Genetic variation in the DNA repair genes is predictive of outcome in lung cancer
1 Section of Cancer Genetics, Institute of Cancer Research, Sutton SM2 5NG, UK, 2 Department of Oncology, University of Cambridge, Cambridge CB2 2RE, UK and 3 GELCAPS Consortium
* To whom correspondence should be addressed at:, Section of Cancer Genetics, Brookes Lawley Building, Institute of Cancer Research, Sutton, Surrey SM2 5NG, UK. Tel: +44 2087224175; Fax: +44 2087224359; Email: richard.houlston{at}icr.ac.uk
Received February 15, 2007; Accepted July 14, 2007
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSSUPPLEMENTARY MATERIALREFERENCES |
|---|
To assess whether DNA repair gene variants influence the clinical behaviour of lung cancer we examined the impact of a comprehensive panel of 109 non-synonymous single-nucleotide polymorphisms (nsSNPs) in 50 DNA repair genes on overall survival (OS) in 700 lung cancer patients. Fifteen nsSNPs were associated with OS, significantly greater than that expected (P = 0.04). SNPs associated with prognosis mapped primarily to two repair pathways—nucleotide excision repair (NER): ERCC5 D1104H (P = 0.004); ERCC6 G399D (P = 0.023), ERCC6 Q1413R (P = 0.025), POLE (P = 0.014) and base excision repair: APEX1 D148E (P = 0.028); EXO1 E670G (P = 0.007); POLB P242R (P = 0.018). An increasing number of variant alleles in EXO1 was associated with a poorer prognosis [hazard ratio (HR) = 1.24; P = 0.0009]. A role for variation in NER and BRCA2/FA pathway genes as determinants of OS was provided by an analysis restricted to the 456 patients treated with platinum-based agents. Our data indicate that the pathway-based approach has the potential to generate prognostic markers of clinical outcome.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSSUPPLEMENTARY MATERIALREFERENCES |
|---|
Lung cancer is a major cause of cancer mortality worldwide, and in the UK it accounts for >33 000 cancer deaths each year (Cancer Research UK, http://www.cancerresearchuk.org/aboutcancer/statistics). While chemotherapy based on platinum regimens is the mainstay treatment for the disease, five-year survival rates from both small (SCLC) and non-small cell lung cancer (NSCLC) remain at only 15% (1). Although the major prognostic determinant for lung cancer is stage at presentation there are differences in survival for patients with same-stage disease. More accurate assessment of prognosis should be helpful in deciding therapeutic options, opening up the possibility of patient-tailored decisions on drug selection. Detecting genes with prognostic relevance also has the potential to aid the identification of pathways that could be targeted for novel therapeutic interventions.
As a potential prognostic factor, the concept of germline variation imparting inter-individual variability in tumour development, progression and metastasis is receiving increasing attention (2–4). Previous studies of lung cancer have generally only evaluated sequence changes in a very restricted number of genes (5–7) and the majority of purported associations remain unconfirmed. Furthermore, most searches have been formulated around the same candidate genes purported to be implicated in defining predisposition to the disease. It is entirely probable that genes in which variation plays a role in defining inter-individual disease expression will impact on the later stages of malignancy rather than on early events associated with inherited predisposition. Moreover, any polymorphic variants identified as having an impact on lung cancer prognosis may also influence the biology of other malignancies.
Based on our understanding of lung cancer biology, genes involved in DNA repair represent attractive candidates for lung cancer prognostic factors as inefficient DNA repair may promote cancer progression by supporting genetic instability and the development of more aggressive tumours (8,9). Furthermore, such variants may also exert an effect through modulating inter-individual response to treatment. For example, platinum compounds exert their effects through the formation of DNA adducts (10) and the level of platinum-DNA adducts correlates with outcome (11,12), hence suboptimal DNA repair within the tumour may lead to the decreased removal of platinum-DNA adducts and therefore, increased response to platinum (13).
