|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
Originally published as JCO Early Release 10.1200/JCO.2005.03.6640 on June 19 2006 © 2006 American Society of Clinical Oncology. Genetic Variations in Radiation and Chemotherapy Drug Action Pathways Predict Clinical Outcomes in Esophageal Cancer
From the Departments of Epidemiology, Pathology, Thoracic and Cardiovascular Surgery, Radiation Oncology, Experimental Radiation Oncology, Experimental Therapeutics, GI Medical Oncology, The University of Texas M.D. Anderson Cancer Center, Houston, TX Address reprint requests to Xifeng Wu, MD, PhD, Department of Epidemiology, Unit 1340, The University of Texas M.D. Anderson Cancer Center, 1155 Pressler Blvd, Houston, TX 77030; e-mail: xwu{at}mdanderson.org
Purpose: Understanding how specific genetic variants modify drug action pathways may provide informative blueprints for individualized chemotherapy. Methods: We applied a pathway-based approach to examine the impact of a comprehensive panel of genetic polymorphisms on clinical outcomes in 210 esophageal cancer patients. Results: In the Cox proportional hazards model, MTHFR Glu429Ala variant genotypes were associated with significantly improved survival (hazard ratio [HR] = 0.56; 95% CI, 0.35 to 0.89) in patients treated with fluorouracil (FU). The 3-year survival rates for patients with the variant genotypes and the wild genotypes were 65.26% and 46.43%, respectively. Joint analysis of five polymorphisms in three FU pathway genes showed a significant trend for reduced recurrence risk and longer recurrence-free survival as the number of adverse alleles decreased (P = .004). For patients receiving platinum drugs, the MDR1 C3435T variant allele was associated with significantly reduced recurrence risk (HR = 0.25; 95% CI, 0.10 to 0.64) and improved survival (HR = 0.44; 95% CI, 0.23 to 0.85). In nucleotide excision repair genes, there was a significant trend for a decreasing risk of death with a decreasing number of high-risk alleles (P for trend = .0008). In base excision repair genes, the variant alleles of XRCC1 Arg399Gln were significantly associated with the absence of pathologic complete response (odds ratio = 2.75; 95% CI, 1.14 to 6.12) and poor survival (HR = 1.92; 95% CI, 1.00 to 3.72). Conclusion: Several biologically plausible associations between individual single nucleotide polymorphisms and clinical outcomes were found. Our data also strongly suggest that combined pathway-based analysis may provide valuable prognostic markers of clinical outcomes.
Esophageal cancer is a highly aggressive malignancy with approximately 14,520 new cases and 13,570 deaths expected in the United States in 2005.1 More than 50% of esophageal cancer patients present with unresectable or metastatic disease, contributing to its dismal overall 5-year survival rate of 14% in the United States.2 Because response rate after standard treatment remains low, basic and translational research to develop new targets for therapy to improve the outcome of these patients is vital.3 Surgery is the most common treatment, yet even after resection, prognosis remains poor.4 As a result, interest has centered on cisplatin- and fluorouracil (FU) -based chemoradiotherapy as the cornerstone of treatment for localized esophageal carcinoma. Chemoradiotherapy can be used as the primary treatment in lieu of surgery or as a preoperative procedure to shrink the tumor, especially if tumor size or location would complicate surgery. Several phase II trials have suggested improved locoregional control and survival,5,6 but the overall picture emerging from the literature remains unclear.7,8 Therefore, identification of patients who will have better clinical outcome after preoperative chemoradiotherapy would allow us to maximize therapeutic benefit and minimize toxicities. Because most clinical parameters cannot successfully predict clinical outcomes, identifying biologic predictors is warranted. Factors that influence drug response include age, sex, ethnicity, comorbid conditions, drug interactions, and genetics.9 The expression levels of critical genes related to drug response also affect clinical outcomes. For example, the mRNA levels of ERCC1, a DNA repair gene, and thymidylate synthase (TS), a target gene of FU, were shown to predict response and survival for gastric cancer patients receiving combination cisplatin and FU therapy.10,11 The recent explosion of literature in pharmacogenetics12,13 suggests