Summary
Hypoxia is a hallmark of solid tumors and has been implicated in poor clinical outcome. In this candidate pathway association study, we found two hypoxia pathway genetic variants in RPA1 and EXO1 correlated with poor survival in advanced-stage lung cancer patients treated with platinum-based chemotherapy but not chemoradiotherapy. This result suggests adding radiotherapy could improve survival in patients harboring these risk genotypes.
Abstract
Hypoxia is a hallmark of solid tumors and has been implicated in the development of advanced disease and poor clinical outcome. In this multi-stage study, we aimed to assess whether genetic variations in hypoxia pathway genes might affect overall survival (OS) in patients with advanced-stage non-small cell lung cancer (NSCLC). We genotyped 598 potentially functional and tagging single nucleotide polymorphisms (SNPs) in 42 genes of the hypoxia pathway in 602 advanced stage NSCLC patients who received platinum-based chemotherapy or chemoradiation (discovery phase). Significant SNPs were validated in an additional 278 advanced stage patients (validation phase). Cox proportional hazard regression analysis was used to evaluate the association of each SNP with OS. Results showed in chemotherapy only group the median survival time (MST) of NSCLC patients with RPA1: rs2270412 AA+GA genotype versus GG genotype was 10.5 versus 12.7 month [P = 0.004, hazard ratio (HR) = 1.42, 95% CI: 1.16–1.74, combined set]. The MST of patients with EXO1: rs9350 GA+AA genotype versus GG genotypes was 13.2 months versus 11.5 months (P = 0.009, HR = 0.70, 95% CI: 0.56–0.87, combined set). Patients harboring two unfavorable genotypes had a 2.02-fold increased risk of death (P = 3.16E−6) and chemoradiation would improve survival for them (HR = 0.75, 95% CI: 0.51–1.10, P = 0.27, combined set). The MST for patients with 0, 1, and 2 unfavorable genotypes was 13.2, 12.7 and 8.9 months, respectively (P = 0.0002, combined set). In summary, two variants in RPA1 and EXO1 were associated with poor survival in NSCLC patients treated by platinum-based chemotherapy. Adding radiotherapy could improve survival in patients harboring these risk genotypes.
Introduction
Lung cancer is the leading cause of cancer death both in males and females all over the world and the incidence and mortality are expected to continue increase for the next few decades in developing countries where the epidemic has just started (1). The overall 5-year survival rate of lung cancer is only 16.8% (2) and the survival is much worse for advanced stage patients with the 5-year and 1-year survival rates of <5 and 33% for non-small-cell lung cancer (NSCLC) patients, respectively (3), which account for the vast majority of lung cancer cases (4).
Several clinical factors are prognostic factors for survival of NSCLC patients including TNM-stage, performance status, sex and co-morbidity (5). Furthermore, biomarkers have been shown to be associated with the survival of NSCLCL patients, such as mutations within DNA repair genes, changes in DNA methylation, the presence of circulating tumor cells (CTCs) and so on (6). Additional biomarkers are needed to predict the prognosis and treatment response in NSCLC patients.
Hypoxia is a hallmark of solid tumors and is associated with advanced disease stage and poor clinical outcome (7). To adapt to hypoxia, tumor cells often have profound molecular alterations. The most well-known molecular change is the up-expression of hypoxia-inducible factors (HIF) (8), which can regulate numerous target gene transcription including genes participating in apoptosis, angiogenesis and cell proliferation. These hypoxia-induced genes can cause resistance to chemotherapy or radiotherapy and lead to poor survival. In addition to HIF gene, some other genes also participate in the process of hypoxia and probably contribute to the survival of cancer patients (9), such as genes in DNA repair system, mitotic spindle checkpoint and so on.
In this study, we determined whether single nucleotide polymorphisms (SNPs) within the hypoxia pathway genes are associated with survival in chemotherapy-treated advanced- state NSCLC patients. To our knowledge, this is the first effort to comprehensively study the role of genetic variants in genes in hypoxia pathway in the survival of NSCLC.
Material and methods
Patient population
In the discovery set, 602 histologically confirmed advanced-stage NSCLC patients were enrolled under an ongoing epidemiologic lung cancer study at MD Anderson Cancer Center between 1995 and 2008. In the validation set, 278 histologically confirmed advanced-stage NSCLC patients were enrolled from the same source and these patients were part of our genome-wide association study of lung cancer risk (10).
