation level is determined by a process comprising the binding of a binding element which is specific to a particular activation state of the particular activatable element.

42. The method in accordance with claim 41, wherein said binding element comprises an antibody.

43. The method in accordance with claim 29, wherein the steps of determining the activation level comprise the use of an instrument for flow cytometry, immunofluorescence, confocal microscopy, immunohistochemistry, immunoelectronmicroscopy, nucleic acid amplification, gene array, protein array, mass spectrometry, patch clamp, 2-dimensional gel electrophoresis, differential display gel electrophoresis, microsphere-based multiplex protein assays, ELISA, and label- free cellular assays to determine the activation level of one or more intracellular activatable element in single cells.

44. A method of choosing a treatment for an individual with a diagnosis or suspected diagnosis of MDS, comprising:

contacting one or more bone marrow or peripheral blood cells from the individual with at least one modulator from the group Lenalidomide, Azacitidine, Decitabine or Vorinostat;

determining a level of DNMT 1, DNMT 3a, or DNMT 3b in the cells;

determining an activation level of one or more activatable elements in the cells from the group comprising cyclin Bl, cyclin A2, Ki-67, cPARP, Amine Aqua or p-H2AX;

determining a level of CD34+ cells;

wherein the determining steps comprise a first result;

repeating the determining steps at a later time point to create a second result;

comparing the first and second results; and

making a decision regarding a therapy based on the results of the comparison.

45. The method in accordance with claim 44, wherein the therapeutic agent is selected from the group consisting essentially of Lenalidomide, Azacitidine, Decitabine, Vorinostat or histone deacetylase inhibitors.

46. The method in accordance with claim 44, further comprising determining if the MDS will

respond to clinical EPO treatment or if the MDS will require RBC transfusion.

47. The method in accordance with claim 44, wherein the modulator is Azacitidine.

48. The method in accordance with claim 44, wherein the modulator is Decitabine.

49. The method in accordance with claim 44, wherein the modulator is Vorinostat or Histone

Deacetylase Inhibitors.

50. The method in accordance with claim 44, wherein the modulator is lenalidomide.

51. The method in accordance with claim 44, further comprising using age-matched controls.

52. The method in accordance with claim 44, wherein said activation level is based on the activation state selected from the group consisting of extracellular protease exposure, novel hetero- oligomer formation, glycosylation state, phosphorylation state, acetylation state, methylation state, biotinylation state, glutamylation state, glycylation state, hydroxylation state, isomerization state, prenylation state, myristoylation state, lipoylation state, phosphopantetheinylation state, sulfation state, ISGylation state, nitrosylation state, palmitoylation state, SUMOylation state, ubiquitination state, neddylation state, citrullination state, deamidation state, disulfide bond formation state, proteolytic cleavage state, translocation state, changes in protein turnover, multi- protein complex state, oxidation state, multi-lipid complex, and biochemical changes in cell membrane.

53. The method in accordance with claim 52, wherein said activation state is a phosphorylation state.

54. The method in accordance with claim 44, wherein said activatable element is selected from the group consisting of proteins, carbohydrates, lipids, nucleic acids and metabolites.

55. The method in accordance with claim 54, wherein said activatable element is a protein capable of being phosphorylated and/or dephosphorylated.

56. The method in accordance with claim 44, wherein said method further comprises determining the presence or absence of one or more cell surface markers, intracellular markers, or combination thereof.

57. The method in accordance with claim 56, wherein the cell surface markers comprise CD 45, CD 235ab, or CD 71.

58. The method in accordance with claim 55, wherein said cell surface markers and said intracellular markers are independently selected from the group consisting of proteins, carbohydrates, lipids, nucleic acids and metabolites.

59. The method in accordance with claim 44, wherein said activation level is determined by a process comprising the binding of a binding element which is specific to a particular activation state of the particular activatable element.

60. The method in accordance with claim 59, wherein said binding element comprises an antibody.

61. The method in accordance with claim 44, wherein the steps of determining the activation level comprise the use of an instrument for flow cytometry, immunofluorescence, confocal microscopy, immunohistochemistry, immunoelectronmicroscopy, nucleic acid amplification, gene array, protein array, mass spectrometry, patch clamp, 2-dimensional gel electrophoresis, differential display gel electrophoresis, microsphere-based multiplex protein assays, ELISA, and label- free cellular assays to determine the activation level of one or more intracellular activatable element in single cells.

