Nucleotide Excision Repair Gene Polymorphisms and Recurrence After Treatment for Superficial Bladder Cancer

1408 Vol. 11, 1408–1415, February 15, 2005 Clinical Cancer Research

Nucleotide Excision Repair Gene Polymorphisms and Recurrence after Treatment for Superficial Bladder Cancer

Jian Gu,1 Hua Zhao,1 Colin P. Dinney,2 Yong Zhu,3 fewer putative high-risk alleles as the reference group, Dan Leibovici,2 Carlos E. Bermejo,2 individuals with six to seven risk alleles and individuals with eight or more risk alleles had higher recurrence risks, with H. Barton Grossman,2 and Xifeng Wu1 hazard ratios of 0.92 (0.54-1.57) and 2.53 (1.48-4.30), 1 2 Departments of Epidemiology and Urology, The University of Texas respectively (P for trend < 0.001). There was also a M.D. Anderson Cancer Center, Houston, Texas, and 3Department of Epidemiology and Public Health, Yale University, New Haven, significant trend for shorter recurrence-free survival time Connecticut with increasing number of variant alleles (log rank test, P = 0.0007). When we stratified the patients according to intravesical Bacillus Calmette-Guerin treatment, we found a ABSTRACT significant trend for shorter recurrence-free survival time in Purpose: Interindividual differences in DNA repair patients with variant alleles of XPA or ERCC6 polymor- capacity not only modify individual susceptibility to carci- phisms who received Bacillus Calmette-Guerin treatment nogenesis, but also affect individual response to cancer (log rank test, P = 0.078 and 0.022, respectively). There were treatment. Nucleotide excision repair (NER) is one of the no significant individual or joint associations between these major DNA repair pathways in mammalian cells involved in polymorphisms and progression. the removal of a wide variety of DNA lesions. Polymorphisms Conclusions: These data suggest that interindividual in NER genes may influence DNA repair capacity and affect differences in DNA repair capacity may have an important clinical outcome of bladder cancer treatment. impact on superficial bladder cancer recurrence. A pathway- Experimental Design: To test the influence of NER gene based approach is preferred to study the effects of individual polymorphisms on superficial bladder cancer outcome polymorphism on clinical outcomes. (recurrence and progression), we conducted a follow-up study of 288 patients with superficial bladder cancer. INTRODUCTION Median follow-up among patients who were recurrence- Bladder cancer is the fifth most commonly diagnosed free at the end of observation was 21.7 months from malignancy in the United States, with more than 55,300 new diagnosis. The specific polymorphic loci examined include cases diagnosed in 2004 (1). Superficial bladder cancer accounts XPA [A/G at 5V untranslated region (UTR)], XPC (poly AT, for >85% of newly diagnosed cases. These cases comprise a Ala499Val, Lys939Gln), XPD (Asp312Asn, Lys751Gln), XPG heterogeneous group of tumors whose individual prognoses are (His1104Asp), ERCC 1 (G/T at 3V UTR), and ERCC6 difficult to predict (2). Nevertheless, following transurethral (Met1097Val, Arg1230Pro). resection, tumor recurrence rate is about 70%, and 10% to 15% of Results: The ERCC6 (Met1097Val) polymorphism had a such recurrences may be associated with progression to more significant impact on recurrence: carriers of at least one invasive forms of cancer (3, 4). Clinical and pathologic variant allele (Val) had a significantly higher recurrence risk prediction variables of disease recurrence have been extensively than carriers of the wild-type allele (Met/Met; hazard ratio, studied. In a patient cohort of 1,529 patients, tumor size, number 1.54; 95% confidence interval, 1.02-2.33). There were no of tumors, and presence of carcinoma in situ have been correlated overall statistically significant differences in the distributions with disease recurrence (5). Administration of intravesical Bacille of the other polymorphisms between patients with and Calmette-Guerin (BCG) has been found to significantly reduce without recurrence. However, when we combined these the risk of recurrence and progression. However, the response variant genotypes, there was a significant trend for an rate for BCG treatment is only 60% to 70%, and about one-third increased recurrence risk with an increasing number of of responders develop recurrence and progression (6, 7). A putative high-risk alleles. Using individuals with five or number of pathologic and molecular markers have been studied as potential predictors of outcome in bladder cancer. Mutant cell cycle and proliferative markers, including Ki-67, cytokeratin 20 Received 6/4/04; revised 1/7/05; accepted 1/13/05. and 34h, and growth factors, such as epidermal growth factor, Grant support: National Cancer Institute grants CA 74880, CA 85576 vascular endothelial growth factor, and fibroblast growth factor (X. Wu), CA 91846 (C. Dinney and X. Wu), and CA 86390 (M. Spitz). receptor-3, have been found to influence recurrence. p53, p21, The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked RB, cadherins cathepsins, metalloproteinases, BCL-2, cyto- advertisement in accordance with 18 U.S.C. Section 1734 solely to keratins, and fibroblast growth factor receptor-3 have been found indicate this fact. to be predictors of tumor progression (8). Requests for reprints: Xifeng Wu, Department of Epidemiology, Unit A possible explanation for the unpredictability of the clinical 1340, The University of Texas MD Anderson Cancer Center, 1155 Pressler Boulevard, Houston, TX 77030. Phone: 713-745-2485; Fax: course of superficial bladder cancer may be related to the fact that 713-792-4657; E-mail: [email protected]. host genetic differences may render some individuals at higher D2005 American Association for Cancer Research. risk than others. Functional polymorphisms may account for