Here, we have adopted a pathway-based approach to evaluate the impact of polymorphic variation in the DNA repair genes on lung cancer prognosis by analysing a comprehensive panel of 109 single-nucleotide polymorphisms (SNPs) in 700 patients with incident disease. All of the SNPs genotyped in the 50 DNA repair genes were non-synonymous SNPs (nsSNPs), altering the encoded amino acid sequence thereby having the potential to affect the function of expressed proteins directly.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSSUPPLEMENTARY MATERIALREFERENCES |
|---|
The median survival time (MST) for the 700 patients was 16.7 months. Factors significantly influencing patient prognosis were stage at presentation (P < 10–5), smoking (P = 0.02), sex (P = 0.02) and radiotherapy (P = 3.0 x 10–4). Patients with SCLC had a MST of 20.4 and 11.4 months if diagnosed with limited and extensive disease, respectively. For NSCLC, MST ranged from 11.9 months in stage IV patients to 49.2 months in the stage I group. As these survival rates are not significantly different to those documented in previously published audits of lung cancer prognosis (Cancer Research UK, http://www.cancerresearchuk.org/aboutcancer/statistics), there is no evidence that ‘healthy study participant’ selection will have biased our analyses. Although administration of platinum-based chemotherapy and family history of lung cancer were associated with prognosis, they did not significantly affect survival when correction was applied for stage, radiotherapy, smoking and sex.
Relationship between SNP genotypes and prognosis
Patients were genotyped for a total of 109 nsSNPs mapping to 50 DNA repair genes [base excision repair (BER): APEX1, EXO1, MBD4, MUTYH, OGG1, SMUG1, TDG, XRCC1, POLB; NER: CCNH, ERCC2, ERCC4, ERCC5, ERCC6, LIG1, MMS19L, POLD1, POLE, RAD23B, XPC; mismatch repair: MLH3, MSH3, MSH4, MSH5, MSH6, PMS1, PMS2; direct reversal of damage: MGMT; homologous recombination genes: BRCA1, BRCA2, RAD52, XRCC2; non-homologous end-joining: XRCC4; genes defective in diseases associated with sensitivity to DNA damaging agents: ATM, BLM, FANCA FANCD2, FANCE, WRN; other conserved DNA damage response genes, ATR, CHEK1, HUS1, HUS1B, RAD1, TP53BP1; other DNA polymerases: POLH, POLI, POLL, POLQ, POLR3B]. Supplementary Material, Table S1 details each of the nsSNPs.
Fifty-nine of the 109 SNPs genotyped had minor allele frequency (MAF) of 5% or higher, 15 were observed at comparatively low frequencies (i.e. having MAF < 1%). Three sets of SNPs (D693N and S1040N mapping to BRCA1 and M1097V, R1213G;Q1413R mapping to ERCC6 and I143V and K178R mapping to MGMT) were in strong linkage disequilibrium (LD; i.e. r2 > 0.9), hence findings for these loci are highly correlated. For each SNP, the relationship between genotype and overall survival (OS) obtained from Cox regression analysis is summarized in Supplementary material, Table S1. Overall, 15 SNPs displayed an association with OS (Table 1), greater than that expected simply by chance (P = 0.04; accounting for LD between SNPs). None of the SNPs displayed significant interactive effects with platinum treatment but for informatively we computed HRs for individual SNPs separately using data from the 456 patients treated with platinum-based chemotherapy (Supplementary material, Table S1). For purposes of clarity, we have restricted our commentary to the pathways for which polymorphic variation in at least one SNP was associated with OS.