All patients were in stages III and IV according to seventh edition of the TNM lung cancer staging system by the International Association for the Study of Lung Cancer (IASLC) (11,12). Patients were treated with first-line platinum-based chemotherapy with or without radiotherapy, and did not receive surgery. Written informed consent was obtained by all study participants and the Institutional Review Board of the University of Texas MD Anderson Cancer Center approved the study. Blood samples were collected and delivered to the laboratory.
Demographic, epidemiologic and clinical data collection
Clinical data including TNM stage, performance status, chemotherapy and radiotherapy were abstracted from the medical records. Demographic information and epidemiologic data including smoking status were collected by trained staff using a structured questionnaire in-person interviews.
The primary endpoint was overall survival (OS), which was calculated from the date of diagnosis to the date to death. The data of OS was obtained from medical records or MD Anderson Tumor Registry.
SNP selection and genotyping
We generated hypoxia pathway gene list using the MSigDB database v4.0 (http://www.broadinstitute.org/gsea/index.jsp). A total of 42 genes were selected after extensive literature review and their relevance to cancer (Supplementary Table 1, available at Carcinogenesis Online). Tagging SNPs were identified using data from the International HapMap Project. Sequences 10 kb before the transcription start site and 10 kb after the transcription end site were included in the tag SNP selection. The Tagger pairwise method (Broad Institute, Cambridge, MA) was used for tagging SNPs selection with an r2 threshold of 0.8 and minor allele frequency of at least 0.05. Potential functional SNPs in the coding region were also included. A total of 598 SNPs were included for genotyping.
Genomic DNA was extracted from peripheral blood samples using a QIAamp DNA Mine Kit (QIAGEN, Valencia, CA). For the discovery phase, Illumina iSelected Chip was used according to the Infinium II assay protocol (Illumina). For the validation phase, Illumina’s HumanHap 317 BeadChip (San Diego, CA) was used for the genotyping. Only SNPs with a sample call rate greater than 95% and samples with a SNP call rate greater than 95% were included in the final analysis.
Statistical analysis
We used Cox’s proportional hazards model to estimate hazard ratios (HRs) and 95% confidence intervals (CIs) for the multivariate survival analyses, adjusting for age, gender, race, pack year, performance status and stage. Three different genetic models were tested: dominant model, recessive model and additive model. The model with the most significant P value was considered the best-fitting model. The relationship between genotype and survival time was evaluated by the Kaplan–Meier survival curves and analyzed using the log-rank test. Meta-analysis was used to combine the results from the discovery and validation sets adjusting for age, gender, race, pack year, performance status and stage. Expression quantitative trait loci (eQTL) analyses was performed using the Genevar database (13) (http://www.sanger.ac.uk/resources/software/genevar/) (14–16). STATA software (version 10, STATA Corporation, College Station, TX) was used in all the analyses. Differences were considered statistically significant at P < 0.05 (two-tailed test).
Result
Patient characteristics
The discovery set had a total of 602 stages III and IV NSCLC patients treated with chemotherapy with or without radiotherapy (Table 1). The median age was 61 years with 54.2% males and 45.8% females. Most patients were Caucasians (78.7%), had adenocarcinoma (51%), in stage IV (56.1%) and with performance status 1 (56.0%). Only 18.8% patients were never smokers, 40.7% patients were former smokers and 40.5% patients were current or recent quitters. The median pack years of smoking was 40 (0.1–156.0). All patients received platinum-based chemotherapy, and 40.4% patients also received radiotherapy. The median follow-up time was 43.9 months and 76.2% patients died at the end of follow-up period with a median survival time (MST) of 11.8 months.
Table 1.
