WO2005098046A2

WO2005098046A2 - Methods for the determination of cell specific biomarkers

Info

Publication number: WO2005098046A2
Application number: PCT/US2005/010940
Authority: WO
Inventors: Shawn Mark O'hara; Denis Smimov
Original assignee: Immunivest Corporation
Priority date: 2004-04-01
Filing date: 2005-03-31
Publication date: 2005-10-20
Also published as: WO2005098046A3

Abstract

The present invention provides methods for the reduction of the considerate amount of white cell background that interferes with meaningful analysis of a patient's food sample when the analysis involves rare cell analysis. Nucleic acid profile analysis of targeted rare cells is obtained from an individual patient's enriched blood sample by subtracting the white cell nucleic acid content from the same enriched sample, prior to positive selection of the target cell content. Subsequent profile analysis of the remaining nucleic acids allow for specific mRNA expression profiles having improved signal-to-noise. The methods are useful in profiling of cells isolated from tissues or body fluids and serves as an adjunct to clinical diagnosis of diverse carcinomas including early stage detection and classification of circulating tumor cells. Monitoring of nucleic acid and protein profiles of cells either in conventional or microarray formats, facilitates management of therapeutic intervention including staging, monitoring response to therapy, confirmation of remission and detection of regression.

Description

Title: Methods for the Determination of Cell Specific Biomarkers.

Inventors: Shawn Mark O'Hara and Denis Smirnov

Background of the Invention

Field of the Invention

This invention relates generally to gene specific amplification, analysis and profiling of cytosolic biomolecules useful in the fields of oncology, diagnostic testing and pharmacogenomics (personalized medicine). The invention is particularly useful in such fields as cancer screening, selecting (identification and stratification of therapy responders / non-responders) and monitoring for chemotherapy treatment, or cancer recurrence. More specifically, the present invention facilitates comprehensive analysis of mRNA and DNA from rare target cells. To accomplish this, the invention acts to subtract the white blood cell (WBC) noise from genetic markers associated with target rare cells.

Description of Related Art

Any given cell will express only a fraction of the total number of genes present in its genome. A portion of the total number of genes that are expressed determine aspects of cell function such as development and differentiation, homeostasis, cell cycle regulation, aging, apoptosis, etc. Alterations in gene expression decide the course of normal cell development and the appearance of disease states, such as cancer. The expression of specific genes will have a profound effect on the nature of any given cell and its response to specific therapies. Accordingly, the methods of analyzing gene expression, as such as those provided by the present invention are important in basic molecular biological research and improved disease management for individuals. Identification of specific genes, especially rare genes, can provide a key to diagnosis, prognosis and treatment for a variety of diseases that reflect these expression levels (Levsky, et al., Single-Cell Gene Expression Profiling, Science, 297:836-840, (2002)). Differential gene expression is a commonly used method of assessing gene expression in a cell. In particular, cDNA microarray analysis compares cDNA target sequence levels obtained from cells or organs from healthy and diseased individuals. These targets are then hybridized to a set of probe fragments immobilized on a membrane. Differences in the resultant hybridization pattern are then detected and related to differences in gene expression of the two sources (US 6,383,749). Competing events such as interactions between non-complementary target sequences nonspecific binding between target and probe and secondary structures in target sequences will interfere with hybridization and result in a decline of the signal- to-noise.

While gene specific primer sets have been used to selectively amplify a specific subset of mRNA from an mRNA library, there exists a clear need to reduce the signal-to-noise ratio in an amplification process which is especially applicable in rare cell detection for diagnostic therapy to encompass both quantitative and qualitative analysis.

Rare cells, such as circulating tumor cells (CTC), represent a surrogate source of tissue in the diagnosis, prognosis and treatment of disease (US 6,645,731 ; US 6,365,362; 10/079,939; 10/269,579). Further, advancements in the detection, phenotyping and genotyping will expand the clinical utility of such cells and may lead to therapies tailored to individual patients. It is generally accepted that the presence of circulating tumor cells (CTC) in a patient's blood provides an early detection system in assessing the need for therapeutic intervention. Highly sensitive assays to allow accurate enumeration of circulating carcinoma cells have shown that the peripheral blood tumor cell load correlate with disease state (Terstappen et al., Peripheral Blood Tumor Cell Load Reflects the Clinical Activity of the Disease in Patients with Carcinoma of the Breast, International J. of Oncology., 17:573-578, 2000).