|
BER genes
BER is one of the principle DNA repair pathways that correct simple DNA base lesions, such as the products of deamination, oxidation and alklyation. Under the Cox proportional hazards model, homozygosity for the apex nuclease (APEX1) D148E variant genotype was significantly associated with a poorer OS (HR = 0.74; 95% CI: 0.56–0.97; Table 1). Kaplan–Meier estimates demonstrated that carriers had a longer MST than patients with the wild-type genotype (MST of 20.4 and 16.3 months, respectively; P = 0.09; Fig. 1). Exonuclease 1 (EXO1) represents another component of BER. Three of the six SNPs mapping to EXO1 gene displayed a significant association with OS, E589K, E670G and H354R. E670G homozygosity was associated with a significantly poorer prognosis (HR = 1.49; 95% CI: 1.12–1.98). Kaplan–Meier estimates of MST were 20.4 and 17.9 months in patients with wild-type and variant homozygous genotypes, respectively; P = 0.09; Fig. 2). Hetero- and homozygosity for E589K and H354R were also associated with a less favourable OS, with HRs 1.23 (95% CI: 0.99–1.51) and 1.49 (95% CI: 1.12–1.99), 1.22 (95% CI: 0.98–1.53) and 1.46 (95% CI: 1.12–1.92), respectively (Table 1). Combining genotypes of SNPs mapping to EXO1 revealed a significant trend of poorer OS with increasing number of variant alleles (HR = 1.24; 95% CI: 1.09–1.45: P = 0.0009; MST associated with possession of >4 risk alleles being 14.8 months compared with 19.8 months with <2 risk alleles).
|
|
DNA polymerase beta (POLB) also participates in BER. Carrier status for the P242R variant of POLB was associated with a poorer prognosis (MST of 12.0 months compared with 16.2 months; HR = 1.73; 95% CI: 1.10–2.72). Polymerase Q (POLQ) introduces mutations at C/G by replicating over abasic sites generated by uracil-DNA glycosylase, and hence has an indirect role in BER. One of the three SNPs mapping to POLQ, T1117R displayed an association with OS; homozygosity being associated with a better prognosis (HR = 0.67; 95% CI: 0.48–0.94; MST 23.4 months compared with 17.3 months).
The association between EXO1 E589K, POLQ T1117R and POLB P242R variants and OS remained significant in patients treated with platinum-based chemotherapy (Table 1).
NER genes
NER is the major cellular system that repairs primarily bulky DNA adducts caused by mutagens, and guanine-cisplatinium adducts formed during chemotherapy. The excision-repair, complementing defective, in Chinese hamster, 5 (ERCC5; XPG) SNP D1104H was associated with OS under a recessive model. Patients homozygous for the ERCC5 1104H genotype had a significantly poorer OS (HR = 1.92; 95% CI: 1.24–2.98). In Kaplan–Meier analysis, individuals with the wild-type genotype had a MST of 24.5 months, whereas the MST was 17.8 and 14.7 months in patients hetero- and homozygous for D1104H variants, respectively (P = 0.15; Fig. 3). The impact of ERCC5 D1104H was most marked in patients receiving platinum-based chemotherapy (Table 1).
|
Four of the five SNPs mapping to excision-repair cross-complementing, group 6 (ERCC6) were associated with differences in OS, but 3 (M1097V, R1213G, Q1413R) were in strong LD. For these SNPs, there was a relationship between heterozygous genotype and survival but not with homozygosity for the rare variant, lessening the validity of the observed association. Carrier status for ERCC6 399D was associated with poorer prognosis (HR = 1.27; 95% CI: 1.03–1.57; MST 12.0 months compared with 16.2 respectively, P = 0.18).
For repair of nuclear DNA, NER utilizes DNA polymerase delta and eplison (POLD and POLE). While none of the SNPs mapping to POLD were associated with lung cancer prognosis, carriers of the POLE A252V had significantly poorer OS (HR = 2.79; 95% CI: 1.23–6.33) although this finding was based on observations from only eight individuals. In Kaplan–Meier analysis, individuals with the wild-type genotype had a MST of 16.2 months, whereas the MST was 12.0 months in A252V carriers (P = 0.18).
Other genes associated with lung cancer prognosis
Other SNPs displaying an association with lung cancer prognosis included those mapping to the Blooms (BLM) and Werner syndrome (WRN) genes, both DNA helicases that interact with human topoisomerases. Carriers of BLM P868L had a poorer prognosis (HR = 1.27; 95% CI: 1.01–1.59; Table 1; MST of 14.4 months compared with 15.8 months in wild-type homozygosity, respectively; P = 0.07; Fig. 4). Similarly, carriers of WRN T324A also had a poorer prognosis (HR = 3.25; 95% CI: 1.19–8.92; MST of 6 months compared with 16.7 months, P = 0.08), albeit based on only six obervations. The other SNP displaying an association with prognosis was H130Q mapping within hydroxyurea-sensitive 1 s. pombe, homolog of b (HUS1B); with homozygosity for 130Q being associated with a 1.32-fold poorer prognosis (95% CI: 1.06–1.65; Table 1; MSTs 12.8 and 17.6 months, respectively).