Characteristics of the NSCLC patients
| Characteristic | Patients (%) | |
|---|---|---|
| Discovery phase | Validation phase | |
| Total, N | 602 | 278 |
| Survival time, median (range), months | 11.8 (0.1–128.7) | 15.8 (1.4–121.6) |
| Follow-up time, median (range), months | 43.9 (34.6–48.5) | 34.5 (27.9–49.1) |
| Age, median (range), years | 61 (28.0–81.0) | 63 (34.0–87.0) |
| Sex, N (%) | ||
| Male | 326 (54.2) | 160 (57.6) |
| Female | 276 (45.8) | 118 (42.4) |
| Race | ||
| Caucasians | 474 (78.7) | 278 (100) |
| Black | 95 (15.8) | 0 (0) |
| Hispanic | 33 (7.0) | 0 (0) |
| Stage, N (%) | ||
| IIIA | 82 (13.6) | 31 (11.2) |
| IIIB | 182 (30.2) | 62 (22.3) |
| IV | 338 (56.1) | 185 (66.5) |
| Performance status, N (%) | ||
| 0 | 144 (23.9) | 58 (20.9) |
| 1 | 337 (56.0) | 133 (47.8) |
| 2–4 | 74 (12.3) | 39 (14.0) |
| Unknown | 47 (7.8) | 48 (17.3) |
| Smoking status, N (%) | ||
| Never | 113 (18.8) | 62 (22.3) |
| Former | 245 (40.7) | 118 (42.4) |
| Current and recent quitter | 244 (40.5) | 98 (35.3) |
| Pack-years, median(range) | 40 (0.1~156.0) | 45 (0.1~150.0) |
| Histology, N (%) | ||
| Adenocarcinoma | 307 (51.0) | 153 (55.0) |
| Squamous cell carcinoma | 126 (20.9) | 55 (19.8) |
| Others | 169 (28.1) | 70 (25.2) |
| Treatment, N (%) | ||
| Chemotherapy only | 359 (59.6) | 150 (54.0) |
| Chemotherapy with radiotherapy | 243 (40.4) | 128 (46.0) |
| Vital status, N (%) | ||
| Dead | 459 (76.2) | 203 (73.0) |
| Alive | 143 (23.8) | 75 (27.0) |
The validation set contained 278 patients (Table 1). The median age was 63 years with 57.6% males and 42.4% females. All patients were Caucasians. Most patients were adenocarcinoma (55.0%), in stage IV (66.5%), and with a performance status of 1 (47.8%). Only 22.3% patients were never smokers, 42.4% were former smokers and 35.3% were current or recent quitters. The median pack years was 45 (0.1~150.0). All patients received platinum-based chemotherapy, and 46.0% patients also received radiotherapy. The median follow-up time was 34.5 months and 73.0% patients were dead at the end of follow-up period with a MST of 15.8 months.
Association of SNPs with OS in the discovery phase
There were 40 SNPs in 20 genes that reached nominal significance (P < 0.05) (Supplementary Table 2, available at Carcinogenesis Online) in the analyses of their associations with OS among a total of 357 analyzable SNPs.
Because the treatment might affect the survival of NSCLC patients, we further analyzed the associations between SNPs and OS stratified by treatment. For patients treated by platinum-based chemotherapy alone, 37 SNPs in 23 genes had a P value of < 0.05. For patients treated by chemotherapy and radiotherapy, 44 SNPs in 18 genes had had a P < 0.05 (Supplementary Table 3, available at Carcinogenesis Online).
Validation of the significant SNPs with OS
We selected all the nominally significant SNPs for validation using our existing genotyping data from genome-wide association study. After adjusted for age, gender, race, pack year, performance status and stage, three SNPs were validated in patients treated by chemotherapy alone, including two highly linked missense SNPs (r2 = 0.8) in EXO1 gene (Table 2). We only analyzed rs9350 in the following analyses because of the strong linkage of rs4658535 with rs9350. Patients with RPA1: rs2270412 AA+GA genotype exhibited significantly worse survival (discovery set: HR = 1.36, 95% CI: 1.06–1.75; validation set: HR = 1.65, 95% CI: 1.12–2.42; combined set: HR = 1.42, 95% CI: 1.16–1.74, P = 0.001) and patients with the missense EXO1: rs9350 AA+AG genotype (discovery set: HR = 0.74, 95% CI: 0.58–0.96; validation set: HR = 0.62, 95% CI: 0.41–0.96; combined set: HR = 0.70, 95% CI: 0.56–0.87, P = 0.001) has significantly better survival (Table 3). In Kaplan–Meier curve analysis, patients with the RPA1: rs2270412 AA+GA genotype had significantly shorter MST than those with the GG genotype (combined set: 10.5 months versus 12.7 months, P = 0.004) (Figure 1A); while patients with the EXO1: rs9350 GG genotype had significantly shorter MST than those with the AA+GA genotype (combined set: 11.5 months versus 13.2 months, P = 0.009) (Figure 1B). These SNPs were not significantly associated with survival in patients receiving chemotherapy and radiotherapy (Table 3).
Table 2.