phenotypic differences that lead to enhanced cancer risk. The Genotyping. Genomic DNA was first isolated from associations between genetic polymorphisms and cancer risk have peripheral blood lymphocytes by proteinase K digestion been extensively reported; however, few studies have focused on followed by isopropanol extraction and ethanol precipitation; polymorphisms as predictors of specific clinical events. the sample was then coded and then subjected to genotyping. One of the pathways for carcinogenesis is related to DNA The specific polymorphic loci examined include XPA (A/G at damage and repair. DNA repair systems are essential for 5V UTR, rs1800975), XPC (poly AT, Ala499Val, Lys939Gln), maintaining genomic integrity. The ability to monitor and repair XPD (Asp312Asn, Lys751Gln), XPG (His1104Asp), ERCC1 (G/T carcinogen-induced DNA damage is an important determinant of at 3V UTR, rs3212986), and ERCC6 (Met1097Val, Arg1230Pro). susceptibility to carcinogenesis. There is considerable evidence Most of the polymorphisms, except for XPA, XPC (poly AT in suggesting that reduced DNA repair capacity may play a role in intron 11), and XPD (Asp312Asn), were determined using cigarette smoking–related cancer risk (9, 10). Interestingly, in TaqMan real-time PCR. The primer and probe sequences, terms of cancer recurrence and survival, efficient DNA repair which are available upon request, were either obtained from capacity is not always a favorable prediction factor. Catto et al. National Cancer Institute’s SNP500 database or designed using (11) reported that reduced expression of two mismatch repair Primer Express Software. The probes were labeled fluorescently genes, hMLH1 and hMSH2, was associated with fewer with either FAM or VIC on the 5V end and a nonfluorescent recurrences of superficial bladder cancer. In addition, Bosken minor groove binder quencher on the 3V end (Applied et al. (12) found that effective host DNA repair capacity was Biosystems, Foster City, CA). Typical amplification mixes (5 associated with poorer survival in patients with non–small cell AL) contained sample DNA (5 ng), 1 TaqMan buffer A, 200 lung cancer who were treated with chemotherapy. Am deoxynucleotide triphosphates, 5 mmol MgCl2, 0.65 units Nucleotide excision repair (NER), one of the major DNA of AmpliTaq Gold, 900 nmol/L each primer, and 200 nmol/L repair pathways in mammalian cells, is involved in the removal of each probe. The thermal cycling conditions consisted of 1 cycle a wide variety of DNA lesions, including bulky carcinogen for 10 minutes at 95jC, 40 cycles for 15 seconds at 95jC, adducts resulting from tobacco exposure. All key NER factors and 1 minute at 60jC. SDS version 2.1 software (Applied have been cloned, and the core of the ‘‘cut-and-paste’’ reaction Biosystems) was used to analyze end-point fluorescence. Water has been reconstituted in vitro (13). Most NER genes are control, internal controls, and previously genotyped samples polymorphic, and some of the polymorphisms have been were included in each plate to ensure accuracy of genotyping. extensively studied in terms of their associations with cancer risk Genotyping for XPA, XPC (poly AT in intron 11), and XPD as well as clinical outcomes in specific cancer types (14–17). (Asp312Asn) were done as described previously using PCR- However, most studies have used a candidate gene approach, RFLP (14–16). investigating one single nucleotide polymorphism (SNP) in a Statistical Analyses. v2 analyses and Fisher’s exact test single gene. In the current study, we applied a pathway-based were used to assess patient characteristics by NER gene approach to systematically examine the individual and joint genotypes when appropriate. Survival analysis methods were impact of most relevant genetic polymorphisms in major NER used to consider the influence of NER gene genotypes on patient genes on clinical outcomes of superficial bladder cancer. recurrence and progression. For each patient, we calculated recurrence-free survival time as the time from diagnosis to MATERIALS AND METHODS recurrence or to the date of last follow-up. We calculated person- Study Population. For this study, 288 patients with years at risk within each genotype category as the sum of superficial bladder cancer were identified from The University recurrence-free survival times of all subjects in that category. of Texas MD Anderson Cancer Center in Houston, TX. Because Overall recurrence-free survival in relation to NER gene 90.6% (261) of the patients were Caucasians, all analyses were genotype was evaluated by Kaplan-Meier survival function and restricted to Caucasians. All cases diagnosed within 1 year log-rank tests. Hazard ratios (HRs) for risk of recurrence were preceding enrollment were histologically confirmed, and were estimated from a multivariate Cox proportional hazards model, previously untreated with chemotherapy or radiotherapy. There with adjustments for age, sex, ethnicity, smoking status, tumor were no age, sex, or ethnic restrictions. Through personal grade, and BCG treatment. Stratified analysis was also done interview, each of the 261 patients were asked to provide detailed according to age, gender, smoking status, and BCG treatment. demographic information using a standardized questionnaire. Stata software (Stata Corp., College Station, TX) was used for Each patient also donated a 40 mL sample of blood for laboratory statistical analysis. analysis. All patients gave written informed consent, and the protocol was approved by the MD Anderson Cancer Center Institutional Review Board. Patients were followed with periodic RESULTS cystoscopic examinations and intravesical treatment. This treat- Characteristics of Subjects. Of the 261 Caucasian ment consisted of either BCG for 6 weekly instillations (induction superficial bladder cancer patients included in this study, BCG) or induction BCG + maintenance BCG according to the 205 (78.54%) were men, and 56 (21.46%) were women. The Southwest Oncology Group protocol (induction BCG followed mean age was 62.37 F 11.63 years. The median follow-up by instillations at 3, 6 and then every 6 months for 3 years). time was 21.7 months (range, 1-74.5 months). Table 1 Tumor recurrence was defined as a newly detected bladder tumor summarizes the detailed distribution of demographic character- following a previous negative follow-up cystoscopy, and tumor istics among patients with or without recurrent disease and progression was defined as the transition from non–muscle- progression for those Caucasian patients with available data. invasive (superficial) to invasive or metastatic disease. There were 127 recurrences and 36 developed progressive