|
Analysis restricted to the patients receiving chemotherapy provided evidence that variation in a number of other DNA repair genes influence prognosis with significant associations being identified between BRCA2 R2034C and FANCE A250T, and OS (Table 1). While the association between BRCA2 R2034C and OS was based on small number of observations, the association with FANCE A502T appears relatively robust.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSSUPPLEMENTARY MATERIALREFERENCES |
|---|
Major strengths of our study are its large size, the fact that it is population-based and has involved the systematic follow-up of patients. We are however, mindful that ideally studies aimed at identifying prognostic markers should be conducted within the context of a clinical trial to limit possible bias. Although bias from non-uniform treatment is a potential confounder in studies of some tumours, the management of lung cancer is relatively uniform in the UK, as there are only a restricted number of chemotherapeutic agents and prognosis is uniformly poor. Support for this assertion is provided by the fact that survival rates observed in our patient cohort were not different with those expected. It therefore seems unlikely that any spurious influences as a consequence of study design will have impacted significantly on our findings. It is well known that the allele frequencies of many SNPs vary among different populations and that population substructure can impact on association results. As our analysis was restricted to white UK patients our study findings are unlikely to be confounded by population stratification (14).
Across all patients, we identified associations between survival and variation in a number of DNA repair genes. The observation that associations were principally with variants of BER and NER genes rather than other DNA repair pathways suggests the statistically significant associations were not random. Although there are differences in the biology of NSCLC and SCLC, the impact of variants on prognosis was independent of histology suggesting effects are generic. Our observation that polymorphic variation in the NER and BER genes influences cancer prognosis is not without precedent. Polymorphic variation in the NER genes has been previously reported as a predictor of OS in advanced lung cancer (15) (16), and polymorphic variation in XRCC1 has been reported to influence survivorship (17). Outside the context of lung cancer, variation in ERCC5 defined by D1104H has been previously reported to be associated with less favourable prognosis in oesophageal cancer (18). Associations identified between variants of WRN and BLM and prognosis are intriguing as case reports have implicated these as susceptibility genes for lung cancer (19,20). Furthermore, mechanistically it is noteworthy that the BLM and WRN helicases can be considered to operate counter to BER.
The nature of our study precluded us from formally evaluating SNPs in relation to response to radiotherapy as this was only administered to a small number of patients. Our dataset did, however, allow us to examine the relationship between OS in over 400 patients receiving platinum therapy. Nucleotide excision is the major pathway involved in the removal of platinum adducts and repair of DNA damage. It has a broad specificity, and no differences have been observed in the excision of adducts induced by cisplatin and structurally diverse platinum-based drugs, such as carboplatin (21). It is therefore intuitive that polymorphic variation in NER-related genes affecting NER functional capacity may predict response to platinum chemotherapy and hence OS. Certainly our data provides a measure of support for such a hypothesis. Similarly, the FA/BRCA pathway regulates the cellular response to cisplatinum and other DNA cross-linking agents (22) and it is noteworthy that we observed an association between variants of BRCA2 and FANCE genes and OS in patients treated with platinum.
Although we have evaluated only nsSNPs with a higher probability of being directly causal, we cannot exclude the possibility that some associations are a consequence of LD. For a number of variants, we identified as impacting on OS there is, however, evidence that they have direct functional effects. The D1104H variant of ERCC5 has been shown to be associated with increased single-strand breakage in lymphocytes (23). For other variants such as EXO1 E670G, ERCC6 R1213G, POLB P242R, POLQ T1117R, BLM P868I and HUS1B H130Q, the only evidence for functionality is provided by bioinformatics analyses. Although predictions about the functional consequences of amino acid changes are not definitive, algorithms such as Sifting Intolerant from Tolerant (SIFT, http://blocks.fhcrc.org/sift/SIFT.html) and Polymorphism Phenotyping (PolyPhen, http://www.bork.embl-heidelberg.de/PolyPhen/) have been demonstrated in benchmarking studies to successfully categorize 
80% of amino acid substitutions (24).