that genetic polymorphisms in genes involved in drug metabolism, drug targets, and DNA repair may contribute significantly to the variability of drug response. Most pharmacogenetic studies have used the candidate gene approach, investigating the effects of only one or a few single nucleotide polymorphisms (SNPs) in a specific gene at a time, such as several studies showing that polymorphisms in ERCC1 and TS were potential prognostic factors for chemotherapy (cisplatin and/or FU) response and survival.14-16 The candidate gene approach is hypothesis driven, uses a priori knowledge of SNP and gene functions, and has yielded informative but often conflicting data. Because chemotherapy drugs exert their effects through a multistep, multigenic cascade, it is unlikely that any one single genetic polymorphism would have such a dramatic effect that could serve as a sole predictive marker for response. In the most likely clinical application, a combination of polymorphisms in genes involved in multiple steps of drug actions would be used as potent predictors of clinical outcomes. In this study, we applied a pathway-based approach to examine the impact of genetic polymorphisms on esophageal cancer treatment outcomes. We focused on patients who received platinum-/FU-based preoperative chemoradiotherapy. We selected a comprehensive panel of SNPs in the major genes involved in drug metabolism, drug disposition, nucleotide excision repair (NER), base excision repair (BER), double strand break (DSB) repair, and folate metabolism, and evaluated the associations of these polymorphisms with clinical outcomes (recurrence and survival) individually and jointly.
Patients This retrospective study included 210 histologically confirmed esophageal cancer patients recruited from The University of Texas M.D. Anderson Cancer Center (Houston, TX) between 1985 and 2003. Diagnosed with resectable adenocarcinoma or squamous cell carcinoma of the esophagus or gastroesophageal junction, these patients were treated with preoperative chemoradiotherapy followed by esophagectomy.
Clinical Data Collection
Genotyping
Statistical Analysis
Characteristics of Patients The median age of the 210 patients was 61 years (range, 32-79 years). There were 182 men (86.67%) and 28 women (13.33%). One hundred ninety patients (90%) were white. Histology distribution was 83% adenocarcinoma (174 cases) and 17% squamous cell carcinoma (36 cases). Fifty-eight percent of patients (123 cases) presented with stage IIA (107 cases) or IIB (16 cases) disease. The remainder presented with stage III (75 cases; 36%) and stage IVA or IVB (12 cases; 6%). Induction chemotherapy followed by chemoradiotherapy was administered to 92 patients (44%), and the remaining 118 (56%) received only preoperative chemoradiotherapy. Among chemotherapy drugs, cisplatin was administered to 146 patients, FU to 201 patients, and paclitaxel to 107 patients. The median follow-up time was 18.6 months. There were 101 deaths (48.1%) and 67 recurrences (31.9%). Overall median survival was 34.8 months.
Associations of Individual SNPs With Clinical Outcomes
Cisplatin Pathway Genes Reduced drug uptake, drug inactivation, increased DNA repair, and disruption of apoptotic pathways all contribute to cisplatin resistance.18 Numerous genes are involved in these pathways, and SNPs in these genes are potential modulators of cisplatin efficacy and toxicity (Table 2).
MDR1 Genes. MDR1 encodes P-glycoprotein (P-gp), a member of the ABC superfamily of membrane transporters. P-gp is involved in the active transport of xenobiotics and many chemotherapy drugs. Genotyping MDR1 Ser892Ala and Ile1144Ile (C3435T), we found that the variant alleles (CT + TT) of C3435T were associated with reduced recurrence rate (HR = 0.25; 95% CI, 0.10 to 0.64) and improved survival (HR = 0.44; 95% CI, 0.23 to 0.85; Table 2). In Kaplan-Meier analysis, individuals with the wild-type genotype (CC) had an MST of 16.77 months, whereas individuals with heterozygous and homozygous variants had median survival times of 29.3 months and 42.3 months, respectively (P = .0296; Fig 2A). When we combined these two MDR1 SNPs, compared with individuals with the Ser892Ala variant alleles (GT + TT) and C3435T high-risk alleles (CC), individuals with the wild-type Ser892Ala allele (GG) and the low-risk C3435T allele (CT + TT) had a significantly reduced recurrence risk (HR = 0.11; 95% CI, 0.02 to 0.59) and better survival (HR = 0.24; 95% CI, 0.06 to 1.03).
