The cancer/testis genes: Review, standardization, and commentary
Matthew J. Scanlan, Andrew J. G. Simpson, and Lloyd J. Old
Ludwig Institute for Cancer Research, New York Branch of Human Cancer Immunology at Memorial Sloan-Kettering Cancer Center, 1275 York Avenue, New York, NY 10021
human, cancer/testis, tumor antigens, mRNA, tissue
Cancer/testis (CT) antigens are immunogenic in cancer patients, exhibit highly tissue-restricted expression, and are considered promising target molecules for cancer vaccines. To date, 44 CT gene families have been identified and their expression studied in numerous cancer types. For example, bladder cancer, non-small cell lung cancer, and melanoma are high CT gene expressors, with 11/20 (55%), 17/33 (51%) and 17/32 (53%) of the CT transcripts examined by RT-PCR detected in 20% or more of the specimens examined, respectively. Breast and prostate cancer can be considered moderate CT gene expressors, with 12/32 (37%) and 6/20 (30%) CT transcripts having an expression frequency >20%, respectively, while renal and colon cancer are low CT gene expressors, with only 3/33 (9%) and 4/25 (16%) CT transcripts having an expression frequency >20%, respectively. In normal tissues, standardized RT-PCR experiments showed that 19/43 CT genes were testis-restricted, 10/43 CT genes were tissue-restricted (mRNA detected in 2 or fewer non-gametogenic tissues), 9/43 CT genes were differentially expressed (mRNA detected in 3-6 non-gametogenic tissues), and 5/43 CT genes were ubiquitously expressed. With the exception of testis-restricted CT transcripts, all remaining CT transcripts were expressed in normal pancreas. In terms of immunogenicity, 14/29 testis/tissue-restricted CT gene families have been shown to induce a cellular and/or humoral immune response in humans. In view of the expanding list of CT genes, a CT gene database was created to standardize CT nomenclature and accumulate relevant data regarding their expression profiles, immunogenicity, function (where known), gene structure and location, and orthologous groups.
Continued progress in the development of antigen-specific cancer vaccines depends in part on the identification of a wider spectrum of immunogenic gene products expressed predominately in human cancer. The foundations for our current efforts to define molecular targets for cancer vaccines were laid in the 1970s and 1980s with the establishment of autologous typing as the methodology to identify small subsets of patients with specific humoral or T cell immunity to their own tumor cells (1). This approach set the stage for the molecular definition of tumor antigens recognized by T cells by Boon and his colleagues (2, 3), and those recognized by antibodies by Pfreundschuh and his colleagues (4). The T cell epitope cloning technique developed by Boon et al. in 1991 (5) led to the discovery of the human tumor antigens MAGEA1 (3), BAGE (6), and GAGE1 (7). It was shown that mRNA transcripts encoding these gene products were present exclusively in normal testis and in a fraction of the samples analyzed for various cancer types. In 1995, Pfreundschuh and colleagues introduced the SEREX (serological expression cloning) approach to identify tumor antigens that elicit high-titer IgG antibody responses in cancer patients (4). Among the genes identified during these initial SEREX analyses of human cancer were SSX2 (4), NY-ESO-1 (8), and SYCP-1 (9), which like MAGE, BAGE, and GAGE, are expressed predominantly in normal testis and in cancer. In recognition of this expression profile shared by otherwise unrelated genes, the terms "cancer-testis (CT) antigen" (8) and "cancer-germline gene" (10) have been introduced to incorporate this heterogeneous group of tumor antigens.
The advent of high throughput mRNA expression analysis, including such techniques as representational difference analysis (RDA), differential display, and cDNA microarray analysis, as well as SAGE (serial analysis of gene expression) and EST (expressed sequence tag) sequencing, has prompted the search for additional gene products with mRNA expression profiles restricted to cancer and testis. For example, RDA led to the cloning of several transcripts, including LAGE-1 (11), MAGEE1/CT10 (12), and SAGE (13). Likewise, the CT genes CTp11/SPAN-X-C1 (14) and MMA-1A (15) were identified by differential display and oligonucleotide array analysis, respectively. Most recently, the bioinformatics based analysis of EST databases was used to identify several CT transcripts, including those derived from BRDT/CT9 (16), PAGE5/CT16 (17), LDHC (18), and TPTE (19). There is a clear distinction between CT genes identified by mRNA expression analysis and those identified through immunological methods; the former group has immunogenic potential, while the latter are known to be immunogenic in cancer patients.
As shown in Table 1, 44 distinct CT "gene" or "antigen" families have been reported in the literature. Certain CT gene families contain multiple members (e.g. MAGEA, GAGE1), as well as splice variants (e.g. XAGE1a, XAGE1b), and currently a total of 89 distinct transcripts are known to be encoded by CT genes. With the exception of the related genes in the MAGE-A, -B, -C and -E families, and the GAGE, XAGE, and PAGE families, there is no general evolutionary linkage between CT genes. Furthermore, the protein products of only 19/44 CT gene families (Table 2) have been demonstrated to elicit an immune response in humans. For this reason, it is more appropriate to consider them as a category of genes or gene products connected by a common expression pattern, rather than as a single family of genes or antigens.
The growing list of CT genes compels the field to implement a standard nomenclature. One can envision a situation similar to that of the "cluster of differentiation" or CD antigens, where hundreds of related molecules are defined. The CT classification system should provide a framework for accommodating a larger number of gene products and their relatedness, replacing the current "-AGE" nomenclature (e.g. MAGE, GAGE, BAGE, PAGE, CAGE, XAGE, CSAGE, HAGE, SAGE). A proposed scheme for the CT family has been suggested (20). Due to a general lack of information regarding the function of CT gene products, this nomenclature is based largely on their chronological order of discovery e.g. MAGEA is CT1; BAGE is CT2 etc. In cases were there are multiple family members, individual members are designated numerically e.g. SSX1 is CT5.1; SSX2 is CT5.2; SSX3 is CT5.3, and so on. In cases were there are multiple isoforms, individual splice variants are designated alphabetically, e.g. XAGE-1a is CT12.1a; XAGE-1b is CT12.1b; and XAGE-3a is CT12.3a. Finally, in deference to the original discovery, this nomenclature lists the given gene identifier in addition to the aforementioned nomenclature (e.g. SYCP1/CT8).
