https://orcid.org/0000-0002-9224-631X
Figures
Abstract
Background
Platinum-based chemotherapy is one of the most common treatments for many cancers; however, the effect of chemotherapy varies from individual to individual. Excision repair cross complementation group 1 (ERCC1) is widely recognized as a key gene regulating nucleotide excision repair (NER) and is closely associated with platinum response. Many studies have yielded conflicting results regarding whether ERCC1 polymorphisms can affect the response to platinum and overall survival (OS). Therefore, it is necessary to perform a meta-analysis of patients with specific races and cancer types.
Methods
Eight databases (EMBASE, PubMed, Cochrane Library, Chinese National Knowledge Infrastructure, Scopus, VIP, China Biology Medicine disc and Wanfang databases) were searched. Results were expressed in terms of odds ratios (ORs), hazard ratios (HRs) and 95% CIs.
Results
In this study, rs11615, rs2298881 and rs3212986 SNPs were studied. In the comparison between CT and TT on the response to platinum, esophageal cancer [I2 = 0%, OR = 6.18, 95% CI(1.89,20.23), P = 0.003] and ovarian cancer [I2 = 0%, OR = 4.94, 95% CI(2.21,11.04), P<0.001] showed that the rs11615 CT genotype predicted a better response. In the comparison between CC and TT, ovarian cancer [I2 = 48.0%, OR = 6.15, 95% CI (2.56,14.29), P<0.001] indicated that the CC genotype predicted a better response. In the meta-analysis of OS, the CC genotype was related to longer OS than TT in ovarian cancer [TT vs CC: I2 = 57.7%, HR = 1.71, 95% CI (1.18, 2.49), P<0.001].
Citation: Wu X, Lu W, Jiang C, Zhang D, Zhang W, Cui Y, et al. (2023) Effect of ERCC1 polymorphisms on the response to platinum-based chemotherapy: A systematic review and meta-analysis based on Asian population. PLoS ONE 18(5): e0284825. https://doi.org/10.1371/journal.pone.0284825
Editor: Mohammad Safiqul Islam, Noakhali Science and Technology University Faculty of Pharmacy, BANGLADESH
Received: December 11, 2022; Accepted: April 6, 2023; Published: May 4, 2023
Copyright: © 2023 Wu et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the manuscript and its Supporting Information files.
Funding: This work was supported by the Beijing Municipal Natural Science Foundation [No. 7212192]. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Platinum-based chemotherapy is the standard treatment for many cancers, such as ovarian cancer, gastric cancer, esophageal cancer, osteosarcoma, etc. Despite the clear clinical benefit of platinum, some patients still exhibit drug resistance and these patients are increasing with increasing treatment lines. Some studies have shown this phenomenon can be explained by the nucleotide excision repair (NER). DNA damage by platinum is the main anti-cancer mechanism of platinum [1, 2], and the damage can be repaired by the NER pathway [3]. Therefore, individual differences in NER capacity will lead to individual differences in response to platinum and thus to different prognoses.
It would certainly benefit patients and clinicians if we could predict the response of patients to platinum before they receive platinum-based chemotherapy. Platinum-sensitive patients receive platinum-containing chemotherapy, while platinum-insensitive patients receive non-platinum chemotherapy or other treatments. It can not only help clinicians to make decisions, but also reduce the pain caused by treatment. Excision repair cross complementation group 1 (ERCC1) component is a mammalian endonuclease that cleaves damaged DNA strands during the repair process of NER and interchain crosslinking. Detection of ERCC1 polymorphisms is expected to predict the efficacy of platinum-based chemotherapy. Two sub-pathways of NER, called global genome repair (GGR) and transcription-coupled repair (TCR), can be mentioned. DNA adducts including those induced by cisplatin are repaired by the nucleotide excision repair (NER) pathway [4]. ERCC1 is a subunit of the damaged 5′-side DNA endonuclease [5]. Following ERCC1-XPF nuclease mediated 5′ incision, the damaged oligonucleotide is excised and appears to bind to one or more components of the repair complex.
Many studies have found conflicting effects of ERCC1 polymorphisms on the platinum response. Gradually, it became clear that the cause of this conflict was ethnic differences. In studies examining the relationship between ERCC1 polymorphisms and the responsiveness of patients with non-small cell lung cancer treated with cisplatin, the results showed that ERCC1 polymorphisms were not associated with platinum responsiveness and there was no evidence of a potential prognostic role in a Caucasian population [6, 7]. While in Asian population, the T allele is a biomarker of low objective response, short progression free survival and overall survival (OS) in patients with gastric, osteosarcoma, ovarian and colorectal cancer [8–11]. So it is necessary to conduct a separate study on the impact of ERCC1 polymorphisms on platinum response and prognosis in Asian populations.