Detoxifying Enzyme Genes. GSTP1 is the primary enzyme responsible for the detoxification of platinum agents. Patients who had the variant Ala114Val alleles (CT) had significantly increased recurrence rate (HR = 2.19; 95% CI, 1.00 to 4.83) and worse survival (HR = 2.10; 95% CI, 1.14 to 3.89) compared with patients with the wild-type genotype. In Kaplan-Meier analysis, individuals with the variant allele had significantly shorter recurrence-free survival (P = .0058) and overall survival time (P = .0289; data not shown) compared with those with the wild-type allele. Myeloperoxidase (MPO) is a metabolic/oxidative enzyme that may be involved in the detoxification of cisplatin through the formation of hypochlorous acid and other reactive oxygen species. Carriers of the variant alleles (CT + CC) of MPO T-764C, a promoter SNP, showed a 1.57-fold increased risk of death (95% CI, 0.89 to 2.75) and a 2.21-fold increased risk of recurrence (95% CI, 1.31 to 6.81). Joint Analysis of Cisplatin Pathway Genes and Outcomes. Tallying the number of variant alleles from cisplatin-related SNPs, we analyzed the associations between total number of putative risk alleles and clinical outcomes. There were significant trends for increasing recurrence risk and decreasing survival with increasing number of putative high-risk alleles (Table 2). For overall survival, compared with individuals with 10 or more putative-risk alleles, individuals with eight to nine risk alleles and those with seven or fewer risk alleles had HRs of 0.62 (95% CI, 0.33 to 1.17) and 0.41 (95% CI, 0.19 to 0.88), respectively (P for trend = .019). NER Genes. NER is the major cellular system for repairing platinum-induced DNA adducts. No significant individual associations with clinical outcomes were found. However, there was a significant trend for better survival with decreasing number of putative high-risk alleles (Table 3). Compared with individuals with five or more putative-risk alleles, individuals with four risk alleles (HR = 0.70; 95% CI, 0.29 to 1.67) and individuals with three or fewer risk alleles (HR = 0.35; 95% CI, 0.16 to 0.73) had a significantly reduced risk of death (P for trend = .008).
Radiation Pathway Genes We genotyped four widely studied SNPs (hOGG1 Ser326Cys, APEX1 Asp148Glu, ADPRT Val762Ala, and XRCC1 Gln399Arg) in the BER pathway and a panel of SNPs in DSB repair genes (data not shown). For XRCC1 Arg399Gln, we found that individuals with the variant alleles (AA + GA) had a significantly increased risk of death (HR = 1.92; 95% CI, 1.00 to 3.72) compared with individuals with the wild type genotype (GG). In Kaplan-Meier analysis, patients with the GG allele had an MST of 57.4 months, whereas the GA and AA genotypes were associated with MSTs of 22.9 months and 13.7 months, respectively (log-rank P = .0002; Fig 2B). Since radiation is more relevant to pathologic response, we also evaluated pathologic complete response (pCR) as an outcome variable. Among NER, BER, and DSB repair gene SNPs, only XRCC1 Arg399Gln was significantly associated with pCR. Individuals with the variant alleles (AA + GA) had significantly worse pCR (odds ratio = 2.75; 95% CI, 1.14 to 6.12) compared with wild-type individuals (GG; data not shown).