SNPs that associated with OS in NSCLC patients
| SNP | Host Gene | chr | Site | Best model | Allelic change |
|---|---|---|---|---|---|
| rs2270412 | RPA1 | 17 | Intron | Dominant | G > A |
| rs9350 | EXO1 | 1 | Missense | Dominant | G > A |
| rs4658535a | EXO1 | 1 | Missense | Dominant | G > A |
ars4658535 showed linkage with rs9350 (r2 = 0.8).
Table 3.
The SNPs associated with OS for NSCLC patients in both discovery and validation groups by chemotherapy only and chemoradiotherapy treatments
| SNP | Genotype | Discovery | Validation | Combined | |||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Dead, N (%) | Alive, N (%) | HR (95% CI)a | P | Dead, N (%) | Alive, N (%) | HR (95% CI)a | P | HR (95% CI)a | P | ||
| Chemotherapy only | |||||||||||
| rs2270412 | GG | 194 (81.51) | 44 (18.49) | 1 (reference) | 65 (73.86) | 23 (26.14) | 1 (reference) | 1 (reference) | |||
| AA+GA | 104 (85.95) | 17 (14.05) | 1.36 (1.06–1.75) | 0.015 | 58 (93.55) | 4 (6.45) | 1.65 (1.12–2.42) | 0.011 | 1.44 (1.17–1.77) | 0.001 | |
| rs9350 | GG | 209 (85.31) | 36 (14.69) | 1 (reference) | 88 (85.44) | 15 (14.56) | 1 (reference) | 1 (reference) | |||
| AA+GA | 89 (78.07) | 25 (21.93) | 0.74 (0.58–0.96) | 0.025 | 35 (74.47) | 12 (25.53) | 0.62 (0.41–0.96) | 0.031 | 0.71 (0.57–0.89) | 0.002 | |
| Chemoradiotherapyb | |||||||||||
| rs2270412 | GG | 109 (65.66) | 57 (34.34) | 1 (reference) | 49 (58.33) | 35 (41.67) | 1 (reference) | 1 (reference) | |||
| AA+GA | 52 (67.53) | 25 (32.47) | 0.80 (0.56–1.15) | 0.226 | 31 (70.45) | 13 (29.55) | 1.35 (0.83–2.19) | 0.221 | 0.97 (0.72–1.29) | 0.819 | |
| rs9350 | GG | 102 (64.15) | 57 (35.85) | 1 (reference) | 61 (61.00) | 39 (39.00) | 1 (reference) | 1 (reference) | |||
| AA+GA | 59 (70.24) | 25 (29.76) | 1.34 (0.95–1.89) | 0.093 | 19 (67.86) | 9 (32.14) | 0.93 (0.54–1.61) | 0.803 | 1.21 (0.91–1.62) | 0.197 | |
aAdjusted by age, gender, race, pack year, performance status and stage.
bChemotherapy combined with radiotherapy concurrently or sequentially.
Figure 1.
Kaplan–Meier survival analysis of the significant SNPs with OS in NSCLC patients. (A) Kaplan–Meier estimates of rs2270412 and OS; (B) Kaplan–Meier estimates of rs9350 and OS.
Predictive markers for OS
The cumulative effect of unfavorable genotype was analyzed by counting the number of unfavorable genotype for each patient. Patients with the RPA1: rs2270412 GG genotype and EXO1: rs9350 AA+GA genotype was considered as reference. Patients treated by chemotherapy alone with either of the unfavorable genotypes (RPA1: rs2270412 AA+GA and EXO1: rs9350 GG) of these two SNPs exhibited a 1.26-fold increased risk of death (95% CI, 0.96–1.65, P = 0.09 in combined dataset), whereas patients with both of the unfavorable genotypes had an over 2-fold increased risk of death (HR = 2.02, 95% CI: 1.50–2.71, P = 3.16E−06 in combined dataset) (Table 4). The same SNPs appeared to exhibit an opposite effect in patients receiving chemotherapy plus radiation: patients with either one of the unfavorable genotypes or both of the unfavorable genotypes as defined above exhibited reduced risk of death with HRs of 0.73 (95% CI, 0.52–1.02, P = 0.063) and 0.78 (95% CI, 0.53–1.16, P = 0.227), respectively (Table 4).
Table 4.
