WO2006085828A1

WO2006085828A1 - Methods for the detection of hepatocellular carcinoma

Info

Publication number: WO2006085828A1
Application number: PCT/SG2006/000026
Authority: WO
Inventors: Chon Kar Leow; Bo San Paul Lai; Heng Hang Tsai; Hon Chiu Eastwood Leung; Joshy George
Original assignee: Agency For Science, Technology And Research
Priority date: 2005-02-11
Filing date: 2006-02-11
Publication date: 2006-08-17

Abstract

The invention provides uses for C3a molecules (e.g., C3a polypeptides, and nucleic acid molecules encoding C3a polypeptides) and other molecules that are differentially expressed in hepatocellular carcinoma samples when compared to non- hepatocellular carcinoma samples.

Description

METHODS FOR THE DETECTION OF HEPATOCELLULAR CARCINOMA

CROSS-REFERENCE TO RELATED APPLICATIONS

This application claims the benefit of U.S. provisional application number 60/651,644, filed February 11, 2005, which is hereby incorporated by reference.

FIELD OF THE INVENTION

The invention is in the field of cancer. More specifically, the invention is in the field of hepatocellular carcinoma.

BACKGROUND OF THE INVENTION

Hepatocellular carcinoma (HCC) is the most common primary malignant tumor of the liver and accounts for more than 70% of liver cancers worldwide (1). Many risk factors have-been associated with the development of HCC, including hepatitis B (HBV) and hepatitis C (HCV) viral infection, cirrhosis, HBV, male gender, exposure to toxins, etc. Death generally occurs due to liver failure associated with cirrhosis, HBV and/or metastasis. HCC is estimated to cause between 250,000-1 million deaths annually worldwide, and the cancer is especially prevalent in Southeast Asia, China, and sub- Saharan Africa. While surgical resection is considered to be the main curative treatment, only 10-15% of cases are suitable for surgery at the time of presentation. This is because either the disease is detected at an advanced stage at presentation or the underlying poor liver functional reserve precludes surgical intervention.

To improve the prognosis of HCC, screening and early detection are widely believed to be an effective form of secondary prevention. Diagnostic tests for HCC rely on radiological examination of liver and include complex imaging procedures such as trans-abdominal ultrasound, computerized tomography, and magnetic resonance imaging (2-7). All these procedures however require a minimal tumor size of 5 mm in diameter in order to be detected unequivocally. Diagnosis of HCC has also included detection of the presence of a liver mass on radiological investigations and the detection of elevated serum alpha fetoprotein (AFP) isoforms (>10 ng/mL) and other serological markers (8- 10). However, elevation of AFP is not exclusive to HCC and has been observed in non- malignant chronic liver disease, such as liver cirrhosis, HBV infection, chronic hepatitis, pregnancy, and cancers such as germ cell cancer (11).

Two dimensional gel electrophoresis analyses of HCC cell lines, auto-antibodies, and liver tissues have identified potential protein biomarkers for HCC [17-22]. Recent advances in mass spectrometry, such as SELDI time-of-flight mass spectrometry, an extension of matrix assisted laser desorption and ionization (MALDI) that uses different surface chemistries for affinity capture of proteins from complex biological samples, followed by mass spectrometry analysis, offer powerful tools for proteomic profiling and characterization.

C3a is one of the anaphylatoxins of the complement system which consists of a family of factors such as C4b, C5b, C6 to C9. C3-deficient mice exhibit high mortality, parenchymal damage, and impaired liver regeneration after partial hepatotectomy (12). Another important function of C3a is in complement-mediated immunological surveillance, and cell surface proteases have been implicated in C3 cleavage (13).

SUMMARY OF THE INVENTION

The invention provides, in one aspect, a method of diagnosing hepatocellular carcinoma in a sample from a subject by: a) detecting the level of expression of a C3a polypeptide or variant, analog, or fragment thereof, or a C3a nucleic acid molecule or complement, variant, analog, or fragment thereof, in the sample; and b) comparing the level of expression detected in step a) to a control, where a decrease in the level of expression of the C3a polypeptide or variant, analog, or fragment thereof, or the C3a nucleic acid molecule or complement, variant, analog, or fragment thereof, in the sample when compared to the control indicates a diagnosis of hepatocellular carcinoma.

In alternative aspects, the invention provides a method of monitoring the progression or regression of hepatocellular carcinoma in a subject by: a) detecting the level of expression of a C3a polypeptide or variant, analog, or fragment thereof, or a C3a nucleic acid molecule or complement, variant, analog, or fragment thereof, in a first sample obtained from the subject at a first time point; b) detecting the level of expression of a C3a polypeptide or variant, analog, or fragment thereof, or a C3a nucleic acid molecule or complement, variant, analog, or fragment thereof, in a subsequent sample obtained from the subject at a subsequent time point; and c) comparing the level of expression detected in steps a) and b), where a differential expression of the C3a polypeptide or variant, analog, or fragment thereof, or the C3a nucleic acid molecule or complement, variant, analog, or fragment thereof, in the subsequent sample, when compared to the first sample, indicates the progression or regression of hepatocellular carcinoma. In various embodiments, a decrease in the level of expression in the subsequent sample when compared to the first sample may indicate progression of hepatocellular carcinoma. In various embodiments, an increase in the level of expression in the subsequent sample when compared to the first sample may indicate regression of hepatocellular carcinoma. In various embodiments, the subsequent sample may be obtained at two or more time points.

In alternative aspects, the invention also provides a method of selecting a subject for a hepatocellular carcinoma therapy by: a) detecting the level of expression of a C3a polypeptide or variant, analog, or fragment thereof, or a C3a nucleic acid molecule or complement, variant, analog, or fragment thereof, in a sample obtained from the subject; and b) comparing the level of expression detected in step a) to a control, where a differential expression of the C3a polypeptide or variant, analog, or fragment thereof, or the C3a nucleic acid molecule or complement, variant, analog, or fragment thereof, in the sample when compared to the control indicates the therapy to be selected.

In alternative aspects, the invention also provides a method of monitoring the efficacy of a hepatocellular carcinoma therapy in a subject by: a) administering the therapy to the subject; b) detecting the level of expression of a C3a polypeptide or variant, analog, or fragment thereof, or a C3a nucleic acid molecule or complement, variant, analog, or fragment thereof, in at least one sample obtained from the subject; and c) comparing the level of expression detected in step b) to a control, where a differential expression of the C3a polypeptide or variant, analog, or fragment thereof, or the C3a nucleic acid molecule or complement, variant, analog, or fragment thereof, in the sample when compared to the control indicates the efficacy of the therapy. In various embodiments, the sample may be obtained prior to administration of the therapy; the sample may be obtained subsequent to administration of the therapy; the therapy may be administered at two or more administration time points; the sample may be obtained at two or more sampling time points. In various embodiments, the method may further include comparing the level of expression of the C3a polypeptide or variant, analog, or fragment thereof, or the C3a nucleic acid molecule or complement, variant, analog, or fragment thereof, in the two or more sampling time points. In various embodiments, a decrease in the level of expression in the sample when compared to the control may indicate an inefficacious therapy. In various embodiments, an increase in the level of expression in the sample when compared to the control may indicate an efficacious therapy.

In alternative aspects, the invention also provides a method of prognosing hepatocellular carcinoma in a sample from a subject by: a) detecting the level of expression of a C3a polypeptide or complement, variant, analog, or fragment thereof, or a C3a nucleic acid molecule or complement, variant, analog, or fragment thereof, in the sample; and b) comparing the level detected in step a) to a control, where a differential expression of the C3a polypeptide or variant, analog, or fragment thereof, or the C3a nucleic acid molecule or complement, variant, analog, or fragment thereof, in the sample when compared to the control indicates the prognosis. In various embodiments, a decrease in the level of expression in the sample when compared to the control may indicate a poor prognosis. In various embodiments, an increase in the level of expression in the sample when compared to the control may indicate a good prognosis.

In various embodiments of the aspects of the invention, the methods may further include detecting trie level of expression of a beta-2 microglobulin or Chain A, human serum albumin mutant R218ρ polypeptide or variant, analog, or fragment thereof or nucleic acid molecule or complement, variant, analog, or fragment thereof. In various embodiments, the method may further include detecting the level of expression of an alpha fetoprotein, cytokeratin 19, des-gamma-carboxy prothrombin (DCP), serum gamma-glutamyl transferase, platelet-derived endothelial growth factor, or insulin-like growth factor binding protein polypeptide or nucleic acid molecule. In various embodiments, the sample may be liver, plasma, or serum; and/or the control may be liver, plasma or serum. In various embodiments, the sample may be, or may be suspected of being, a HCC sample. In various embodiments, the control may be a non-HCC sample. In various embodiments, the level of expression may be detected using a peptide, a small molecule, or an antibody (e.g., a detectably labeled antibody) that specifically binds to the C3a polypeptide or variant, analog, or fragment thereof. In various embodiments, the C3a nucleic acid molecule may be a mRNA molecule. In various embodiments, the level of expression may be detected using a probe or primer (e.g., a detectably labeled probe or primer) that hybridizes to the C3a nucleic acid molecule or complement, variant, analog, or fragment thereof. In various embodiments, the methods may further include generating a polypeptide or nucleic acid molecule expression profile; and/or the level of expression may be detected using a high. throughput assay. In various embodiments, the subject (e.g., . a human) may have, may be suspected of having, or may be at risk for having hepatocellular carcinoma.

In alternative aspects, the invention also provides a method of screening a candidate compound for treating hepatocellular carcinoma by: a) contacting a test system with a test compound; b) detecting the level of expression of a C3a polypeptide or variant, analog, or fragment thereof, or a C3a nucleic acid molecule or complement, variant, analog, or fragment thereof, in the test system; c) detecting the level of expression of the C3a polypeptide or variant, analog, or fragment thereof, or the C3a nucleic acid molecule or complement, variant, analog, or fragment thereof in a control system that is not exposed to the test compound; and d) comparing the level of expression in step b) and step c), where differential expression of the C3a polypeptide or variant, analog, or fragment thereof, or the C3a nucleic acid molecule or complement, variant, analog, or fragment thereof in the comparison indicates that the test compound is a candidate compound for treating hepatocellular carcinoma.

In various embodiments, the method may further include detecting the level of expression of a beta-2 microglobulin or Chain A, human serum albumin mutant R218p polypeptide oi variant, analog, or fragment thereof or nucleic acid molecule or complement, variant, analog, or fragment thereof; and/or may further include detecting the level of expression of an alpha fetoprotein, cytokeratin 19, des-gamma-carboxy prothrombin (DCP), serum gamma-glutamyl transferase, platelet-derived endothelial growth factor, and/or insulin-like growth factor binding protein polypeptide or nucleic acid molecule. In various embodiments, the level of expression may be detected using an antibody (e.g., a detectably labeled antibody) that specifically binds to the C3a polypeptide or variant, analog, or fragment thereof. In various embodiments, the C3a nucleic acid molecule may be a mRNA. In various embodiments, the level of expression may be detected using a probe or primer (e.g., a detectably labeled probe or primer) that hybridizes to the C3a nucleic acid molecule or complement, variant, analog, or fragment thereof. In various embodiments, the method may further include generating a polypeptide or nucleic acid molecule expression profile; or may including detecting the level of expression using a high throughput assay. In various embodiments, the test system or control system may be an animal model for hepatocellular carcinoma or may be a hepatocellular carcinoma cell line.

In alternative aspects, the invention also provides a composition including an addressable collection of two or more C3a or beta-2 microglobulin polypeptide or variants, analogs, or fragments thereof, or two or more C3a or beta-2 microglobulin nucleic acid molecule or complements, variants, analogs, or fragments thereof, that are differentially expressed in hepatocellular carcinoma. In various embodiments, the composition may further include a Chain A, human serum albumin mutant R218p polypeptide or variant, analog, or fragment thereof or nucleic acid molecule or complement, variant, analog, or fragment thereof; and/or may include an alpha fetoprotein, cytokeratin 19, des-gamma-carboxy prothrombin (DCP), serum gamma- glutamyl transferase, platelet-derived endothelial growth factor, and/or insulin-like growth factor binding protein polypeptide or nucleic acid molecule. In various embodiments, the nucleic acid molecules or the polypeptides may be attached to a solid support (e.g., a microarray). In alternative aspects, the invention also provides use of the composition in the preparation of a medicament for diagnosis, prognosis, and/or monitoring the progression or regression of hepatocellular carcinoma, and/or for monitoring the efficacy of a therapy, and/or for selecting a subject for a therapy for hepatocellular carcinoma.

In alternative aspects, the invention also provides use of a C3a polypeptide or variant, analog, or fragment thereof, or a C3a nucleic acid molecule or complement, variant, analog, or fragment thereof, in the preparation of a medicament for diagnosis, prognosis, and/or monitoring the progression or regression of hepatocellular carcinoma, and/or for monitoring the efficacy of a therapy, and/or selecting a subject for a therapy for hepatocellular carcinoma wherein the C3a polypeptide or nucleic acid molecule is differentially expressed in hepatocellular carcinoma. In various embodiments, the use may further include a beta 2 microglobulin or Chain A, human serum albumin mutant R218p polypeptide or variant, analog, or fragment thereof or nucleic acid molecule or complement, variant, analog, or fragment thereof; and/or an alpha fetoprotein, cytokeratin 19, des-gamma-carboxy prothrombin (DCP), serum gamma-glutamyl transferase, platelet-derived endothelial growth factor, and/or insulin-like growth factor binding protein polypeptide or nucleic acid molecule.

In alternative aspects, the invention also provides a method of preventing degradation of a C3a polypeptide or variant, analog, or fragment thereof, in a sample containing a matrix metalloprotease 9 polypeptide or nucleic acid molecule, by contacting the matrix metalloprotease 9 polypeptide or nucleic acid molecule with a matrix metalloprotease 9 inhibitor, such that the matrix metalloprotease 9 polypeptide or nucleic acid molecule is prevented from degrading the C3a polypeptide. The sample may be a liver tissue, liver cells, plasma, serum, or other body fluid sample. The sample may be, or may be suspected of being, a HCC sample. The level of expression may be detected using an antibody, peptides, or small molecules that specifically binds to the C3a polypeptide or variant, analog, or fragment thereof. The antibody may be detectably labeled.

In alternative aspects, the invention also provides a method of screening a candidate compound for treating hepatocellular carcinoma by contacting a C3a polypeptide or variant, analog, or fragment thereof, with a matrix metalloprotease 9 polypeptide in the presence or in the absence of a test compound (e.g., a matrix metalloprotease 9 inhibitor) and detecting the level of expression of the C3a polypeptide or variant, analog, or fragment thereof in the presence and in the absence of the test compound, where an increase in the expression of the C3a polypeptide or variant, analog, or fragment thereof in the presence of the test compound, relative to the expression of the C3a polypeptide or variant, analog, or fragment thereof in the absence of the test compound, indicates that the test compound is a candidate compound for treating hepatocellular carcinoma. The level of expression may be detected using an antibody that specifically binds to the C3a polypeptide or variant, analog, or fragment thereof. The antibody may be detectably labeled.

In alternative aspects, the invention also provides a method of preventing degradation of a C3a polypeptide in a subject in need thereof, by administering a matrix metalloprotease 9 inhibitor to the subject. In alternative aspects, the invention also provides the use of a matrix metalloprotease 9 inhibitor for the preparation of a medicament for preventing degradation of a C3a polypeptide in. a subject in need thereof. The subject (e.g., a human) may have, or may be suspected of having, or may be at risk for having hepatocellular carcinoma.

By "C3a" is meant a complement component 3 molecule. In some embodiments, a C3a molecule may include a polypeptide, or a fragment, analog, or variant thereof, or a nucleic acid molecule encoding or corresponding to a complement component 3 precursor polypeptide or a fragment, analog, or variant thereof. In some embodiments, a C3a nucleic acid molecule may include C3a coding or may include C3a non-coding genomic sequences, or may include C3a coding and non-coding genomic sequences such as those crossing exon-intron boundaries. In some embodiments, a C3a nucleic acid molecule may include C3a exon sequences, C3a intron sequences, C3a promoter sequences, C3a enhancer sequences etc. In some embodiments, a C3a polypeptide or nucleic acid molecule according to the invention may include, without limitation, human C3a precursor sequences identified by the following Accession numbers: NP_000055, P01024, C3HU, or as described herein, or fragments, analogs, or variants thereof, or sequences from other species, for example, those identified by the following Accession numbers: NPJ333908 (mouse), P01026 (rat), P01027 (mouse), P12387 (domestic guinea pig), C3GP (domestic guinea pig), C3RT (rat), C3MS (mouse), or fragments, analogs, or variants thereof. In some embodiments, a C3a polypeptide includes C3a polypeptide and fragments as described herein or as set forth in for example PCT publication no. WO 98/58960 or U.S. Patent No. 6,682,740 or in Accession Nos.: AAA73037 (synthetic), AAA72712 (synthetic), AAG00532 (rat), P01025 (pig), AAS31171, AAS31170, AAS31169, AAS31168, AAS31167, AAS31166, AAS31165, AAS31164, AAS31163, AAS31162, AAS31161, AAS31160, AAS31159, AAS31158, AAS31157, AAS31156, AAS31155, AAS31154, AAS31153, AAS31152, AAS31151, AAS31150, AAS31149, AAS31148, AAS31147, AAS31146, AAS31145, AAS31144, AAS31143, or AAS31142. In some embodiments, a C3a polypeptide includes a 68 amino acid residue fragment of C3a after C-terminal truncation of 9 amino acids from the full-length C3a protein.

