|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
Originally published as JCO Early Release 10.1200/JCO.2004.04.900 on June 1 2004 © 2004 American Society of Clinical Oncology.
The Use and Development of Germline Polymorphisms in Clinical OncologyUSC/Norris Comprehensive Cancer Center, Los Angeles, CA In the future, we will invariably see much more on germline polymorphisms and their application in clinical oncology. One reason for such excitement is that germline polymorphisms can be easily assayed from DNA in readily available normal tissue, such as peripheral blood, thus precluding the need for tumor tissue. Moreover, the methods for such assays are relatively simple, as well as economical. But, what are germline polymorphisms and why do we think that they may be important in cancer patients? Germline polymorphisms are inherited genetic variants that are present in all cells of the body. About 1.4 million such variants in differing frequencies have been identified thus far. In most cases, variations present as single nucleotide changes, as in the case of XPD and XRCC-1 genes. In other genes, they can take place as sequence variations or deletions. In the thymidylate synthase (TS) gene, for instance, polymorphic changes can take place as 28-bp sequence tandem repeats (two to even nine repeats) in the promoter region, or a 6-bp sequence deletion in the 3'-UTR of the gene. Interestingly, a potentially functional single nucleotide change variant within the tandem repeat of the TS gene has recently been described (a polymorphism within a polymorphism). In the case of the UGT1A1 gene, a common polymorphism leads to an additional two base pair repeat (TA) repeat in the TATA sequence of the UGT1A1 promoter, in addition to single nucleotide changes. In many cases, the functional significance of these polymorphisms is unknown. Depending on their location along the gene sequence, polymorphisms may affect gene transcription, translation, mRNA stability, protein activity, or do nothing. If gene expression or protein activity of these enzymes plays a critical role in the efficacy or toxicity to a given agent—via impact on metabolism, influx, and efflux of anticancer-drugs, processes influencing cell viability such as DNA-repair mechanisms, and so on—then their respective gene polymorphisms may possess predictive or prognostic value for clinical outcome. Moreover, multiple polymorphisms can exist within a single gene, be in linkage disequilibrium, as well as interact with each other. Pharmacogenetic variability in drug metabolizing enzyme systems is a major determinant of variations in predicting the outcome of therapy, in terms of both tumor response and host toxicity. Depending on whether the germline polymorphism is causing an increase or decrease function of the metabolizing protein, the patients receiving chemotherapy may be undertreated with no toxicity or overtreated with excessive toxicity. For instance, we know that patients with Gilbert's disease (associated with UGT1A1 polymorphism) will have excessive toxicity with the recommended dose of irinotecan due to impaired ability of detoxifying the active metabolite of the drug. Thus, the goal of pharmacogenetic screening before chemotherapy would be to identify patients such as these who may have increased toxicity. This is of particular importance in cancer chemotherapy, as these agents generally have a narrow margin of safety. Important limitations for the potential use of genotypic analysis as predictors of clinical outcome must be acknowledged. First, a genotype represents a static value unable to change in response to a new situation, such as exposure to chemotherapy. Also, it may not reflect changes in tumor DNA, such as loss of heterozygosity, thus limiting its predictive strength. For example, loss of heterozygosity of 18p in the TS gene contains tandem repeats. Thus, analysis of tumor DNA still has an important role for attaining more accurate information in regards to expression or function. A number of promising pharmacogenetic candidates for prediction of chemotherapy efficacy and toxicity, as well as survival, have been identified, especially in patients with gastrointestinal malignancies (Table 1).1 Most have been evaluated retrospectively, based on a relatively small number of patients. The differences in drug effects among variant genotypes have often shown to be dramatic, as in the case of DPD and UGT1A1. Thus, the inclusion of analyses of known functional genetic polymorphisms that may impact on both efficacy and toxicity of the agents we use should become standard procedure in future clinical studies, and hopefully clinical practice.