Cumulative effect of unfavorable genotype associated with OS in NSCLC patients
| Number of unfavorable genotypesa | Discovery (N = 359) | Validation (N = 243) | Meta-analysis | |||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Dead, N (%) | Alive, N (%) | HR (95% CI)b | P | Dead, N (%) | Alive, N (%) | HR (95% CI)b | P | HR (95% CI)b | P | |
| Chemotherapy only | ||||||||||
| 0 | 62 (79.49) | 16 (20.51) | 1 (reference) | 19 (63.33) | 11 (36.67) | 1 (reference) | 1 (reference) | |||
| 1 | 159 (81.12) | 37 (18.88) | 1.10 (0.82–1.5) | 0.521 | 62 (82.67) | 13 (17.33) | 2.00 (1.13–3.51) | 0.017 | 1.26 (0.96–1.65) | 0.090 |
| 2 | 77 (90.59) | 8 (9.41) | 1.89 (1.32–2.7) | 0.001 | 42 (93.33) | 3 (6.67) | 2.68 (1.47–4.88) | 0.001 | 2.07 (1.52–2.81) | 3.53E−06 |
| Chemoradiotherapyc | ||||||||||
| 0 | 44 (72.13) | 17 (27.87) | 1 (reference) | 13 (65.00) | 7 (35.00) | 1 (reference) | 1 (reference) | |||
| 1 | 80 (62.50) | 48 (37.50) | 0.67 (0.45–0.99) | 0.044 | 42 (58.33) | 30 (41.67) | 0.91 (0.48–1.72) | 0.780 | 0.73 (0.52–1.02) | 0.063 |
| 2 | 37 (68.52) | 17 (31.48) | 0.61 (0.38–0.98) | 0.040 | 25 (69.44) | 11 (30.56) | 1.37 (0.68–2.77) | 0.377 | 0.78 (0.53–1.16) | 0.227 |
aThe unfavorable genotypes: GA+AA of rs2270412, GG of rs9350; rs4658535 was not included due to linkage with rs9350 (r2 = 0.8).
bAdjusted by age, gender, race, pack year, performance status and stage.
cChemotherapy combined with radiotherapy concurrently or sequentially.
In Kaplan–Meier curve analysis, compared to patients with no unfavorable genotype (RPA1: rs2270412 GG genotype and EXO1: rs9350 AA+GA genotype), who had a MST of 13.2 months in combined discovery and validation sets, patients treated by chemotherapy alone with either of the unfavorable genotypes (RPA1: rs2270412 AA+GA or EXO1: rs9350 GG) and both unfavorable genotypes had a MST of 12.7 months and 8.9 months, respectively (Figure 2). For patients treated with chemotherapy plus radiation, survival and unfavorable genotype analyses of the two SNPs did not generate significant results in the combined group (Supplementary Figure 1, available at Carcinogenesis Online; Table 4) suggesting the two predictive variants are treatment specific to platinum-based therapy only.
Figure 2.
Kaplan–Meier survival analysis for cumulative effect of unfavorable genotypes (UFG) on OS in NSCLC patients treated with chemotherapy only.
e-QTL analysis
To explore the potential function and underlying mechanism of the identified loci, we performed expression quantitative trait loci (eQTL) analysis using an online bioinformatics tool. Results from Genevar database showed that the genotype of rs2270412 was associated with altered expression of the RPA1 gene based on data from the Multiple Tissue Human Expression Resource (MuTHER) Study, which analyzed gene expression in adipose and skin tissues as well as lymphoblastoid cell lines (LCLs) derived from 196 Caucasian female twins. Compared with the AA genotype, the GG genotype of rs2270412 was associated with lower expression of the RPA1 gene in all tissue types (adipose, skin and LCLs); the correlation coefficient (rho) ranged from 0.23 to 0.34, and the P value ranged from 0.002 to 0.051 (Supplementary Figure 2, available at Carcinogenesis Online).
Discussion
Hypoxia is a hallmark of solid tumors and is associated with advanced disease stage and poor clinical outcome. In this study, we analyzed a large panel of hypoxia pathway gene SNPs and found one SNP each in exonuclease 1 (EXO1) and human replication protein A 1 (RPA1) gene associated with survival in late stage NSCLC patients receiving platinum-based chemotherapy.