By "beta-2 microglobulin" is meant a beta-2 microglobulin polypeptide or a fragment, analog, or variant thereof, or a nucleic acid molecule encoding or corresponding to a beta-2 microglobulin polypeptide or a fragment, analog, or variant ' thereof. In some embodiments, a beta-2 microglobulin nucleic acid molecule may include beta-2 microglobulin coding or may include beta-2 microglobulin non-coding genomic sequences, or may include beta-2 microglobulin coding and non-coding genomic sequences such as those crossing intron-exon boundaries. In some embodiments, a beta-2 microglobulin nucleic acid molecule may include beta-2 microglobulin exon sequences, beta-2 microglobulin intron sequences, beta-2 microglobulin promoter sequences, beta-2 microglobulin enhancer sequences etc. In some embodiments, a beta-2 microglobulin polypeptide or nucleic acid molecule may include, without limitation, a human beta-2 microglobulin sequences identified by the following Accession numbers: NP_004039, AAH64910, AAH32589, (human, precursor), or as described herein, or fragments, analogs, or variants thereof, or sequences from other species, for example, those identified by the following Accession numbers: AAH85164.1 (mouse), NP_033865.2 (mouse), NP_776318.1 (cow), NP_036644.1 (rat), etc. or fragments, analogs, or variants thereof. By "promoter" is meant minimal sequence sufficient to direct transcription. Promoters usually lie 5' from the sequence to be read and regulate the transcription rate of a gene. Also included in the invention are those promoter elements which are sufficient to render promoter-dependent gene expression controllable for cell-type specific, tissue-specific or inducible by external signals or agents; such elements may be located in the 5' or 3' regions of the native gene. "Enhancers" can be counted among the activators and differ from other regulation elements in that they usually lie at a greater distance from the promoter 5' or 3' and can increase the transcription activity in a position-independent manner.

A "protein," "peptide" or "polypeptide" is any chain of two or more amino acids, including naturally occurring or non-naturally occurring amino acids or amino acid analogs, regardless of post-translational modification (e.g., glycosylation or phosphorylation). An "amino acid sequence", "polypeptide", "peptide" or "protein" of the invention may include peptides or proteins that have abnormal linkages, cross links and end caps, non-peptidyl bonds or alternative modifying groups. Such modified peptides are also within the scope of the invention. The term "modifying group" is intended to include structures that are directly attached to the peptidic structure (e.g., by covalent coupling), as well as those that are indirectly attached to the peptidic structure (e.g., by a stable non-covalent association or by covalent coupling to additional amino acid residues, or mimetics, analogs or derivatives thereof, which may flank the core peptidic structure). For example, the modifying group can be coupled to the amino- terminus or carboxy-terminus of a peptidic structure, or to a peptidic or peptidomimetic region flanking the core domain. Alternatively, the modifying group can be coupled to a side chain of at least one amino acid residue of a peptidic structure, or to a peptidic or peptido- mimetic region flanking the core domain (e.g., through the epsilon amino group of a lysyl residue(s), through the carboxyl group of an aspartic acid residue(s) or a glutamic acid residue(s), through a hydroxy group of a tyrosyl residue(s), a serine residue(s) or a threonine residue(s) or other suitable reactive group on an amino acid side chain). Modifying groups covalently coupled to the peptidic structure can be attached by means and using methods well known in the art for linking chemical structures, including, for example, amide, alkylamino, carbamate or urea bonds. A polypeptide according to the invention includes a C3a, beta-2 microglobulin, or human serum albumin mutant R218p polypeptide or fragment, variant or analog thereof.

The terms "nucleic acid" or "nucleic acid molecule" encompass both RNA (plus and minus strands) and DNA, including cDNA, genomic DNA, and synthetic (e.g., chemically synthesized) DNA. The nucleic acid may be double-stranded or single- stranded. Where single-stranded, the nucleic acid may be the sense strand or the antisense strand. A nucleic acid molecule may be any chain of two or more covalently bonded nucleotides, including naturally occurring or non-naturally occurring nucleotides, or nucleotide analogs or derivatives. By "RNA" is meant a sequence of two or more covalently bonded, naturally occurring or modified ribonucleotides. One example of a modified RNA included within this term is phosphorothioate RNA. By "DNA" is meant a sequence of two or more covalently bonded, naturally occurring or modified deoxyribonucleotides. By "cDNA" is meant complementary or copy DNA produced from an RNA template by the action of RNA-dependent DNA polymerase (reverse transcriptase). Thus a "cDNA clone" means a duplex DNA sequence complementary to an RNA molecule of interest, carried in a cloning vector. An "oligonucleotide" as used herein is a single stranded molecule which may be used in hybridization or amplification technologies. In general, an oligonucleotide may be any integer from about 15 to about 100 nucleotides in length, but may also be of greater length.

A "probe" or "primer" is a single-stranded DNA or RNA molecule of defined sequence that can base pair to a second DNA or RNA molecule that contains a complementary sequence (the target). The stability of the resulting hybrid molecule depends upon the extent of the base pairing that occurs, and is affected by parameters such as the degree of complementarity between the probe and target molecule, and the degree of stringency of the hybridization conditions. The degree of hybridization stringency is affected by parameters such as the temperature, salt concentration, and concentration of organic molecules, such as formamide, and is determined by methods that are known to those skilled in the art. Probes or primers may hybridize to the target molecule, e.g., a C3a, beta-2 microglobulin, or human serum albumin mutant R218p nucleic acid molecule as described herein or known in the art under conditions of high stringency. Probes or primers specific for the nucleic acid sequences described herein, or portions thereof, may vary in length by any integer from at least 8 nucleotides to over 500 nucleotides, including any value in between, depending on the purpose for which, and conditions under which, the probe or primer is used. For example, a probe or primer may be 8, 10, 15, 20, or 25 nucleotides in length, or may be at least 30, 40, 50, or 60 nucleotides in length, or may be over 100, 200, 500, or 1000 nucleotides in length. Probes or primers specific for the C3a, beta-2 microglobulin, or human serum albumin mutant R218p nucleic acid molecules described herein or known in the art may have greater than any integer between 20-30% sequence identity, or at least any integer between 55-75% sequence identity, or at least any integer between 75-85% sequence identity, or at least any integer between 85-99% sequence identity, or 100% sequence identity to the nucleic acid sequences described herein. Probes or primers can be detectably-labeled, either radioactively or non-radioactively, by methods that are known to those skilled in the art. Probes or primers can be used for methods involving nucleic acid hybridization, such as nucleic acid sequencing, nucleic acid amplification by the polymerase chain reaction, single stranded conformational polymorphism (SSCP) analysis, restriction fragment polymorphism (RFLP) analysis, Southern hybridization, northern hybridization, in situ hybridization, electrophoretic mobility shift assay (EMSA), microarray, and other methods that are known to those skilled in the art. Probes or primers may be derived from genomic DNA or cDNA, for example, by amplification, or from cloned DNA segments, or may be chemically synthesized. A nucleic acid molecule according to the invention includes a C3a, beta-2 microglobulin, or human serum albumin mutant R218p nucleic acid molecule or complement, fragment, variant or analog thereof.

By "differential expression" or "differentially expressed" is meant increased, upregulated or present, or decreased, downregulated or absent, polypeptide or nucleic acid molecule expression as detected by the absence, presence, or change (up or down) in the amount of a polypeptide or nucleic acid molecule of interest in a sample. For example, the change may be detected by comparison of the decrease or downregulation in the expression level of a polypeptide or nucleic acid molecule of interest in a HCC sample when compared to a non-HCC sample. The absolute change of expression of the polypeptide or nucleic acid molecule in a HCC sample when compared to a non-HCC sample is not important, as long as the change is reproducible, and measurable by for example performing standard statistical analyses. Differential expression may include qualitative or quantitative changes in polypeptide or nucleic acid molecule expression. In some embodiments, a "differentially expressed" gene may be regulated at the nucleic acid or polypeptide level or may be alternatively spliced, leading to related expression products which may manifest in altered mRNA levels, cellular partitioning, etc. In some embodiments, the change (up or down) in polypeptide or nucleic acid molecule expression may be at least 1-fold or at least 1.5-fold or may be over 2.0, 2.5, 3.0, 3.5, 4.0, 4.5, 5.0, 5.5, 6.0, 6.5, 7.0, 7.5, 8.0, 8.5, 9.0, 9.5, or over 10.0-fold. In some embodiments, the change (up or down) in polypeptide or nucleic acid molecule expression may be 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90%, or 100%.

By "detecting" it is intended to include determining the presence or absence, or quantifying the amount, of the level of expression of a polypeptide or nucleic acid molecule of interest, such as a C3a, beta-2 microglobulin or Chain A, human serum albumin mutant R218p molecule. The term thus refers to the use of the materials, compositions, and methods of the present invention for qualitative and quantitative determinations. Detecting may include comparing the expression level of a polypeptide or nucleic acid molecule of interest, such as a C3a, beta-2 microglobulin or Chain A, human serum albumin mutant R218p molecule, between a test sample and a control sample or standard, e.g., detecting differential expression. In some embodiments, detecting may include quantifying a change (increase or decrease) of any value between 10% and 90%, or of any value between 30% and 60%, or over 100% (e.g., a change of about 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90%, or 100%) of a nucleic acid molecule or polypeptide of interest, when compared to a control. In other embodiments, detecting may include quantifying a change of any value between 1 to 5 fold or more (e.g., at least 1-fold or at least 1.5-fold or may be over 2.0, 2.5, 3.0, 3.5, 4.0, 4.5, 5.0, 5.5, 6.0, 6.5, 7.0, 7.5, 8.0, 8.5, 9.0, 9.5, or over 10.0-fold) of a polypeptide or nucleic acid molecule of interest when compared to a control. The quantification may be represented in arbitrary units, depending on the assay used. For example, the quantification may be represented in arbitrary units from a scintillation counter, densitometer, ELISA reader, luminometer, etc.

"Hepatocellular carcinoma" or "HCC" is cancer that arises from hepatocytes, the major cell type of the liver. It is the most common form of primary tumor of liver arising from hepatocytes. The phrase "suspected of being cancerous" as used herein means a HCC sample believed by one of ordinary skill in the art to contain HCC tissue or cells or extracts thereof as for example determined by standard diagnostic or other techniques (e.g., liver biopsy, including radiological biopsy by means of a radiological scan, laparoscopy, or open surgical biopsy, liver . autopsy, histologic staining, microscopic analysis, immunoassay, ultrasound, computed tomography, magnetic resonance imaging, hepatic arteriography, etc.). HCC therapy may include liver transplantation, surgical resection, ethanol injection, radiofrequency ablation, transarterial chemoembolization, etc.

"Non-HCC" refers to a sample demonstrated by for example standard diagnostic or other techniques (e.g., liver biopsy, liver autopsy, histologic staining, microscopic analysis, immunoassay, ultrasound, computed tomography, magnetic resonance imaging, hepatic arteriography, etc.) to contain no HCC cells or no evidence of HCC. In some embodiments, a non-HCC sample may include tissue or cells or extracts thereof from a subject having a liver disorder, such as hepatitis (e.g., HBV or HCV) infection, cirrhosis, exposure to aflatoxins, etc. In some embodiments, a non-HCC sample may include tissue or cells or extracts thereof from a subject diagnosed with a cancer that is not determined to be HCC. In some embodiments, a non-HCC sample may include tissue or cells or extracts thereof from a normal subject, e.g., a subject not known to have a pathological condition associated with or resulting in liver dysfunction.

A "sample" can be any organ, tissue, cell, or cell extract isolated from a subject. For example, a sample can include, without limitation, cells or tissue (e.g., from a biopsy or autopsy) from bone, brain, breast, colon, muscle, nerve, ovary, prostate, retina, skin, skeletal muscle, intestine, testes, heart, liver, kidney, stomach, pancreas, uterus, adrenal gland, tonsil, spleen, soft tissue, peripheral blood, whole blood, red cell concentrates, platelet concentrates, leukocyte concentrates, blood cell proteins, blood plasma, platelet- rich plasma, a plasma concentrate, a precipitate from any fractionation of the plasma, a supernatant from any fractionation of the plasma, blood plasma protein fractions, purified or partially purified blood proteins or other components, serum, semen, mammalian colostrum, milk, urine, stool, saliva, placental extracts, amniotic fluid, a cryoprecipitate, a cryosupernatant, a cell lysate, mammalian cell culture or culture medium, products of fermentation, ascitic fluid, proteins present in blood cells, solid tumours, isolated from a mammal with a hepatocellular carcinoma, or any other specimen, or any extract thereof, obtained from a patient (human or animal), test subject, or experimental animal. In some embodiments, the tissue may be from a normal (healthy) subject; a subject having a hepatocellular carcinoma; a subject having a cancer that is not hepatocellular carcinoma; a subject infected with a hepatitis virus (e.g., HBV or HCV); a subject having a liver disorder e.g., cirrhosis, or a subject having a normal liver e.g., not diagnosed with or suspected of having a liver disorder. In some embodiments, it may be desirable to separate cancerous cells from non-cancerous cells in a sample. A sample may also include, without limitation, products produced in cell culture by normal or transformed cells (e.g., via recombinant DNA or monoclonal antibody technology). A sample may also include, without limitation, any organ, tissue, cell, or cell extract isolated from a non-mammalian subject, such as an insect or a worm. A "sample" may also be a cell or cell line created under experimental conditions, that is not directly isolated from a subject. A sample can also be cell-free, artificially derived or synthesised. A sample may be from a cell or tissue known to be cancerous, suspected of being cancerous, or believed non-cancerous (e.g., normal or control). In some embodiments, a sample refers to liver tissue or cells or extracts thereof. A "test sample" includes a sample obtained from a subject having, or suspected of having, or at risk for having a hepatocellular carcinoma. A "test sample"also includes a sample from a hepatocellular carcinoma cell line or animal model. A "control sample" includes a sample obtained from a normal (healthy) subject; a subject having a cancer that is not hepatocellular carcinoma; a subject infected with a hepatitis virus (e.g., HBV or HCV); a subject having a liver disorder e.g., cirrhosis, or a subject having a normal liver e.g., not diagnosed with or suspected of having or at risk for having a liver disorder. A control sample may also include a sample that contains substantially the same level of C3a, beta-2 microglobulin, or human serum albumin mutant R218p nucleic acid molecule or polypeptide normally present in normal (healthy) subject; a subject having a cancer that is not hepatocellular carcinoma; a subject infected with a hepatitis virus (e.g., HBV or HCV); a subject having a liver disorder e.g., cirrhosis, or a subject having a normal liver e.g., not diagnosed with or suspected of having or at risk for having a liver disorder. Multiple test and/or control samples may be obtained from the same subject. Samples may be obtained at the same time or at multiple different time points.

A "control" includes a sample or standard obtained for use in determining baseline expression or activity. Accordingly, a control may be obtained by a number of means including from non-cancerous cells or tissue e.g., from non-cancerous cells surrounding a HCC carcinoma of a subject; from a subject not having a cancer; from a subject not suspected of being at risk for a cancer; from a subject not having a hepatitis infection (e.g., HBV or HCV), or from cells or cell lines derived from such subjects, or extracts thereof. A control also includes a standard, e.g., previously established standard. Accordingly, any test or assay conducted according to the invention may be compared with the standard and it may not be necessary to obtain a control sample for comparison each time.

As used herein, a subject may be a human, non-human primate, rodent (e.g., rat, mouse, hamster, guinea pig, etc.) cow, horse, pig, sheep, goat, dog, cat, fly, worm, etc. The subject may be a clinical patient, a clinical trial volunteer, an experimental animal, etc. The subject may be suspected of having or being at risk for having HCC, be diagnosed with HCC, or be a control subject e.g., a subject that is confirmed to not have HCC, or that is confirmed to not have a hepatitis infection. Diagnostic methods for HCC and the clinical delineation of HCC diagnoses are known to those of ordinary skill in the art, and include biopsy including radiological biopsy by means of a radiological scan, laparoscopy, or open surgical biopsy, liver autopsy, histologic staining, microscopic analysis, immunoassay, ultrasound, computed tomography, magnetic resonance imaging, hepatic arteriography, etc.

A "composition" as used herein includes a plurality of the nucleic acid molecules described herein, including complements, analogs, variants, and fragments thereof. A composition as used herein also includes a plurality of polypeptides encoded by the nucleic acid molecules described herein, and analogs, variants, and fragments thereof. A composition as used herein also includes a plurality of polypeptides capable of specifically binding to the polypeptides or nucleic acid molecules described herein (e.g., antibodies). The composition may include any combination of the nucleic acid molecules described herein, including complements, analogs, variants, and fragments thereof, or polypeptides encoded by these nucleic acid molecules. In some embodiments, the composition may include subsets of the nucleic acid molecules or polypeptides described herein. These nucleic acid molecules or polypeptides may for example be used with a substrate (e.g, a solid substrate or a liquid substrate) in a variety of applications, including the diagnosis of HCC, or monitoring the progression of HCC.

By "addressable collection" is meant a combination of nucleic acid molecules or polypeptides capable of being detected by, for example, the use of hybridization techniques or antibody binding techniques or by any other means of detection known to those of ordinary skill in the art.