Despite the strong biological plausibility and consistency with literature for several individual associations as discussed herein, some of these associations may be false positives as a result of the inherent pitfalls of the approach. Specifically, there is the issue of adjustment for multiple comparisons. We assessed 109 polymorphisms in 50 genes so a conservative adjustment would dictate that we adjust all associations for 109 comparisons. However, because more than one polymorphism was tested in some genes, the results are not independent; therefore for SNPs mapping to EXO1 the statistical threshold for global significance is 0.0002. Clearly, at this level the P-values associated with individual SNPs within this gene are not globally significant. Combined analyses of multiple variants in the same gene or pathways are theoretically less affected by the issue of multiple testing. While this strategy was productive with respect to multiple SNPs in EXO1, no further significant pathway/gene associations were observed when applied to other loci.
The magnitude of any difference in prognosis associated with individual SNPs is likely to be at best modest. Hence, stipulating significance levels of 10–4 or less for an analysis to have 80% power to demonstrate a 5% difference in survival, which is clinically relevant, would require at least 4800 patient samples to be analysed even if the frequency of the at risk genotype is 50%. For less frequent genotypes, samples sizes would be impossibly large. On this basis, the imposition of very stringent P-values is questionable creating the serious issue of generating a raft of type II errors (25).
Irrespective of the degree of statistical support for individual SNPs, the observation that the SNPs associated with OS map to a restricted set of pathway genes provides a measure of robustness to our findings. Genetic markers such as those we have identified are unlikely to replace conventional markers. They do, however, have potential to assist in distinguishing different outcome patterns among patients with the same-stage disease as 5–10% differences in prognosis are relevant in lung cancer. Furthermore, our findings provide additional insight into the biological determinants of response to treatment and prognosis, thereby opening up the possibility of a rational, targeted approach to cancer treatment based on a combination of genotype and tumour characteristics of an individual.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSSUPPLEMENTARY MATERIALREFERENCES |
|---|
Patients
The current study is based on 700 patients with lung cancer ascertained through the Genetic Lung Cancer Predisposition Study (GELCAPS) a population-based study of lung cancer. Further details about the design and conduct of the study are described in previously published material (26). Briefly all incident lung cancer cases over the age of 18 years, diagnosed between March 1999 and July 2004, referred to 140 clinical centres were eligible for entry into GELCAPS. Upon completion of study recruitment, 30 of the participating centres provided clinical details on the outcome of the recruited lung cancer patients. A standardized proforma was used to collect information on diagnosis, stage, treatment and survival. Records were requested based on their date of accrual, with those accrued at the beginning of the study being requested first. Follow-up data completeness was 88% with mean follow-up time of 16.75 months (range, 0–66.3 months). During the follow-up time there were 455 (65%) deaths in the cohort. All patients were self-reported as having white ethnicity. The mean age at diagnosis in patients was 64.2 years.
A total of 175 patients had SCLC (25%), somewhat less than half (43%) of these presenting with limited disease. Of the 525 patients with NSCLC, 13% had stage I, 14% had stage II, 43% had stage III and 29% had stage IV disease at presentation. All patients with limited stage SCLC received chemotherapy, the majority being treated with a combination of radical radiotherapy (60%) and platinum-based chemotherapy (88%). The main treatment modality for SCLC patients with extensive disease was primarily platinum-based chemotherapy (91%) with only a restricted number receiving radical radiotherapy (12%). Patients with early stage NSCLC (stage I and II disease) were mainly treated with surgical resection of the primary tumour (57%), while about one-third received chemotherapy, primarily platinum-based (32%) and radical radiotherapy (38%). Chemotherapy was the mainstay treatment modality for patients with stage III and IV NSCLC mainly with platinum-based agents (66 and 70%, respectively).