In view of the expanding list of CT genes and their significance in relation to antigen-specific cancer vaccines, we have recently created a CT gene database to accumulate relevant data at a single web interface (21). The CT Gene Database focuses on several characteristics of CT genes, including their expression profiles, immunogenicity, function (where known), gene structure and location, and orthologous groups. As an introduction to the database, this review examines the expression and immunogenicity of the current list of 44 CT gene families.
gene expression in cancer
A search of the EST (22) and SAGE (23) databases was used to verify CT gene expression in cancer. Of the 44 CT gene families, 34 were found to be represented by at least one EST derived from non-germ cell associated tumors. With the exception of tumor cell lines and germ cell tumors, no homologous ESTs for SAGE/CT14, NA88/CT18, CAGE/CT26, HOM-TES-85/CT28, LDHC/CT32, MORC/CT33, SPO11/CT35, NY-SAR-35/CT37, FTHL17/CT38, and TPTE/CT44 were present in more than 4,000 EST libraries derived from tumor tissue. However, a small number of SAGE tags corresponding to these ten CT transcripts were found in SAGE libraries derived from various non-germ cell related malignant tissues, indicating that these genes have a low expression frequency and/or a low transcript copy number in tumor tissue. Thus, it was possible to verify the expression of all founding members of each CT gene family in cancer using publicly available transcript data.
A survey of publications describing the expression frequencies of CT genes was used to generate an RT-PCR based expression profile for 41 of the 44 founding members of each CT gene family in 16 different cancer types (RT-PCR expression data in cancer was not available for IL13RA/CT19, CSAGE/CT24.1, and BORIS/CT27). As shown in Table 3, the scope of this expression profile varies with tumor type. Bladder cancer, breast cancer, colon cancer, non-small cell lung cancer, prostate cancer, renal cancer, and melanoma have been evaluated for the expression of more than 20 different CT gene families, while esophageal cancer, gastric cancer, head and neck cancer, ovarian cancer, leukemia/lymphoma, hepatocellular carcinoma, and sarcoma have been evaluated for the expression of 10-20 different CT families. On the other hand, brain and pancreatic cancer have been less widely studied, with the expression of fewer than ten CT gene families having been analyzed in these tumor types. It should be noted that the mRNA expression frequencies shown in Table 3 were obtained from numerous sources under non-standardized RT-PCR conditions. Therefore, it would be incorrect to directly compare the level of expression of individual CT genes in cancer using these data. However, they do permit generalized comparisons of CT expression in different tumor types.
Published mRNA expression frequencies of CT genes in various cancers
Excluding brain and pancreatic cancers, which have been insufficiently studied with regard to CT transcript expression, the tumor types listed in Table 3 can be placed into three groups based on the number of CT genes expressed and their expression frequency: (i) High CT expressors: tumors which express more than 50% of the CT antigens examined at a expression frequency greater than 20% (a frequency of 20% of the specimens is above the average expression frequency of 16.1% for all CT transcripts shown in Table 3, n=305, SD ± 18.2); (ii) Moderate CT expressors: tumors which express between 30-50% of the CT transcripts examined at an expression frequency greater than 20%; and (iii) Low CT expressors: tumors which express less than 30% of the CT transcripts examined at an expression frequency greater than 20%. As mentioned above, seven tumor types (bladder cancer, breast cancer, colon cancer, non-small lung cancer, melanoma, prostate cancer and renal cancer) have been tested for expression of at least 20 CT transcript families. Of these, bladder cancer, non-small cell lung cancer, and melanoma are high expressors, where 11/20 (55%), 17/33 (51%) and 17/32 (53%) of the CT transcripts examined by RT-PCR were detected in 20% or more of the specimens examined, respectively. In turn, breast and prostate cancer can be considered moderate expressors, where 12/32 (37%) and 6/20 (30%) of the CT transcripts examined had an expression frequency of greater than 20% in these tumor types, respectively. On the other hand, renal cancer and colon cancer are low CT expressors, in that only 3/33 (9%) and 4/25 (16%) of the CT genes examined had an expression frequency of greater than 20% in these tumor types, respectively. These variations in the observed incidences of CT expression may, in part, stem from the fact that most colon and renal cancer specimens evaluated were from primary lesions, whereas most melanomas were metastatic lesions. Whether these differences in CT expression have significance with regard to tumor etiology or pathology needs to be explored. The conclusions of this analysis pointing to high CT expression in non-small cell lung cancer and melanoma, as well as infrequent expression in colon cancer and renal cancer, are in agreement with other recent overviews (24). On the other hand, the finding of frequent CT expression in bladder cancer is new. This adds another tumor type to the list of cancers (e.g. melanoma and lung cancer) that might be exceptionally suitable for treatment by immunotherapy with polyvalent CT antigen-based vaccines. Although these data provide a generalized view of the incidence of CT expression in cancer, individual tumor subtypes were not evaluated (e.g. adenocarcinoma vs. squamous cell carcinoma). Such analyses are infrequent (16, 25), though critical to fully understand the phenomenon of CT expression.