By "matrix metalloprotease 9" or "MMP-9" is meant a matrix metalloprotease 9 polypeptide or a fragment, analog, or variant thereof that is capable of proteolyrically degrading or digesting a C3a molecule. A matrix metalloprotease 9 also includes a nucleic acid molecule encoding or corresponding to a matrix metalloprotease 9 polypeptide or a fragment, analog, or variant thereof, that is capable of proteolytically degrading or digesting a C3a molecule. In some embodiments, a matrix metalloprotease 9 molecule may include, without limitation, a matrix metalloprotease 9 molecule identified by Accession number P14780 or AAA51539.

By "preventing degradation" of a C3a polypeptide is meant an increase in the level of expression, or an increase in the half-life, of a C3a polypeptide in a sample in the presence of a matrix metalloprotease 9 inhibitor, relative to a control, e.g., in the absence of a matrix metalloprotease 9 inhibitor. Such an increase may of any value between 10% and 90%, e.g., 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90%, or over 100%, or may be at least 1-fold or at least 1.5-fold or may be a change of any value between 2-fold to 10-fold, inclusive, e.g, 3 fold, 5 fold, 7 fold, or more e.g., 50 fold or 100-fold when compared to the control. The exact amount of increase is not critical, as long as it is statistically significant.

By "matrix metalloprotease 9 inhibitor" is meant a molecule, e.g., a polypeptide, nucleic acid molecule, or small molecule, that is capable of decreasing the expression or activity of a matrix metalloprotease 9 polypeptide or nucleic acid molecule. For example, a matrix metalloprotease 9 inhibitor may degrade, or cause to be degraded, a matrix metalloprotease 9 polypeptide or nucleic acid molecule, or may prevent a matrix metalloprotease 9 polypeptide from degrading a C3a polypeptide, by for example, binding to its active site and inhibiting its protease acitivity. A matrix metalloprotease 9 inhibitor may be naturally occurring or artificially synthesized. A matrix metalloprotease 9 inhibitor may be, without limitation, an antisense oligonucleotide, a triple-strand forming oligonucleotide, or a siRNA molecule directed against a matrix metalloprotease 9 nucleic acid sequence, a TIMP (tissue inhibitor of metalloproteinase) molecule, a hydroxamic acid (e.g., Marimastat), etc.

BRIEF DESCRIPTION OF THE DRAWINGS

Figure 1. Gel view of spectra of low molecular weight copper binding proteins in serum samples. Each strip represents the protein profile of one sample. Molecular weight ranges from 2500 Da to 15000 Da. MW: molecular weight. The arrows depict up- regulated 5904 Da marker in HCC, and down-regulated 8132 Da marker in HCC. Sample class: Normal, hepatocellular carcinoma (HCC), primary colorectal carcinoma (colon cancer), benign hepatitis B virus carrier (HBV), nasopharyngeal carcinoma (NPC).

Figures 2A-B. Natural log transformation of SELDI spectra. A) Before log transformation. B) log transformation. Figures 3A-B. Global mean normalization of SELDI data. A) Before global mean normalization. B) After global mean normalization.

Figure 4. Clustering of serum samples based on nine serum protein peaks. Scale bar of expression is at the left. Labels at the right denote sample type. N or Normal denotes normal samples. X or H denotes HCC samples.

Figure 5. Classification and regression tree (CART)-based segregation of serum samples. Each node shows the molecular weight of a selected protein peak and the cutoff value. X5904 < 0.015 denotes the cutoff of the protein peak signal at 5904 Da is at 0.015 intensity unit. Samples are segregated according to cutoff value at each node.

Figures 6A-B. Preliminary screening of binding and elution conditions for copper binding proteins on CU2+ chip surface with increasing concentrations of imidazole. A) Normal serum, and B) HCC serum. Arrows indicate target peaks.

Figures 7A-C. Correlation of protein band and SELDI peak of eluents from Cu2+ NTA column. A: SDS gel electrophoretic diagram of eluents from Cu2+-NTA column. B: SELDI profile of eluent of normal after Cu2+-NTA column. C: SELDI profile of eluent of tumor sample after Cu2+-NTA column. M: Markers. N: Normal sample. T: Tumor sample.

Figures 8A-C. Sequence analysis of tryptic peptide of 8.1 kDa gel band. A) Sequence spectrum of one tryptic peptide. B) Sequences of two peptides from Tandem mass spectrometry and Mascot search. C) Sequence alignment and theoretical mass calculation of complement C3a and sequenced fragments (underlined).

Figure 9. Western blot analysis of serum samples using polyclonal anti-C3a antibody. Lanes label denote the sample types. Normal: sera of normal individuals; ENT: Ear-nose-throat symptoms but non-NPC. HCC: sera of hepatocellular carcinoma patients at one day before surgery; . Colon: sera of primary colorectal carcinoma patients; HBV: sera of HBV carrier patients; NPC: sera of nasopharyngeal carcinoma patients, C3a std: 20 ng and 50 ng of C3a pure standard C3a protein. Each lane except for standard C3a lanes represents an individual patient sample.

Figure 10. Receiver operating characteristic ROC curve analysis of C3a band signal in Western Blot analyses of serum samples with antibody against C3a. Area under the ROC curve (AUC) (stepped line on the left) was 0.91. The diagonal line is the 45-degree diagonal of the ROC space.

Figures 11A-B. A) Two tryptic peptide sequences of 5.9 kDa gel band, B) alignment of human serum albumin with the sequenced peptides (underlined sequences).

Figures 12A-D. A) Gel Pattern of purified 7.5 kDa differential protein B) Amino acid sequences of two tryptic peptides from 7.5 kDa gel band by using MS/MS tandem mass spectrometry. C) Alignment of human beta-2 microglobulin (B2M) with sequenced fragments (underlined sequences). D) Western blot of B2M by using anti-B2M polyclonal antibody.

Figures 13A-E. C3a sequences. A) Polypeptide sequence of C3a precursor; B, C) Nucleotide sequence of C3a precursor, as set forth in Accession No. NP_000055; D) Polypeptide sequence of C3a; E) Polypeptide sequence of C3a C-terminal truncated fragment.

Figure 14. Degradation of C3a protein by MMP-9. Lane M: Mark 12 protein standards. Lane 1: 25 ng MMP-7 alone. Lane 2: 25 ng MMP-9 alone. Lane 3: 25 ng MMP-Il alone. Lane 4: 25 ng MMP-13 alone. Lane 5: 50 ng of C3a alone. Lane 6: 8 ng of gelatin alone. Lane 7: 8 ng of gelatin with 25 ng MMP7. Lane 8: 8 ng of gelatin with 25 ng MMP-9. Lane 10: 8 ng of gelatin with 25 ng of MMP-Il. Lane 11: 50 ng of C3a with 25 ng of MMP-7. Lane 12: 50 ng of C3a with 25 ng of MMP-9. Lane 13: 50 ng of C3a with 25 ng of MMP-Il. Lane 14: 50 ng of C3a with 25 ng of MMP-13.

DETAILED DESCRIPTION OF THE INVENTION

Phenotypic changes in cancer may be due to cellular changes at the molecular level. Thus, some genes may be differentially expressed, e.g., expressed, overexpressed, under-expressed, or not expressed in tumor cells relative to non-tumor cells. Selecting one or more of differentially expressed HCC genes, nucleic acid molecules, and/or polypeptides and creating an "expression profile" assists in predictable and accurate diagnosis and prognosis, and design of efficacious therapeutics.

The invention provides, in part, molecular markers for HCC. In some embodiments, the molecular markers described herein can distinguish HCC samples from control samples e.g., samples from normal (healthy) subjects, or subjects with chronic liver disease, other cancers, or other diseases. In alternative embodiments, the molecular markers described herein can distinguish HCC samples and chronic liver disease samples from samples from normal (healthy) subjects, or subjects with other cancers, or other diseases. Thus, the invention provides, in part, C3a molecules (e.g., C3a polypeptides and nucleic acid molecules, and complements, fragments, analogs, or variants thereof) that are differentially expressed in samples derived from subjects having HCC, when compared to non-HCC samples, on non-liver disease samples. The invention also provides, in part, beta-2 microglobulin or Chain A, human serum albumin mutant R218p, Accession No. 1HE3A, molecules (e.g., beta-2 microglobulin or Chain A, human serum albumin mutant R218p polypeptides and nucleic acid molecules, and fragments, analogs, or variants thereof) that are differentially expressed in samples from subjects having HCC, when compared to non-HCC samples.

C3a molecules may be detected alone or in combination with one or more other molecules (e.g., beta-2 microglobulin or Chain A, human serum albumin mutant R218p, Accession No. 1HK3A). In some embodiments, C3a molecules may be detected in combination with one or more markers for HCC such as alpha fetoprotein (10), cytokeratin 19, des-gamma-carboxy prothrombin (DCP), serum gamma-glutamyl transferase, platelet-derived endothelial growth factor, and insulin-like growth factor binding proteins. In some embodiments, an "expression profile" may be generated by simultaneously evaluating the expression level of a C3a, beta-2 microglobulin, and/or Chain A, human serum albumin mutant R218p molecule, in combination with any one or more of alpha fetoprotein, cytokeratin 19, des-gamma-carboxy prothrombin (DCP), serum gamma-glutamyl transferase, platelet-derived endothelial growth factor, and insulin-like growth factor binding protein molecules, or other molecules correlated with a cancer.

The invention also provides, in alternative aspects, methods for detecting C3a, beta-2 microglobulin or Chain A, human serum albumin mutant R218p, either individually, or in combination with other molecules, for HCC diagnosis or prognosis (e.g., good or poor long term survival after surgery); to assess HCC progression or regression; to assess subjects at risk for HCC; to monitor subjects in clinical trials for HCC therapeutics; to assess the efficacy and/or toxicity of HCC therapeutics; and/or to identify candidate compounds for HCC therapy, with high predictive accuracy.

Various alternative embodiments and examples of the invention are described herein. These embodiments and examples are illustrative and should not be construed as limiting the scope of the invention.

Nucleic Acid Molecules, Polypeptides, And Test Compounds

Compounds according to the invention include, without limitation, molecules substantially identical to C3a polypeptides and nucleic acid molecules, beta-2 microglobulin polypeptides and nucleic acid molecules, or human serum albumin mutant R218p polypeptides and nucleic acid molecules, as described herein or known to a skilled person, as well as complements, analogs, fragments, and variants thereof. In some embodiments of the invention, compounds of the invention include antibodies or antibody fragments that specifically bind to C3a, beta-2 microglobulin, or Chain A, human serum albumin mutant R218p, polypeptides and nucleic acid molecules, as described herein or known to a skilled person, as well as complements, analogs, fragments, and variants thereof. An antibody "specifically binds" an antigen when it recognises and binds the antigen, for example, a C3a, beta-2 microglobulin, or Chain A, human serum albumin mutant R218p polypeptide or nucleic acid molecule, but does not substantially recognise and bind other reference molecules in a sample, for example, an antigen that is not substantially identical to C3a, beta-2 microglobulin, or Chain A, human serum albumin mutant R218p polypeptides and nucleic acid molecules. Such an antibody has, for example, an affinity for the antigen which is at least 5, 10, 100, 1000 or 10000 times greater than the affinity of the antibody for another reference molecule in a sample. In alternative embodiments of the invention, compounds of the invention may include polypeptides having substantially the same molecular weights as indicated in Table 3 when analyzed from HCC vs. non-HCC samples as described in the Examples herein. A "substantially identical" sequence is an amino acid or nucleotide sequence that differs from a reference sequence only by one or more conservative substitutions, as discussed herein, or by one or more non-conservative substitutions, deletions, or insertions located at positions of the sequence that do not destroy the biological function of the polypeptide or nucleic acid molecule, or that do not destroy the detectability (e.g., by hybridization or specific binding) of the polypeptide or nucleic acid molecule. Such a substantially identical sequence can be any value from 10% to 99%, or more generally at least 10%, 20%, 30%, 40%, 50, 55% or 60%, or at least 65%, 15%, 80%, 85%, 90%, or 95%, or as much as 96%, 91%, 98%, or 99% identical when optimally aligned at the amino acid or nucleotide level to the reference sequence (e.g., C3a, beta-2 microglobulin, or. Chain A, human serum albumin mutant R218p sequence as described herein) used for comparison using, for example, the Align Program (26) or CLUSTAL X (27-31). For polypeptides, the length of comparison sequences may be at least 2, 5, 10, or 15 amino acids, or at least 20, 25, or 30 amino acids. In alternative embodiments, the length of comparison sequences may be at least 35, 40, or 50 amino acids, or over 60, 80, or 100 amino acids. For nucleic acid molecules, the length of comparison sequences may be at least 5, 10, 15, 20, or 25 nucleotides, or at least 30, 40, or 50 nucleotides. In alternative embodiments, the length of comparison sequences may be at least 60, 70, 80, or 90 nucleotides, or over 100, 200, or 500 nucleotides. Sequence identity can be readily measured using publicly available sequence analysis software (e.g., Sequence Analysis Software Package of the Genetics Computer Group, University of Wisconsin Biotechnology Center, 1710 University Avenue, Madison, Wis. 53705, or BLAST software available from the National Library of Medicine, or as described herein). Examples of useful software include the programs Pile-up and PrettyBox. Such software matches similar sequences by assigning degrees of homology to various substitutions, deletions, substitutions, and other modifications.

Alternatively, or additionally, two nucleic acid sequences may be "substantially identical" if they hybridize under high stringency conditions. In some embodiments, high stringency conditions are, for example, conditions that allow hybridization comparable with the hybridization that occurs using a DNA probe of at least 500 nucleotides in length, in a buffer containing 0.5 M NaHPO₄, pH 7.2, 7% SDS, 1 mM EDTA, and 1% BSA (fraction V), at a temperature of 65⁰C, or a buffer containing 48% formamide, 4.8x SSC₅ 0.2 M Tris-Cl, pH 7.6, Ix Denhardt's solution, 10% dextran sulfate, and 0.1% SDS, at a temperature of 42⁰C. (These are typical conditions for high stringency northern or Southern hybridizations.) Hybridizations may be carried out over a period of about 20 to 30 minutes, or about 2 to 6 hours, or about 10 to 15 hours, or over 24 hours or more. High stringency hybridization is also relied upon for the success of numerous techniques routinely performed by molecular biologists, such as high stringency PCR, DNA sequencing, single strand conformational polymorphism analysis, and in situ hybridization. In contrast to northern and Southern hybridizations, these techniques are usually performed with relatively short probes (e.g., usually about 16 nucleotides or longer for PCR or sequencing and about 40 nucleotides or longer for in situ hybridization). The high stringency, conditions used in these techniques are well known to those skilled in the art of molecular biology, and examples of them can be found, for example, in Ausubel et al. (32), which is hereby incorporated by reference.

A "variant" is a nucleic acid molecule that is a recognized variation of a nucleic acid molecule or expression product thereof. Splice variants may be determined for example by using computer programs, e.g, BLAST. Allelic variants have in general a high percent identity to the nucleic acid molecule of interest. "Single nucleotide polymorphism" (SNP) refers to a change in a single base as a result of a substitution, insertion or deletion. The change may be a conservative change, such as a transition (purine for purine) or a non-conservative change, such as a transversion (purine to pyrimidine or vice versa) and may or may not result in a change in an encoded amino acid. A polypeptide variant may be encoded by a variant nucleic acid molecule. However a variant nucleic acid molecule may not necessarily encode a variant polypeptide (e.g., if there is no change in the encoded amino acid as a result of the nucleotide(s) change). A polypeptide variant may or may not be recognized to be associated with a disease.

An "analog" is a nucleic acid molecule or polypeptide that has been subjected to a chemical modification. Nucleic acid analogs can include substitution of a non-traditional base such as queosine or of an analog such as hypoxanthine, or other substitutions known in the art. Polypeptide analogs can include substitution of an amino acid analog, such as alpha amino acid analogs, beta homoamino acids, phenylpropionic acid analogs, phenylbutyric acid analogs, or other substitutions known in the art. Analogs in general retain the biological activities of the naturally occurring molecules but may confer advantages such as longer lifespan or enhanced activity.

By "complementary" or "complement" is meant that two nucleic acids, e.g., DNA or RNA, contain a sufficient number of nucleotides which are capable of forming Watson-Crick base pairs to produce a region of double-strandedness between the two nucleic acids. Thus, adenine in one strand of DNA or RNA pairs with thymine in an opposing complementary DNA strand or with uracil in an opposing complementary RNA strand. It will be understood that each nucleotide in a nucleic acid molecule need not form a matched Watson-Crick base pair with a nucleotide in an opposing complementary strand to form a duplex. A nucleic acid molecule is "complementary" to another nucleic acid molecule, or is a "complement" of that other nucleic acid molecule, if it hybridizes, under conditions of high stringency, with the second nucleic acid molecule. The "complement" of a C3a, beta-2 microglobulin or Chain A, human serum albumin mutant R218p, nucleic acid molecule may in some embodiments include a nucleic acid molecule that is complementary over the full length of the sequence of a C3a, beta-2 microglobulin or Chain A, human serum albumin mutant R218p, nucleic acid molecule.