Table 1 Host characteristics and clinical variables for superficial bladder cancer Variables Recurrence, n (%) No recurrence, n (%) P Progression, n (%) No progression, n (%) P Sex Male 101 (62.73) 60 (37.27) 35 (19.77) 142 (80.23) Female 26 (55.32) 21 (44.68) 0.359 1 (2.08) 47 (97.92) 0.003 Age, mean (SD) 63.21 (11.29) 60.65 (10.50) 0.056 65.08 (9.48) 61.80 (11.29) 0.143

Smoking status Never 36 (63.16) 21 (36.84) 9 (14.75) 52 (85.25) Ever 80 (58.39) 57 (41.61) 0.538 24 (16.00) 126 (84.00) 0.821 Pack-years, mean (SD) 27.37 (33.92) 35.62 (52.91) 0.348 42.05 (47.38) 29.78 (42.67) 0.284

Focal 1 47 (54.65) 39 (45.35) 10 (10.42) 86 (89.58) 2 23 (76.67) 7 (23.33) 61 (8.75) 26 (81.25) 3 30 (58.82) 21 (41.18) 11 (20.00) 44 (80.00) Other 27 (65.85) 14 (34.15) 0.167 9 (21.43) 33 (78.57) 0.263

Grade 1 8 (80.00) 2 (20.00) 1 (9.09) 10 (90.91) 2 48 (57.83) 35 (42.17) 9 (10.23) 79 (89.77) 3 70 (61.40) 44 (38.60) 0.392 25 (20.16) 99 (79.84) 0.121

Histology Papillary 102 (60.36) 67 (39.64) 19 (10.38) 164 (89.62) Other 25 (64.10) 14 (35.90) 0.665 17 (40.48) 25 (59.52) <0.001

Carcinoma in situ Yes 40 (65.57) 21 (34.43) 18 (26.47) 50 (73.53) No 87 (59.18) 60 (40.82) 0.390 18 (11.46) 139 (88.54) 0.005

Tumor size (cm) 0-2 27 (56.25) 21 (43.75) 8 (16.33) 41 (83.67) 2-5 24 (54.55) 20 (45.45) 3 (6.25) 45 (93.75) >5 18 (72.00) 7 (28.00) 3 (10.00) 27 (90.00) Unknown 56 (62.92) 33 (37.08) 0.454 22 (22.92) 74 (77.08) 0.056