Multiple Testing Using Benjamini-Hochberg Method
The two most significant findings of this study are that (1) there were biologically plausible associations between certain individual polymorphisms and clinical outcomes that support the potential of genetic profiles to predict clinical outcomes, such as time to recurrence and survival, and (2) combined analyses of two or more related SNPs reveal otherwise undetectable effects of individual SNPs on clinical outcome, highlighting the need for the pathway-based approach in association studies. Using the candidate gene approach, thousands of studies have reported associations between genetic polymorphisms and disease susceptibility. However, given the limited impact of single polymorphisms on complex etiologies, as well as the relatively small sample sizes of most previous studies, it has been estimated that only approximately a quarter of previously reported associations were real positives.19,20 A similar theoretical problem besieges the multitude of recent publications reporting associations between genetic polymorphisms and clinical cancer outcomes, which are far too complex to be significantly impacted by any one gene and any one SNP.14,21-23 The pathway-based approach applied in this study is hypothesis-driven and uses a priori knowledge of potentially functional SNPs and gene functions. Analyzing each SNP individually and jointly with other alleles in its pathway, we found biologically plausible associations with outcomes. Furthermore, by evaluating the combined impact of multiple polymorphisms, we were able to identify minor associations that would not have been detected with the candidate gene approach. We found protective roles of the variant alleles of both MTHFR C677T and A1298C SNPs in recurrence and survival; moreover, patients with variant alleles at both loci had a significantly reduced recurrence risk and better survival. The improved clinical outcomes conferred by MTHFR variant genotypes are biologically plausible and consistent with literature. The variant alleles of both SNPs result in reduced MTHFR enzymatic activity.24,25 Reduced MTHFR activity increases the amount of folate—specifically 5,10-MTHF, a substrate for both TS and MTHFR. 5,10-MTHF forms a tertiary complex with TS and FU, enhancing the action of FU. Animal experiments have suggested that the efficacy of FU is enhanced when 5,10-MTHF is administered.26 A recent meta-analysis of 3,300 patients with advanced colorectal cancer randomly assigned in 19 clinical trials concluded that the combination of folate with FU results in significantly higher tumor response rate and overall survival compared with FU alone.27 Our results, showing that these two MTHFR SNPs conferred better outcomes, are therefore biologically plausible.
Analyzing cisplatin pathway genes, we found a significant association between the variant GSTP1 Ala114Val genotype and increased risks of recurrence and death. GSTP1 codes GST-pi (
Although MDR1 is not a major transporter for platinum drugs, a previous esophageal cancer study suggested that MDR1 had prognostic value in patients receiving cisplatin based chemoradiotherapy.33 Harpole et al showed that high expression levels of GST- We also found significant individual associations between XRCC1 Arg399Gln and clinical outcomes. Patients with the variant genotypes had a significantly worse pCR, increased risk of death, and shorter survival times (Fig 2B). Two previous studies made similar observations in lung cancer and colorectal cancer patients receiving platinum-based chemotherapy.22,37 Biologically, XRCC1 mainly repairs single-base damage. Previous reports suggested that the variant allele (Gln) results in deficient DNA repair and increased DNA and chromosome damage.38,39 The worse pCR and survival conferred by the variant allele may be due to increased genetic instability and heterogeneity of tumors, because ample evidence has linked chromosome abnormality with poor prognosis.40 Interestingly, XRCC1 Arg399Gln polymorphism may also predict overall survival among patients who had pCR. This could have important clinical implication, because approximately 30% to 40% of patients still had poor survival despite complete pathologic response. These patients may already have occult metastasis due to the unfavorable tumor characteristics. Due to the limited number of minorities in our study, we restricted our