Seven tumor types (esophageal cancer, gastric cancer, head and neck cancer, leukemia/lymphoma, hepatocellular carcinoma, ovarian cancer, and sarcoma) have been examined for the expression of 10-20 CT transcript families. Of these seven tumor types, only hepatocellular carcinoma can be considered a high CT expressor. In hepatocellular carcinoma 8/10 CT genes examined were expressed in more than 20% of the specimens examined. Esophageal cancer, head and neck cancer, ovarian cancer, and sarcoma were identified as moderate CT transcript expressors, where 4/12 (33%), 5/15 (33%), 7/16 (43%), and 7/19 (37%) of the CT transcripts examined were detected at a frequency of greater than 20%, respectively. Both gastric cancer and leukemia/lymphoma are low CT transcript expressors, with 2/10 and 3/16 (approximately 20%) of the CT transcripts examined being detected at a frequency of at least 20%, respectively. Since a relatively small number of CT transcripts were examined in these seven tumor types, a more accurate assessment of the degree of CT expression in these cancers will depend upon further RT-PCR analyses involving additional CT transcripts.
It should be emphasized that these data correspond to mRNA expression frequencies, which do not necessarily directly correlate with protein expression frequencies. For example, only 2/14 or 14% of breast cancers were positive for NY-ESO-1/CT6.1 expression as measured by immunohistochemical analyses using the NY-ESO-1/CT6.1 specific monoclonal antibody ES121 (26). This study is in contrast to the NY-ESO-1/CT6.1 mRNA expression frequency of 30% (10/33 specimens) in breast cancer (8). Conversely, 9/11 (81%) invasive ductal carcinomas of the breast were positive for PLU-1/CT31 protein expression as judged by Western blotting using polyclonal antisera specific for PLU-1, which is consistent with its mRNA expression frequency [6/7 or 86% of breast cancer specimens (27)]. Discrepancies between gene and protein expression frequencies may reflect variations in tissue sampling, tumor heterogeneity or different levels of sensitivity of detection. Studies similar to those described by Jungbluth and colleagues (28), in which MAGEA1/CT1.1 transcript expression and MAGEA1/CT1.1 protein expression were directly compared in the same lung cancer specimens, are very useful in distinguishing between these possibilities. Unfortunately, the list of available monoclonal antibodies to CT proteins that can be used in an immunohistochemical or biochemical analysis of protein expression is relatively limited, but includes mAbs E975 and ES121 to NY-ESO-1/CT6.1 (29), mAb MA454 to MAGEA1/CT1.1 (30), and mAb CT7-33 to MAGEC1/CT7.1 (31), as well as mAbs that recognize multiple members of CT antigen families, such as mAb 57B to MAGEA/CT1 (32) and mAb E3AS to SSX/CT5 (33). Thus, monoclonal antibodies are currently available for only four of the protein products encoded by the 44 CT transcript families. The generation of additional monoclonal antibodies specific for CT proteins is essential for defining their tissue and tumor distribution.
transcript expression in normal tissues
A standardized, quantitative end-point RT-PCR assay was used to evaluate the expression profile of 44 genes representing founding members of each CT family in a series of normal tissues. End-point RT-PCR, as opposed to real-time PCR, was chosen as the assay system since it is the most widespread method used for analyzing CT transcript expression. In this standardized assay, the PCR primers and annealing temperatures were those described in the corresponding primary publications [see the CT Gene Database (21)]. A total of 35 PCR amplification cycles were performed in all experiments. The cDNA templates consisted of commercially available, normalized cDNA panels (MTC panels I and II, BD Biosciences, Palo Alto CA) encompassing 16 normal tissues (including the three gametogenic tissues ovary, testis and placenta), obtained from multiple (3-500) disease-free individuals. A "hot start" PCR master mix (Platinum PCR Super Mix, Invitrogen, Carlsbad CA), containing recombinant Taq DNA polymerase complexed with Taq antibody (20 U/ml), and deoxyribonucleotide triphosphates (200 µM), and MgCl2 (1.5 mM), was used in all experiments. The resultant PCR products were electrophoresed in 2% agarose/Tris-Acetate-EDTA (TAE) gels and quantified using a standard curve of signal intensities generated from simultaneous electrophoresis of molecular mass markers (Quanti-ladder, OriGene Technologies, Rockville MD). Signal intensities were calculated using digital image capture equipment (Kodak Image Station 440, Eastman Kodak, New Haven CT) and corresponding image analysis software (Kodak ID 3.6). Due to image saturation, the upper limit of quantification was 2.0 µg, while the sensitivity of the assay was limited to 0.01 µg cDNA. Gel images are available at the CT Gene Database (21).
As summarized in Table 4, standardized RT-PCR expression analysis segregated the CT family into four categories on the basis of normal tissue expression: (i) testis-restricted transcripts, (ii) tissue-restricted CT genes expressed in two or fewer of the non-gametogenic tissues tested, (iii) differentially expressed CT genes expressed in three to six of the 13 non-gametogenic tissues tested, and (iv) ubiquitously expressed CT genes. Of the 43 CT transcripts amplified (HCA661/CT30 could not be amplified), 19 were expressed exclusively in normal testis, as judged by this assay, with final levels of cDNA ranging from more than 2 µg to a minimal level of 0.01 µg produced by 35 amplification cycles. A subset of ten CT transcripts was tissue restricted, being expressed in testis (>2.0-0.02 µg cDNA product), ovary (0.01 µg cDNA product), and/or placenta (0.1-0.01 µg cDNA product), plus one or two non-gametogenic tissues. In all nine of these cases, pancreas was one of the two additional non-gametogenic tissues (1.2-0.01 µg cDNA product), while liver (0.19-0.01 µg cDNA product) or spleen (0.01 µg cDNA product) was the other non-gametogenic tissue found to express CT genes in this tissue-restricted category. An additional subset of nine CT transcripts was differentially expressed, with mRNA detected in three to six of 13 non-gametogenic tissues. Again, corresponding mRNA was detected in pancreas for each of these CT genes (0.07-0.01 µg cDNA product). Other tissues that express these differentially expressed CT transcripts include liver, spleen, thymus, kidney, lung, prostate and brain. Overall, with the exceptions of SPANXC/CT11.3 expression in pancreas (1.2 µg cDNA product detected) and XAGE-1a/CT12.1a expression in lung (>2.0 µg cDNA product detected), the expression levels of tissue-restricted and differentially expressed subsets of CT genes in non-gametogenic tissues were quite low, with 0.11-0.01 µg cDNA product detected. It is unclear whether these low-levels of mRNA expression correspond to significant quantities of protein. For example, although MAGEA1/CT1.1 mRNA was detected in pancreas (0.02 µg cDNA product detected), and NY-ESO-1/CT6.1 mRNA was detected in both pancreas (0.02 µg cDNA product detected) and liver (0.11 µg cDNA product detected), no corresponding protein has ever been reported in these tissues by immunohistochemistry (26, 34).