Charting gene expression patterns of rare cell events (e.g. CTC) through microarray analysis of gene expression levels would be a desirable indicator of tumor properties in other diseases such as lymphomas, acute leukemia, breast cancer, prostate cancer, lung cancer and liver cancer etc. However, to discover and adapt this genetic information for patient management use requires resolution of inherent significant signal-to-noise issues in present state-of-the-art technology.

One of the most pressing goals of rare cell detection research is to identify a set of markers that allow efficient detection and prognosis of these cells in the peripheral blood of patients, having these cells. In addition to simply detecting the presence in peripheral blood, some markers can also provide useful information about the tissue of origin and potentially serve as a predictor of clinical outcome for a patient and a selection guide for the most efficient therapeutic agent. Continuing detection and characterization can help to track a treatment progress of the cancer patients. The expression of the marker gene as minimal or absent in the blood cells other than the target rare cells provides for a clear signal.

A reliable method of standardized WBC subtraction of nucleic acid noise from the target genetic markers provides an unmeet need in the analysis of gene expression. This is especially true where fast hybridization, highly specific binding of targets to complementary probes, and substantially improved signal-to-noise ratios are used in rare cell detection and analysis. Consequently, the present invention has additional importance when assessing gene expression as it relates to cancer and disease related states as well as in rare circulating endothelial cell (CEC) events associated with cardiovascular disease (see US App. 10/079,939 and US App. 09/904,472 both of which are fully incorporated by reference herein).

Summary of the Invention

The present invention provides methods for detecting genetic information of rare cells in a biological sample, which methods generally comprise: a. obtaining a biological sample containing a mixed population of cells from an individual suspected of having target rare cells; b. fractionating said biological sample to obtain a fraction suspected of containing said rare cells; c. assessing said fraction for a first gene profile; d. separating said rare cells from said fraction whereby a depleted fraction is devoid of said rare cells; e. determining a second gene profile of said depleted fraction; and f. subtracting said second gene profile from said first gene profile to obtain said genetic information from said rare cells.

In a preferred embodiment of the invention, the method involves selecting the rare cells from a group consisting of cancer cells, epithelial cells, endothelial cells, activated T-lymphocyte cells, dendritic cells and combinations thereof.

The present invention also provides methods for the reduction of the considerable amount of white cell background that interferes with meaningful analysis of a patient's blood sample when the analysis involves rare cell analysis. Nucleic acid profile analysis of targeted rare cells is obtained from an individual patient's enriched blood sample by subtracting the white cell nucleic acid content from the same enriched sample, prior to positive selection of the target cell content. Subsequent profile analysis of the remaining nucleic acids allow for specific mRNA expression profiles having improved signal-to-noise. The methods of the invention are useful in profiling of cells isolated from tissues or body fluids and serves as an adjunct to clinical diagnosis of diverse carcinomas including early stage detection and classification of circulating tumor cells. Monitoring of nucleic acid and protein profiles of cells either in conventional or microarray formats, facilitates management of therapeutic intervention including staging, monitoring response to therapy, confirmation of remission and detection of regression.

Brief Description of the Drawings

Figure 1 illustrates the mRNA expression levels of CK19, PSA, PSM, AR, Hepsin, HK2, PSGR, MGB1 and MGB2 in the mRNA libraries from 23 samples of CTC enriched from 9 metastatic cancer patients.

Figure 2 illustrates the mRNA expression levels of 37 genes listed and demonstrated the fundamental problem with current mRNA analysis of Ficoll/Percoll or immunomagnetically enriched CTC/CEC in that WBC or nonspecific binding of WBC confounds and limits the breadth and depth of genes that can be measured in a meaningful manor. As can be see only 12 of the 37 genes of interest were able to be measured with out any interference from the donor WBC population (AR, CEA, CK5, CK19, EGFR, ER-b, HK2, MGB1 , MGB2, PSA, PSGR, PSM, TROP2). As a result of this only a sub set there of 9 genes were then applied to advanced prostate cancer as shown in Figure 1. The other 24 genes could not be measures (NKX3A-CK10) due levels of these genes expressed in WBC nonspecifically bound to the epithelial specific immunomagnetic beads.