Ethical approval for the study was obtained from the London Multi-Centre Research Ethics Committee (MREC/98/2/67) and relevant local ethical committees in accordance with the tenets of the Declaration of Helsinki. All patients provided informed consent.
Selection of candidate genes, SNP selection and genotyping
SNPs mapping to DNA repair genes were identified by interrogation of a database of potential functional SNPs and published work on resequencing of DNA repair genes (27, http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db = http://www.icr.ac.uk/cancgen/molgen/MolPopGen_PICS_database.htm;). To predict the putative impact of missense variants on protein function, we applied two in silico algorithms, PolyPhen (http://www.bork.embl-heidelberg.de/PolyPhen/) and SIFT (http://blocks.fhcrc.org/sift/SIFT.html). PolyPhen predicts the functional impact of amino acid changes by considering evolutionary conservation, physicochemical differences and the proximity of the substitution to predicted functional domains and/or structural features (27). PolyPhen scores were designated probably damaging, possibly damaging, potentially damaging, borderline or benign according to the classification proposed by Xi et al. (24). SIFT predicts the functional importance of amino acid substitutions based on the alignment of orthologous and/or paralogous protein sequences. SIFT scores were classified as intolerant, potentially intolerant, borderline or tolerant according to the classification proposed by Xi et al. (24) and Ng and Henikoff (28). Genotyping was conducted by means of Illumina Sentrix Bead Arrays (Illumina, San Diego, USA) or via Taqman technology implemented on ABI700HT platform (Applied Biosystems, Foster City, USA) according to manufacturers’ protocols (details on request).
Statistical methods
OS of patients was the end point of the analysis. Survival time was calculated from the date of diagnosis of lung cancer to the date of death. Patients who were not deceased were censored at the date of last contact. Median follow-up time was computed among censored observations only. Cox regression analysis (29) was used to estimate HRs while adjusting for age, sex, treatment, histology, family history, smoking and stage. Platinum and surgery did not satisfy the proportional hazards assumption required for the Cox model. Therefore, we used a stratified Cox proportional hazards model, stratifying on platinum and surgery. Likelihood ratio testing for the inclusion of covariates and interaction terms was performed to determine the best-fitting model. For each SNP, HRs were generated using common allele homozygotes as the reference group. For polymorphisms with fewer than five minor allele homozygotes, minor allele homozygote genotypes were combined with heterozygotes. The Kaplan–Meier function (29) and log rank tests (29) were used to assess OS in relation to individual polymorphisms. In addition, we evaluated OS as a function of the number of ‘risk alleles’ involved in each gene or DNA repair system. In this analysis, risk was trichotomized into low, medium and high-risk categories on the basis of individuals carrying <3, 3–4 and >5 risk alleles. Due to the exploratory nature of this study, we report nominal statistical associations. All statistical analyses were undertaken using S-Plus (Version 8, Insightful Corporation, USA). To assess the level of LD between SNPs, we calculated the pair-wise LD measure r2 between markers mapping to the same gene using a Markov chain Monte Carlo approach implemented in PHASE (30). The power to demonstrate a relationship between SNP genotype and OS was estimated using sample size formulae for comparative binomial trials (31).
|
SUPPLEMENTARY MATERIAL |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSSUPPLEMENTARY MATERIALREFERENCES |
|---|
Supplementary Material is available at HMG Online.
|
ACKNOWLEDGEMENTS |
|---|
We are grateful to patients for their participation. Work was undertaken with support from Cancer Research UK, NCRN, HEAL and Sanofi-Aventis. Athena Matakidou was the recipient of a clinical research fellowship from the Allan J Lerner Fund. The authors are indebted to Richard Coleman, Christina Fleischmann, Nick Hearle, Rosalind Mutch, Mobshra Qureshi, Elaine Ryder-Mills, Hayley Spendlove and Remben Talaban for sample ascertainment and preparation.
Conflict of Interest statement. None declared.
|
FOOTNOTES |
|---|
List of GELCAPS Consortium collaborators available on request. 