Normal tissue expression profiles of CT genes as determined by RT-PCR
A subset of five CT transcripts, IL13RA/CT19, TSP50/CT20, SPA-17/CT22, OY-TES-1/CT23, and PLU-1/CT31, were ubiquitously expressed, with mRNA being detected in a majority of the non-gametogenic tissues tested. As shown in Figure 1, PLU-1/CT31 mRNA was detected at low levels (0.05 µg cDNA product detected or less) in all tissues except testis (0.2 µg). Thus, it appears that PLU-1/CT31 is universally expressed at low levels in non-gametogenic tissues. With regard to TSP50/CT20, the corresponding mRNA was detected at high levels (>1.0 µg cDNA product detected) in testis, kidney, liver, and pancreas; at a moderate level in thymus (0.11 µg); and at low levels in all other normal tissues (0.05 µg or less). Transcripts encoding IL13RA/CT19, SPA17/CT22, and OY-TES-1/CT23 were detected at high levels in testis (>2.0 µg cDNA product detected), and at moderate to high levels (0.1-2.0 µg cDNA product detected) in 11 of the 13 non-gametogenic tissues examined. Thus in addition to their high level of expression in testis, SPA17/CT22, OY-TES-1/CT23, and IL13RA/CT19 were also expressed at high levels in many non-gametogenic tissues, and therefore on the basis of mRNA expression, their inclusion in the CT gene category is suspect.
Expression profile of the ubiquitously expressed CT genes IL13RA/CT19, TSP50/CT20, OY-TES-1/CT23, and PLU-1/CT31. A standardized, quantitative end-point RT-PCR assay was used to evaluate the mRNA expression profile of IL13RA/CT19, TSP50/CT20, OY-TES-1/CT23, and PLU-1/CT31 in a series of 16 normal tissues (MTC panels I and II, BD Biosciences, Palo Alto CA). The level of PLU-1/CT31 expression was low in all tissues with the exception of testis. TSP50/CT20 transcripts were detected at high levels in testis, kidney, liver, and pancreas; at a moderate level in thymus and at low levels in all other normal tissues. Transcripts encoding IL13RA/CT19, SPA17/CT22, and OY-TES-1/CT23 were detected at high levels in 11/13 non-gametogenic tissues.
The frequent detection of CT transcripts in normal pancreas was unexpected. In order to validate these results, additional pancreatic cDNA was prepared as previously described (17) and subjected to the same RT-PCR analysis for CT gene expression as described above. As shown in Figure 2, 16 of the 24 CT transcripts originally detected in normal pancreas were also detected in this second sample of normal pancreas cDNA. In addition to the universally expressed CT transcripts, SPA17/CT22, OY-TES-1/CT23, and IL13RA/CT19, mRNA expression in normal pancreas was also confirmed for MAGEA1/CT1.1, XAGE-1a/CT12.1a, HAGE/CT13, PAGE-5/CT16A, TSP50/CT20, BORIS/CT27, D40/CT29, PLU-1/CT31, SGY-1/CT34, NY-SAR-35/CT37, NXF2/CT39, TAF7L/CT40, and TDRD1/CT41.1. Conversely, mRNA encoding NY-ESO-1/CT6.1, MAGEC1/CT7.1, SYCP1/CT8, SPANXC/CT11.3, ADAM2/CT15, CAGE/CT26, TEX15/CT42, and FATE/CT43 was not detected in this second cDNA preparation from normal pancreas. These genes, whose expression could not be confirmed in a second source of normal pancreas, do not necessarily correspond to those previously detected in pancreas at exceptionally low levels (<0.05 µg cDNA product detected). For example, MAGEA1/CT1A was detected at low levels in both the first and second pancreatic cDNA preparations (0.02 µg cDNA product detected), while SPANXC/CT11.3 was originally detected in pancreas at a high level (1.2 µg cDNA product detected), but was not detected in the second pancreatic cDNA (Figure 1). It is possible that these differences in CT transcript expression in the pancreas reflect differences in the cellular composition of the tissue used in the different pancreatic RNA preparations and/or differences among tissue donors. Nevertheless, these data indicate that CT transcripts are frequently expressed in the pancreas.