Detailed Description of the Invention

Because of the considerable amount of white cell background in mRNA profile analysis of rare cells, methods are presented to provide meaningful analysis of patient blood samples containing rare circulating target cells such as CTC and/or CEC. These methods provide an individual patient-matched comparison between the combined genetic information (CTC, CEC and blood sample WBC) and genetic information after CTC and/or CEC depletion (WBC blood sample only). For example in assessing the genetic profile of those patients diagnosed with a particular cancer and suspected of having circulating tumor or endothelial cells, eliminating the WBC background in a sample provides a meaningful reduction in the noise component enabling substantially improved analysis of genetic signals from the rare cells. Currently known methods are limited to amplification of individual genes that provide the least background noise, and do not consider other genes that have substantial background interference yet may be relevant to a particular disease material. Thus by subtracting an individual patient's WBC gene profile from the same individual patient's CTC and/or CEC gene profile, a meaningful detection system is described to allow assessment of relevant specific genetic information. This can be performed on a single gene basis with, multiplex gene analysis or global analysis of transcriptome/proteome/genome such as with massively parallel probe arrays.

Using the method of the present invention, expression profiling of genetic information is improved with the subtraction of background genetic information obtained from the same individual patient's WBC. This genetic profile is subtracted from the same blood sample, leaving only the genetic information from the target cells to further analyze. More specifically, an enriched fraction of whole blood is immunomagnetically enriched as previously described (US 6,365,362; US 6,645,731 ; US 10/079,939; US 10/269,579). The target cells are positively selected using antibodies specific to target cell antigens which are most often surface antigens. The remaining fraction, containing the depleted target cells, is assayed separately and compared to the same enriched patient blood sample fraction, prior to positive selection by array analysis or by RT-PCR etc.

Further the subtraction of WBC noise (e.g. nonspecifically enriched cells carried over due to process) provides a unique target cell specific panel of genes. These genes are consistently found in disease groups suggesting an important role in the diagnosis and management of diseases linked to the circulating rare cells. More specifically, diseases such as colorectal cancer, breast cancer prostate cancer and any combinations thereof can be screened for unique after early detection.

As used herein, the following terms are defined as follows:

"Cytoplasmic biomolecules" includes cellular target molecules of interest such as, but not limited to, protein, polypeptides, glycoprotein, oligosaccharide, lipids, electrolytes, RNA, DNA and the like, that is located in the cytoplasmic compartment of a cell. Upon contacting a cell with a permeabilization compound and subsequent cell separation, the cytoplasmic biomolecules are present in the supernatant for downstream analysis. All soluble cytoplasmic biomolecules, for example, the entire cytoplasmic RNA library or target components capable of traversing the membrane pores can be isolated and analyzed. In a preferred embodiment, the focus is on the analysis of transcribed mRNA and translated proteins, for example in CTC, as indicators of oncogenic transformations of interest in the management of cancer diagnosis and therapy.

"Membrane biomolecules" includes any extracellular, intra-membrane, or intracellular domain molecule of interest that is associated with or imbedded in the cell membranes including, but not limited to, the outer cell membrane, nuclear membrane, mitochondrial and other cellular organelle membranes. Upon permeabilization with a permeabilization compound of this invention, the targeted membrane biomolecules are normally not solubilized or removed from the membrane, i.e. the membrane biomolecules remain associated with the permeabilized cell. Membrane biomolecules include, but are not limited to, proteins, glycoproteins, lipids, carbohydrates, nucleic acids and combinations thereof, that are associated with the cellular membrane, including those exposed on the external or extracellular surface of the outer membrane as well as those that are present on the internal surface of the outer membrane, and those proteins associated with the nuclear, mitochondrial and all other intracellular organelle membranes. Membrane biomolecules also include cytoskeletal proteins.

Morphology in reference to cell structure is used as customarily defined, pertaining to cell and nuclear topology and surface characteristics including intracellular or surface markers or epitopes permitting staining with histochemical reagents or interaction with detectably labeled binding partners such as antibodies. In addition morphology shall include the entire field of "morphometry" defined as: quantitative measure of chromatin distribution within the nucleus.