A "fragment" may be any portion of a nucleic acid molecule or polypeptide as described herein that is capable of being differentially expressed or detected in an assay or screening method according to the invention. For example, a nucleic acid molecule fragment, e.g., a C3a nucleic acid molecule fragment or C3a precursor molecule fragment, a beta-2 microglobulin molecule fragment, or a Chain A, human serum albumin mutant R218p molecule fragment, may be any value between 8 and 2000 nucleotides in length, e.g., at least 8, 10, 15, 20, or 25 nucleotides in length, or at least 30, 40, 50, or 60 nucleotides in length, or over 100, 200, 500, or 1000 nucleotides in length. A polypeptide fragment, e.g., an immunogenic fragment of a C3a molecule or C3a precursor molecule, a beta-2 microglobulin molecule fragment, or a Chain A, human serum albumin mutant R218p molecule fragment, may be any value between 8 and 500 amino acids in length, e.g., at least 8, 10, 15, 20, or 25 amino acids in length, or at least 30, 40, 50, 60, 70, 80, 90, or 100 amino acids in length. In some embodiments, a C3a polypeptide fragment includes a 68 amino acid residue fragment of C3a after C-terminal truncation of 9 amino acids.

Various genes and nucleic acid sequences of the invention may be recombinant sequences. The term "recombinant" means that something has been recombined, so that when made in reference to a nucleic acid construct the term refers to a molecule that is comprised of nucleic acid sequences that are joined together or produced by means of molecular biological techniques. The term "recombinant" when made in reference to a protein or a polypeptide refers to a protein or polypeptide molecule which is expressed, using a recombinant nucleic acid construct created by means of molecular biological techniques. The term "recombinant" when made in reference to genetic composition refers to a gamete .or progeny with new combinations of alleles that did not occur in the parental genomes. Recombinant nucleic acid constructs may include a nucleotide sequence which is ligated to, or is manipulated to become ligated to, a nucleic acid sequence to which it is not ligated in nature, or to which it is ligated at a different location in nature. Referring to a nucleic acid construct as 'recombinant' therefore indicates that the nucleic acid molecule has been manipulated using genetic engineering, i.e. by human intervention. Recombinant nucleic acid constructs may for example be introduced into a host cell by transformation. Such recombinant nucleic acid constructs may include sequences derived from the same host cell species or from different host cell species, which have been isolated and reintroduced into cells of the host species. Recombinant nucleic acid construct sequences may become integrated into a host cell genome, either as a result of the original transformation of the host cells, or as the result of subsequent recombination and/or repair events.

As used herein, "heterologous" in reference to a nucleic acid or protein is a molecule that has been manipulated by human intervention so that it is located in a place other than the place in which it is naturally found. For example, a nucleic acid sequence from one species may be introduced into the genome of another species, or a nucleic acid sequence from one genomic locus may be moved to another genomic or extrachromasomal locus in the same species. A heterologous protein includes, for example, a protein expressed from a heterologous coding sequence or a protein expressed from a recombinant gene in a cell that would not naturally express the protein.

A compound is "substantially pure" when it is separated from the components that naturally accompany it. Typically, a compound is substantially pure when it is at least 10%, 20%, 30%, 40%, 50%, or 60%, more generally 70%, 75%, 80%, or 85%, or over 90%, 95%, or 99% by weight, of the total material in a sample. Thus, for example, a polypeptide that is chemically synthesized, produced by recombinant technology, isolated by known purification techniques, will be generally be substantially free from its naturally associated components. A substantially pure compound can be obtained, for example, by extraction from a natural source; by expression of a recombinant nucleic acid molecule encoding a polypeptide compound; or by chemical synthesis. Purity can be measured using any appropriate method such as column chromatography, gel electrophoresis, HPLC, etc. A nucleic acid molecule is substantially pure or "isolated" when it is not immediately contiguous with (i.e., covalently linked to) the coding sequences with which it is normally contiguous in the naturally occurring genome of the organism from which the DNA of the invention is derived. Therefore, an "isolated" gene or nucleic acid molecule is intended to mean a gene or nucleic acid molecule which is not flanked by nucleic acid molecules which normally (in nature) flank the gene or nucleic acid molecule (such as in genomic sequences) and/or has been completely or partially purified from other transcribed sequences (as in a cDNA or RNA library). For example, an isolated nucleic acid of the invention may be substantially isolated with respect to the complex cellular milieu in which it naturally occurs. In some instances, the isolated material will form part of a composition (for example, a crude extract containing other substances), buffer system or reagent mix. In other circumstance, the material may be purified to essential homogeneity, for example as determined by PAGE or column chromatography such as HPLC. The term therefore includes, e.g., a recombinant nucleic acid incorporated into a vector, such as an autonomously replicating plasmid or virus; or into the genomic DNA of a prokaryote or eukaryote, or which exists as a separate molecule (e.g., a cDNA or a genomic DNA fragment produced by PCR or restriction endonuclease treatment) independent of other sequences. It also includes a recombinant nucleic acid which is part of a hybrid gene encoding additional polypeptide sequences. Preferably, an isolated nucleic acid comprises at least about 40%, 50%, 60%, 70%, 80%, 90%, 95%, or 99% (on a molar basis) of all macromolecular species present. Thus, an isolated gene or nucleic acid molecule can include a gene or nucleic acid molecule which is synthesized chemically or by recombinant means. Recombinant DNA contained in a vector are included in the definition of "isolated" as used herein. Also, isolated nucleic acid molecules include recombinant DNA molecules in heterologous host cells, as well as partially or substantially purified DNA molecules in solution. In vivo and in vitro RNA transcripts of the DNA molecules of the present invention are also encompassed by "isolated" nucleic acid molecules. Such isolated nucleic acid molecules are useful in the manufacture of the encoded polypeptide, as probes for isolating homologous sequences (e.g., from other mammalian species), for gene mapping (e.g., by in situ hybridization with chromosomes), or for detecting expression of the gene in tissue (e.g., human tissue, such as peripheral blood), such as by Northern blot analysis.

Polypeptide compounds can be prepared by, for example, replacing, deleting, or inserting an amino acid residue at any position of a peptide or a peptide analog, for example, a peptide as described herein, with other conservative amino acid residues, i.e., residues having similar physical, biological, or chemical properties. It is well known in the art that some modifications and changes can be made in the structure of a polypeptide without substantially altering the biological function of that peptide, to obtain a biologically equivalent polypeptide. In one aspect of the invention, polypeptides of the present invention also extend to biologically equivalent peptides that differ from a portion of the sequence of the polypeptides of the present invention by conservative amino acid substitutions. As used herein, the term "conserved amino acid substitutions" refers to the substitution of one amino acid for another at a given location in the peptide, where the substitution can be made without substantial loss of the relevant function. In making such changes, substitutions of like amino acid residues can be made on the basis of relative similarity of side-chain substituents, for example, their size, charge, hydrophobicity, hydrophilicity, and the like, and such substitutions may be assayed for their effect on the function of the peptide by routine testing. Conservative changes can also include the substitution of a chemically derivatized moiety for a non-derivatized residue, by for example, reaction of a functional side group of an amino acid. Peptides or peptide analogs can be synthesized by standard chemical techniques, for example, by automated synthesis using solution or solid phase synthesis methodology. Automated peptide synthesizers are commercially available and use techniques well known in the art. Peptides and peptide analogs can also be prepared using recombinant DNA technology using standard methods such as those described in, for example, Sambrook, et al. (33) or Ausubel et al. (32). Computer programs such as LASERGENE software (DNASTAR, Madison Wis.), MACVECTOR software (Genetics Computer Group, Madison Wis.) and RasMol software (www.umass.edu/microbio/rasmol) may be used to determine which and how many amino acid residues in a particular portion of the protein may be substituted, inserted, or deleted without abolishing biological or immunological activity.

Compounds (e.g., C3a, beta-2 microglobulin or Chain A, human serum albumin mutant R218p molecules) according to the invention may be detectably labeled for example to facilitate use in diagnostic or other assays. A wide variety of detectable labels and conjugation techniques are known by those skilled in the art and may be used in various nucleic acid molecule and polypeptide assays. The nucleic acid molecules, proteins, antibodies and other compounds according to the invention may be labeled by joining them, either covalently or noncovalently, with a detectable label. By "detectably labeled" is meant any means for marking and identifying the presence of a molecule, e.g., an oligonucleotide probe or primer, a gene or fragment thereof, a cDNA molecule, or a polypeptide. Methods for detectably-labeling a molecule are well known in the art and include, without limitation, radioactive labeling (e.g., with an isotope such as ³²P or ³⁵S) and nonradioactive labelling such as, enzymatic labeling (for example, using horseradish peroxidase or alkaline phosphatase), chemiluminescent labeling, fluorescent labeling (for example, using fluorescein), bioluminescent labeling, or antibody detection of a ligand attached to the probe. Also included in this definition is a molecule that is detectably labeled by an indirect means, for example, a molecule that is bound with a first moiety (such as biotin) that is, in turn, bound to a second moiety that may be observed or assayed (such as fluorescein-labeled streptavidin). Labels also include digoxigenin, luciferases, and aequorin. Synthesis of labeled molecules performed by using labels such as ³²P- dCTP, Cy3-dCTP or Cy5-dCTP or ³⁵S-methionine. Compounds according to the invention may also be directly labeled by chemical conjugation to amines, thiols and other groups present in the molecules using reagents such as BIODIPY or FITC (Molecular Probes, Eugene, OR, USA). A detectable label may be detected and quantified using for example spectroscopic, immunological, biochemical, scintillation counting, or other techniques.

Monitoring changes in nucleic acid molecule or polypeptide e.g., gene or protein, expression may also be advantageous when screening candidate HCC therapeutics. Often candidate compounds are screened and prescreened for the ability to interact with a major target without regard to other effects they may have on cells or in the subject to be treated, such as toxicity, which prevent the development and use of the potential, compound. Thus, the methods of the invention may be used to identify candidate compounds suitable for HCC therapy.

In general, candidate or test compounds are identified from large libraries of both natural products or synthetic (or semi-synthetic) extracts or chemical libraries according to methods known in the art. Those skilled in the field of drug discovery and development will understand that the precise source of test extracts or compounds is not critical to the methods of the invention. Accordingly, virtually any number of chemical extracts or compounds can be screened using the exemplary methods described herein. Examples of such extracts or compounds include, but are not limited to, plant-, fungal-, prokaryotic- or animal-based extracts, fermentation broths, and synthetic compounds, as well as modification of existing compounds. Numerous methods are also available for generating random or directed synthesis (e.g., semi-synthesis or total synthesis) of any number of chemical compounds, including, but not limited to, saccharide-, lipid-, peptide-, and nucleic acid-based compounds. Synthetic compound libraries are commercially available. Alternatively, libraries of natural compounds in the form of bacterial, fungal, plant, and animal extracts are commercially available from a number of sources, including Biotics (Sussex, UK), Xenova (Slough, UK), Harbor Branch Oceanographic Institute (Ft. Pierce, FL, USA), and PharmaMar, MA, USA. Furthermore, if desired, any library or compound is readily modified using standard chemical, physical, or biochemical methods. Candidate compounds useful for treating HCC may also be identified by assessing variations in the expression of a C3a, beta-2 microglobulin, or Chain A, human serum albumin mutant R218ρ molecule, prior to and after contacting HCC cells or tissues with candidate pharmacological agents for the treatment of HCC. The cells may be grown in culture (e.g. from a HCC cell line) or may be obtained from a subject, (e.g. in a clinical trial of candidate pharmaceutical agents to treat HCC). Alterations in C3a, beta-2 microglobulin, or Chain A, human serum albumin mutant R218p molecule expression in HCC cells or -tissues tested before and after contact with a candidate pharmacological agent to treat HCC, indicate progression, regression, or stasis of the HCC thereby indicating efficacy of candidate agents and concomitant identification of candidate compounds for therapeutic use in HCC. Candidate compounds may also be screened for toxicity, specificity, etc.

When a crude extract is found to modulate expression levels of any of the nucleic acid molecules or polypeptides of the invention, further fractionation of the positive lead extract is necessary to isolate chemical constituents responsible for the observed effect. Thus, the goal of the extraction, fractionation, and purification process is the careful characterization and identification of a chemical entity within the crude extract having the modulatory activities. The same assays described herein for the detection of activities in mixtures of compounds can be used to purify the active component and to test derivatives thereof. Methods of fractionation and purification of such heterogeneous extracts are known in the art. If desired, compounds shown to be useful agents for treatment are chemically modified according to methods known in the art.

In one aspect, the invention provides nucleic acid molecule or polypeptide arrays including C3a, beta-2 microglobulin, or Chain A, human serum albumin mutant R218ρ molecules. In some embodiments, the invention provides nucleic acid molecule or polypeptide arrays including C3a, in combination with one or more, or as many as all, of beta-2 microglobulin or Chain A, human serum albumin mutant R218p, alpha fetoprotein, cytokeratin 19, des-gamma-carboxy prothrombin (DCP), serum gamma- glutamyl transferase, platelet-derived endothelial growth factor, and insulin-like growth factor binding molecules.

Arrays refer to ordered arrangements of at least two nucleic acid molecules or polypeptides on a substrate, which can be any rigid or semi-iigid support to which two nucleic acid molecules or polypeptides may be attached. In some embodiments, a substrate may be a liquid medium. Substrates include membranes, filters, chips, slides, wafers, fibers, beads, gels, capillaries, plates, polymers, and microparticles etc. Because the nucleic acid molecules or polypeptides are located at specified locations on the substrate, the hybridization or binding patterns and intensities create a unique expression profile, which can be interpreted in terms of expression levels of particular genes and can be correlated with HCC progression, regression, therapy, or can be used to screen test or candidate compounds, etc.

High density nucleic acid or polypeptide arrays are also referred to as "microarrays," and may for example be used to monitor the presence or level of expression of a large number of genes or polypeptides or for detecting sequence variations, mutations and polymorphisms. Arrays and microarrays generally require a solid support (for example, nylon, glass, ceramic, plastic, silica, aluminosilicates, borosilicates, metal oxides such as aluminum and nickel oxide, various clays, nitrocellulose, etc.) to which the nucleic acid molecules or polypeptides are attached in a specified 2-dimensional arrangement, such that the pattern of hybridization or binding to a probe is easily determinable. In some embodiments, at least one of the nucleic acid molecules or polypeptides is a control, standard, or reference molecule, such as a housekeeping gene or portion thereof (e.g., PBGD, GAPDH), that may assist in the normalization of expression levels or assist in the determining of nucleic acid quality and binding characteristics; reagent quality and effectiveness; hybridization success; analysis thresholds and success, etc.

Nucleic acid molecules or polypeptide probes may be derived from compounds as described herein for example C3a, beta-2 microglobulin, or Chain A, human serum albumin mutant R218p molecules and the compositions of the invention may be used as elements on a microarray to analyze expression profiles. In some embodiments, more than one probe derived from compounds as described herein for example C3a, beta-2 microglobulin, or Chain A, human serum albumin mutant R218p molecules may be used, for example, probes including overlapping sequences (i.e., probes having common sequences of for example 4 or more contiguous amino acids or nucleotide bases) or probes directed to different sections of the C3a, beta-2 microglobulin, or Chain A, human serum albumin mutant R218p sequences. For the purpose of such arrays, "nucleic acid molecules" may include any polymer or oligomer of nucleosides or nucleotides (polynucleotides or oligonucleotides), which include pyrimidine and purine bases, preferably cytosine, thymine, and uracil, and adenine and guanine, respectively. In some embodiments, the microarray substrate, may be coated with a compound to enhance synthesis of the nucleic acid molecule on the substrate as disclosed in, for example, U.S. Pat. No. 4,458,066. In some embodiments, probes may be synthesized directly on the substrate in a predetermined ordered arrangement. Methods for storing, querying and analyzing microarray data have for example been disclosed in, for example, United States Patent No. 6,484,183; United States Patent No. 6,188,783; and Holloway, A.J., 2002; each of which is incorporated herein by reference.

In an alternative aspect, the invention provides nucleic acid or polypeptide microarrays including C3a, in combination with one or more, or as many as all, of beta-2 microglobulin, Chain A, human serum albumin mutant R218p, alpha fetoprotein, cytokeratin 19, des-gamma-carboxy prothrombin (DCP), serum gamma-glutamyl transferase, platelet-derived endothelial growth factor, and insulin-like growth factor binding molecules or other selected nucleic acid or polypeptide sequences correlated with HCC. The number of sequences in the microarray may for example be any integer between 2 and 1 x 10⁵, such as at least 10², 10³, 10⁴, or 10⁵. The size of the sequences may vary depending on the intended use, and can be determined by a skilled person. For example, the nucleic acid sequences may range from 15 to 5000 bases or more, or any integer between this range.

In an alternative aspect of the invention, libraries may be constructed of bacterial strains each of which bears a plasmid expressing a different C3a, beta-2 microglobulin, or Chain A, human serum albumin mutant R218p nucleic acid molecule, under control of an inducible promoter. Open reading frames are amplified using polymerase chain reaction and cloned into a vector that enables their expression as for example N-terminal his- tagged polypeptides. These amplicons are also used to construct hybridization micro arrays and enable targeted gene disruption, reducing expenses. A suitable expression host (e.g. E. coli) is selected, and genes encoding particular biochemical activities are identified by screening arrayed pools of his-tagged proteins.