62. The method in accordance with claim 59, further comprising separating low risk MDS from high risk MDS.

63. The method in accordance with claim 59, further comprising determining if the MDS will respond to clinical EPO treatment or if the MDS will require RBC transfusion.

64. The method in accordance with claim 17, further comprising analyzing the results of the

determining steps in SPADE.

65. The method in accordance with claim 24, further comprising analyzing the results of the

determining steps in SPADE.

66. The method in accordance with claim 38, further comprising analyzing the results of the

determining steps in SPADE.

67. The method in accordance with claim 56, further comprising analyzing the results of the

determining steps in SPADE.

68. A method of screening a drug that is in development as a candidate therapeutic to treat an MDS patient, comprising:

contacting one or more bone marrow or peripheral blood cells from an individual with a diagnosis or suspected diagnosis of MDS with a candidate therapeutic either in vivo or in vitro;

determining a level of DNMT 1, DNMT 3a, or DNMT 3b in the cells;

determining an activation level of one or more activatable elements in the cells from the group comprising cyclin Bl, cyclin A2, Ki-67, cPARP, Amine Aqua or p-H2AX;

determining the level of CD34+ cells;

wherein the determining steps comprise a first result;

repeating the determining steps at a later time point to create a second result;

comparing the first and second results; and

determining if the candidate therapeutic is effective as a treatment for MDS.

69. The method in accordance with claim 68, further comprising using age-matched controls.

70. The method in accordance with claim 68, wherein said activation level is based on the activation state selected from the group consisting of extracellular protease exposure, novel hetero- oligomer formation, glycosylation state, phosphorylation state, acetylation state, methylation state, biotinylation state, glutamylation state, glycylation state, hydroxylation state, isomerization state, prenylation state, myristoylation state, lipoylation state, phosphopantetheinylation state, sulfation state, ISGylation state, nitrosylation state, palmitoylation state, SUMOylation state, ubiquitination state, neddylation state, citrullination state, deamidation state, disulfide bond formation state, proteolytic cleavage state, translocation state, changes in protein turnover, multi- protein complex state, oxidation state, multi-lipid complex, and biochemical changes in cell membrane.

71. The method in accordance with claim 70, wherein said activation state is a phosphorylation state.

72. The method in accordance with claim 68, wherein said activatable element is selected from the group consisting of proteins, carbohydrates, lipids, nucleic acids and metabolites.

73. The method in accordance with claim 72, wherein said activatable element is a protein capable of being phosphorylated and/or dephosphorylated.

74. The method in accordance with claim 68, further comprising determining the presence or absence of one or more cell surface markers, intracellular markers, or combination thereof.

75. The method in accordance with claim 74, wherein the cell surface markers comprise CD 45, CD 235ab, or CD 71.

76. The method in accordance with claim 74, wherein said cell surface markers and said intracellular markers are independently selected from the group consisting of proteins, carbohydrates, lipids, nucleic acids and metabolites.

77. The method in accordance with claim 68, wherein said activation level is determined by a process comprising the binding of a binding element which is specific to a particular activation state of the particular activatable element.

78. The method in accordance with claim 77, wherein said binding element comprises an antibody.

79. The method in accordance with claim 68, wherein the steps of determining the activation level comprise the use of an instrument for flow cytometry, immunofluorescence, confocal microscopy, immunohistochemistry, immunoelectronmicroscopy, nucleic acid amplification, gene array, protein array, mass spectrometry, patch clamp, 2-dimensional gel electrophoresis, differential display gel electrophoresis, microsphere-based multiplex protein assays, ELISA, and label- free cellular assays to determine the activation level of one or more intracellular activatable element in single cells.

80. The method in accordance with claim 68, further comprising analyzing the results of the

determining steps in SPADE.

81. The method of any of claims 29, 44, and 68, wherein the first result is determined at between about 15 minutes to about 36 hours.

82. The method of any of claims 29, 44, and 68, wherein the second result is determined at between about 10 hours to about 80 hours.