The terms genomic and proteomic are used as conventionally defined. "Functional" is herein used as an adjective for an empirically detectable biological characteristic or property of a cell such as "functional cellomic" which more broadly encompasses both genomic and proteomic as well as other target categories including, but not limited to, "glyconomic" for carbohydrates and "lipidomic" for cellular lipids. The resultant cell characteristics provide profiles permitting differentiation of normal from transformed cells.

"Contacting" means bringing together, either directly or indirectly, a compound or reagent into physical proximity of a cell. The cell and/or compounds can be present in any number of buffers, salts, solutions, etc. Contacting includes, for example, placing the reagent solution into a tube, microtiter plate, microarray, cell culture flask, or the like, for containing the cell(s). The microtiter plate and microarray formats further permit multiplexed assays for simultaneously analyzing a multiplicity of cellular target compounds or components including, but not limited to, nucleic acids and proteins.

"Permeabilization compound, reagent, or composition" means any reagent that forms small pores in the cell membranes, comprising the lipid-cholesterol bilayer, while maintaining sufficient membrane, cytoplasmic and nuclear structure such that subsequent phenotypic analysis can be carried out on the permeabilized cell(s). For example, saponin is a known "pore-forming" compound that complexes with cell membrane components thereby forming numerous trans-membrane pores of about 8 nm size in the cell wall or membrane, thus allowing outward diffusion of small soluble cytosolic constituents, such as enzymes, proteins, glycoproteins, globulins, electrolytes, and the like, and internal equilibration with extracellular reagent components, such as electrolytes.

"Immunomagnetic beads" are magnetically labeled nanoparticles or microparticles also having covalently attached binding reagents (e.g. antibodies) with substantially selective affinity for surface markers or epitopes on cells, thereby achieving selective capture of magnetically labeled cells when exposed to a magnetic field such as generated in high gradient magnetic separation system (HGMS). Other terms used herein for methodologies, reagents and instruments are as conventionally defined and known to persons skilled in the art.

Description of Preferred Embodiments

As has been indicated in the foregoing discussion, a more comprehensive and practical form of cancer diagnosis must also include analysis of intra- and extra-cellular membrane antigens as well as analysis of cellular RNA content and DNA content in the same cell or cell population (US 6,365,362).

One of the many applications of this type of cell analysis is in cancer diagnostics. Many clinicians believe that cancer is an organ specific disease when confined to its early stages. The disease becomes systemic by the time it is first detected using methods currently available. Accordingly, evidence to suggest the presence of tumor cells in the circulation would provide a first line detection mechanism that could either replace, or function in conjunction with other tests such as mammography or measurements of prostate specific antigen. By analyzing cellular phenotype (protein and RNA) and genotype through specific markers for these cells, the organ origin of such cells may readily be determined, e.g., breast, prostate, colon, lung, ovarian or other non- hematopoietic cancers. Thus in situations where protein RNA and genome can be analyzed, especially where no clinical signs of a tumor are available, it is possible to identify the presence of a specific tumor as well as the organ of origin. As these profiles define cell function, they also indicate what the most appropriate therapy type and course should be when used in cancer cell detection. Further in monitoring cases where there is no detectable evidence of circulating tumor cells as with post operative surgery or other successful therapies, it may be possible to determine from a further clinical study whether further treatment is necessary.

Generally, the profiling of any targeted rare event after subtraction of an enriched sample is considered in this invention. Accordingly, hormones, proteins, peptides, lectins, oligonucleotides, drugs, chemical substances, nucleic acid molecules (such as RNA and/or DNA), bioparticles such as cells, apoptotic bodies, cell debris, nuclei, mitochondria, viruses, bacteria, and the like would be included in the embodiment of this invention. Enrichment of the target event can be accomplished by any means known in the art, but preferably immunomagnetic enrichment. After subtraction of the combined cytoplasmic biomolecule population in the enriched sample from the biomolecule population in the rare event, a profile analysis of the remaining signals is used as a descriptive index of the rare event.