The invention also provides databases including the nucleic acid and polypeptide sequences described herein, as well as gene expression information in various cancerous and non-cancerous liver and liver cell line samples. Such databases may be used to access information that may aid in diagnosis, prognosis, or other HCC-related methods of the invention. A database as used herein includes any electronic form of the compounds (e.g., nucleic acid and polypeptide sequences) of the invention, and information regarding these compounds, and includes computer readable media and any suitable form for storing the information.

The invention also provides kits including for example one or more of the nucleic acid molecules or polypeptides of the invention (or complements, analogs, variants, or fragments thereof), an appropriate buffer, appropriate reagents for detection, and appropriate controls. For example, a kit may include probes or primers (which may or may not be detectably labeled) suitable for hybridization or amplification, or may include antibodies or ligands suitable for specific binding. A kit may also include written or electronic instructions.

Assays

Compounds, compositions, and reagents (e.g., arrays or microarrays) according to the invention may be used to detect and/or quantify differential expression of C3a, beta-2 microglobulin and/or Chain A, human serum albumin mutant R218p molecules. In some embodiments, C3a molecules, in combination with one or more, or as many as all, of beta-2 microglobulin, Chain A, human serum albumin mutant R218p, alpha fetoprotein, cytokeratin 19, des-gamma-carboxy prothrombin (DCP), serum gamma-glutamyl transferase, platelet-derived endothelial growth factor, or insulin-like growth factor binding molecules, may be detected. Down-regulation or under-expression or absence may be indicated by lack of a positive response to a standard assay or test, while upregulation or overexpression or presence may be indicated by an enhanced response to a standard assay or test. Over or under-expression or absence or presence may also be determined by a comparison of expression levels C3a, beta-2 microglobulin or Chain A, human serum albumin mutant R218p, molecules in a test sample compared to a control sample obtained for example from one or more individuals known not to have HCC, or compared to an established standard. The control sample may be from an individual or from a population pool that is determined to not have HCC, or from an individual or a population pool that is not at risk for HCC.

Polypeptide levels may be detected and/or quantified directly, or may be detected indirectly by for example detecting and/or quantified related nucleic acid molecules such as a mRNA transcript corresponding to the polypeptide. Multiple nucleic acid molecules and polypeptides may be assayed separately or simultaneously.

Expression levels of nucleic acid molecules such as mRNA transcripts or related DNA or RNA products may be detected and/or quantified using hybridization techniques such as Northern blots, in situ hybridization e.g., to HCC and non-HCC tissue arrays; amplification techniques such as RT-PCR, differential display, or other PCR techniques; mass spectrometry techniques such as MALDI or SELDI; SAGE techniques; array or chip technology, etc.

Expression levels of polypeptides may be detected and/or quantified using immunoassays such as ELISA, immunoprecipitation, Western blots, in situ imaging, or immunohistochemistry; chromatographic techniques such as affinity chromatography; mass spectrometry techniques such as MALDI, SELDI, liquid or gas chromatography- mass spectrometry (LC-MS, HPLC-MS, or GC-MS), tandem mass spectrometry; protein separation techniques such as two dimensional gel electrophoresis; nuclear magnetic resonance techniques, array or chip technology, etc. For example, polypeptide levels may be quantified by contacting a sample with an antibody or antibody fragment that specifically binds a C3a, beta-2 microglobulin or Chain A, human serum albumin mutant R218p molecule, detecting whether the antibody binds a molecule in the sample, and quantifying the level of C3a, beta-2 microglobulin or Chain A, human serum albumin mutant R218p molecule bound. The sample may be a test sample or a control sample, and the amount of C3a, beta-2 microglobulin or Chain A, human serum albumin mutant R218p molecule bound in the test sample may be compared to the C3a, beta-2 microglobulin or Chain A, human serum albumin mutant R218p molecule bound in the control sample. The antibody may be detectably labeled to facilitate quantification.

In some embodiments, to provide a basis for the detection of HCC, a non-HCC or standard C3a, beta-2 microglobulin or Chain A, human serum albumin mutant R218p nucleic acid molecule, e.g., mRNA, or polypeptide expression profile may be established. This may be accomplished by combining a sample obtained from normal or non-HCC subjects or from non-cancerous tissue from a subject with HCC, with for example a probe under conditions for hybridization or amplification. Standard hybridization may be quantified by comparing the values obtained using non-HCC subjects or non-cancerous tissue with values from an experiment in which a known amount of a substantially purified C3a, beta-2 microglobulin or Chain A, human serum albumin mutant R218p sequence is used. Standard values obtained in this manner may be compared with values obtained from samples from patients who are symptomatic for or diagnosed with HCC. Deviation from standard values toward those associated with HCC is used to diagnose HCC.

Microarrays may be used to examine the expression of all the genes in a tissue or cell such as a liver cell or a HCC cell or from a serum sample. Thus, C3a, beta-2 microglobulin, or Chain A, human serum albumin mutant R218p nucleic acid molecules may be attached to a solid support, hybridized with single stranded detectably-labeled cDNAs (corresponding to a "complementary" orientation), and quantified using an appropriate method such that a signal is detected at each location at which hybridization has taken place. The intensity of the signal would then reflect the amount of gene expression. Similarly, protein microarrays may be used according to methods known in the art. Comparison of results from different cells or tissue, for example, hepatocellular carcinoma cells or tissue, hepatitis virus infected cells or tissue, non-tumor cells or tissue, normal cells or tissue, cirrhotic liver cells or tissue, or any combination thereof would elucidate differing levels of expression of specified genes from the different sources.

Such assays, as well as other assays known in the art, may be used to diagnose HCC or to determine the prognosis for HCC in a subject. For example, diagnostic and prognostic assays may be used to detect and compare polypeptide or nucleic acid molecule expression from a test sample from a subject, for example, a subject suspected of having HCC or at risk for having HCC, to a control sample or standard. By analyzing changes, e.g., under-expression, in patterns of C3a, beta-2 microglobulin and/or Chain A, human serum albumin mutant R218p molecule, e.g., mRNA or polypeptide, expression, HCC can be diagnosed at earlier stages, for example, before the subject is symptomatic. Similarly, prognostic evaluations may be performed to assist in determination of optimal treatment regimens.

Qualitative or quantitative methods for detection and comparison are known in the art, and any suitable method may be used. Generally, an absence or decrease in the expression level of a C3a molecule in a test sample relative to a standard, or a presence or¹ increase in the expression level of a beta 2 microglobulin molecule or a Chain A, human serum albumin mutant R218p in a test sample relative to a standard, will be indicative of a HCC diagnosis or of a poor HCC prognosis. Conversely, a presence or increase in the expression level of a C3a molecule in a test sample relative to a standard, or an absence or decrease in the expression level of a beta 2 microglobulin molecule or a Chain A, human serum albumin mutant R218ρ in a test sample relative to a standard, will be indicative of a non-HCC diagnosis or of a good prognosis.

The diagnostic assays may also include the detection of expression levels of other molecules e.g., HCC markers, tumor markers, cirrhosis markers, hepatitis virus infection markers or compounds that bind hepatitis virus molecules, at levels generally accepted to be diagnostic. The diagnostic assays may be used in combination with existing HCC diagnostic methods such as biopsy including radiological biopsy by means of a radiological scan, laparoscopy, or open surgical biopsy, liver autopsy, histologic staining, microscopic analysis, immunoassay, contrast ultrasound, computed tomography, magnetic resonance imaging, hepatic arteriography, pre-existing hepatitis virus infection or cirrhosis, exposure to aflatoxin and other risk factors for HCC, age, functional status, etc.

The prognostic assays may be used in combination with other prognostic indicia such as determination of the anatomy or aggressiveness of a HCC. For example, size of the tumor or largest lesion, number of lesions, whether the lesions are unilobar or bilobar, vascular invasion, metastasis, histology, or mutation rate (e.g., by gene array) may be determined.

Assays as described herein or known in the art may also be used to monitor the progression or regression of HCC in a subject. For example, once the presence of HCC is diagnosed or established in a subject, the assays may be repeated on a regular basis to determine if the level of expression of C3a, beta 2 microglobulin, and/or Chain A, human serum albumin mutant R218ρ molecules in the subject begins to approximate that which is observed in a non-HCC subject. Thus, compounds (e.g., C3a, beta-2 microglobulin or Chain A, human serum albumin mutant R218p molecules) according to the invention may be used to monitor the progression or regression of HCC in an individual subject over a period of time.

Progression or regression of HCC may be determined by comparison of two or more different HCC samples taken at multiple different times from a subject (e.g., at least 2, 3, 4, or 5 or more time points) over the course of days to months. For example, progression or regression may be evaluated by assessing expression of sets of two or more, or as many as all, of C3a, beta-2 microglobulin, or human serum albumin mutant R218p nucleic acid molecules in a HCC tissue sample from a subject before, during, and following treatment for HCC. In some embodiments, progression or regression may be evaluated by assessing expression of sets of C3a molecules, in combination with one or more, or as many as all, of beta-2 microglobulin, Chain A, human serum albumin mutant R2l8p, alpha fetoprotein, cytokeratin 19, des-gamma-carboxy prothrombin (DCP), serum gamma-glutamyl transferase, platelet-derived endothelial growth factor, and insulin-like growth factor binding molecules.

The results obtained from successive assays i.e., assays conducted on samples obtained at multiple time points may be used to monitor the course of HCC over a period ranging from several days to months. Generally, an absence or decrease in the expression level of a C3a molecule in an earlier test sample relative to a subsequent or later test sample, or a presence or increase in the expression level of a beta 2 microglobulin molecule or a Chain A, human serum albumin mutant R218p in an earlier test sample relative to a subsequent or later test sample, will indicate that the HCC is progressing and that therapeutic interventions, if any, are not proving efficacious. Conversely, presence or increase in the expression level of a C3a molecule in an earlier test sample (e.g., the first diagnostic sample) relative to a subsequent or later test sample, or an absence or decrease in the expression level of a beta 2 microglobulin molecule or a Chain A, human serum albumin mutant R218p in an earlier test sample relative to a subsequent or later test sample, will indicate that the HCC is regressing

The assays of the invention may be used to monitor the efficacy of a HCC therapy in for example animal models, in clinical trials, or to monitor the treatment of an . individual patient or groups of patients, for example before, during, and following treatment for HCC. The assays may be conducted on samples obtained at multiple time points over the course of days, months, or years. The assays may thus be used to determine whether therapeutic interventions, if any, are efficacious. The therapy or therapies may be similarly monitored over the course of days, months, or years. For therapies with known side effects, compounds according to the invention may be employed to improve the therapeutic regimen. For example, dosages that causes changes in C3a, beta-2 microglobulin or Chain A, human serum albumin mutant R218p mRNA or polypeptide expression levels that represent efficacious treatment may be determined, and expression profiles associated with the onset of undesirable side effects may be avoided. This approach may be more sensitive and rapid than waiting for the subject to show inadequate improvement, or to manifest side effects, before altering the course of treatment. In another aspect of the invention, pre- and post-treatment alterations in expression of C3a, beta-2 microglobulin or Chain A, human serum albumin mutant R218p polypeptide or nucleic acid molecules in HCC cells or tissues may be used to assess treatment parameters including, but not limited to: dosage, method of administration, timing of administration, and combination with other known treatments for HCC.

The assays of the invention may also be used to select a subject for a HCC therapy. For example, a more aggressive therapy may be selected for a subject with rapid progressing HCC while more conservative therapies may be selected for a subject in which HCC is static or regressing. HCC therapies include total hepatectomy with concomitant liver transplantation, partial hepatectomy, tumor ablation e.g., by alcohol (e.g., ethanol) injection, radiofrequency ablation, cryoablation, transarterial chemoembolization, chemotherapy, radiotherapy, such as proton beam radiotherapy, carbon ion radiotherapy, intensity modulated radiotherapy, etc. It will be understood that selection of any particular therapy is at the discretion of the medical practitioner, and that any of the results of the assays described herein may be used by a medical practitioner in determining optimal treatment regimens for example after diagnosis or after therapeutic intervention. For example, the assay results may assist in determining whether to implement radical or less radical treatment protocols, or whether to continue or discontinue treatment protocols. In some cases, it may be advisable to use potent therapies even though they may have deleterious side effects.

In some aspects, any one or more of the compounds provided herein may be used in therapeutic applications. For example, C3a, beta-2 microglobulin or Chain A, human serum albumin mutant R218p compounds provided herein may be used as therapeutic targets for the identification of agents, that modulate their expression levels and/or activity, that may be used to treat HCC.

In alternative aspects, compounds according to the invention may be used in assays such as those described herein to screen for candidate compounds useful for the treatment of HCC. For example, candidate compounds that increase or decrease C3a, beta-2 microglobulin or Chain A, human serum albumin mutant R218p molecule expression levels may be potential HCC therapeutic compounds. In alternative aspects, compounds that interfere with the ability of a MMP-9 polypeptide to degrade a C3a polypeptide may be used in assays such as those described herein to screen for candidate compounds useful for the treatment of HCC. Compounds identified as being of potential therapeutic, prophylactic, diagnostic, or other value may be subsequently analyzed using HCC cell lines (e.g., SK-Hepl, SMMC-7721, FHCC-98, or as commercially available from ATCC, Manassas, VA, USA) or animal models for HCC (e.g., ground squirrels, ducks or woodchucks exposed to HBV; rodent models of transplanted xenogeneic hepatocytes (see, for example, Dandri, M. et. al. (34); Petersen, J. et. al. (35)); rodent models of diethylnitrosamine administration, etc., or as commercially available from The Jackson Laboratories, Bar Harbor, ME, USA.

In alternative aspects, a HCC therapeutic compound includes a compound capable of inhibiting or decreasing the expression level or activity of a MMP-9 molecule, e.g., a MMP-9 inhibitor. Accordingly, the invention provides methods for treating or preventing HCC by administering a MMP-9 inhibitor to a subject in need thereof, or . provides use of a MMP-9 inhibitor for the preparation of a medicament for treating or preventing HCC.

Pharmaceutical & Veterinary Compositions, Dosages, And Administration

Compounds for treating HCC (e.g., a MMP-9 inhibitor) or can be provided alone or in combination with other compounds (for example, nucleic acid molecules, small molecules, peptides, or peptide analogues), in the presence of a liposome, an adjuvant, or any pharmaceutically acceptable carrier, in a form suitable for administration to mammals, for example, humans, cattle, sheep, etc. If desired, treatment may be combined with more traditional and existing therapies for HCC. By "treatment" is meant treating and preventing and accordingly, both prophylactic and therapeutic uses are within the scope of the invention. Compounds for treating HCC may be provided chronically or intermittently. "Chronic" administration refers to administration of the compound(s) in a continuous mode as opposed to an acute mode, so as to maintain the initial therapeutic effect (activity) for an extended period of time. "Intermittent" administration is treatment that is not consecutively done without interruption, but rather is cyclic in nature Conventional pharmaceutical practice may be employed to provide suitable formulations or compositions to administer the compounds to subjects suffering from or presymptomatic for HCC. Any appropriate route of administration may be employed, for example, parenteral, intravenous, subcutaneous, intramuscular, intracranial, intraorbital, ophthalmic, intraventricular, intracapsular, intraspinal, intrathecal, intracisternal, intraperitoneal, intranasal, aerosol, topical, or oral administration. Therapeutic formulations may be in the form of liquid solutions or suspensions; for oral administration, formulations may be in the form of tablets or capsules; and for intranasal formulations, in the form of powders, nasal drops, or aerosols.

Methods well known in the art for making formulations are found in, for example, "Remington's Pharmaceutical Sciences" (19^th edition), ed. A. Gennaro, 1995, Mack Publishing Company, Easton, Pa. Formulations for parenteral administration may, for example, contain excipients, sterile water, or saline, polyalkylene glycols such as polyethylene glycol, oils of vegetable origin, or hydrogenated napthalenes. Biocompatible, biodegradable lactide polymer, lactide/glycolide copolymer, or polyoxyethylene-polyoxypropylene copolymers may be used to control the release of the compounds. Other potentially useful parenteral delivery systems for include ethylene-vinyl acetate copolymer particles, osmotic pumps, implantable infusion systems, and liposomes. Formulations for inhalation may contain excipients, for example, lactose, or may be aqueous solutions containing, for example, ρolyoxyethylene-9-lauryl ether, glycocholate and deoxycholate, or may be oily solutions for administration in the form of nasal drops, or as a gel. For therapeutic or prophylactic compositions, the compounds are administered to an individual in an amount sufficient to stop or slow HCC, or to prevent or reduce the degradation of a C3a polypeptide or nucleic acid molecule.

An "effective amount" of a compound for treating HCC includes a therapeutically effective amount or a prophylactically effective amount. A "therapeutically effective amount" refers to an amount effective, at dosages and for periods of time necessary, to achieve the desired therapeutic result, such as treatment of HCC or prevention or reduction of degradation of a C3a polypeptide or nucleic acid molecule. A therapeutically effective amount of a compound may vary according to factors such as the disease state, age, sex, and weight of the individual, and the ability of the compound to elicit a desired response in the individual. Dosage regimens may be adjusted to provide the optimum therapeutic response. A therapeutically effective amount is also one in which any toxic or detrimental effects of the compound are outweighed by the therapeutically beneficial effects. A "prophylactically effective amount" refers to an amount effective, at dosages and for periods of time necessary, to achieve the desired prophylactic result, such as prevention of HCC or prevention or reduction of degradation of a C3a polypeptide or nucleic acid molecule. Typically, a prophylactic dose is used in subjects prior to or at an earlier stage of disease, so that a prophylactically effective amount may be less than a therapeutically effective amount. A preferred range for therapeutically or prophylactically effective amounts of a compound may be any integer from 0.1 nM-0.1M, 0.1 nM-0.05M, 0.05 nM-15 M or 0.01 nM- lOμM.