RPA1, also known as RPA70, is important in DNA repair, replication and recombination. Phosphorylated RPA will promote DNA repair. It can bind to single strand (ssDNA) and interact with several proteins, such as p53 to form p53/RPA70N complex (17), which was shown as the primary factor for the resistance to apoptosis of hypoxic cancer cells (18). RPA1 is involved in cancer development and progression. Previous studies have shown that tumor cells had higher expression of RPA1 protein (19) and mRNA (20) than normal cells. Lower expression of RPA1 correlated with a lesser probability of survival in bladder urothelial carcinoma (21), and higher RPA1 expression was associated with worse outcome in esophageal carcinoma (19) and colon cancer (22). We report here that chemotherapy-treated NSCLC patients with RPA1: rs2270412 AA+GA genotype had a significantly worse survival than those with the GG genotype. Moreover, in e-QTL analysis, the GG genotype had lower expression of RPA1, consistent with prior studies that lower expression of RPA1 was associated with better survival and higher expression with worse survival in other cancers (19,20,21).
EXO1 encodes a multifunctional exonuclease that belongs to XPG/Rad2 family (23) and plays an important role in DNA damage repair, replication and homologous recombination (24). A recent study showed that a missense SNP (rs1047840, 1765G > A, Glu589Lys) in EXO1 was associated with relapse-free survival in head and neck squamous cell carcinoma patients (25) and a previous study reported another missense SNP (rs735943, His354Arg) in EXO1 was associated with OS in pancreatic cancer patients (26). In our current study, we also evaluated these two SNPs in our discovery phase but did not find significant associations. However, we found another missense SNP (rs9350, Leu757Pro) associated with OS in NSCLC patients receiving chemotherapy only. All these data suggest that EXO1 plays an important role in modulating the clinical course and chemotherapy response of cancer patients. The mechanisms underlying the link of EXO1 and prognosis in NSCLC and other cancers remain to be determined.
An interesting observation was that there appeared to be an opposite effect on survival by these two SNPs in NSCLC patient treated with chemotherapy alone or chemoradiation. These data suggest that in addition to clinicopathological variables, genetic information may need to be considered when deciding whether patients should receive chemotherapy alone or chemoradiation. For those patients with unfavorable genotypes for chemotherapy outcome, adding radiation to chemotherapy may improve patient outcomes.
In summary, to our knowledge, this is the first study to comprehensively evaluate SNPs in hypoxia pathway genes and survival of NSCLC patients. RPA1: rs2270412 AA+GA genotype and EXO1: rs9350 GG genotype were associated with poor survival in NSCLC patients treated by platinum-based chemotherapy alone. Adding radiotherapy could improve survival in patients harboring these unfavorable genotypes. If validated, these data may have clinical implications in personalized chemotherapy/chemoradiation of late state NSCLC patients.
Supplementary material
Supplementary data are available at Carcinogenesis online.
Funding
This work was supported in part by grants from the Cancer Prevention and Research Institute of Texas (RP1300502) and National Cancer Institute (P50 CA070907 and R01 CA176568). Additional funding was provided by MD Anderson institutional support for the Center for Translational and Public Health Genomics, Duncan Family Institute for Cancer Prevention and Risk Assessment.
Conflict of Interest Statement: None declared.
Supplementary Material
Abbreviations
- CTC
circulating tumor cell
- eQTL
expression quantitative trait loci
- MST
median survival time
- NSCLC
non-small cell lung cancer
- OS
overall survival
- SNP
single nucleotide polymorphism
References
- 1. Torre L.A., et al. (2015)Global cancer statistics, 2012. CA. Cancer J. Clin., 65, 87–108. [DOI] [PubMed] [Google Scholar]
- 2. Ettinger D.S., et al. (2014)Non-small cell lung cancer, version 1.2015. J. Natl. Compr. Canc. Netw., 12, 1738–1761. [DOI] [PubMed] [Google Scholar]
- 3. Polo V., et al. (2014)Maintenance strategies in stage IV non-small-cell lung cancer (NSCLC): in which patients, with which drugs? Ann. Oncol., 25, 1283–1293. [DOI] [PubMed] [Google Scholar]
- 4. Athey V.L., et al. (2012)Early diagnosis of lung cancer: evaluation of a community-based social marketing intervention. Thorax, 67, 412–417. [DOI] [PubMed] [Google Scholar]