The fluid sample includes, without limitation, cell-containing bodily fluids, peripheral blood, bone marrow, urine, saliva, sputum, semen, tissue homogenates, nipple aspirates, and any other source of rare cells that is obtainable from a human subject.

One method of providing for a more comprehensive diagnosis, embodied in the present invention, is the profiling of nucleic acids uniquely identified in circulating rare cells that are found in whole blood in a rapid, dependable, and standardized procedure. To this end, a whole blood sample is obtained to magnetically enrich the cytoplasmic biomolecules from a cell or population of cells from an individual patient to yield a fraction containing WBC and rare cells. The rare cells are positively selected, and the remaining enriched fraction is assayed on an array. This array is subtracted from the initially enriched sample to yield a genetic profile of the rare cell. Gene expression targets (mRNA) for identifying tissue of origin, diagnosis, prognosis, therapy target characterization and monitoring include but are not limited to cells derived from cancers of the breast, prostate, lung, colon, ovary, kidney, bladder, and the like for the purpose of detection and monitoring of sensitive or resistant genes expressing markers such as mammoglobin 1 (MGB1), mammoglobin 2 (MGB2), prolactin inducible protein (PIP), carcinoembryonic antigen (CEA), prostate specific antigen (PSA), prostate specific membrane antigen (PSMA), glandular kallikrein 2 (hK2), androgen receptor (AR), prostasin, Hespin (HPN), DD3, Her-2/Neu, BCL2, epidermal growth factor receptor (EGFR), tyrosine kinase-type receptor (HER2), thymidylate synthetase (TS), vascular endothelial growth factor VEGF, pancreatic mucin (Mud ), guanylyl cyclase c (GC-C), phosphatidylinositol 3 kinase (PIK3CG), protein kinase B gamma (AKT), excision repair protein (ERCC1), alpha-1 globin (F6), macrophage inhibitory cytokin-1 (G6), dihydropyrimidine dehydrogenase (DPYD), insulin growth factor receptor (IGF2) estrogen receptors alpha and beta (ER), progesterone receptor (PR), aromatase (cyp19), Telomerase (TERT), general epithelial tissue specific genes, cytokeratin 19 (CK19), cytokeratin 5 (CK5), cytokeratin 8 (CK8), cytokeratin 10 (CK10), cytokeratin 20 (CK20), epithelial cell adhesion molecule (EpCAM), mucins including mucin 1 (MUC1 ), topoisomerases, urokinase plasminogen activator (uPA), urokinase plasminogen activator receptor (uPAR), matrix metalloproteinases (MMP), general white blood cell specific mRNA, alpha-1-globin, CD16, CD45, and CD31 , and the like. This list is intended to illustrate the general diversity of arrays of mRNA-specific genes that could be assembled to differentiate cells from diverse origins, types and diseases, and is not intended to be comprehensive.

From a previously disclosed invention commonly assigned herewith, US Patent No. 6,365,362 and US App. Serial No.10/079,939 (both of which are incorporated by reference herein), circulating epithelial cells can be enriched relative to leukocytes to the extent of at least 2,500 fold to around 10,000 fold. Immunomagnetic selection of circulating epithelial cells in blood is followed by nucleotide analysis embodied in this invention. The enrichment is only one example of many methods known in the art for selecting specific populations of cells to be used in the embodiment of this invention.

Immunomagnetic enrichment of circulating tumor cells provides a 4 to 5 log unit reduction in leukocytes, the typical range of CTC to leukocytes is .1-10 CTC per 10⁴ leukocytes. The low number of CTC's and the leukocyte carried over during the CTC enrichment process pose significant detection restrictions in the signal-to-noise, constraining the choice of genes and gene expression profiling methods (Figure 2). Therefore, subtraction of the leukocytes from the sample would minimize the affect on the signal-to-noise.

Positive selection of cytoplasmic biomolecules such as target rare cells is accomplished through immunomagnetic selection with an antibody specific for the target cell. The nucleic acid content of the remaining sample is profiled and subtracted from the profile of the initial sample, prior to positive selection.