BCG treatment None 62 (64.58) 34 (35.42) 15 (14.15) 91 (85.85) Induction 49 (65.33) 26 (34.67) 18 (22.78) 61 (77.22) Maintenance 16 (43.24) 21 (56.76) 0.049 3 (7.69) 36 (92.31) 0.084

disease during the follow-up period. Between recurrent and without recurrence, there were no statistically significant nonrecurrent patients, there were no statistically significant differences in the distributions of the other polymorphisms. differences in any of the characteristics except BCG treatment When stratified analysis was done according to age, gender, (P = 0.049). Clinical factors associated with increased and smoking status, we found that the effect of the ERCC6 progression risk included aggressive histologic subtypes (Met1097Val) polymorphism was more evident in men (HR, (P < 0.001), the presence of carcinoma in situ (P = 0.005), 1.90; 95% CI, 1.18-3.05) compared with women (HR, 1.14; and tumor size (P = 0.056). In addition, substantially more 95% CI, 0.40-3.22) and in older individuals (HR, 1.99; 95% men developed progression than women (P = 0.003). Those CI, 1.14-3.45) compared with younger individuals (HR, 1.20; with recurrent or progressive disease tend to be older, 95% CI, 0.62-2.35). In addition, only older individuals with at although the differences were not statistically significant least one variant XPD (Asp312Asn) allele had a significantly (P = 0.056 and 0.143, respectively). reduced risk for recurrence (HR, 0.49, 95% CI, 0.28-0.87). Genotypes and Recurrence. To test the associations of Individuals with at least one variant ERCC6 allele (Arg1230Pro) NER gene polymorphisms with recurrence of superficial had a significantly increased risk for recurrence (HR, 2.21; bladder cancer, we applied a pathway-based approach to 95% CI, 1.13-4.32). genotype a comprehensive panel of SNPs in the major genes Most importantly, when we combined these variant known to be involved in the NER pathway. Table 2 shows the genotypes, there was a significant trend for an increasing description of each SNP and the association of each individual recurrence risk with an increasing number of putative high-risk SNP with risk of recurrence as analyzed by Cox proportional alleles. Because the two XPD polymorphisms, three XPC hazards model. Only the ERCC6 (Met1097Val) polymorphism polymorphisms, and two ERCC6 polymorphisms were in strong had a significant impact on recurrence: carriers of at least one linkage disequilibrium (data not shown), we only included XPC variant allele (Val) had a significantly higher recurrence risk (Ala499Val), XPD (Asp312Asn), and ERCC6 (Met1097Val) in the than carriers of the wild-type allele [Met/Met; HR, 1.54; 95% combined analyses. There were a total of eight polymorphisms in confidence interval (CI), 1.02-2.33]. Between patients with and eight genes included in the combined analysis (Table 3).

Compared with carriers with five or fewer putative high-risk Table 3 Combined effects of polymorphisms in NER genes on alleles, those with six to seven risk alleles, and with eight superficial bladder cancer recurrence or more risk alleles had higher recurrence risks with HRs of HRy Log rank 0.92 (0.54-1.57) and 2.53 (1.48-4.30), respectively (P for Alleles* Recurrence No recurrence (95% CI) test trend < 0.001). Low risk 31 24 1 (0-5) Table 2 Multivariate analysis of the variant genotypes of NER genes Medium risk 44 35 0.92 (0.54-1.57) and superficial bladder cancer recurrence (6-7) High risk 37 10 2.53 (1.48-4.30) HR (95% CI), (8-16) multivariate P for trend <0.001 0.0007 Genotypes* No recurrence Recurrence analysisy *SNPs include XPA + XPC (Ala499Val) + XPD (Asp312Asn) + XPG XPA (5V UTR) + RAD23B + CCNH + ERCC1 + ERCC6 (Met1097Val). AA 13 15 yAdjusted for age, gender, tumor grade, pack years, and BCG AG 35 53 1.14 (0.62-2.19) treatment. GG 24 49 1.30 (0.70-2.40)