- 5. Goldstraw P., et al. (2011)Non-small-cell lung cancer. Lancet, 378, 1727–1740. [DOI] [PubMed] [Google Scholar]
- 6. Nalejska E., et al. (2014)Prognostic and predictive biomarkers: tools in personalized oncology. Mol. Diagn. Ther., 18, 273–284. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Vaupel P., et al. (1989)Blood flow, oxygen and nutrient supply, and metabolic microenvironment of human tumors: a review. Cancer Res., 49, 6449–6465. [PubMed] [Google Scholar]
- 8. Ortmann B., et al. (2014)Cell cycle progression in response to oxygen levels. Cell. Mol. Life Sci., 71, 3569–3582. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Koi M., et al. (2011)Tumor hypoxia and genetic alterations in sporadic cancers. J. Obstet. Gynaecol. Res., 37, 85–98. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Amos C.I., et al. (2008)Genome-wide association scan of tag SNPs identifies a susceptibility locus for lung cancer at 15q25.1. Nat. Genet., 40, 616–622. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Groome P.A., et al. (2007)The IASLC Lung Cancer Staging Project: validation of the proposals for revision of the T, N, and M descriptors and consequent stage groupings in the forthcoming (seventh) edition of the TNM classification of malignant tumours. J. Thorac. Oncol., 2, 694–705. [DOI] [PubMed] [Google Scholar]
- 12. Rami-Porta R., et al. (2007)The IASLC Lung Cancer Staging Project: proposals for the revision of the T descriptors in the forthcoming (seventh) edition of the TNM classification for lung cancer. J. Thorac. Oncol., 2, 593–602. [DOI] [PubMed] [Google Scholar]
- 13. Yang T.P., et al. (2010)Genevar: a database and Java application for the analysis and visualization of SNP-gene associations in eQTL studies. Bioinformatics, 26, 2474–2476. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Stranger B.E., et al. (2012)Patterns of cis regulatory variation in diverse human populations. PLoS Genet., 8, e1002639. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Grundberg E., et al. (2012)Mapping cis- and trans-regulatory effects across multiple tissues in twins. Nature Genet., 44, 1084–1089. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Grundberg E., et al. (2013)Global analysis of DNA methylation variation in adipose tissue from twins reveals links to disease-associated variants in distal regulatory elements. Am. J. Hum. Genet., 93, 876–890. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Kaustov L., et al. (2006)p53 transcriptional activation domain: a molecular chameleon? Cell Cycle, 5, 489–494. [DOI] [PubMed] [Google Scholar]
- 18. Madan E., et al. (2012)p53 Ser15 phosphorylation disrupts the p53-RPA70 complex and induces RPA70-mediated DNA repair in hypoxia. Biochem. J., 443, 811–820. [DOI] [PubMed] [Google Scholar]
- 19. Dahai Y., et al. (2013)A relationship between replication protein A and occurrence and prognosis of esophageal carcinoma. Cell Biochem. Biophys., 67, 175–180. [DOI] [PubMed] [Google Scholar]
- 20. Hoxha M., et al. (2014)Relevance of telomere/telomerase system impairment in early stage chronic lymphocytic leukemia. Genes. Chromosomes Cancer, 53, 612–621. [DOI] [PubMed] [Google Scholar]
- 21. Levidou G., et al. (2011)Prognostic significance of replication protein A (RPA) expression levels in bladder urothelial carcinoma. BJU Int., 108 (2 Pt 2), E59–E65. [DOI] [PubMed] [Google Scholar]
- 22. Givalos N., et al. (2007)Replication protein A is an independent prognostic indicator with potential therapeutic implications in colon cancer. Mod. Pathol., 20, 159–166. [DOI] [PubMed] [Google Scholar]
- 23. Schaetzlein S., et al. (2013)Mammalian Exo1 encodes both structural and catalytic functions that play distinct roles in essential biological processes. Proc. Natl. Acad. Sci. USA, 110, E2470–E2479. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Desai A., et al. (2014)Exonuclease 1 is a critical mediator of survival during DNA double strand break repair in nonquiescent hematopoietic stem and progenitor cells. Stem Cells, 32, 582–593. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Nogueira G.A., et al. (2015)Association between genetic polymorphisms in DNA mismatch repair-related genes with risk and prognosis of head and neck squamous cell carcinoma. Int. J. Cancer, 137, 810–818. [DOI] [PubMed] [Google Scholar]
- 26. Dong X., et al. (2009)Significant associations of mismatch repair gene polymorphisms with clinical outcome of pancreatic cancer. J. Clin. Oncol., 27, 1592–1599. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.