Profile analysis of cytoplasmic RNA (and other RNA such as mtRNA and hnRNA), DNA, and protein based analysis techniques. These include all types of cDNA, RNA and protein microarrays for profile analyses, mass spectrometry, fluorescent in situ hybridization (FISH), single nucleotide polymorphism (SNP), all genomic-based amplification techniques such as PCR and the like, microsatellite analysis, restriction fragment length polymorphism (RFLP, ALFP), SAGE, DD-RT-PCR, and the like.

Such analyses can be conducted on as few as 1-10 RNA molecules for each and any RNA sequence type, but preferably on tens of thousands up to millions copies of targets to enable detection of subtle alterations in cellular translation or transcription profiles as indicators of disease states in a clinical setting. Other functional cell profiles of releasable and non-releasable cellular components, such as proteins, glycoproteins, lipoproteins, oligoglycosides and the like, can similarly be generated by analyzing the two fractions by conventional microarray, HPLC, electrophoretic methods including the high- resolution 2D electrophoresis method, or antibody array profiling.

The following examples are provided to exemplify the practicality of the disclosed invention and to demonstrate the impact of the invention on diagnostic techn logy. These examples are not intended to limit the scope of the invention. In addition, the disclosures of each patent, patent application, and publication cited or described in this document are incorporated herein by reference in the ntirety.

EXAMPLE 1

mRNA EXPRESSION OF MULTIPLE GENES IN CTCs

The characterization of CTC is further improved over cell enumeration as it is feasible to profile nucleic acid content in these cells by ]n vitro transcription based RT-PCR expression obtained from patients with hormone refractory prostate cancer (HRPC). Expression of 37 genes with potential utility for epithelial cell characterization was evaluated from antisense RNA (aRNA) libraries constructed from immunomagnetically enriched CTC from 7.5 ml of blood samples from healthy donors and HRPC.

The results showed no expression in 13 of 37 genes in the control group. Of the genes expressed in the CTC from the 23 blood specimens drawn from 9 metastatic prostate cancer patients were CK 19 18/23 (78%), PSA 20/23 (89%), PSM 17/23 (74%), AR 16/23 (70%), hK2 7/23 (30%), EGFR 4/23 (17%), and PSGR 2/23 (9%). The number of CTC in these samples ranged from 4 to 283 per 7.5 mL blood (mean 87, median 89). Some of the genes had a low level of expression in the control samples and were expressed at higher levels in the patient samples. In all 23 samples CK19, EpCAM or Muc- 1 was expressed . Due to background expression in the controls, expression of 13 of the 37 genes including HER-2, p53 and BCL-2 could not be measured in CTCs (Figure 1 )

From these results, aRNA libraries can be constructed from CTCs and gene expression profiles of CTCs were obtained in HRPC. This can enhance characterization of HRPC and facilitate the development of more effective therapies in HRPC.

EXAMPLE 2

Assessment of IVIicroarray Chip Selectivity Several clinical sample types and a model cell line system were assessed. Affymetrix Focus 8,700 gene microarray chips were evaluated using two test systems. One system is composed of actual patient samples where CTC and WBC were predetermined by Flow. The other test system is a reconstituted cell line model system (LN-CAP/ZR75 mixture) having known copy numbers of nine different CTC mRNA species.

With the cell line model system, gene expression was detected down to range of about 140-800 copies of specific mRNA present, following immunomagnetic enrichment. This sensitivity result approximately equals the Affymetrix claimed sensitivity of 1/10⁵. Thus if 10 or more CTC are present in a sample, the sensitivity translates into an ability to detect a substantial number of gene sequences when present at about 50-100 copies per cell (i.e. 50-100 copies/cell x 10 CTC = 500-1000 copy signal) and suggests the successful a pplication to rare cell events in blood such as circulating tumor or endothelial cell mRNA profiling.

The second test system utilized clinical containing samples from patients with known cancers. Hybridization with samples from patients with advanced prostate cancer (650 CTCs) and colon cancer (105 CTCs) revealed a set of genes that are upregulated in CTC samples, after subtraction of the depleted background .