The authors wish it to be known that, in their opinion, the last two authors should be regarded as joint First Authors.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSSUPPLEMENTARY MATERIALREFERENCES |
|---|
-
Jemal A., Thomas A., Murray T., Thun M. Cancer statistics, 2002. CA Cancer J. Clin. (2002) 52:23–47.
[Abstract/Free Full Text] - Hunter K. Host genetics influence tumour metastasis. Nat. Rev. Cancer (2006) 6:141–146.[CrossRef][Web of Science][Medline]
-
Hunter K.W., Crawford N.P. Germ line polymorphism in metastatic progression. Cancer Res. (2006) 66:1251–1254.
[Abstract/Free Full Text] -
Lenz H.J. The use and development of germline polymorphisms in clinical oncology. J. Clin. Oncol. (2004) 22:2519–2521.
[Free Full Text] -
Gurubhagavatula S., Liu G., Park S., Zhou W., Su L., Wain J.C., Lynch T.J., Neuberg D.S., Christiani D.C. XPD and XRCC1 genetic polymorphisms are prognostic factors in advanced non-small-cell lung cancer patients treated with platinum chemotherapy. J. Clin. Oncol. (2004) 22:2594–2601.
[Abstract/Free Full Text] -
Wang Y.C., Chen C.Y., Chen S.K., Chang Y.Y., Lin P. p53 codon 72 polymorphism in Taiwanese lung cancer patients: association with lung cancer susceptibility and prognosis. Clin. Cancer Res. (1999) 5:129–134.
[Abstract/Free Full Text] -
Isla D., Sarries C., Rosell R., Alonso G., Domine M., Taron M., Lopez-Vivanco G., Camps C., Botia M., Nunez L., et al. Single nucleotide polymorphisms and outcome in docetaxel-cisplatin-treated advanced non-small-cell lung cancer. Ann. Oncol. (2004) 15:1194–1203.
[Abstract/Free Full Text] -
Nakamura H., Saji H., Idiris A., Kawasaki N., Hosaka M., Ogata A., Saijo T., Kato H. Chromosomal instability detected by fluorescence in situ hybridization in surgical specimens of non-small cell lung cancer is associated with poor survival. Clin. Cancer Res. (2003) 9:2294–2299.
[Abstract/Free Full Text] - Rosell R., Pifarre A., Monzo M., Astudillo J., Lopez-Cabrerizo M.P., Calvo R., Moreno I., Sanchez-Cespedes M., Font A., Navas-Palacios J.J. Reduced survival in patients with stage-I non-small-cell lung cancer associated with DNA-replication errors. Int. J. Cancer (1997) 74:330–334.[CrossRef][Web of Science][Medline]
- Johnson S., Stephenson J., O’Dwyer P. Cisplatin and its Analogues (2001) 6th edn. Philadelphia: Lippincott Williams and Wilkins.
-
Bosken C.H., Wei Q., Amos C.I., Spitz M.R. An analysis of DNA repair as a determinant of survival in patients with non-small-cell lung cancer. J. Natl. Cancer Inst. (2002) 94:1091–1099.
[Abstract/Free Full Text] - Jordan P., Carmo-Fonseca M. Molecular mechanisms involved in cisplatin cytotoxicity. Cell Mol. Life Sci. (2000) 57:1229–1235.[CrossRef][Web of Science][Medline]
- Shellard S.A., Fichtinger-Schepman A.M., Lazo J.S., Hill B.T. Evidence of differential cisplatin-DNA adduct formation, removal and tolerance of DNA damage in three human lung carcinoma cell lines. Anticancer Drugs (1993) 4:491–500.[Medline]
- Genome-wide association study of 14,000 cases of seven common diseases and 3,000 shared controls. Nature (2007) 447:661–678.[CrossRef][Medline]
-
Zhou W., Gurubhagavatula S., Liu G., Park S., Neuberg D.S., Wain J.C., Lynch T.J., Su L., Christiani D.C. 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. (2004) 10:4939–4943.