Currently, the most studied SNPs are rs11615, rs2298881 and rs3212986. Because the current study showed that the results of ERCC1 SNPs as a predictor of platinum treatment response and prognosis were contradictory, we performed a systemic review and meta-analysis to assess the evidence of effects of ERCC1 rs11615 C>T, rs2298881C>A and rs3212986 C>A SNPs on the efficacy of platinum-based chemotherapy and overall survival.
Materials and methods
1. Literature search
The systematic review was performed according to the Meta-Analysis of Observational Studies in Epidemiology (i.e., MOOSE), Preferred Reporting Items for Systematic Reviews and Meta-analysis (PRISMA) statement and protocol. The systematic review protocol was registered with the international prospective register of systematic reviews (PROSPERO): CRD42022339301.
Two reviewers (CH J and ZL Z) independently searched databases including EMBASE, PubMed, Cochrane Library, Chinese National Knowledge Infrastructure (CNKI), Scopus, VIP, China Biology Medicine disc (CBM) and Wanfang from inception to June 2022. The search strategy is displayed in the S1 Table. The search time was from the establishment of the database to June 10, 2022. Two reviewers (CH J and DN Z) independently screened the titles and abstracts of all retrieved records to exclude irrelevant studies. The remaining studies were assessed by reading the full text. Any disagreement was resolved by consensus or by involving an arbiter (XQ W).
2. Study selection
Inclusion criteria: (1) Population: Asian patients who were included in detection (primary detection) or being followed for recurrent disease and had cancer recurrence detected in histopathology. Patients who were receiving chemotherapy including cisplatin, carboplatin, oxaliplatin or nedaplatin. (2) Intervention: ERCC1 polymorphisms, including rs11615 (CT vs CC vs TT), rs3212986 (AA vs CA vs CC) and rs2298881 (AA vs CA vs CC). (3) Control: each phenotypic data was served as a comparator with the other two (rs11615: CT vs CC vs TT; rs3212986: AA vs CA vs CC and rs2298881: AA vs CA vs CC). (4) Outcome: Primary outcome is served as the response to platinum-based chemotherapy in each SNP. After receiving platinum containing chemotherapy, patients with CR or PR were defined as having good response, while those with SD or PD were defined as having poor response. The secondary outcome was defined as OS for each SNP. OS was defined as the time from the observation point to death from any cause. (5) Study design: cohort and case control studies.
Exclusion criteria: All studies with mutated bases in gene fragments that did not meet the requirements listed in the inclusion criteria were excluded. We also excluded studies reporting data on laboratory developments tests as well as reviews, letters to editors, editorials, study protocols, case reports, brief correspondence, and articles published in languages other than English or Chinese. The references of all papers included were scanned for additional studies of interest.
3. Data extraction
Three investigators independently extracted the following information from the included articles: author name, publication year, country, chemotherapy regimen, number of patients, cancer species, number of good responses (CR+PR) and poor response (SD+PD), hazard ratio and 95% confidence interval between different genotypes and overall survival. The Newcastle‒Ottawa Quality Assessment Scale was used to assess the quality of each of the included studies. Any discrepancy was resolved by discussion or by involving an arbiter (XQ W).
4. Trial sequential analysis (TSA)
We assessed the risk of false positives or false negatives by TSA in a meta-analysis. Sequential monitoring boundaries were established to limit Type I error to 5% and Type II error to 20% to detect responsiveness to platinum and OS of patients under various models. When cumulative Z curves crossed conventional boundaries and trial sequential monitoring boundaries and the sample size reached the accrued information size (AIS), the expected intervention effect may reach a sufficient level of evidence. TSA was completed using the Stata, version 13.0, R package "foreign" and "ldBounds".
5. Statistical analysis
The effect estimates of interest were the OR (95%CI) in the meta-analysis of response to platinum and HR (95%CI) in the meta-analysis of overall survival. Studies exploring the response to platinum and overall survival of cancers were analyzed separately, and subgroup analyses were performed according to different cancer species. Assessment of heterogeneity was performed using Higgins I2; I2 greater than 50% and a P value less than 0.10 suggested significant heterogeneity, and a random effect model was applied; I2 less than 50% and a P value more than 0.10 suggested tolerable heterogeneity, and a fixed effect model was applied. Sensitivity analysis was performed by sequentially omitting each study to examine the robustness of the results. Potential publication bias was evaluated using a funnel plot (number of studies ≥ 10) and Begg’s and Egger’s tests. If significant publication bias existed, the trim and fill method was performed to validate the robustness of the meta-analysis results. Sensitivity analysis adopts the row-by-row division method. All statistical analyses were calculated via Stata, version 13.0 (StataCorp, College Station, TX). P values < 0.05 and Bonferroni-corrected p < 0.017 in studies with the response to platinum as the outcome (Results 2) and p values <0.05 and Bonferroni-corrected p < 0.025 in studies with OS as the outcome (Results 3), were defined as statistically significant, except those for heterogeneity.