EXAMPLE 3

Microarray Expression Analysis in Genes with No Detectable Expression Following Depletion of CTCs

Analysis of genes detected prior to depletion of the WBC fraction and not after subtraction of the WBC genetic information resulted in sets of genes identified only through their expression in CTCs exclusively from breast, prostate, or colorectal cancers. Gene sets were also identified in exclusive combinations of cancer patients (i.e. breast and colorectal, breast and prostate, prostate and colorectal), and in general expressed in all three cancers. Affymetrix Focus 8,700 gene microarray chips were used after individual patient WBC subtraction by immunomagnetic selection. Table 1 shows 322 genes identified from individ ual patients diagnosed with cancer. Each Affymetrix chip contains over 8000 full-length human transcripts that are commercially available for screening. Patients diagnosed with breast cancer showed 86 positives unique for breast cancer. Patients with diagnosed prostate cancer had 60 positives unique for prostate cancer, and patients with colorectal cancer had 74 positives unique for colorectal cancer. Further, 32 genes were positive for both breast and prostate cancers, 17 genes were positive for breast and colorectal cancer, 10 genes were positive for prostate and colorectal cancer, and 43 genes were positive for breast prostate and colorectal.

Table 1: Genetic profile of genes not detected in the depleted WBC portion.

Table 2 condenses the number of these selected genes to a number that is easily used in rapid screening. Gene numbers for breast cancer (10), prostate (7), colorectal (7), and combinations thereof (7) showed the most prominent signal-to-noise separation and, thus, were appropriate in number and type for profile analysis. These combinations provide a collection of genes that could have diagnostic/prognostic significance in the treatment of cancer. Table 2: Reduction in the number of genes to limit each panel to a workable number for rapid screening.

EXAMPLE 4

Microarray Expression Analysis in Genes Detectable after CTC Depletion

Analysis of genes detected prior to depletion of the WBC fraction and after subtraction of the WBC resulted in sets of genes substantially attenuated in the depleted portion. Gene sets were the CTC levels are at least 3 fold greater than the CTC depleted WBC detectable signal are shown in Table 3. As with Example 3, the same patient groups (breast, prostate, colorectal) were compared.

Table 3: Genetic profile where at least a 3 fold reduction in the individual gene signal was detected in the WBC-depleted portion.

EXAMPLE 5

Immunophenotyping of Leukocyte Carryover

In order to characterize leukocyte subsets carried over during immunomagnetic enrichment, EpCAM immunomagnetic selection was followed by further immunomagnetic selection using subset specific antigens to obtain proportional comparisons of leukocyte subsets. Amplification of the selected transcripts from EpCAM selected cells showed substantial signal interference from leukocyte contamination with epithelial cells.

Immunomagnetic selection of leukocytes subsets was used to obtain RNA from subset populations using CellSearch® Cancer Assay (Immunicon Corporation, Huntingdon Valley, Pennsylvania). Magnetic beads coated with antibodies directed against epithelial cells were used to isolate circulating epithelial cells from blood samples in the presence of an appropriate magnetic field. RNA was liberated from theses cells for amplification by RT-PCT (Table 4). This method is useful for obtaining satisfactory signal-to-noise in assessing genes found only in epithelial cells and not in leukocytes.

Table 4: Relative WBC background expression level in selected gene transcripts.

As shown in Table 4, a fraction of leukocytes are selected in addition to magnetically isolating epithelial cells. For every 7.5 ml of blood, 2,000 to 5,000 leukocytes are also selected with EpCAM immunomagnetic selection (about 0.005% to 0.01% of the leukocyte population). This small percentage contributes to background RNA interference after amplification of the total recovered pool. Information as to whether these specific genes are limited to leukocyte subsets or are universally retained throughout the leukocyte population would provide further insight into any analysis of their expression.

After EpCAM immunomagnetic selection, leukocyte subsets were selected by populations specific for CD3, CD4, CD8, CD14, CD15, CD20, and CD56. Resultant cell counts were determined, and the purity of selected population assessed using FACSCalibur flow cytometer. The collected cells were resuspended in 2.5 ml PBS for RNA analysis. The results show that all major leukocyte subsets are present after EpCAM immunomagnetic selection. The proportions of leukocyte subsets- present in the carry-over, shifts from the expected proportion in average^ human blood (i.e. lymphocytes/monocytes to granulocytes is 40% to 60%, respectively) to an increase in lymphocytes/monocytes, possibly due to art increase in B-cells and monocytes (i.e. lymphocytes/monocytes to> granulocytes is now 60% to 40%, respectively). This shift is present after EpCAM immunoselection in both normal donor blood samples and prostate* blood samples.