Verification of CT expression in normal pancreas by RT-PCR. End-point RT-PCR assay was used to verify the mRNA expression of 25 CT transcripts in a second sample of pancreatic cDNA (see text). The presence of mRNA encoding MAGEA1/CT1.1, XAGE-1a/CT12.1a, HAGE/CT13, PAGE-5/CT16A, IL13RA/CT19, TSP50/CT20, SPA17/CT22, OY-TES-1/CT23, BORIS/CT27, D40/CT29, PLU-1/CT31, SGY-1/CT34, NY-SAR-35/CT37, NXF2/CT39, TAF7L/CT40, and TDRD1/CT41.1 was confirmed in normal pancreas. Lane assignments are as follows: 1, OY-TES-1/CT23; 2, MAGEC1/CT7.1; 3, NY-ESO-1/CT6.1; 4, SYCP1/CT8; 5, BORIS/CT27; 6, ADAM2/CT15; 7, IL13RA/CT19; 8, PLU-1/CT31; 9, TDRD1/CT41.1; 10, TEX15/CT42; 11, MAGEA1/CT1.1; 12, TSP50/CT20; 13, NXF2/CT39; 14, SGY-1/CT34; 15, CAGE/CT26; 16, TAF7L/CT42; 17, D40/CT29; 18, PAGE-5/CT16.1; 19, NY-SAR-35/CT37; 20, SPA17/CT22; 21, HAGE/CT13; 22, SAGE/CT14; 23, SPANXC/CT11.3; 24, XAGE-1a/CT12.1a; 25, FATE/CT43.
transcripts encoding immunogenic proteins
As their name implies, CT antigens are immunogenic gene products. Nineteen of the 44 CT families encode proteins associated with an immune response in humans, and can be considered bona fide CT antigens (Table 2). Immune recognition of the majority of these CT antigens is cancer-related, occurring spontaneously in cancer patients, but not in cancer-free individuals. The exceptions to this are humoral immune responses to MAGEB1/CT3.1 in systemic lupus erythematosus patients (35) and to SPA17/CT22 in vasectomized men (36). The process of CT antigen discovery in cancer, by either T cell epitope cloning or by serological analyses, has separated the cancer-related CT antigens into two groups, those defined by T lymphocyte recognition (MAGEA/CT1, BAGE/CT2, GAGE/CT4, NA88/CT18) and those defined by antibody recognition (SSX/CT5, NY-ESO-1/CT6, MAGEC/CT7, SYCP1/CT8, SPANXB1/CT11B, CTAGE/CT21, OY-TES-1/CT23, CAGE/CT26, HOM-TES-85/28, HCA661/CT30, NY-SAR-35/CT37, FATE/CT43, TPTE/CT44). However, there is increasing evidence of a coordinated immune response to these antigens in cancer patients. Both antibody and T-lymphocyte (CD4 and CD8) responses have been shown for three of these CT families, MAGEA/CT1, SSX/CT5 and NY-ESO-1/CT6 (3, 29, 37, 38, 39). Characterization of the T cell response to other serologically-defined CT antigens is a high priority. This should not only increase the number of vaccine targets, but also provide further evidence of the coordinated nature of the immune response to cancer.
With regard to the tissue expression patterns of CT antigens, 14 of the 19 immunogenic CT gene products have highly tissue-restricted mRNA expression profiles in normal tissues, i.e., they are expressed in two or fewer non-gametogenic tissues. These data emphasize the specificity of the host immune response to cancer and underscore the utility of immunogenic methods of gene discovery. The exceptions to the tissue restriction/immunogenicity paradigm are the differentially expressed genes CAGE/CT26 and FATE/CT43, and the ubiquitously expressed genes SPA17/CT22 and OY-TES-1/CT23 (in the present study HCA661/CT30 could not be amplified by RT-PCR). In these cases, expression in multiple normal tissues does not appear to be tolerogenic, suggesting that these antigens are expressed at an immunologically irrelevant level in normal tissues, and their expression in the context of cancer elicits an immune response.
Based on their immunogenicity and restricted expression, CT antigens are ideal for use in cancer vaccines. To date, approximately 90 CT transcripts have been identified, thus providing a substantial arsenal for active, antigen-specific immunotherapy of cancer. Many of the antigens have been proven to be immunogenic in cancer patients. The fact that cancer patients with spontaneous immune reactivity to CT antigens still present with progressive disease begs the question as to whether naturally immunoreactive antigens are valid targets for cancer vaccines. What we do not know is whether cancer would progress more rapidly in the absence of an immune response. The influence of vaccine-induced immunity on tumor growth without pre-existing immunity to CT antigens is, of course, the way to answer this question. Two features of immunogenic CT gene products (e.g. NY-ESO-1) greatly favor efforts to develop protective vaccines. One, the spontaneous immunogenicity of these antigens in a small subset of patients indicates their potential immunogenicity in a wider spectrum of patients, and two, the analysis of the humoral and cellular immune responses to these antigens in patients with spontaneous immunity has given us useful tools for monitoring the immunogenicity of CT vaccines and for determining the relative immunogenicity of different vaccine constructs.
As opposed to the conventional view that CT genes are expressed only in gametogenic tissues, the present survey shows CT mRNA transcripts to be present in other normal tissues, most notably, pancreas. In most cases, this does not detract from their immunotherapeutic potential, since expression levels in non-gametogenic tissues are exceptionally low and unlikely to result in the presence of immunologically relevant levels of MHC/CT peptide complexes. However, such hypotheses need to be tested, perhaps by direct measurement of the concentrations of MHC/CT peptide complexes in normal tissues (40).
Many questions regarding CT genes remain, including the biological function of their protein products, and the elucidation of the factors that control their expression in normal tissues and cancer. Epigenetic events such as DNA methylation and histone acetylation are known to influence CT gene expression (41, 42). Genome-wide, random hypomethylation in cancer has been viewed as a triggering event for CT gene expression, but hypomethylation itself may be part of a larger gene program characteristic of cancer. More specifically, the appearance of gametic traits in cancer, such as the expression of CT genes, global hypomethylation, cellular immortality, migratory behavior, and ectopic production of gonadotrophins, indicates a strong link between gametogenesis and carcinogenesis, and raises the possibility that cancer has acquired many of its traits by usurping the genetic program directing gametogenesis (43). The testing of such hypotheses awaits the determination of CT gene product function and a detailed analysis of genomic regulatory elements and their methylation status in normal and malignant tissues. Indeed, this gene set is highly unusual in that research into their utility in cancer therapy is far more advanced than that into their function. Nonetheless, this is in the grand tradition of the empirical nature of vaccine development that started with Pasteur and which still pervades the entire field today.