It is to be noted that dosage values may vary with the severity of the condition to be alleviated. For any particular subject, specific dosage regimens may be adjusted over time according to the individual need and the professional judgement of the person administering or supervising the administration of the compositions. Dosage ranges set forth herein are exemplary only and do not limit the dosage ranges that may be selected by medical practitioners. The amount of active compound(s) in the composition may vary according to factors such as the disease state, age, sex, and weight of the individual. Dosage regimens may be adjusted to provide the optimum therapeutic response. For example, a single bolus may be administered, several divided doses may be administered over time or the dose may be proportionally reduced or increased as indicated by the exigencies of the therapeutic situation. It may be advantageous to formulate parenteral compositions in dosage unit form for ease of administration and uniformity of dosage. In general, compounds of the invention should be used without causing substantial toxicity. Toxicity of the compounds of the invention can be determined using standard techniques, for example, by testing in cell cultures or experimental animals and determining the therapeutic index, i.e., the ratio between the LD50 (the dose lethal to 50% of the population) and the LDlOO (the dose lethal to 100% of the population). In some circumstances however, such as in severe disease conditions, it may be necessary to administer substantial excesses of the compositions.

EXAMPLES Samples and Methods Sample Collection and Handling

Institutional research and ethics committee approval for the study and informed consent from all patients were obtained. Clinical data of HCC patients were as summarized in Table 1. Blood was drawn from patients with HCC (n=49, ten females and thirty nine males, median age: 53, age range: 24-71) secondary to HBV infection one day prior to curative surgery. Blood was also drawn from normal individuals (n=27, non-HBV carrier, twelve females and fifteen males, median age: 47, age range 25-69), benign hepatitis B virus carrier (HBV positive) patients (n=13, four females and nine males, median age: 43, age range: 22-64), nasopharyngeal carcinoma (NPC) patients (n =10, two females and eight males, median age: 45, age range: 30-72), primary colon cancer patients (n=3, two females and one male, median age: 52, age range: 31-79), and patients with ear-nose-throat symptoms only (n=6, two females and four males, median age: 41, age range: 16-61). All cancer patients had histological confirmation of their conditions. All samples were collected and processed in identical manner as follows. All samples were collected into plain collection tube without anticoagulant. The clotted samples were immediately spun at 1000 xg for 15 minutes at 4oC. The supernatant was withdrawn, snap-frozen in liquid nitrogen and then stored at -80oC for later use. Serum samples went through less than two freeze-thaw cycles of -80oC before SELDI analyses. Protein was quantified using microBCA assays (Pierce Inc. Rockford, IL) according to vendor's instruction with bovine serum albumin (BSA) as standard.

Table 1.Clinical data of HCC patients.

Serum protein profiling using SELDI technology

Normal (n=9), HBV positive (n=13), NPC (n=10), primary colon cancer (n=3), and HCC (n=49) serum samples were used in the discovery phase study. Four microliters of serum were diluted five-fold with phosphate buffer saline (PBS) pH 7.4 to 20 μl. Diluted sera were vortexed briefly for 20 sec and spun down at 13,000 x g at room temperature for 20 seconds before use. Copper II (Cu²⁺) ion-loaded immobilized metal ion affinity capture (IMAC3) ProteinChip Arrays were prepared according to the vendor's instruction (Ciphergen Biosystems Inc., Fremont, CA). The array surface was equilibrated twice with 5 μl of PBS pH 7.4 with 0.1 M NaCl. Five microliters of diluted serum were loaded onto Cu²⁺ surface for 30 min in a humidity chamber on a rotation shaker for.120 rpm at room temperature. Bound proteins were washed once with 5 μl of PBS pH 7.4 with 0.1 M NaCl and twice with HPLC water. Half microlitre of α-4-cyano- cinnamic acid (saturated in 50% acetonitrile and 0.1 % trifluoro acetic acid) were added twice to the samples before reading in PBS II mass reader. Each sample received 25 laser shots on average. Mass readers recorderd masses ranging from 1 to 20 kDa. The specific settings in the detection process were as follows: first pass signal/noise ratio at 5, second pass at 2, minimum peak threshold at 10%, and mass error at 0.2%. Averaged values of spectra from the same patients were fed into machine learning algorithms for feature selection and class prediction analyses. 8132 Da and 5904 Da protein peaks were calibrated internally using beta-endorphin (3456Da) and cytochrome C (12230 Da) as standards after the imidazole elution screening. The mass accuracy of the mass reader was 200 parts per million (ppm).

Data Formatting:

The data set consisted of 46 normal spectra, 179 HCC spectra, 52 HBV spectra, 40 NPC spectra, and 12 colon cancer spectra. We selected a subset of five spectra from each class in the data set and kept them for independent testing of the classification procedure. The testing step was repeated five times by using other sets of five spectra in each round. The remaining data set was used to discover the best features that could distinguish HCC samples from non-HCC samples. All mass spectra were normalized to have the same total count of ion current. Common protein clusters in Excel format were generated by using the Biomarker WizardTM software (Ciphergen Biosystems, Fremont, CA).

Preprocessing step

A histogram plot of the samples indicated the distribution skewed (Figure 2A). A natural log transformation of the data was performed (Figure 2B) to improve overall distribution of signal intensities within samples, and zero entries in the data set were replaced by a very small value (0.00001) so that the log-transformation could be made. A box plot of the data for every sample indicated considerable variation of the data (Figure 3A). A global mean normalization of the data was performed by subtracting the mean of the log transformed data from the measured signal values for every patient, to reduce variation of the averaged intensity of peak clusters among samples. This procedure made protein profiles of all patients comparable. (Figure 3B).

Feature Selection and Classification

We did not adopt the traditional split-sample training-and-testing scheme in view of the potential serious bias resulted from few normal samples. Instead, a leave-one-out cross validation (LOOCV) procedure was adopted in feature selection from a comparison of two classes of samples [23]. This could reduce the false positive discovery rate [24]. Duplicated spectra from one sample were removed from the whole data set at each round. Protein peaks in the remaining sample set were ranked according to the p-values after Wilcoxon rank-sum nonparametric test. For the first round of feature selection, two differentially expressed peak values in normal and HCC groups with the highest statistical differences (p = 0.00067) were selected, expressed peak values in normal and HCC groups with the highest statistical differences (p = 0.00067) were selected. The two selected features were used to predict the class of the spectra of the withheld sample. The next step was to withhold spectra from another sample. The whole process was repeated until all samples had been withheld once. This gave the overall sensitivity and specificity of the classification using two features. Progressive increase in the number of peak selected was conducted by adding one feature at a time until the highest prediction accuracy was obtained. Top ranking proteins in the classification algorithm were added until a good accuracy of prediction was attained. A non-supervised hierarchical clustering (14) of normal and HCC samples using nine protein peaks based on one minus Pearson's correlation metric and average linkage was plotted in Figure 4.

To minimize algorithm-based selection bias on the data set, we decided to discriminate samples using different algorithms such as diagonal linear discriminant analysis (DLDA), K-nearest neighbor (KNN), and support vector machine (SVM) with a Gaussian kernel (15). As SVM was used primarily as a binary classifier, we developed all possible binary classifiers to discriminate five categories of samples (normal, HCC, colon cancer, HBV, and NPC). Therefore, a total of ten different binary classifiers were made from comparing every class to the other four different classes. The appropriate support vector machine algorithm was trained by choosing the appropriate group of data samples. Hyper parameters of SVM were optimized using a grid search. The above binary classifiers were combined to make a multi-class predictor in the following manner: We applied all the ten different classifiers for every test sample. Each classifier gave a label to the sample. As every class was represented by four different classifiers, it should get the correct label four times. In fact that should be the only one label which obtained four votes. Hence the multi-class predictor used a voting scheme to predict the class. The trained scheme was subsequently used to test for accuracy of prediction on a set of 25 new samples (five per class). This test procedure was repeated five times with five other sets of independent samples. Testing this classifier on an independent data set gave a very good performance.

Multiple sample classification using classification and regression trees^' (CART)

A different discrimination approach such as classification and regression tree (CART) algorithm was used to find a consensus set of markers and to confirm the results of previous discrimination algorithms. A classification binary tree from the total data set was learned [16], No preprocessing step was employed for this purpose. All possible intensity values of every peak were tested to split subsets (nodes) of measurement into two descendent subsets. Each terminal subset was assigned a class label. The resulting partition tree of nodes corresponded to the classifier [25].

The robustness of the selected markers was tested by classifying the samples using CART algorithm. The features selected by CART also had a lot of overlap with the features selected by SVM based classification system. A classification binary tree was created from all the data sets (16). We did the normal preprocessing step as described herein. For every peak, we tried all possible intensity values that could split the groups of samples into two groups. The ideal splitting was measured by an impurity measure. The impurity of a node was a relative frequency of the classes in that node: i(t)=φ(p₁, p₂,..,pj) where the pj(j=1,... ,J) were the relative frequencies of the J different classes in that node. Ideally the impurity function should have the following properties:

1. Should be a maximum when the observations were distributed evenly over all classes

2. Should be a minimum when all observations belonged to a single class

3. Should be a symmetric function of p₁...p_J

The impurity function we used was the entropy and was defined as:

We defined the quality of a split s in node t as the reduction of impurity achieved. Ai(s, t)=i(t)-π(l)i(l)-π(r)i(r)

Where / was the left child of t, r was the right child of t. π(l). was the proportion of cases sent to the left and π(r) was the proportion of cases sent to right. The entropy from each intensity point was computed to check the homogeneity of two groups after splitting into two child nodes. The children nodes went through the same splitting rules as above until a decision tree was established that could segregate all samples into different groups. SELDI guidance of protein purification on IMAC-copper surface

Diluted normal serum (NI) and HCC serum (S103) and IMAC-Cu²⁺ arrays were prepared as described herein. Diluted sera were spotted onto eight spots of IMAC-Cu²⁺ arrays. After binding in a humidity chamber for 30 min at room temperature, protein solutions were removed using Kimwipe™ paper. 5 μl of PBS pH7.4 and 0.1M NaCl with 0, 5, 10, 25, 50, 100, 200, or 500 mM of imidazole were added to each spot to wash away non-bound proteins. Solutions on spot were removed by using Kimwipe™ papers followed by two washes of 5 μl of HPLC water. Two rounds of 0.5 μl of α-4-cyano- cinnamic acid (saturated in 50% acetonitrile and 0.1 % trifluoroacetic acid) were added to samples before reading in PBS II mass reader. An average of 25 laser shots was applied to each 'sample. Masses ranging from 1 to 20 kDa were recorded. Target peaks intensity reflected competition of imidazole with target proteins onto the copper surface.

Purification of differentially expressed serum proteins

All the. steps were done in room temperature except otherwise stated. Fifty μl of diluted serum was mixed with 50 μl of PBS with 0.1 M NaCl and 1% triton X-100. The samples were vortexed for 20 sec in room temperature. K30 gel filtration spin column (Ciphergen Biosystems Inc. Fremont, CA) was equilibrated with 1 ml of PBS pH7.4 with 0.1M NaCl overnight at room temperature. Diluted serum (25 μl ) was applied to K-30 spin column. The presence of low molecular weight target protein was checked by spotting 1 μl of eluent onto an IMAC-Cu²⁺ array. Fractions with target peaks were pooled and further purified on a Cu²⁺ column. Ni-NTA His bind resin was from Novagen (Madison, WI). One ml of Ni-NTA His bind resin slurry (Novagen, Madison, WI) was poured into 0.4 x 8 cm Poly-prep column (Bio-Rad, Hercules, CA). The resin was first stripped off bound nickel by washing with the stripping buffer. After two washes with 3 ml of HPLC water, the resin was reloaded with Cu²⁺ ion by adding 2 ml of 100 mM CuSO₄ twice. The column was further equilibrated with two ml of PBS pH 7.4 with 0.1 M NaCl (binding buffer). Pooled fractions of K-30 eluent were loaded onto Cu²⁺-NTA column. Bound proteins were first washed with 1.5 ml of binding buffer. Eluents were collected for each 0.5 ml fraction after each wash. The second wash consisted of 1.5 ml binding buffer with 5 mM imidazole. The third wash was done by using 1.5 ml of binding buffer with 25 mM imidazole. The final wash was performed by rinsing the column with 1.5 ml binding buffer with 100 mM imidazole. Bound copper II ion and residual proteins were stripped by using binding buffer with 0.1M EDTA. Buffers in all eluent fractions were changed by spinning in YM3 spin membrane with addition of half the volume of 0.5x PBS. Protein quantitation of eluent were done by using Nano-Orange (Molecular Probes, Eugene, OR). Aliquots of eluent were subject to 4-12% gradient SDS-PAGE (Novex, Carlsbad, CA) analyses. The same tube of purified protein eluents were also subjected to SELDI analyses to correlate molecular weight and purity of protein bands in SDS-PAGE and protein peaks in mass reader.

Identification of purified copper binding low molecular weight proteins

Protein bands were revealed by using SilverQuest™ (Invitrogen Life Technologies, Carlsbad, CA). Target protein bands were excised and destained according to the vendor's instruction. Gel plugs were sent to Proteomic Research Services for trypsin digestion and peptide sequencing. Samples were subjected to proteolytic digestion on a ProGest workstation as follows: ammonium bicarbonate was added to each sample. Reduction was performed with DTT. Samples were allowed to cool to room temperature. Alkylation was performed with iodoacetamide. Samples were incubated at 37oC overnight in the presence of trypsin. Formic acid was added to stop the reaction. Samples that proved inconclusive following MALDI/MS were analyzed by nanoLC/MS/MS on a Micromass q-TOF2 mass spectrometer. 15 μL of hydrolysate were processed on a 15 μm Clδ column at a flow rate of 200 nL/min. MS/MS data were searched using a local version of Mascot software (Matrix Science Inc, Boston, MA). The parameters for all LC/MS/MS (Mascot) searches were as follows: type of search: MS/MS ion search; enzyme: Trypsin; fixed modification: carbamidomethyl (C); variable modifications: oxidation (M), acetyl (N-term), pyro-glu (N-term-Q), pyro-glu (N-term E); mass values: monoisotopic; protein mass: unrestricted; peptide mass tolerance: ±100 ppm; fragment mass tolerance: ±0.1 Da; max missed cleavages: 1. Western blotting

Sample classes and size were as follows: Normal (n = 28), ear-nose-throat symptoms (n = 8), hepatitis B virus-related HCC (n = 49), primary colorectal carcinoma (n = 3), benign Hepatitis B virus carrier (n = 13), and nasopharyngeal carcinoma (n = 12). Twenty five micrograms of serum total proteins of each sample was mixed with Laemmli loading buffer. Diluted sera or pure C3a proteins standards (2 ng to 50 ng, Calbiochem, La Jolla, CA) were boiled at for 5 min. at 95⁰C, followed by rapid cooling in ice. SDS PAGE was run by using 15 % SDS gel with IxMES buffer. Protein bands were transferred to PVDF membranes (Amersham Biosciences, Pitscataway, NJ) by using semi-dry blotting apparatus at 15V for 1 hr. Polyclonal antiserum against C3a (Calbiochem, La JoIIa, CA) was applied at 1:1000 dilution. Secondary anti-rabbit IgG antibody conjugated with horse radish peroxidase (Amersham Biosciences, Pitscataway, NJ) was applied at 1:10000 dilutions. Similar protocol was used for Western blot analysis of beta-2 microglobulin. Positive band signal was developed by using ECL chemiluminescent kit (Amersham Biosciences, Pitscataway, NJ). Positive signals were captured on X-ray films (Fuji Super Rx, Tokyo, Japan) and the images were scanned using image scanner (Amersham Bioscience, Pitscataway, NJ). Samples were allowed to run through the whole process two times and average values were taken. Volumes of specific 8.1 kDa band signal were analyzed by using sigma scan software (Jandel). Receiver operating characteristic curve was generated using statistical software SPSS version 12.0.

Identification of the protease responsible for the degradation ofC3a.

Matrix Metalloprotease 7 (MMP-7), MMP-9, MMP-Il, MMP-13 were from Sigma. C3a was from Calbiochem. Gelatin was from Sigma. MMP digestion buffer was made of 50 mM Tris ρH7.4, 10 mM CaCl₂, 150 mM NaCl. Master mixes for gelatin and C3a were made and then aliquoted with 200 ng of gelatin or C3a in 47 uL. The reaction mixtures were prewarmed at 37⁰C for 30 min. Three microlitres (100 ng) of MMP-7, MMP-9, MMP-Il, and MMP-13 were added and mixed. Protein substrates were allowed to digest for 1 hour in 37⁰C. 2 uL of gelatin reaction mixture or 12.5 uL of other reaction mixtures were aliquoted and loaded on Bis-Tris 4-12% polyacrylamide gel with MES buffer and subsequently stained with SilverQuest staining kit according to the vendor's instruction.