analyses to white patients. It is well known that the allele frequencies of many SNPs vary among different populations. It would be interesting to test these polymorphisms in minority patients if the sample size is large enough. Due to the exploratory nature of this study, we reported only nominal statistical associations. Despite the strong biologic plausibility and consistency with literature from several individual associations as discussed herein, some of these associations may be false positives as a result of the inherent pitfalls of the candidate gene approach. As an initial attempt to address multiple comparisons and the false-positive issue, we estimated the FDR using the bum function in S-PLUS. The FDR of 26% was relatively high. However, several of the individual associations remained significant after FDR-adjustment at the 5% alpha level, and the majority of individual findings remained significant after FDR-adjusting at the 15% alpha level. In addition, the combined analyses of multiple variant alleles in the same pathway should be influenced less by multiple testing than by individual analyses. The individual associations reported in this article should be interpreted with caution, and larger studies are warranted to confirm these individual findings. Nevertheless, even for those true associations, it is unlikely that any individual SNP would have sufficient power to predict clinical outcomes in a disease as complex as cancer. Results from our study highlight the potential of the pathway-based approach as a more powerful way to identify clinical outcome predictors. Combined analyses of two or more SNPs in the same pathway showed that the magnitude of impact correlated with decreasing number of putative risk alleles. These joint effects were much greater than the impacts of individual SNPs. We did not have sufficient statistical power to analyze the effect of each polymorphism on treatment response because of the limited number of responders for individual treatments (FU or cisplatin). Therefore, whether the markers identified in this study are merely global indicators of clinical outcome or specific treatment response indicators needs to be explored in larger studies. Although we limited our analyses of each polymorphism to patients receiving either FU or cisplatin depending on the hypothesized involvement of the gene in the distinct drug pathway, we could not conclude that these markers affect only patients receiving respective drugs. The analyses of the same markers in noncisplatin or non–FU-treated patients will address this question, but we did not have enough patients in these categories in this study. Nevertheless, the results of this study may provide a glimpse of future individualized therapy based on the genomic profile of each individual. For example, a patient-specific chemotherapeutic regimen may be possible on the basis of the patient’s make-up of genetic markers. For patients with potential resistance to cisplatin, a noncisplatin treatment may be warranted. However, before we can design any such clinical trials based on genomic profiles, results from this hypothesis-generating retrospective study cannot be overinterpreted and need to be verified in larger, independent prospective studies. Moreover, the magnitude of reported associations in this study, individually and jointly, is far from clinically applicable. It is unrealistic to exhaustively genotype all identified SNPs with potential functional significance in all genes involved in the many pathways of drug action. The future of cancer pharmacogenetics may have to move more toward pharmacogenomics, using a whole-genome approach. The completion of the human haplotype maps may be a key step to achieving individualized cancer therapy.
The authors indicated no potential conflicts of interest.
published online ahead of print at www.jco.org on June 19, 2006. Supported by National Cancer Institute (National Institutes of Health) Grants No. CA74880 and CA91846, and an M.D. Anderson Cancer Center multidisciplinary research program grant for esophageal cancer. Terms in blue are defined in the glossary, found at the end of this article and online at www.jco.org. Authors’ disclosures of potential conflicts of interest and author contributions are found at the end of this article.