1. Old LJ. Cancer immunology: the search for specificity--G. H. A. Clowes Memorial lecture. Cancer Res 1981; 41: 361-75. (PMID: 7004632)
2. Traversari C, van der Bruggen P, Van den Eynde B, Hainaut P, Lemoine C, Ohta N, Old L, Boon T. Transfection and expression of a gene coding for a human melanoma antigen recognized by autologous cytolytic T lymphocytes. Immunogenetics 1992; 35: 145-52. (PMID: 1537606)
3. van der Bruggen P, Traversari C, Chomez P, Lurquin C, De Plaen E, Van den Eynde B, Knuth A, Boon T. A gene encoding an antigen recognized by cytolytic T lymphocytes on a human melanoma. Science 1991; 254: 1643-7. (PMID: 1840703)
4. Sahin U, Tureci O, Schmitt H, Cochlovius B, Johannes T, Schmits R, Stenner F, Luo G, Schobert I, Pfreundschuh M. Human neoplasms elicit multiple specific immune responses in the autologous host. Proc Natl Acad Sci USA 1995; 92: 11810-13. (PMID: 8524854)
5. Van den Eynde B, Lethe B, Van Pel A, De Plaen E, Boon T. The gene coding for a major tumor rejection antigen of tumor P815 is identical to the normal gene of syngeneic DBA/2 mice. J Exp Med 1991; 173: 1373-84. (PMID: 1903428)
6. Boel P, Wildmann C, Sensi ML, Brasseur R, Renauld JC, Coulie P, Boon T, van der Bruggen P. BAGE: a new gene encoding an antigen recognized on human melanomas by cytolytic T lymphocytes. Immunity 1995; 2: 167-75. (PMID: 7895173)
7. Van den Eynde B, Peeters O, De Backer O, Gaugler B, Lucas S, Boon T. A new family of genes coding for an antigen recognized by autologous cytolytic T lymphocytes on a human melanoma. J Exp Med 1995; 182: 689-98. (PMID: 7544395)
8. Chen YT, Scanlan MJ, Sahin U, Tureci O, Gure AO, Tsang S, Williamson B, Stockert E, Pfreundschuh M, Old LJ. A testicular antigen aberrantly expressed in human cancers detected by autologous antibody screening. Proc Natl Acad Sci USA 1997; 94: 1914-18. (PMID: 9050879)
9. Tureci O, Sahin U, Zwick C, Koslowski M, Seitz G, Pfreundschuh M. Identification of a meiosis-specific protein as a member of the class of cancer/testis antigens. Proc Natl Acad Sci USA 1998; 95: 5211-16 (PMID: 9560255)
10. Coulie PG, Karanikas V, Lurquin C, Colau D, Connerotte T, Hanagiri T, Van Pel A, Lucas S, Godelaine D, Lonchay C, Marchand M, Van Baren N, Boon T. Cytolytic T-cell responses of cancer patients vaccinated with a MAGE antigen. Immunol Rev 2002; 188: 33-42. (PMID: 12445279)
11. Lethe B, Lucas S, Michaux L, De Smet C, Godelaine D, Serrano A, De Plaen E, Boon T. LAGE-1, a new gene with tumor specificity. Int J Cancer 1998; 76: 903-8. (PMID: 9626360)
12. Güre AO, Stockert E, Arden KC, Boyer AD, Viars CS, Scanlan MJ, Old LJ, Chen YT. CT10: A new cancer-testis (CT) antigen homologous to CT7 and the MAGE family, identified by representational-difference analysis. Int J Cancer 2000; 85: 726-32. (PMID: 10699956)
13. Martelange V, De Smet C, De Plaen E, Lurquin C, Boon T. Identification on a human sarcoma of two new genes with tumor-specific expression. Cancer Res 2000; 60: 3848-55. (PMID: 10919659)
14. Zendman AJ, Cornelissen IM, Weidle UH, Ruiter DJ, van Muijen GN. CTp11, a novel member of the family of human cancer/testis antigens. Cancer Res 1999; 59: 6223-9. (PMID: 10626816)
15. de Wit N, Weidle UH, Ruitter DJ, van Muijen GN. Expression profiling of MMA-1 and splice variant MMA-1B: new cancer/testis antigens identified in human melanoma. Int J Cancer 2002; 98: 547-553. (PMID: 11920614)
16. Scanlan MJ, Altorki NK, Güre AO, Williamson B, Jungbluth A, Chen YT, Old LJ. Expression of cancer-testis antigens in lung cancer: definition of bromodomain testis-specific gene (BRDT) as a new CT gene, CT9. Cancer Lett 2000; 150: 155-64. (PMID: 10704737)
17. Scanlan MJ, Gordon CM, Williamson B, Lee SY, Chen YT, Stockert E, Jungbluth A, Ritter G, Jager D, Jager E, Knuth A, Old LJ. Identification of cancer/testis genes by database mining and mRNA expression analysis. Int J Cancer 2002; 98: 485-92. (PMID: 11920606)
18. Koslowski M, Tureci O, Bell C, Krause P, Lehr HA, Brunner J, Seitz G, Nestle FO, Huber C, Sahin U. Multiple splice variants of lactate dehydrogenase C selectively expressed in human cancer. Cancer Res 2002; 62: 6750-5. (PMID: 12438276)
19. Dong XY, Su YR, Qian XP, Yang XA, Pang XW, Wu HY, Chen WF. Identification of two novel CT antigens and their capacity to elicit antibody response in hepatocellular carcinoma patients. Br J Cancer 2003; 89: 291-7. (PMID: 12865919)