The rapid growth of therapeutic choices and strategies in the management of cancer is allowing effective alternative treatment options. This makes the incorporation of pharmacogenomic studies into clinical trials, with the goal of potential tailoring of chemotherapy, a more urgent and fulfilling task. Although a large amount of new data on relationships between clinical outcome of current therapeutic strategies and interindividual genetic markers are being published, the clinical efficacy and toxicity for the individual patient of particular chemotherapy agents are still unpredictable. We are just beginning to use our knowledge of the genome and the newest high through-put technology to acquire more information to develop individualized therapies. The crucial next step will be the rapid transfer of these newly acquired data from the research laboratory into the clinic, with prospective clinical trials to confirm the results of pharmacogenetic testing. The paper by Gurubhagavatula et al2 describes the association of XPD and XRCC-1 polymorphisms with overall survival in 103 of 112 patients with advanced non–small-cell lung cancer (NSCLC), stages IIIA-IV, treated with platinum-based chemotherapy at a single institution. This is a subset of a much larger cohort of NSCLC patients enrolled in a molecular epidemiologic study. The majority received platinum as first-line therapy. Not surprisingly, significantly more patients with stage III disease received radiation compared with patients with stage IV disease (who received mostly palliative radiation). This retrospective analysis shows that XPD codon 312 and XRCC-1 codon 399 variants were correlated with overall survival in their cohort of patients, even after adjusting for stage and performance status. The paper demonstrates that genomic polymorphisms in DNA repair genes may play a role in clinical outcome in patients with NSCLC. Interestingly, both genes have been shown to be associated with clinical outcome with colorectal cancer. This article is a wonderful example of the potential, limitations, frustrations, and need for a better functional understanding in our quest to elucidate the role of germline polymorphisms in clinical oncology. The retrospective nature of this study places obvious limitations in regards to availability/reliability of clinical response data or time to progression, as acknowledged by the authors. The interpretation of the findings is further complicated by significant treatment differences between subgroups; for example, some patients received radiation therapy, although such treatment did not seem to account for the reported associations. One remains unsure whether the survival differences seen are as a result of differential response to platinum, radiation, or tumor aggressiveness. Given the paucity of nonplatinum treated patients with advanced NSCLC, an alternative strategy could be to perform genotypic analyses among early stage patients in the larger cohort and their time to recurrence, which could yield some clues regarding tumor aggressiveness, unless they also received adjuvant platinum therapy. Based on our knowledge of germline polymorphisms in these two genes, three questions arise: Why did the authors examine only one genomic polymorphism in each gene? Why only two genes, when there is an extensive pathway of base and nucleotide excision repair known? And did the findings concur with the functional understanding of these polymorphisms? Though the authors allude to the need for inclusion of additional genes (pathway approach), the answers to these questions is unclear. In the article by Gurubhagavatula et al,2 the candidate approach seems to have been used; an approach that is hypothesis-driven and focused on only one genomic polymorphism in the each of the genes studied.3 This approach does not account for a number of additional known polymorphisms with functional significance, as well as others where their functionality is unclear. If the gene of interest is considered a critical gene, it should ideally be comprehensively studied, understanding the complete sequence of the gene and all its potentially relevance polymorphisms, such as XPD-156, 312, 7513-5 and XRCC1-194, 280, 3996-8). Studies focusing on a single polymorphism as a predictive or prognostic marker have been promising, but at times have led to conflicting data. Often, these polymorphic variants are associated with small changes in enzyme activity compared to the wild type. Thus, it would be unusual that a single polymorphic change turns out to be a significant predictive marker of efficacy or toxicity. In the future, a combination of polymorphisms of genes in metabolic and biologic pathways may become more powerful predictors of response, survival, and toxicity, especially in view of