XPC (intron 11, poly AT) Tumor recurred in 56.4% of patients with zero to five / 30 59 variant alleles, in 55.7% of patients with six to seven variant /+ 37 41 0.73 (0.47-1.14) +/+ 9 23 1.13 (0.66-1.97) alleles, and in 78.7% of patients with eight or more variant alleles (P = 0.001). Figure 1 presents the Kaplan-Meier survival XPC (Ala499Val) functions for overall recurrence-free survival among superficial CC 45 62 bladder cancer patients by number of variant alleles. According CT 28 48 1.39 (0.90-2.15) to the differences in the recurrence-free curves by variant allele TT 3 10 1.57 (0.76-3.26) numbers and on the log-rank test (P = 0.0007), there was a XPC (Lys939Gln) significant trend for shorter recurrence-free survival time with an AA 26 45 increasing number of variant alleles. Individuals with eight or AC 43 55 0.82 (0.52-1.30) more variant alleles had a significantly shorter recurrence-free CC 9 21 1.07 (0.60-1.94) survival time (median, 6.9 months) than patients with zero to XPD (Asp312Asn) five variant alleles (median, 16.6 months). For individuals with GG 28 52 six to seven variant alleles, the recurrence-free survival time was GA 39 60 0.73 (0.47-1.11) intermediate (median, 13.6 months). AA 11 12 0.62 (0.30-1.28) XPA and ERCC6 Genotypes and Time to Recurrence in XPD (Lys751Gln) Bacillus Calmette-Guerin Treatment Groups. When we AA 24 47 stratified patients according to intravesical BCG treatment, we AC 44 61 0.85 (0.54-1.32) found that patients with variant XPA or ERCC6 alleles who CC 10 15 0.81 (0.40-1.62) received BCG treatment had shorter recurrence-free survival XPG (Asp1104His) (Fig 2, P = 0.078 and Fig 3, P = 0.022, respectively), but this GG 45 73 trend was not evident in patients who did not receive BCG GC + CC 33 50 1.04 (0.69-1.57) treatment (P = 0.939 and 0.678, respectively, data not shown). The median recurrence-free survival time was 10.1 months for 249 RAD23B (Ala Val) patients with the XPA GG genotype, 12.9 months for patients CC 53 84 CT + TT 25 36 0.85 (0.55-1.32) with the AG genotype, and >17.3 months for patients with the AA genotype (Fig. 2). The median recurrence-free survival time CCNH (Val270Ala) was 6.0 months for patients with the ERCC6 (Met1097Val) GG TT 44 79 genotype, 9.67 months for patients with the AG genotype, and TC + CC 34 42 0.78 (0.52-1.19) 22.93 months for patients with the AA genotype (Fig. 3). ERCC1 (3V UTR) Genotypes and Progression. Disease progression oc- GG 39 70 curred in 36 patients during the follow-up period. After adjusting GT + TT 39 50 0.85 (0.56-1.29) for age, gender, tumor grade, pack-years, and BCG treatment, we did not observe any significant individual or joint associations ERCC6 (Met1097Val) AA 52 69 between the polymorphisms we examined and disease progres- AG + GG 25 51 1.54 (1.02-2.33) sion (Table 4).

ERCC6 (Arg1230Pro) GG 63 97 DISCUSSION GC + CC 14 24 1.42 (0.83-2.45) The data from this study, which offers a pathway-based *For those polymorphisms with very few homozygous variant analysis of multiple polymorphic genes, suggest that individuals alleles, only the combined results of the heterozygous and homozygous variant alleles are shown. with higher numbers of genetic variations in the DNA repair yHR adjusted for age, gender, tumor grade, pack years, and BCG pathway have increased risks for superficial bladder cancer treatment. recurrence, confirming the importance of taking a pathway-based

Fig. 1 Kaplan-Meier survival function for recurrence among bladder cancer patients by number of risk alleles.

approach. In addition, in patients who received BCG treatment, biologically plausible. For example, we found that that XPD we found associations between XPA and ERCC6 polymorphisms and ERCC6 polymorphisms conferred more evident effects on and recurrence-free survival; this finding may have important recurrence risk in older individuals and smokers, which might clinical implications for predicting BCG response. be because these two groups typically have accumulated more There have been many studies about the associations of DNA damage than younger individuals and nonsmokers. genetic polymorphisms with clinical outcomes in cancer (18–20), However, because stratified analysis often deals with smaller most of which have focused on one SNP in a certain gene resulting sample sizes, these observations warrant further corroboration. in often contradictory results. Most importantly, because the More significantly, although each individual polymorphism prognosis of cancer patients likely involves multistep, multigenic had a minor effect on recurrence, when combining the putative pathways, it is unlikely that any one single genetic polymorphism risk alleles, individuals with higher numbers of putative risk would have a dramatic effect on clinical outcome. In addition, alleles had a 2.53-fold increased risk of recurrence in a association studies of clinical outcomes often have small sample multivariate model. In addition, there was a significant sizes compared with case-control studies. The limits of the association between increased variant alleles and shorter candidate gene approach study mandate a better approach for such overall recurrence-free survival time for superficial bladder association studies. The pathway-based approach would reflect cancer patients (P = 0.0007). the combined effects of a panel of polymorphisms in a molecular Another interesting finding of this study is that poly- pathway and thus increase the chance of finding minor effects by morphisms in XPA and ERCC6 genes were associated with BCG evaluating a combination of polymorphisms in genes involved in response. It is well established that deficient DNA repair multiple pathways. capacity increases cancer susceptibility. However, little is known We found significant associations between the ERCC6 about the effects of deficient DNA repair capacity on treatment polymorphism and increased recurrence risk and some other outcome. Several studies have shown that polymorphisms in interesting associations in stratified analysis, which are DNA repair genes modulate response to cisplatin-based therapy

Fig. 2 Kaplan-Meier survival function for recurrence among bladder cancer patients by BCG treatment with XPA genotypes.