Amplification of genes overexpressed in epithelial cells, yet still expressed in certain leukocytes may be relevant in disease diagnosis and treatment. The background noise from the leukocyte component contributes substantial interference to the amplification of these genes as they are expressed on isolated epithelial cells. The relative expression in leukocyte subsets and the carryover of these subsets are considerations in any genetic interpretation of circulating epithelial cells, especially after WBC subtraction.

These examples are several of many possible gene sets obtained through the embodiment of the present invention which can be exclusively expressed in specific cancer types like these (breast, prostate, or colorectal cancer), and potentially serve as cancer-specific CTC markers. Genetic information describing two or more cancer types may also serve as cancer-specific markers, but may further provide insight into a common thread between surveyed cancer types in the research and development of anti-cancer agents.

Accordingly, it is to be appreciated that the foregoing preferred embodiments of the present invention are not intended to be limitative of its scope, and that one skilled in the art will be able to conceive of various variations and modifications of such particular embodiments, all of which should be considered to be within the scope of the invention, which is limited solely by the following claims.

Claims

What is claimed is:

1. A method for detecting genetic information of rare cells in a biological sample comprising: g. obtaining a biological sample containing a mixed population of cells from an individual suspected of having target rare cells; h. fractionating said biological sample to obtain a fraction suspected of containing said rare cells; i. assessing said fraction for a first gene profile; j. separating said rare cells from said fraction whereby a depleted fraction is devoid of said rare cells; k. determining a second gene profile of said depleted fraction; and

I. subtracting said second gene profile from said first gene profile to obtain said genetic information from said rare cells.

2. The method of claim 1 whereby said rare cells are from a group consisting of cancer cells, epithelial cells, endothelial cells, activated T-lymphocyte cells, dendritic cells and combinations thereof.

3. The method of claim 1 whereby said fraction is a white blood cell region from a density-partitioned blood sample.

4. The method of claim 1 whereby said separating is an immunomagnetic enrichment of said rare cell populations from said fraction.

5. The method of claim 1 whereby said assessing is by detection of hybridized genetic material in said fraction with an array of known genetic markers on a first fixed support.

6. The method of claim 1 whereby said determining is by detection of hybridized genetic material in said depleted fraction with said array of known genetic markers on a second fixed support.

7. The method of claim 1 whereby said subtracting is a direct comparative analysis of individual genes within said gene profile.

8. The method of claim 1 whereby said genetic information is indicative o^"f cancer, cardiovascular disease, autoimmune diseases and combinations thereof.

9. A system for detecting genetic information of rare cells in a biological sample comprising: a. means for obtaining a biological sample containing a mixed population of cells from an individual suspected of having rare cells; b. means for fractionating said biological sample to obtain a fraction suspected of containing said rare cells; c. means for assessing said fraction for a first gene profile; d. means for separating said rare cells from said fraction whereby a depleted fraction is devoid of said rare cells; e. means for determining a second gene profile of said depleted fraction; and f. means for subtracting said second gene profile from said first gene profile to obtain said genetic information from said rare cells.

10. The system of claim 9 whereby said rare cells are from a group consisting of cancer cells, epithelial cells, endothelial cells, activated T-lymphocyte cells, dendritic cells and combinations thereof.

11. The system of claim 9 whereby said fractionating means is centrifugation which forms a density-partitioned blood sample.

12. The system of claim 9 whereby said assessing means is a first microarry chip.

13. The system of claim 9 whereby said separating means is- an immunomagnetic particle antigenically linked to said rare cell.

14. The system of claim 9 whereby said determining means is a second microarry chip.

15. The system of claim 9 whereby said subtracting means b y a comparision between fluorescent hybridization intensity signals of individual genes on said first microarray chip and said second microarray chip by the group consisting of manual inspection, automated fluorescent analysis, and combinations thereof.

16. The system of claim 9 whereby said genetic information is a diagnostic tool in assessing cancer, cardiovascular disease, autoimmune diseases and combinations thereof.