1. Jemal A, Murray T, Ward E, et al: Cancer statistics, 2005. CA Cancer J Clin 55:10-30, 2005 2. Enzinger PC, Mayer RJ: Esophageal cancer. N Engl J Med 349:2241-2251, 2003 3. Polednak AP: Survival of US black and white patients with squamous cell cancer of the esophagus. J Natl Med Assoc 96:87-92, 2004[Medline] 4. Lerut T, Coosemans W, De Leyn P, et al: Treatment of esophageal carcinoma. Chest 116:463S-465S, 1999 5. Heath EI, Burtness BA, Heitmiller RF, et al: Phase II evaluation of preoperative chemoradiation and postoperative adjuvant chemotherapy for squamous cell and adenocarcinoma of the esophagus. J Clin Oncol 18:868-876, 2000 6. Stewart J, Hoff R, Johnson SJ: Improved survival with neoadjuvant therapy and resection for adenocarcinoma of the esophagus. Ann Surg 218:571-576, 1993[Medline] 7. Brenner B, Ilson DH, Minsky BD: Treatment of localized esophageal cancer. Semin Oncol 31:554-565, 2004[CrossRef][Medline] 8. Minsky BD: Combined modality therapy for esophageal cancer. Semin Oncol 30:46-55, 2003[Medline] 9. Evans WE, Relling MV: Pharmacogenomics: Translating functional genomics into rational therapeutics. Science 286:487-491, 1999 10. Metzger R, Leichman CG, Danenberg KD, et al: ERCC1 mRNA levels complement thymidylate synthase mRNA levels in predicting response and survival for gastric cancer patients receiving combination cisplatin and fluorouracil chemotherapy. J Clin Oncol 16:309-316, 1998 11. Lenz HJ, Leichman CG, Danenberg KD, et al: Thymidylate synthase mRNA level in adenocarcinoma of the stomach: A predictor for primary tumor response and overall survival. J Clin Oncol 14:176-182, 1996[Abstract] 12. Ulrich CM, Robien K, McLeod HL: Cancer pharmacogenetics: Polymorphisms, pathways and beyond. Nat Rev Cancer 3:912-920, 2003[CrossRef][Medline] 13. Weinshilboum R: Inheritance and drug response. N Engl J Med 348:529-537, 2003 14. Zhou W, Gurubhagavatula S, Liu G, et al: Excision repair cross-complementation group 1 polymorphism predicts overall survival in advanced non-small cell lung cancer patients treated with platinum-based chemotherapy. Clin Cancer Res 10:4939-4943, 2004 15. Pullarkat ST, Stoehlmacher J, Ghaderi V, et al: Thymidylate synthase gene polymorphism determines response and toxicity of 5-FU chemotherapy. Pharmacogenomics J 1:65-70, 2001[Medline] 16. Villafranca E, Okruzhnov Y, Dominguez MA, et al: Polymorphisms of the repeated sequences in the enhancer region of the thymidylate synthase gene promoter may predict downstaging after preoperative chemoradiation in rectal cancer. J Clin Oncol 19:1779-1786, 2001 17. Benjamini Y, Hochberg Y: Controlling the false discovery rate: A practical and powerful approach to multiple testing. J Royal Stat Soc Ser B 57:289-300, 1995 18. Siddik ZH: Cisplatin: Mode of cytotoxic action and molecular basis of resistance. Oncogene 22:7265-7279, 2003[CrossRef][Medline] 19. Ioannidis JP, Ntzani EE, Trikalinos TA, et al: Replication validity of genetic association studies. Nat Genet 29:306-309, 2001[CrossRef][Medline] 20. Lohmueller KE, Pearce CL, Pike M, et al: Meta-analysis of genetic association studies supports a contribution of common variants to susceptibility to common disease. Nat Genet 33:177-182, 2003[CrossRef][Medline] 21. Nagasubramanian R, Innocenti F, Ratain MJ: Pharmacogenetics in cancer treatment. Annu Rev Med 54:437-452, 2003[CrossRef][Medline] 22. Gurubhagavatula S, Liu G, Park S, et al: XPD and XRCC1 genetic polymorphisms are prognostic factors in advanced non-small-cell lung cancer patients treated with platinum chemotherapy. J Clin Oncol 22:2594-2601, 2004 23. Stoehlmacher J, Park DJ, Zhang W, et al: A multivariate analysis of genomic polymorphisms: Prediction of clinical outcome to 5-FU/oxaliplatin combination chemotherapy in refractory colorectal cancer. Br J Cancer 91:344-354, 2004[Medline] 24. Frosst P, Blom HJ, Milos R, et al: A candidate genetic risk factor for vascular disease: A common mutation in methylenetetrahydrofolate reductase. Nat Genet 10:111-113, 1995[CrossRef][Medline] 25. Weisberg I, Tran P, Christensen B, et al: A second genetic polymorphism in methylenetetrahydrofolate reductase (MTHFR) associated with decreased enzyme activity. Mol Genet Metab 64:169-172, 1998[CrossRef][Medline] 26. Carlsson G, Hafstrom LO, Spears CP, et al: 5-fluorouracil (5-FU) and 5, 10-methylene tetrahydrofolate (5, 10-CH2FH4) as adjuvant therapy in an experimental rodent colon carcinoma model. Anticancer Res 17:3671-3674, 1997[Medline] 27. The Meta-Analysis Group in Cancer: Modulation