Fig. 3 Kaplan-Meier survival function for recurrence among bladder cancer patients by BCG treatment with ERCC6 genotypes.

(18–20) because NER is the major pathway repairing cisplatin- oligomers containing 8-oxodG appear extracellularly in cell induced DNA damage. Although BCG intravesical instillation culture supernatant, plasma, or urine (29). XPA was previously is the dominant and most effective method for treating reported to be actively involved in the dual incision step of superficial bladder cancer and preventing recurrence, its repairing ROS-induced 5V,8-purine cyclodeoxynucleosides, mechanism of action is not yet fully understood (21). The which is formed by intramolecular cross-linking between the antitumor effects of BCG are mostly related to local C-8 position of adenine or guanine and the 5V position of immunologic events, which might include two pathways. In 2-deoxyribose (30). ERCC6 (CSB) protein plays an the first pathway, after incubation with BCG, normal and important role in transcription-coupled and global genome- malignant urothelial cells produce an array of cytokines, DNA repair, as well as base excision repair of some types including interleukin-1, tumor necrosis factor-a, and inter- of oxidative damage, including the aforementioned 8-oxodG leukin-6. The cytotoxic response to tumor necrosis factor-a has (31). Therefore, the interindividual variability in DNA repair features of both apoptotic and necrotic cell death. One capacity that affects ROS-induced DNA damage is likely to immediate cause of cell death is believed to be the production impact the treatment effect of BCG. Optimal DNA repair of oxygen-free radicals by mitochondria following tumor capacity will lead to decreased ROS-induced DNA damage, necrosis factor-a action. Moreover, several lines of evidence which subsequently results in less tumor cell death and suggest that tumor necrosis factor-a stimulation induce worse clinical outcome. On the contrary, suboptimal DNA extensive oxidative DNA damage; for instance, Nathan et al. repair capacity will lead to increased ROS-induced DNA (22) found that tumor necrosis factor-a induced wide damage, which subsequently translates into more tumor cell chromosome damage in DNA repair-deficient cells. In the death and better clinical outcome. In this regard, it is second possible pathway, BCG stimulation of human peripheral biologically plausible that the G allele of the XPA blood mononuclear cells leads to the generation of tumor polymorphism is associated with a poorer outcome (shorter cytotoxic natural killer cells. IFN-g and interleukin 2 are time to recurrence after BCG treatment), because a previous essential for BCG-induced cytotoxicity. Some recent studies study showed that the G allele, compared with the A allele, suggest that the antitumor effect of IFN-g could possibly be was associated with more efficient DNA repair capacity mediated by induction of reactive oxidative species (ROS; refs. (14). The functional significance of ERCC6 polymorphisms 23, 24). Watanabe et al. (23) reported that IFN-g-induced has not been reported in any studies; however, because generation of ROS leads to apoptosis in primary hepatocytes. patients with G alleles had worse outcomes after BCG Besides local immune response, BCG treatment also induces a treatment, it would be interesting to find out whether the G systemic activation of the immune system and consequent allele is associated with better DNA repair capacity. systemic production of oxygen free-radicals (25). It is known Since 1976, BCG has remained the most effective that in many cases, ROS are the primary cause of tumor cell treatment against superficial bladder cancer leading to im- death (26). Therefore, ROS production might be an important proved recurrence and progression-free survival, yet about one- pathway in the mediation of BCG treatment response. third of patients do not respond to BCG treatment, and another ROS causes a wide range of DNA damage, and the one-third develop recurrence and progression despite initial accumulation of irreparable DNA damage leads to cell death. response. Early identification of BCG nonresponders may The majority of ROS-induced DNA damage is repaired by base allow these patients to be saved from recurrence and excision repair. However, NER seems to also have a function in progression by timely radical cystectomy. The functional the repair of oxidative lesions (27). Reardon et al. (28) showed impacts on BCG response of SNPs in XPA and ERCC6,as the removal of 8-oxodG by NER. A recent study suggests that observed in our study, may help identify BCG nonresponders NER represents a plausible mechanism by which 8-oxodG and early.

This is the first molecular epidemiologic study to of clinical outcomes. However, the magnitude of the associations investigate the association between genetic polymorphisms in found in our study is far from clinically relevant. In the future, DNA repair genes and risk of bladder cancer recurrence. The more comprehensive genotyping and haplotyping are warranted pathway-based approach is more promising in finding predictors for using genomic polymorphisms as clinically applicable predictive markers.