20. Chen YT, Gure AO, Tsang S, Stockert E, Jager E, Knuth A, Old LJ. Identification of multiple cancer/testis antigens by allogeneic antibody screening of a melanoma cell line library. Proc Natl Acad Sci USA 1998; 95: 6919-23. (PMID: 9618514)
24. Scanlan MJ, Gure AO, Jungbluth AA, Old LJ, Chen YT. Cancer/testis antigens: an expanding family of targets for cancer immunotherapy. Immunol Rev 2002; 188: 22-32. (PMID: 12445278)
25. Tajima K, Obata Y, Tamaki H, Yoshida M, Chen YT, Scanlan MJ, Old LJ, Kuwano H, Takahashi T, Takahashi T, Mitsudomi T. Expression of cancer/testis (CT) antigens in lung cancer. Lung Cancer 2003; 42: 23-33. (PMID: 14512184)
26. Jungbluth AA, Chen YT, Stockert E, Busam KJ, Kolb D, Iversen K, Coplan K, Williamson B, Altorki N, Old LJ. Immunohistochemical analysis of NY-ESO-1 antigen expression in normal and malignant human tissues. Int J Cancer 2001; 92: 856-60. (PMID: 11351307)
27. Barrett A, Madsen B, Copier J, Lu PJ, Cooper L, Scibetta AG, Burchell J, Taylor-Papadimitriou J. PLU-1 nuclear protein, which is upregulated in breast cancer, shows restricted expression in normal human adult tissues: a new cancer/testis antigen? Int J Cancer 2002; 101: 581-8. (PMID: 12237901)
28. Jungbluth AA, Stockert E, Chen YT, Kolb D, Iversen K, Coplan K, Williamson B, Altorki N, Busam KJ, Old LJ. Monoclonal antibody MA454 reveals a heterogeneous expression pattern of MAGE-1 antigen in formalin-fixed paraffin embedded lung tumours. Br J Cancer 2000; 83: 493-7. (PMID: 10945497)
29. Stockert E, Jager E, Chen YT, Scanlan MJ, Gout I, Karbach J, Arand M, Knuth A, Old LJ. A survey of the humoral immune response of cancer patients to a panel of human tumor antigens. J Exp Med 1998; 187: 1349-54. (PMID: 9547346)
30. Chen YT, Stockert E, Chen Y, Garin-Chesa P, Rettig WJ, van der Bruggen P, Boon T, Old LJ. Identification of the MAGE-1 gene product by monoclonal and polyclonal antibodies. Proc Natl Acad Sci USA 1994; 91: 1004-8. (PMID: 8302824)
31. Jungbluth AA, Antonescu CR, Busam KJ, Iversen K, Kolb D, Coplan K, Chen YT, Stockert E, Ladanyi M, Old LJ. Monophasic and biphasic synovial sarcomas abundantly express cancer/testis antigen NY-ESO-1 but not MAGE-A1 or CT7. Int J Cancer 2001; 94: 252-6 (PMID: 11668506)
32. Kocher T, Schultz-Thater E, Gudat F, Schaefer C, Casorati G, Juretic A, Willimann T, Harder F, Heberer M, Spagnoli GC. Identification and intracellular location of MAGE-3 gene product. Cancer Res 1995; 55: 2236-9. (PMID: 7757970)
33. Dos Santos NR, Torensma R, de Vries TJ, Schreurs MW, de Bruijn DR, Kater-Baats E, Ruiter DJ, Adema GJ, van Muijen GN, van Kessel AG. Heterogeneous expression of the SSX cancer/testis antigens in human melanoma lesions and cell lines. Cancer Res 2000; 60: 1654-62. (PMID: 10749136)
34. Jungbluth AA, Busam KJ, Kolb D, Iversen K, Coplan K, Chen YT, Spagnoli GC, Old LJ. Expression of MAGE-antigens in normal tissues and cancer. Int J Cancer 2000; 85: 460-5. (PMID: 10699915)
35. McCurdy DK, Tai LQ, Nguyen J, Wang Z, Yang HM, Udar N, Naiem F, Concannon P, Gatti RA. MAGE Xp-2: a member of the MAGE gene family isolated from an expression library using systemic lupus erythematosus sera. Mol Genet Metab 1998; 63: 3-13. (PMID: 9538511)
36. Lea IA, Adoyo P, O'Rand MG. Autoimmunogenicity of the human sperm protein Sp17 in vasectomized men and identification of linear B cell epitopes. Fertil Steril 1997; 67: 355-61. (PMID: 9022615)
37. Ayyoub M, Stevanovic S, Sahin U, Guillaume P, Servis C, Rimoldi D, Valmori D, Romero P, Cerottini JC, Rammensee HG, Pfreundschuh M, Speiser D, Levy F. Proteasome-assisted identification of a SSX-2-derived epitope recognized by tumor-reactive CTL infiltrating metastatic melanoma. J Immunol 2002; 168: 1717-22 (PMID: 11823502)
38. Jager E, Nagata Y, Gnjatic S, Wada H, Stockert E, Karbach J, Dunbar PR, Lee SY, Jungbluth A, Jager D, Arand M, Ritter G, Cerundolo V, Dupont B, Chen YT, Old LJ, Knuth A. Monitoring CD8 T cell responses to NY-ESO-1: correlation of humoral and cellular immune responses. Proc Natl Acad Sci USA 2000; 97: 4760-5. (PMID: 10781081)
39. Peptide database of T-cell defined tumor antigens. URL: http://www.cancerimmunity.org/peptidedatabase/Tcellepitopes.htm
40. Denkberg G, Cohen CJ, Lev A, Chames P, Hoogenboom HR, Reiter Y. Direct visualization of distinct T cell epitopes derived from a melanoma tumor-associated antigen by using human recombinant antibodies with MHC-restricted T cell receptor-like specificity. Proc Natl Acad Sci USA 2002; 99: 9421-6. (PMID: 12093904)