Results

Protein profiling of low molecular weight copper binding serum proteins

A panel of forty five sera samples was assayed with IMAC-Cu²⁺ arrays. The breakdown of sera samples from patients were as follows: forty nine hepatoceullar carcinoma, thirteen HBV, three primary colon cancer, ten nasopharyngeal carcinoma, and nine normal volunteer as a control group. Proteins with molecular weight of up to 20 kDa were revealed by using PBS II mass spectrometer. The coefficient of variation of signal intensity from repeat runs of the same sample was 29%. Three differential peaks specific to HCC sera were visualized by aligning all the spectra in a gel view format (Figure 1). Two proteins (5.3 and 5.9 kDa) were up-regulated while another one (8.1 kDa) was down-regulated. These three differential proteins peaks were not seen in non-liver cancer (colon and nasopharyngeal) or chronic liver disease (HBV carrier) samples. This led to the conclusion that the panels of differential protein markers were specific to HCC but not hepatitis B infection nor general tumorigenesis.

Classifier accuracy and selected features

We selected a subset of samples for independent testing of the classifier. Using the remaining set of samples we developed ten binary classifiers. We increased the number of features (selected based on p- values of Wilcoxon rank-sum non-parametric tests) so that we get a minimal training set error rate. In all the ten binary classifiers using the features listed below we could get 100% accuracy when we used a leave-one-out procedure. The best performance was observed using SVM, achieving 100% sensitivity and 89% specificity (Table 2). All forty nine HCC samples were identified correctly in leave-one-out training and testing cycles. Eight out of nine normal samples were classified as normal samples. In comparing HBV with HCC samples, KNN and SVM performed equally well (Table 2), with 98% sensitivity and 85% specificity. However, prediction using CART yielded the lowest specificity and sensitivity in both comparisons

Table 2. Sensitivity and specificity of classification using different bioinformatics algorithms.

1: DIDA: Diagonal linear Discriminant Analysis; 2, KNN: K-Nearest Neighbor: 3: CART: Classification and Regression Tree Analysis; 4: SVM: Support Vector Machine.

(Table 2). Protein peak signatures for binary classifiers (which resulted in 100% classification accuracy) are listed in Table 3 Each binary classifier contained four to ten protein peak signatures. The number of selected protein peak signatures in the set was optimized. Increase or decrease in the number of selected protein peak signatures reduced the sensitivity of class prediction by at least three percent. The protein peaks of 5036, 5S69, and 5904 Da occurred in all three binary classifiers when comparing classes of HCC, normal, and HBV. The protein peaks of 8132, 4066 (doubly charged protein peak of the 8132 Da peak), and 7552 Da occurred in two binary classifiers. The number of features selected did not correlate with the number of samples tested in each class, i.e., a class with least number of spectra did not lead to lowest or highest number of selected features. Colon cancer had the least number of analyzed spectra. The number of protein peak signatures selected varied from four (NPC vs. colon cancer) to ten (HCC vs. colon cancer) suggesting that the final number of selected features depended on the spatial distance between two classes of samples.

Table 3. Selected protein peak signatures for class prediction of normal. HBV. and HCC serum samples.

* protein peaks with increasing p-values are listed from left to right. mix values are in Da.

Classifier performance on independent test data

Selected features were then used to evaluate the test data set consisting of five spectra of every class in each round. Five spectra from each class were kept out of the training set until selected features were obtained. These five independent spectra were used to test the accuracy of prediction of the set of selected features. This cycle repeated for five times. For colon cancer we had only five spectra in the test set, there was a high possibility that the same spectrum was used as test set. All spectra from normal patients were classified as normal samples. All spectra from HCC patients were classified as HCC. Twenty one of twenty five NPC spectra were classified correctly as NPC but four of them did not get all four votes as expected. They obtained three votes for NPC and three votes for HBV. All spectra from the colon cancer samples were classified as colon cancer. All the spectra from HBV samples were classified as HBV. It should be noted that we could increase the specificity of the classification by trading off the sensitivity by insisting on all four votes for a sample to be classified to a particular group.

To test if different machine learning algorithms resulted in similar selected protein peak features, we used CART to reanalyze the spectra. An independent machine learning algorithm (classification tree) resulted in selection of similar set of proteins peaks (Figure 5), i.e., twelve out of thirteen clusters selected by CART were also selected in the SVM-based classification. The only peak that was used in CART but not in SVM was 6641 Da. Doubly charged molecules of the same protein generated half m/z values, for example the 4066 Da peak was a doubly charged molecule of the 8132 Da peak. Therefore, thirteen clusters may reflect only nine unique proteins. The predicted accuracy of this method was evaluated using re-substitution method. After the classifier was constructed, all the samples were run through the classifier. The proportion of the cases mis-classified gave a measure of the accuracy. None of the tumor patients were classified as normal patients, although some tumors were wrongly classified, confirming that the protein markers selected by SVM algorithm were also selected by using classification tree algorithm.

Comparison of selected features from i) hierarchical clustering, ii) SVM algorithm, and iii) CART algorithm revealed that 5904 and 8132 Da peaks were selected by all three algorithms to segregate normal from HCC samples. Because of the repeatedly selected characteristics, the 5904 Da (up-regulated in HCC) and 8132 (down-regulated in HCC), Da peaks were chosen for further purification and primary structural analyses using tandem mass spectrometry. Comparison of selected features from i) SVM algorithm, and ii) CART algorithm revealed that 5336, 5804, and 5868 Da peaks were selected by both algorithms. The 5868 Da peak may be the Cu²⁺ adduct of 5804 Da peak. However, confirmation test needed to be done after purification and sequencing by using tandem mass spectrometry.

SELDI guidance of protein purification on IMAC-copper surface

In order to optimize the conditions for off-line column chromatography purification of particular Cu²⁺ binding proteins, we did a fast screening of binding conditions with gradient increase of washing stringency to bound proteins on IMAC-Cu²⁺ surface. Normal serum sample (Nl) and HCC serum sample (S103) were allowed to bind onto multiple spots of a chip. Each spot was washed with buffer with different stringency to bound proteins on the IMAC-Cu²⁺ surface. As shown in Figure 6A, the intensity of the protein peak with molecular weight of 8132 Da in normal serum decreased from over 60 to 8.3 intensity unit when imidazole concentration increased from 0 mM to 25 mM. All bound proteins were competed out in the presence of 250 mM imidazole. Therefore, the optimal binding and elution condition for the 8132 Da peak onto the IMAC-Cu²⁺ surface was PBS with 0.1 M NaCl, and PBS with 0.1 M NaCl and 25 mM imidazole, respectively. Therefore, the same binding and elution conditions were used for off-line purification of these two proteins in liquid chromatography. The intensity of the 5904 Da peak in HCC serum decreased from 60.8 to 10.4 intensity unit when imidazole concentration increased from 0 mM to 10 mM (Figure 6B). In the presence of 25 mM imidazole, the intensity decreased further to 2.2 intensity unit. Therefore, both 8132 and 5904 Da peak could be effectively competed out of binding to Cu²⁺-NTA surface by using a buffer with 25 mM imidazole.

Purification of 8132 and 5904 Da proteins by Cu ^+'NTA column chromatography

We took the advantage of the fact that the Cu²⁺-NTA surface of Proteinchip array and Cu²⁺-NTA resin shared identical surface characteristics. Low MW proteins after size exclusion chromatography were allowed to bind to Cu²⁺-NTA resin. With the SELDI guidance information, we eluted 8132 and 5904 Da proteins effectively at 25 mM imidazole. Eluents at 25 mM imidazole were loaded onto SDS-PAGE. As seen in Figure 7A, a major band at 8 kDa was observed in normal sample (lane N) but not in HCC serum (lane C), whereas a 5 kDa band was observed only in HCC serum but not in normal serum. The same tubes of eluent solutions were spotted on a hydrophobic surface and tested for protein purity. A single peak of 8132 Da was observed for the 25 mM imidazole eluents of normal serum (Figure 7B). However, a major peak (5904 Da) and a minor peak (5332 Da) were observed in the 25 mM imidazole eluents of HCC serum (Figure 7B). Therefore we concluded that the 8 kDa gel band in the SDS PAGE corresponded to the 8132 Da protein peak in the SELDI profiles. The exact MW of 8132 and 5904 Da peak were analyzed using internal calibration with beta-endorphine (3456 Da) and cytochrome C (12230 Da) as standards. The mass accuracy of the mass reader was 200 ppm. This setting translated to a variation of one to two Da in assigning molecular mass of the 8132 or 5904 Da peak

Identification of 8132 and 5904 Da proteins

The 8 kDa and 6 kDa gel bands were excised and destained before peptide sequencing. Two tryptic peptides from each gel band were sequenced by tandem mass spectrometry. Figure 8A shows the mass spectrometry spectrum of the 1023 Da tryptic peptide from the 8 kDa gel band. The sequence was read as FISLGEACK. Another tryptic peptide (803 Da) was sequenced and the read out was SVQLTEK (Figure '8B). Combined score of these peptides was 73, which was higher than the threshold for identification with confidence level of 95%. Database search of these two peptides matched to human complement component 3 precursor (Accession. number NP__000055). Sequence analysis of C3a complement full length protein (77 amino acid residues) yielded a theoretical mass of 9190 Da. However, deletion of nine amino acids from the C-terminal yielded a theoretical mass of 8132 Da, which was exactly the same observed mass in SELDI profiles suggesting that the observed 8132 Da protein fragment was the 68 amino acid residue fragment of C3a after C-terminal truncation of 9 amino acids. Primary structure analysis showed that the sequence contained two set of consecutive di- cysteine residues and a histidine residues at the C-terminal of the peptide. These may be the potential Cu²⁺ binding sites that bind to Cu²⁺ ion during SELDI analysis and Cu²⁺ column purification.

Two tryptic peptides of the 5.9 kDa gel band were read as QTALVELVK and LVAASQAALGL (Figure HA). The sequences matched Chain A, Human Serum Albumin Mutant R218p Complexed With Thyroxine (Accession number 1HK3A) and AFP (gi27692693) with total Mascot score of 106 in both matches. Validation of down regulation of complement C3a fragment by Western blotting

Polyclonal antibody against human complement C3a was used in Western blotting of serum samples to confirm the presence of C3a C-terminal truncated fragment in normal but not HCC serum samples. As shown in Figure 9, positive signals at 8 kDa region were observed in an enlarged set of independent normal, colon cancer, ENT (ear nose throat), NPC, and HBV serum samples but were decreased in pre-operative HCC serum samples. There was an average of five-fold decrease in HCC sample compared to normal samples (Table 4). C3a protein level was lowest (more than six- fold lower than normal) at one month after surgery in poor prognosis samples. Good prognosis samples showed slightly higher C3a level (five-fold lower than normal) at one month after surgery than the poor prognosis samples. C3a protein level continued to increase at six months after surgery to near before surgery level (Table 4).

Table 4: Summary of Western blot analysis of C3a protein in serum samples

Fold change*; ratio of averaged value of the class to the averaged value of normal samples. Negative values denote down-regulation whilst positive values denote up-regulation.

We further investigated the analyses of area under ROC curve from Western blot signal of normal, HCC, and HBV samples. The area under ROC curve analyses of C3a band signal in Western blot was 0.942 [95%confidence interval, 0.873-1.012, P<0.001] for all cases (Figure 10). Sensitivity and specificity of C3a band volume smaller than 108736 pixels was 94.4% and 87.5%, respectively. In summary, Western analysis showed similar trend as SELDI analysis that C3a was down-regulated in HCC but not in HBV, colon cancer, NPC, ENT, and normal serum samples.

We checked the mRNA level of C3a in HCC tumor tissues and adjacent normal tissues from our data of oligonucleotide array and cDNA microarray. Approximately 30 % of the HCC tumor tissue samples showed more than two-fold down-regulation of C3 precursor message. In this particular study, the remaining samples did not exhibit statistically significant correlating differential expression of C3a messenger RNA, suggesting that in some cases, regulation of expression of serum C3a may be at the post-transcriptional, translational, or post-translational level.

Identification and validation of identity of 7.5 Wa protein

In an independent run of purification of differential proteins using Cu²⁺ NTA column, we omitted the washing step with 5 niM imidazole buffer. Eluents from 25 mM imidazole were subjected to SDS PAGE. A 7.5 kDa band was observed in tumor serum but not in normal serum (Figure 12A). The gel band was excised and destained. Two fragments of the tryptic digest were sequenced (Figure 12B). These sequences matched to beta-2 microglobulin (Accession number NP_004039) (Figure 12C). Commercially available polyclonal antibody against beta-2 microglobulin was used to confirm the identity of the 7.5 kDa fragment in Western blot assay. HCC and HBV serum samples showed higher intensity of beta-2 microglobulin band than colon cancer and NPC samples whilst normal sera showed the lowest intensity (Figure 12D). This assay confirmed that the 7.5 kDa band was beta-2 microglobulin fragment.

Identification of the protease responsible for the degradation ofC3a.

C3a was specifically digested by MMP-9 but not MMP-7, MMP-Il, or MMP-13. Half of C3a was digested in 1 hour (lane 12 Figure 14) whilst more than 90% of C3a remained in other MMP tubes (lane 5, lane 11, lane 13, and lane 14 of Figure 14). The results showed that MMP-9 was the active enzyme responsible for the degradation of C3a protein suggesting that MMP-9 may play a role in degradation of C3a in the blood stream of HCC patients.

REFERENCES

The following publications are incorporated by reference herein. 1. Parkin, D. M., Pisani, P., and Ferlay, J. (1999). Global cancer statistics. CA Cancer J. Clin. 49, 33-64.

2. Takigawa Y, Sugawara Y, Yamamoto J, Shimada K, Yamasaki S, Kosuge T, Makuuchi M. New lesions detected by intraoperative ultrasound during liver resection for hepatocellular carcinoma. Ultrasound Med Biol 2001;27:151-156.

3. Zhou KR, Yan FH, Tu BW. Arterial phase of biphase enhancement spiral CT in diagnosis of small hepatocellular carcinoma. Hepatobiliary Pancreat Dis Int 2002;l:68- 71.

4. Zhao H, Zhou KR, Yan FH. Role of multiphase scans by multirow-detector helical CT in detecting small hepatocellular carcinoma. World J Gastroenterol 2003;9:2198-2201.

5. Yan F, Zhou K, Shen J. [Comparison of enhancement patterns of multi-phase scan of dynamic MRI and dynamic CT in small hepatocellular carcinoma]. Zhonghua Zhong Liu Za Zhi 2001;23:413-416.

6. Lin J, Zhou KR, Chen ZW, Wang JH, Yan ZP, Wang YX. 3D contrast-enhanced MR portography and direct X-ray portography: a correlation study. Eur Radiol 2003;13:1277-1285.

7. Liu R, Wang J, Zhou K, Yan Z. [CT, MR and DSA imaging in the follow-up of patients with hepatocellular carcinoma well filled with lipiodol after transcatheter arterial chemoembolization]. Zhonghua Gan Zang Bing Za Zhi 2002;10:145.

8. Yu, A. S., and Keeffe, E. B. (2003). Management of hepatocellular carcinoma. Rev. Gastroenterol. Disord. 3, 8-24. 9. Johnson PJ, Poon TC, Hjelm NM, Ho CS, Blake C, Ho SK. Structures of disease- specific serum alpha-fetoprotein isoforms. Br J Cancer 2000;83:1330-1337.

10. Poon TQ Mok TS, Chan AT, Chan CM, Leong V, Tsui SH, Leung TW, et al. Quantification and utility of monosialylated alpha-fetoprotein in the diagnosis of hepatocellular carcinoma with nondiagnostic serum total alpha-fetoprotein. Clin Chem 2002,48:1021-1027.

11. Bosl, G. J., and Head, M. D. (1994). Serum tumor marker half-life during chemotherapy in patients with germ cell tumors. Int. J. Biol. Markers 9, 25-28.

12. Eisen MB, Spellman PT, Brown PO, Botstein D. Cluster analysis and display of genomewide expression patterns. Proc Natl Acad Sci U S A 1998;95:14863-14868.

13. Strey CW, Markiewski M, Mastellos D, Tudoran R, Spruce LA, Greenbaum LE, Lambris JD. The proinflammatory mediators C3a and C5a are essential for liver regeneration. J Exp Med 2003;198:913-923.

14. Frade R. Structure and functions of proteases which cleave human C3 and are expressed on normal or tumor human cells: some are involved in tumorigenic and metastatic properties of human melanoma cells. Immunopharmacology 1999;42:39-45.

15. Cortes C, Vapnik V. Support-vector networks. Machine Learning 1995;20,:273- 297.

16. Breiman L. FJ, Olshen L., and Stone C: In: Classification and Regression Trees. Boca Raton, FL: CRC, 1984. 17. Seow, T. K., Liang, R. C, Leow, C. K., Chung, M. C. Proteomics 2001, 1, 1249-1263.