of fluorouracil by leucovorin in patients with advanced colorectal cancer: An updated meta-analysis. J Clin Oncol 22:3766-3775, 2004 28. Ban N, Takahashi Y, Takayama T, et al: Transfection of glutathione S-transferase (GST)-pi antisense complementary DNA increases the sensitivity of a colon cancer cell line to adriamycin, cisplatin, melphalan, and etoposide. Cancer Res 56:3577-3582, 1996 29. Goto S, Iida T, Cho S, et al: Overexpression of glutathione S-transferase pi enhances the adduct formation of cisplatin with glutathione in human cancer cells. Free Radic Res 31:549-558, 1999[Medline] 30. Shiga H, Heath EI, Rasmussen AA, et al: Prognostic value of p53, glutathione S-transferase pi, and thymidylate synthase for neoadjuvant cisplatin-based chemotherapy in head and neck cancer. Clin Cancer Res 5:4097-4104, 1999 31. Piedrafita FJ, Molander RB, Vansant G, et al: An alu element in the myeloperoxidase promoter contains a composite SP1-thyroid hormone-retinoic acid response element. J Biol Chem 271:14412-14420, 1996 32. Van Schooten FJ, Boots AW, Knaapen AM, et al: Myeloperoxidase (MPO) –463G 33. Harpole DH Jr, Moore MB, Herndon JE II, et al: The prognostic value of molecular marker analysis in patients treated with trimodality therapy for esophageal cancer. Clin Cancer Res 7:562-569, 2001 34. Drescher S, Schaeffeler E, Hitzl M, et al: MDR1 gene polymorphisms and disposition of the P-glycoprotein substrate fexofenadine. Br J Clin Pharmacol 53:526-534, 2002[CrossRef][Medline] 35. Hoffmeyer S, Burk O, von Richter O, et al: Functional polymorphisms of the human multidrug-resistance gene: Multiple sequence variations and correlation of one allele with P-glycoprotein expression and activity in vivo. Proc Natl Acad Sci U S A 97:3473-3478, 2000 36. Hitzl M, Drescher S, van der Kuip H, et al: The C3435T mutation in the human MDR1 gene is associated with altered efflux of the P-glycoprotein substrate rhodamine 123 from CD56+ natural killer cells. Pharmacogenetics 11:293-298, 2001[CrossRef][Medline] 37. Stoehlmacher J, Ghaderi V, Iobal S, et al: A polymorphism of the XRCC1 gene predicts for response to platinum based treatment in advanced colorectal cancer. Anticancer Res 21:3075-3079, 2001[Medline] 38. Lunn RM, Langlois RG, Hsieh LL, et al: XRCC1 polymorphisms: Effects on aflatoxin B1-DNA adducts and glycophorin A variant frequency. Cancer Res 59:2557-2561, 1999 39. Duell EJ, Wiencke JK, Cheng TJ, et al: Polymorphisms in the DNA repair genes XRCC1 and ERCC2 and biomarkers of DNA damage in human blood mononuclear cells. Carcinogenesis 21:965-971, 2000 40. Albertson DG, Collins C, McCormick F, et al: Chromosome aberrations in solid tumors. Nat Genet 34:369-376, 2003[CrossRef][Medline] Submitted August 1, 2005; accepted March 24, 2006. This article has been cited by other articles:
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
TOP
ABSTRACT
INTRODUCTION
2 and Fisher’s exact tests. Applying the Cox proportional hazard model with the study end point event set as the date of the patient’s first recurrence or overall death, we estimated the hazard ratios (HRs) associated with genetic polymorphisms. If a patient was lost to follow-up or if the study ended before recurrence or death, the outcome variable was flagged as censored. HRs were estimated by fitting the Cox model while adjusting for age, sex, radiation dosage, chemoradiotherapy sequence, clinical stage, chemotherapy regimens, histology, tumor location, pathology, and histologic viability. The Kaplan-Meier survival function and log-rank tests were used to assess clinical outcome in relation to individual polymorphisms. In addition, we evaluated risk as a function of number of risk alleles involved in each pathway. For this analysis, risk was trichotomized into low, medium, and high categories. Because of the small number of minority patients, all analyses were restricted to white patients. The above statistical analyses were completed with the STATA software (version 8, College Station, TX). We used the bum function in S-PLUS (Insightful Corp, Seattle, WA) to estimate the false-discovery rate (FDR). The Benjamini-Hochberg method was used to calculate the FDR-adjusted P values.
), which is actively involved in the detoxification of cisplatin
(alpha), µ (mu), 
(theta). 
A reduces MPO activity and DNA adduct levels in bronchoalveolar lavages of smokers. Cancer Epidemiol Biomarkers Prev 13:828-833, 2004