41. De Smet C, De Backer O, Faraoni I, Lurquin C, Brasseur F, Boon T. The activation of human gene MAGE-1 in tumor cells is correlated with genome-wide demethylation. Proc Natl Acad Sci USA 1996; 93: 7149-53. (PMID: 8692960)
42. Weiser TS, Guo ZS, Ohnmacht GA, Parkhurst ML, Tong-On P, Marincola FM, Fischette MR, Yu X, Chen GA, Hong JA, Stewart JH, Nguyen DM, Rosenberg SA, Schrump DS. Sequential 5-Aza-2 deoxycytidine-depsipeptide FR901228 treatment induces apoptosis preferentially in cancer cells and facilitates their recognition by cytolytic T lymphocytes specific for NY-ESO-1. J Immunother 2001; 24: 151-61. (PMID: 11449072)
43. Old LJ. Cancer/Testis (CT) antigens - a new link between gametogenesis and cancer. Cancer Immun 2001; 1: 1. (PMID: 12747762)
44. Van den Eynde BJ, van der Bruggen P. T cell defined tumor antigens. Curr Opin Immunol 1997; 9: 684-693. (PMID: 9368778)
45. Lurquin C, De Smet C, Brasseur F, Muscatelli F, Martelange V, De Plaen E, Brasseur R, Monaco AP, Boon T. Two members of the human MAGEB gene family located in Xp21.3 are expressed in tumors of various histological origins. Genomics 1997; 46: 397-408. (PMID: 9441743)
46. Tureci O, Chen YT, Sahin U, Gure AO, Zwick C, Villena C, Tsang S, Seitz G, Old LJ, Pfreundschuh M. Expression of SSX genes in human tumors. Int J Cancer 1998; 77: 19-23. (PMID: 9639388)
47. Zendman AJ, Van Kraats AA, Weidle UH, Ruiter DJ, Van Muijen GN. The XAGE family of cancer/testis-associated genes: alignment and expression profile in normal tissues, melanoma lesions and Ewing's sarcoma. Int J Cancer 2002; 99: 361-9. (PMID: 11992404)
48. Moreau-Aubry A, Le Guiner S, Labarriere N, Gesnel MC, Jotereau F, Breathnach R. A processed pseudogene codes for a new antigen recognized by a CD8(+) T cell clone on melanoma. J Exp Med 2000; 191: 1617-24. (PMID: 10790436)
49. Yuan L, Shan J, De Risi D, Broome J, Lovecchio J, Gal D, Vinciguerra V, Xu HP. Isolation of a novel gene, TSP50, by a hypomethylated DNA fragment in human breast cancer. Cancer Res 1999; 59: 3215-21. (PMID: 10397268)
50. Eichmuller S, Usener D, Dummer R, Stein A, Thiel D, Schadendorf D. Serological detection of cutaneous T-cell lymphoma-associated antigens. Proc Natl Acad Sci USA 2001; 98: 629-34. (PMID: 11149944)
51. Lim SH, Wang Z, Chiriva-Internati M, Xue Y. Sperm protein 17 is a novel cancer-testis antigen in multiple myeloma. Blood 2001; 97: 1508-10. (PMID: 11222401)
52. Ono T, Kurashige T, Harada N, Noguchi Y, Saika T, Niikawa N, Aoe M, Nakamura S, Higashi T, Hiraki A, Wada H, Kumon H, Old LJ, Nakayama E. Identification of proacrosin binding protein sp32 precursor as a human cancer/testis antigen. Proc Natl Acad Sci USA 2001; 98: 3282-7. (PMID: 11248070)
53. Cho B, Lim Y, Lee DY, Park SY, Lee H, Kim WH, Yang H, Bang YJ, Jeoung DI. Identification and characterization of a novel cancer/testis antigen gene CAGE. Biochem Biophys Res Commun 2002; 292: 715-26. (PMID: 11922625)
54. Tureci O, Sahin U, Koslowski M, Buss B, Bell C, Ballweber P, Zwick C, Eberle T, Zuber M, Villena-Heinsen C, Seitz G, Pfreundschuh M. A novel tumour associated leucine zipper protein targeting to sites of gene transcription and splicing. Oncogene 2002; 21: 3879-88. (PMID: 12032826)
55. Takimoto M, Wei G, Dosaka-Akita H, Mao P, Kondo S, Sakuragi N, Chiba I, Miura T, Itoh N, Sasao T, Koya RC, Tsukamoto T, Fujimoto S, Katoh H, Kuzumaki N. Frequent expression of new cancer/testis gene D40/AF15q14 in lung cancers of smokers. Br J Cancer 2002; 86: 1757-62. (PMID: 12087463)
56. Wang Y, Han KJ, Pang XW, Vaughan HA, Qu W, Dong XY, Peng JR, Zhao HT, Rui JA, Leng XS, Cebon J, Burgess AW, Chen WF. Large scale identification of human hepatocellular carcinoma-associated antigens by autoantibodies. J Immunol 2002; 169: 1102-9. (PMID: 12097419)
57. Lee SY, Obata Y, Yoshida M, Stockert E, Williamson B, Jungbluth AA, Chen YT, Old LJ, Scanlan MJ. Immunomic analysis of human sarcoma. Proc Natl Acad Sci USA 2003; 100: 2651-6. (PMID: 12601173)
58. Loriot A, Boon T, De Smet C. Five new human cancer-germline genes identified among 12 genes expressed in spermatogonia. Int J Cancer 2003; 105: 371-6. (PMID: 12704671)
59. Luo G, Huang S, Xie X, Stockert E, Chen YT, Kubuschok B, Pfreundschuh M. Expression of cancer-testis genes in human hepatocellular carcinomas. Cancer Immun 2002; 2: 11. (PMID: 12747756 )
Matthew J. Scanlan
Ludwig Institute for Cancer Research
New York Branch at Memorial Sloan-Kettering Cancer Center
1275 York Avenue
New York, NY 10021
Fax: + 1 212 717-3100