Table 4 Multivariate analysis of the variant genotypes of NER genes REFERENCES and superficial bladder cancer progression 1. Jemal A, Tiwari RC, Murray T, et al. Cancer statistics, 2004. OR (95% CI), CA Cancer J Clin 2004;54:8–29. multivariate 2. Chopin DK, Gattegno B. Superficial bladder tumors. Eur Urol 2002; Genotype* No progression Progression analysisy 42:533–41. XPA (5V UTR) 3. Kurth K, Denis L, Sylvester R, de Pauw M. The natural history and AA 22 7 the prognosis of treated superficial bladder cancer. EORTC GU group. AG 76 17 0.50 (0.16-1.60) Prog Clin Biol Res 1992;378:1–7. GG 70 12 0.42 (0.13-1.38) 4. Pashos CL, Botteman MF, Laskin BL, Redaelli A. Bladder cancer: epidemiology, diagnosis, and management. Cancer Pract 2002;10: XPC (intron 11, poly AT) 311–22. / 80 16 5. Millan-Rodriguez F, Chechile-Toniolo G, Salvador-Bayarri J, Palou J, /+ 73 10 0.47 (0.16-1.36) Vicente-Rodriguez J. Multivariate analysis of the prognostic factors of +/+ 27 8 1.93 (0.62-5.99) primary superficial bladder cancer. J Urol 2000;163:73–8.

499 6. Merz VW, Marth D, Kraft R, Ackermann DK, Zingg EJ, Studer UE. XPC (Ala Val) Analysis of early failures after intravesical instillation therapy with CC 96 20 Bacille Calmette-Guerin for carcinoma in situ of the bladder. Br J Urol CT 68 13 0.70 (0.27-1.82) 1995;75:180–4. TT 12 2 0.93 (0.17-5.05) 7. Schenk-Braat EAM, Bangma CH. Immunotherapy for superficial XPC (Lys939Gln) bladder cancer. Cancer Immunol Immunother 2004; [Epub ahead of AA 65 12 print]. AC 88 15 0.83 (0.30-2.28) 8. Helpap B, Schmitz-Drager BJ, Hamilton PW, et al. Molecular CC 26 8 2.32 (0.68-7.92) pathology of non-invasive urothelial carcinomas (part I). Virchows Arch 2003;442:309–16. XPD (Asp312Asn) 9. Goode EL, Ulrich CM, Potter JD. Polymorphisms in DNA repair GG 71 15 genes and associations with cancer risk. Cancer Epidemiol Biomarkers GA 89 18 1.19 (0.48-2.99) Prev 2002;11:1513–30. AA 22 3 1.04 (0.24-4.50) 10. Amos CI, Xu W, Spitz MR. Is there a genetic basis for lung cancer susceptibility? Recent Results Cancer Res 1999;151:3–12. XPD (Lys751Gln) 11. Catto JW, Xinarianos G, Burton JL, Meuth M, Hamdy FC. AA 62 16 Differential expression of hMLH1 and hMSH2 is related to bladder AC 94 16 0.75 (0.30-1.92) cancer grade, stage and prognosis but not microsatellite instability. Int J CC 25 4 1.16 (0.29-4.60) Cancer 2003;105:484–90. XPG (Asp1104His) 12. Bosken CH, Wei Q, Amos CI, Spitz MR. An analysis of DNA repair GG 105 20 as a determinant of survival in patients with non-small-cell lung cancer. GC + CC 76 16 1.01 (0.42-2.40) J Natl Cancer Inst 2002;94:1091–9. 13. de Laat WL, Jaspers NG, Hoeijmakers JH. Molecular mechanism of RAD23B (Ala249Val) nucleotide excision repair. Genes Dev 1999;13:768–85. CC 122 26 14. Wu X, Zhao H, Wei Q, et al. XPA polymorphism associated with CT + TT 56 9 0.64 (0.23-1.75) reduced lung cancer risk and a modulating effect on nucleotide excision repair capacity. Carcinogenesis 2003;24:505–9. 270 CCNH (Val Ala) 15. Marin MS, Lopez-Cima MF, Garcia-Castro L, Pascual T, Marron TT 109 23 MG, Tardon A. Poly (AT) polymorphism in intron 11 of the XPC DNA TC + CC 70 12 0.98 (0.41-2.37) repair gene enhances the risk of lung cancer. Cancer Epidemiol Biomarkers Prev 2004;11:1788–93. ERCC1 (3V UTR) GG 96 19 16. Spitz MR, Wu X, Wang Y, et al. Modulation of nucleotide excision GT + TT 90 16 1.17 (0.47-2.67) repair capacity by XPD polymorphisms in lung cancer patients. Cancer Res 2001;61:1354–7. ERCC6 (Met1097Val) 17. Volker M, Mone M, Karmakar P, et al. Sequential assembly of the AA 111 22 nucleotide excision repair factors in vivo. Mol Cell 2001;8:213–24. AG + GG 66 13 0.86 (0.35-2.14) 18. Gurubhagavatula S, Liu G, Park S, et al. XPD and XRCC1 genetic polymorphisms are prognostic factors in advanced non-small-cell lung ERCC6 (Arg1230Pro) cancer patients treated with platinum chemotherapy. J Clin Oncol GG 147 27 2004;22:2594–601. GC + CC 31 8 1.32 (0.44-3.96) 19. Zhou W, Gurubhagavatula S, Liu G, et al. Excision repair cross- *For those polymorphisms with very few homozygous variant complementation group 1 polymorphism predicts overall survival in alleles, only the combined results of the heterozygous and homozygous advanced non-small cell lung cancer patients treated with platinum-based variant alleles are shown. chemotherapy. Clin Cancer Res 2004 Aug 1;10:4939–43. yOR adjusted for age, gender, tumor grade, pack years, and BCG 20. Nagasubramanian R, Innocenti F, Ratain MJ. Pharmacogenetics in treatment. cancer treatment. Annu Rev Med 2003;54:437–52.