combination therapies. DNA repair is a classical example because nucleotide and base excision is a symphony of enzymes working together to repair DNA defects. Enzymes such as XPD, ERCC-1, XPA, XRCC-1, and others all depend on each others' function; alterations of any of these genes can affect DNA repair capacity in a clinically relevant level. Establishing base excision and nucleotide excision repair as the biologic pathway of platinum sensitivity, all genes involved in this pathway should be investigated. XRCC-1 and XPD are only two among many. The limitations of the candidate polymorphism approach are that it is more likely helpful in the clinical setting and is based on specific polymorphic genes, assuming a functional significant polymorphism. The pathway approach takes advantage of association between genes within a metabolic and biologic pathway, and is hypothesis-driven. However, it can be limited by the understanding of the biochemical function of the genes in the pathway and their interaction. Moreover, data analyses are much more complex and require larger sample sizes. Recently, a new statistical method has been developed to address the pathway approach using genomic polymorphisms, which will help to test these polymorphisms in clinical trials.9 In the future, a comprehensive genotyping (haplotyping) together with a functional understanding of the polymorphic variants are essential in using and developing genomic polymorphism as possible predictive and prognostic makers in clinical oncology. Data, such as those from Gurubhagavatula et al,2 provide important clues and directions for future studies. These results should be validated in prospective clinical trials with all relevant genes encompassing metabolic and biologic pathways. In the future, identification of functional polymorphisms associated with clinical outcome and toxicity will complement data from gene expression arrays as well as the proteinomic approaches. A unique, yet untapped potential use of functional polymorphisms is the identification of new regulatory mechanisms of genes critical for the efficacy of chemotherapeutic agents. The identification of new transcription factors or genomic variations changing the structure/function of a critical protein may lead to potential discovery of novel targets, and the development of more efficient and less toxic regimens. Author's Disclosures of Potential Conflicts of Interest The following authors or their immediate family members have indicated a financial interest. No conflict exists for drugs or devices used in a study if they are not being evaluated as part of the investigation. Acted as a consultant within the last 2 years: Heinz-Josef Lenz, Chiron, Response- Genetics, Genentech, Pfizer, Roche, Novartis, Sanofi. Performed contract work within the last 2 years: Heinz-Josef Lenz, National Cancer Institute, Southwestern Oncology Group. Received more than $2,000 a year from a company for either of the last 2 years: Heinz-Josef Lenz, Roche, Sanofi, BMS, ImClone, Genentech, Pfizer, Novartis, H3 Pharma.
REFERENCES 1. Park D, Stoehlmacher J, Lenz H-J: Tailoring chemotherapy in advanced colorectal cancer. Curr Opin Pharmacol 3:378-385, 2003[CrossRef][Medline]
2. Gurubhagavatula S, Liu G, Park W, et al: XPD and XRCC-1 genetic polymorphisms are prognostic factors in advanced non–small-cell lung cancer patients treated with platinum chemotherapy. J Clin Oncol 22:2594-2601, 2004
3. Park DJ, Stoehlmacher J, Zhang W, et al: A Xeroderma pigmentosum group D gene polymorphism predicts clinical outcome to platinum-based chemotherapy in patients with advanced colorectal cancer. Cancer Res 61:8654-8658, 2001 4. Lunn RM, Helzlsouer KJ, Parshad R, et al: XPD polymorphisms: Effects on DNA repair proficiency. Carcinogenesis 21:5551-5555, 2000
5. Spitz MR, Wu X, Wang Y, et al: Modulation of nucleotide excision repair capacity by XPD polymorphisms in lung cancer patients. Cancer Res 61:1354-1357, 2001 6. Siciliano MJ, Carrano AV, Thompson LH: Assignment of a human DNA-repair gene associated with sister chromatid exchange to chromosome 19. Mutat Res 174:303-308, 1986[CrossRef][Medline]
7. Thompson LH, Broockman KW, Jones NJ, et al: Molecular cloning of the human XRCC1 gene, which corrects defective DNA strand break repair and sister chromatid exchange. Mol Cell Biol 10:6160-6171, 1990 8. Stoehlmacher J, Ghaderi V, Iqbal S, et al: A polymorphism of the XRCC1 gene predicts for response to platinum based treatment in advanced colorectal cancer. Anticancer Res 21:3075-3080, 2001[Medline] 9. Conti DV, Cortessis V, Molitor J, et al: Bayesian modeling of complex metabolic pathways. Hum Hered 56:83-93, 2003[Medline]
Related Article
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2004 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