[Abstract/Free Full Text] - Ryu J.S., Hong Y.C., Han H.S., Lee J.E., Kim S., Park Y.M., Kim Y.C., Hwang T.S. Association between polymorphisms of ERCC1 and XPD and survival in non-small-cell lung cancer patients treated with cisplatin combination chemotherapy. Lung Cancer (2004) 44:311–316.[CrossRef][Web of Science][Medline]
- Yoon S.M., Hong Y.C., Park H.J., Lee J.E., Kim S.Y., Kim J.H., Lee S.W., Park S.Y., Lee J.S., Choi E.K. The polymorphism and haplotypes of XRCC1 and survival of non-small-cell lung cancer after radiotherapy. Int. J. Radiat. Oncol. Biol. Phys. (2005) 63:885–891.[CrossRef][Web of Science][Medline]
-
Wu X., Gu J., Wu T.T., Swisher S.G., Liao Z., Correa A.M., Liu J., Etzel C.J., Amos C.I., Huang M., et al. Genetic variations in radiation and chemotherapy drug action pathways predict clinical outcomes in esophageal cancer. J. Clin. Oncol. (2006) 24:3789–3798.
[Abstract/Free Full Text] - Takemiya M., Shiraishi S., Teramoto T., Miki Y. Bloom’s syndrome with porokeratosis of Mibelli and multiple cancers of the skin, lung and colon. Clin. Genet. (1987) 31:35–44.[Web of Science][Medline]
-
Yamanaka A., Hirai T., Ohtake Y., Kitagawa M. Lung cancer associated with Werner’s syndrome: a case report and review of the literature. Jpn. J. Clin. Oncol. (1997) 27:415–418.
[Abstract/Free Full Text] - Chaney S.G., Vaisman A. Specificity of platinum-DNA adduct repair. J Inorg. Biochem. (1999) 77:71–81.[CrossRef][Web of Science][Medline]
-
Chirnomas D., Taniguchi T., de la Vega M., Vaidya A.P., Vasserman M., Hartman A.R., Kennedy R., Foster R., Mahoney J., Seiden M.V., et al. Chemosensitization to cisplatin by inhibitors of the Fanconi anemia/BRCA pathway. Mol. Cancer Ther. (2006) 5:952–961.
[Abstract/Free Full Text] -
Vodicka P., Kumar R., Stetina R., Sanyal S., Soucek P., Haufroid V., Dusinska M., Kuricova M., Zamecnikova M., Musak L., et al. Genetic polymorphisms in DNA repair genes and possible links with DNA repair rates, chromosomal aberrations and single-strand breaks in DNA. Carcinogenesis (2004) 25:757–763.
[Abstract/Free Full Text] - Xi T., Jones I.M., Mohrenweiser H.W. Many amino acid substitution variants identified in DNA repair genes during human population screenings are predicted to impact protein function. Genomics (2004) 83:970–979.[CrossRef][Web of Science][Medline]
-
Perneger T.V. What’s wrong with Bonferroni adjustments. Bmj (1998) 316:1236–1238.
[Free Full Text] - Matakidou A., Eisen T., Bridle H., O’Brien M., Mutch R., Houlston R.S. Case-control study of familial lung cancer risks in UK women. Int. J. Cancer (2005) 116:445–450.[CrossRef][Web of Science][Medline]
-
Rudd M.F., Williams R.D., Webb E.L., Schmidt S., Sellick G.S., Houlston R.S. The predicted impact of coding single nucleotide polymorphisms database. Cancer Epidemiol. Biomarkers Prev. (2005) 14:2598–2604.
[Abstract/Free Full Text] -
Ng P.C., Henikoff S. Accounting for human polymorphisms predicted to affect protein function. Genome Res. (2002) 12:436–446.
[Abstract/Free Full Text] - Klein J., Moeschberger M. Survival Analysis:Techniques for Censored and Truncated Data (1997) 1st edn. Berlin: Springer.
- Stephens M., Smith N.J., Donnelly P. A new statistical method for haplotype reconstruction from population data. Am. J. Hum. Genet. (2001) 68:978–989.[CrossRef][Web of Science][Medline]
-
Farrington C.P., Manning G. Test statistics and sample size formulae for comparative binomial trials with null hypothesis of non-zero risk difference or non-unity relative risk. Stat. Med. (1990) 9:1447–1454.[Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||