21. Bohle A, Brandau S. Immune mechanisms in Bacillus Calmette- 27. Mitra S, Boldogh I, Izumi T, Hazra TK. Complexities of the DNA Guerin immunotherapy for superficial bladder cancer. J Urol 2003; base excision repair pathway for repair of oxidative DNA damage. 170:964–9. Environ Mol Mutagen 2001;38:180–90. 22. Nathan I, Dizdaroglu M, Bernstein L, et al. Induction of oxidative 28. Reardon JT, Bessho T, Kung HC, Bolton PH, Sancar A. In vitro DNA damage in u937 cells by TNF or anti-Fas stimulation. Cytokine repair of oxidative DNA damage by human nucleotide excision 2000;12:881–7. repair system: possible explanation for neurodegeneration in 23. Watanabe Y, Suzuki O, Haruyama T, Akaike T. Interferon-g induces xeroderma pigmentosum patients. Proc Natl Acad Sci U S A 1997;94: reactive oxygen species and endoplasmic reticulum stress at the hepatic 9463–8. apoptosis. J Cell Biochem 2003;89:244–53. 29. Gackowski D, Rozalski R, Roszkowski K, Jawien A, Foksinski M, 24. Pearl-Yafe M, Halperin D, Halevy A, Kalir H, Bielorai B, Olinski R. 8-Oxo-7,8-dihydroguanine and 8-oxo-7,8-dihydro-2V-deoxy- Fabian I. An oxidative mechanism of interferon induced priming of guanosine levels in human urine do not depend on diet. Free Radic Res the Fas pathway in Fanconi anemia cells. Biochem Pharmacol 2003; 2001;35:825–32. 65:833–42. 30. Kuraoka I, Bender C, Romieu A, Cadet J, Wood RD, Lindahl T. 25. Mitropoulos D, Deliconstantinos G, Zervas A, Giannopoulos A, Removal of oxygen free-radical-induced 5V,8-purine cyclodeoxynucleo- Kyriakou G, Dimopoulos C. Oxidative stress of red blood cells during sides from DNA by the nucleotide excision-repair pathway in human Bacillus Calmette-Guerin intravesical instillations. In Vivo 2000;14: cells. Proc Natl Acad Sci U S A 2000;97:3832–7. 721–4. 31. Licht CL, Stevnsner T, Bohr VA. Cockayne syndrome group B 26. Ueda S, Masutani H, Nakamura H, Tanaka T, Ueno M, Yodoi J. cellular and biochemical functions. Am J Hum Genet 2003;73: Redox control of cell death. Antioxid Redox Signal 2002;4:405–14. 1217–39.

Downloaded from clincancerres.aacrjournals.org on September 28, 2021. © 2005 American Association for Cancer Research. Nucleotide Excision Repair Gene Polymorphisms and Recurrence after Treatment for Superficial Bladder Cancer

Jian Gu, Hua Zhao, Colin P. Dinney, et al.

Clin Cancer Res 2005;11:1408-1415.

Updated version Access the most recent version of this article at: http://clincancerres.aacrjournals.org/content/11/4/1408

Cited articles This article cites 30 articles, 6 of which you can access for free at: http://clincancerres.aacrjournals.org/content/11/4/1408.full#ref-list-1

Citing articles This article has been cited by 9 HighWire-hosted articles. Access the articles at: http://clincancerres.aacrjournals.org/content/11/4/1408.full#related-urls

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://clincancerres.aacrjournals.org/content/11/4/1408. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.