18. Seow, T. K., Ong, S. E., Liang, R. C-, Ren, E. C, et al. Electrophoresis 2000, 21, 1787-1813.

19. Ou, K., Seow, T. K., Liang, R. C, Ong, S. E., et al. Electrophoresis 2001, 22, 2804-2811.

20. Park, K. S., Cho, S. Y., Kim, H., Paik, Y. K. Int. J. Cancer 2002, 97, 261- 265.

21. Le Naour, F., Brichory, F., Misek, D. E., Brechot, C, et al. MoI. Cell. Proteomics 2002, I, 197-203.

22. Poon, T. C, Chan, A. T., Zee, B., Ho, S. K., et al. Oncology 2001, 61, 275- 283.

23. Molinaro, A. M., Simon, R., and Pfeiffer, R.M. Prediction error estimation: a comparison of resampling methods. Bioinformatics. 2005 Aug l;21(15):3301-7.

24. Dangond, F., Hwang, D., Camelo, S., Pasinelli, P., et al. Physiol. Genomics 2004, 16, 229239.

25. Vlahou, A., Schorge, J. O., Gregory, B. W., Coleman, R. L. J Biomed. Biotechnol. 2003, 2003, 308-314.

26. Myers and Miller, CABIOS, 1989, 4:11-17 27. Higgins,D.G., Thomρson,J.D. and Gibson,T.J. (1996) Using CLUSTAL for multiple sequence alignments. Methods Enzymol., 266, 383-402.

28. Jeannmougin,F., Thompson,J.D., Gouy,M., Higgins,D.G. and Gibson,T.J. (1998) Multiple sequence alignment with Clustal X. Trends B iochem. ScL, 23, 403-405.

29. Higgins,D.G. and Sharp,P.M. (1988) CLUSTAL: a package for performing multiple sequence alignment on a microcomputer. Gene, 73, 237-244.

30. Thompson,J.D., Gibson,TJ., Plewniak,F., Jeanm ougin,F. and Higgins,D.G. (1997) The ClustalX windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acids Res., 25, 4876-4882.

31. Thomρson,J.D., Higgins,D.G. and Gibson,TJ. (1994) CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, positions-specific gap penalties and weight matrix choice. Nucleic Acids Res., 22, 4673- 4680. . . .

32. Ausubel et al., Current Protocols in Molecular Biology, John Wiley & Sons, New York, N.Y., 1998.

33. Sambrook, et al, Molecular Cloning: A Laboratory Manual. 2^nd ed., Cold Spring Harbor Laboratory, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y., 1989.

34. Dandri, M. et.al. Repopulation of mouse liver with human hepatocytes and in vivo infection with hepatitis B. virus. Hepatology 33:981-988, (2001). 35. Petersen, J. et.al. Liver repopulation with xenogenic hepatocytes in B and tT cell deficient mice leads to chronic hepadnavirus infection and clonal growth of hepatocellular carcinoma. Proc. Natl. Acad. Sci. 95:310-315 (1998).

OTHER EMBODIMENTS

Although various embodiments of the invention are disclosed herein, many adaptations and modifications may be made within the scope of the invention in accordance with the common general knowledge of those skilled in this art. Such modifications include the substitution of known equivalents foτ any aspect of the invention in order to achieve the same result in substantially the same way. Accession numbers, as used herein, may refer to Accession numbers from multiple databases, including GenBank, the European Molecular Biology Laboratory (EMBL), the DNA Database of Japan (DDBJ), or the Genome Sequence Data Base (GSDB), for nucleotide sequences, and including the Protein Information Resource (PIR), SWISSPROT, Protein Research Foundation (PRF), and Protein Data Bank (PDB) (sequences from solved structures), as well as from translations from annotated coding regions from nucleotide sequences in GenBank, EMBL, DDBJ, or RefSeq, for polypeptide sequences. Accession numbers, as used herein, may also refer to Accession numbers from databases such as UniGene, OMIM, LocusLink, or HomoloGene. Numeric ranges are inclusive of the numbers defining the range. In the specification, the word "comprising" is used as an open-ended term, substantially equivalent to the phrase "including, but not limited to", and the word "comprises" has a corresponding meaning. Citation of references herein shall not be construed as an admission that such references are prior art to the present invention. All publications are incorporated herein by reference as if each individual publication, were specifically and individually indicated to be incorporated by reference herein and as though fully set forth herein. The invention includes all embodiments and variations substantially as hereinbefore described and with reference to the examples and drawings.

Claims

WHAT IS CLAIMED IS:

1. A method of diagnosing hepatocellular carcinoma in a sample from a subject comprising: a) detecting the level of expression of a C3a polypeptide or variant, analog, or fragment thereof, or a C3a nucleic acid molecule or complement, variant, analog, or fragment thereof, in the sample; and b) comparing the level of expression detected in step a) to a control, wherein a decrease in the level of expression of the C3a polypeptide or variant, analog, or fragment thereof, or the C3a nucleic acid molecule or complement, variant, analog, or fragment thereof, in the sample when compared to the control indicates a diagnosis of. hepatocellular carcinoma.

2. A method of monitoring the progression or regression of hepatocellular carcinoma in a subject comprising: a) detecting the level of expression of a C3a polypeptide or variant, analog, or fragment thereof, or a C3a nucleic acid molecule or complement, variant, analog, or fragment thereof, in a first sample obtained from the subject at a first time point; b) detecting the level of expression of a C3a polypeptide or variant, analog, or fragment thereof, or a C3a nucleic acid molecule or complement, variant, analog, or fragment thereof, in a subsequent sample obtained from the subject at a subsequent time point; and c) comparing the level of expression detected in steps a) and b), wherein a differential expression of the C3a polypeptide or variant, analog, or fragment thereof, or the C3a nucleic acid molecule or complement, variant, analog, or fragment thereof, in the subsequent sample, when compared to the first sample, indicates the progression or regression of hepatocellular carcinoma.

3. The method of claim 2 wherein a decrease in the level of expression in the subsequent sample when compared to the first sample indicates progression of hepatocellular carcinoma.

4. The method of claim 2 wherein an increase in the level of expression in the subsequent sample when compared to the first sample indicates regression of hepatocellular carcinoma,

5. The method of any one of claims 2 through 4 wherein the subsequent sample is obtained at two or more time points.

6. A method of selecting a subject for a hepatocellular carcinoma therapy comprising: a) detecting the level of expression of a C3a polypeptide or variant, analog, or fragment thereof, or a C3a nucleic acid molecule or complement, variant, analog, or fragment thereof, in a sample obtained from the subject; and b) comparing the level of expression detected in step a) to a control, wherein a differential expression of the C3a polypeptide or variant, analog, or fragment thereof, or the C3a nucleic acid molecule or complement, variant, analog, or fragment thereof, in the sample when compared to the control indicates the therapy to be selected.

7. A method of monitoring the efficacy of a hepatocellular carcinoma therapy in a subject comprising: a) administering the therapy to the subject; b) detecting the level of expression of a C3a polypeptide or variant, analog, or fragment thereof, or a C3a nucleic acid molecule or complement, variant, analog, or fragment thereof, in at least one sample obtained from the subject; and c) comparing the level of expression detected in step b) to a control, wherein a differential expression of the C3a polypeptide or variant, analog, or fragment thereof, or the C3a nucleic acid molecule or complement, variant, analog, or fragment thereof, in the sample when compared to the control indicates the efficacy of the therapy.

8. The method of claim 7 wherein the sample is obtained prior to administration of the therapy

9. The method of claim 7 wherein the sample is obtained subsequent to administration of the therapy.

10. The method of claim 7 wherein the therapy is administered at two or more administration time points.

11. The method of any one of claims 7 through 10 wherein the sample is obtained at two or more sampling time points. :

12. The method of any one of claims 7 through 11 further comprising, comparing the level of expression of the C3a polypeptide or variant, analog, or fragment thereof, or the C3a nucleic acid molecule or complement, variant, analog, or fragment thereof, in the two or more sampling time points.

13. The method of any one of claims 7 through 12 wherein a decrease in the level of expression in the sample when compared to the control indicates an inefficacious therapy.

14. The method of any one of claims 7 through 12 wherein an increase in the level of expression in the sample when compared to the control indicates an efficacious therapy.

15. A method of prognosing hepatocellular carcinoma in a sample from a subject comprising: a) detecting the level of expression of a C3a polypeptide or complement, variant, analog, or fragment thereof, or a C3a nucleic acid molecule or complement, variant, analog, or fragment thereof, in the sample; and b) comparing the level detected in step a) to a control, wherein a differential expression of the C3a polypeptide or variant, analog, or fragment thereof, or the C3a nucleic acid molecule or complement, variant, analog, or fragment thereof, in the sample when compared to the control indicates the prognosis.

16. The method of claim 15 wherein a decrease in the level of expression in the sample when compared to the control indicates a poor prognosis.

17. The method of claim 15 wherein an increase in the level of expression in the sample when compared to the control indicates a good prognosis.

18. The method of any one of claims 1 through 17 further comprising detecting the level of expression of a beta-2 microglobulin or Chain A, human serum albumin mutant R218p polypeptide or variant, analog, or fragment thereof or nucleic acid molecule or complement, variant, analog, or fragment thereof.

19. The method of any one of claims 1 through 18 further comprising detecting the level of expression of an alpha fetoprotein, cytokeratin 19, des-gamma-carboxy prothrombin (DCP), serum gamma-glutamyl transferase, platelet-derived endothelial growth factor, or insulin-like growth factor binding protein polypeptide or nucleic acid molecule.

20. The method of any one of claims 1 through 19 wherein the sample is liver, plasma, or serum.

21. The method of any one of claims 1 through 20 wherein the control is liver, plasma, or serum.

22. The method of any one of claims 1 through 21 wherein the sample is, or is suspected of being, a HCC sample.

23. The method of any one of claims 1 through 22 wherein the control is a non-HCC sample.

24. The method of any one of claims 1 through 23 wherein the level of expression is detected using an antibody, peptides, or small molecules that specifically binds to the C3a polypeptide or variant, analog, or fragment thereof.

25. The method of claim 24 wherein the antibody is detectably labeled.

26. .The method of any one of claims 1 through 23 wherein the C3a nucleic acid molecule is a mRNA.

27. The method of any one of claims 1 through 23 wherein the level of expression is detected using a probe or primer that hybridizes to the C3a nucleic acid molecule or complement, variant, analog, or fragment thereof.

28. The method of claim 27 wherein the probe or primer is detectably labeled.

29. The method of any one of claims 1 through 27 further comprising generating a polypeptide or nucleic acid molecule expression profile.

30. The method of any one of claims 1 through 29 wherein the level of expression is detected using a high throughput assay.

31. The method of any one of claims 1 through 30 wherein the subject has, is suspected of having, or is at risk for having hepatocellular carcinoma.

32. The method of any one of claims 1 through 31 wherein the subject is a human.

33. A method of screening a candidate compound for treating hepatocellular carcinoma comprising: a) contacting a test system with a test compound; b) detecting the level of expression of a C3a polypeptide or variant, analog, or fragment thereof, or a C3a nucleic acid molecule or complement, variant, analog, or fragment thereof, in the test system; c) detecting the level of expression of the C3a polypeptide oi variant, analog, or fragment thereof, or the C3a nucleic acid molecule or complement, variant, analog, or fragment thereof in a control system that is not exposed to the test compound; and d) comparing the level of expression in step b) and step c), wherein differential expression of the C3a polypeptide or variant, analog, or fragment thereof, or the C3a nucleic acid molecule or complement, variant, analog, or fragment thereof in the comparison indicates that the test compound is a candidate compound for treating hepatocellular carcinoma.

34. The method of claim 33 further comprising detecting the level of expression of a beta-2 microglobulin or Chain A, human serum albumin mutant R218p polypeptide or variant, analog, or fragment thereof or nucleic acid molecule or complement, variant, analog, or fragment thereof.

35. The method of claim 33 or 34 further comprising detecting the level of expression of an alpha fetoprotein, cytokeratin 19, des-gamma-carboxy prothrombin (DCP), serum gamma-glutamyl transferase, platelet-derived endothelial growth factor, or insulin-like growth factor binding protein polypeptide or nucleic acid molecule.

36. The method of any one of claims 33 through 35 wherein the level of expression is detected using an antibody that specifically binds to the C3a polypeptide or variant, analog, or fragment thereof.

37. The method of claim 36 wherein the antibody is detectably labeled.

38. The method of any one of claims 33 through 35 wherein the C3a nucleic acid molecule is a mRNA.

39. The method of any one of claims 33 through 35 wherein the level of expression is detected using a probe or primer that hybridizes to the C3a nucleic acid molecule or complement, variant, analog, or fragment thereof.

40. The method of claim 39 wherein the probe or primer is detectably labeled.

41. The method of any one of claims 33 through 40 further comprising generating a , polypeptide or nucleic acid molecule expression profile.

42. The method of any one of claims 33 through 41 wherein the level of expression is detected using a high throughput assay.

43. The method of any one of claims 33 through 42 wherein the test system or control system is an animal model for hepatocellular carcinoma or a hepatocellular carcinoma cell line.

44. A composition comprising an addressable collection of two or more C3a or beta-2 microglobulin polypeptide or variants, analogs, or fragments thereof, or two or more C3a or beta-2 microglobulin nucleic acid molecule or complements, variants, analogs, or fragments thereof, that are differentially expressed in hepatocellular carcinoma.

45. The composition of claim 44 further comprising a Chain A, human serum albumin mutant R218p polypeptide or variant, analog, or fragment thereof or nucleic acid molecule or complement, variant, analog, or fragment thereof.

46. The composition of claim 44 or 45 further comprising a alpha fetoprotein, cytokeratin 19, des-gamma-carboxy prothrombin (DCP), serum gamma-glutamyl transferase, platelet-derived endothelial growth factor, and insulin-like growth factor binding protein polypeptide or nucleic acid molecule.

47. The composition of any one of claims 44 through 46 wherein the nucleic acid molecules or the polypeptides are attached to a solid support.

48. The composition of claim 47 wherein the solid support comprises a microarray.

49. Use of the composition of any one of claims 44 through 47 in the preparation of a medicament for diagnosis, prognosis, monitoring the progression or regression of hepatocellular carcinoma, or for monitoring the efficacy of a therapy, or selecting a subject for a therapy for hepatocellular carcinoma.

50. Use of a C3a polypeptide or variant, analog, or fragment thereof, or a C3a nucleic acid molecule or complement, variant, analog, or fragment thereof, in the preparation of a medicament for diagnosis, prognosis, monitoring the progression or regression of hepatocellular carcinoma, or for monitoring the efficacy of a therapy, or selecting a subject for a therapy for hepatocellular carcinoma wherein the C3a polypeptide or nucleic acid molecule is differentially expressed in hepatocellular carcinoma.

51. The use of claim 50 further comprising a beta 2 microglobulin or Chain A, human serum albumin mutant R218p polypeptide or variant, analog, or fragment thereof or nucleic acid molecule or complement, variant, analog, or fragment thereof.

52. The use of claim 50 or 51 further comprising an alpha fetoprotein, cytokeratin 19, des-gamma-carboxy prothrombin (DCP), serum gamma-glutamyl transferase, platelet- derived endothelial growth factor, or insulin-like growth factor binding protein polypeptide or nucleic acid molecule.

53. A method of preventing degradation of a C3a polypeptide or variant, analog, or fragment thereof, in a sample containing a matrix metalloprotease 9 polypeptide or nucleic acid molecule, the method comprising contacting the matrix metalloprotease 9 polypeptide or nucleic acid molecule with a matrix metalloprotease 9 inhibitor, whereby the matrix metalloprotease 9 polypeptide or nucleic acid molecule is prevented from degrading the C3a polypeptide.

54. The method of claim 53 wherein the sample is liver tissue, liver cells, cell extract, plasma, serum, or other body fluids.

55. The method of claim 53 or 54 wherein the sample is, or is suspected of being, a HCC sample.

56. The method of any one of claims 53 through 55 wherein the level of expression is detected using an antibody, peptides, or small molecules that specifically binds to the C3a polypeptide or variant, analog, or fragment thereof.

57. The method of claim 56 wherein the antibody is detectably labeled.

58. A method of screening a candidate compound for treating hepatocellular carcinoma comprising: a) contacting a C3a polypeptide or variant, analog, or fragment thereof, with a matrix metalloprotease 9 polypeptide in the presence or in the absence of a test compound; and b) detecting the level of expression of the C3a polypeptide or variant, analog, or fragment thereof in the presence and in the absence of the test compound, wherein an increase in the expression of the C3a polypeptide or variant, analog, or fragment thereof in the presence of the test compound, relative to the expression of the C3a polypeptide or variant, analog, or fragment thereof in the absence of the test compound, indicates that the test compound is a candidate compound for treating hepatocellular carcinoma.

59. The method of claim 54 wherein the test compound is a matrix metalloprotease 9 inhibitor.

60. The method of claims 58 through 59 wherein the level of expression is detected using an antibody that specifically binds to the C3a polypeptide or variant, analog, or fragment thereof.

61. The method of claim 60 wherein the antibody is detectably labeled.

62. A method of preventing degradation of a C3a polypeptide in a subject in need thereof, the method comprising administering a matrix metalloprotease 9 inhibitor to said subject.

63. The method of claim 62 wherein the subject has, is suspected of having, or is at risk for having hepatocellular carcinoma.

64. The method of claim 62 or 63 wherein the subject is a human.

65. Use of a matrix metalloprotease 9 inhibitor for the preparation of a medicament for preventing degradation of a C3a polypeptide in a subject in need thereof. 66. The use of claim 65 wherein the subject has, is suspected of having, or is at risk for having hepatocellular carcinoma.

64. The use of claim 65 or 66 wherein the subject is a human.