Results
1. Literature research
We identified 42 studies in CBM, 228 studies in Embase, 274 studies in the PubMed database, 32 studies in the Cochrane Library, 218 studies in the CNKI database, 86 studies in the VIP database, 20 studies in Scopus, and 306 studies in the Wanfang database. The flowchart of study selection is shown in Fig 1, and 27 studies were included in our meta-analysis. Among these studies, 26 studies reported the relationship between the ERCC1 rs11615 (C>T) polymorphism and platinum reactivity, 8 of which reported the effect of the rs11615 (C>T) polymorphism on OS. Six studies reported the relationship between the rs229981 (C>A) polymorphism and the response to platinum, 3 of which focused on the influence of the rs229981 (C>A) polymorphism on OS. Ten studies observed an association between the rs3212986 (C>A) polymorphism and the response to platinum, 6 of which also reported the rs3212986 (C>A) polymorphism and OS. The characteristics of these studies are summarized in Table 1. Newcastle‒Ottawa Scale (NOS) assessments are shown in Table 2.
2. Response to platinum‑based chemotherapy
2.1 The rs11615 polymorphism.
2.1.1 Dominant model (CT+TT vs CC). The rs11615 polymorphism has a wild-type C allele and a variant T allele. A total of 26 studies [12–37], involving 4889 patients were included. There were 4 studies on ovarian cancer, 5 on osteosarcoma, 2 on esophageal cancer, 6 on gastric cancer, 6 on lung cancer, 1 on nasopharyngeal, 1 on cervical cancer and 1 on colon cancer. CT+TT vs CC: I2 = 50.4%, random effect model was used, OR = 0.81, 95% CI (0.66,0.99), P = 0.041, as shown in Fig 2(A). Sensitivity analysis showed unreliable results (Fig 2(B)). The results of Meta-regression showed that different cancer types may be the source of heterogeneity, especially colon cancer and osteosarcoma, with p-values of 0.088 and 0.089, respectively. Subsequently, subgroup analysis was performed according to cancer type, which showed that the ERCC1 rs11615 CT+TT genotype in ovarian cancer was related to a better platinum response. CT+TT vs CC: I2 = 0%, subtotal OR = 0.77, 95%CI(0.60, 0.97), P = 0.030, while esophageal, lung, gastric, nasopharyngeal, cervical cancer and osteosarcoma were not related. The results of the subgroup analysis are shown in Fig 2(C). It should be noted that there was only one study in the colon, nasopharyngeal and cervical cancer groups.
(a) Results of the response to platinum-based chemotherapy in all cancer types. (b) Results of the sensitivity analysis. (c) Results of subgroup analysis of the response to platinum-based chemotherapy according to cancer type. (d) The funnel plot of publication bias. (e) TSA on the response to platinum. The x-axis ticks refer to the studies added sequentially into the analysis. The blue curve was the adjusted Z-curve, and it crossed the dark-green horizontal line (conventional boundary for α = 0.05). The vertical red line shows the required information size—the needed sample size for a robust result with sufficient statistical power. The analysis included 4889 participants, which exceeded the required information size (n = 4733).
The funnel plot was basically symmetrical (Fig 2(D)), and Begg’s test and Egger’s test showed P = 0.243 and 0.251, respectively, indicating no publication bias. The TSA results showed that the enrolled studies were adequate (Fig 2(E)).
2.1.2 Heterozygous model (CT vs TT). Number of studies and people involved in the study as described in 2.1.1. The ERCC1 rs11615 polymorphism CT genotype was related to the response to platinum-based chemotherapy. Heterozygous model (CT vs TT): I2 = 52.0%, random effect model was used, OR = 1.43, 95% CI(1.07,1.92), P = 0.016, as shown in Fig 3(A). In the subgroup analysis, esophageal cancer [I2 = 0%, OR = 6.18, 95% CI (1.89, 20.23), P = 0.003] and ovarian cancer [I2 = 0%, OR = 4.94, 95% CI (2.21, 11.04), P<0.001] showed that the rs11615 CT genotype predicted a better response to platinum. In gastric cancer [I2 = 8.0%, OR = 1.00, 95%CI(0.75,1.32), P = 0.975], lung cancer [I2 = 0%, OR = 0.78, 95%CI(0.37,1.64), P = 0.509], colon cancer [OR = 1.81, 95%CI(0.55,6.01), P = 0.331], nasopharyngeal [OR = 1.25,95%CI(0.06,26.87)], cervical cancer [OR = 0.39,95%CI(0.02,9.58)] and osteosarcoma [I2 = 35.5%, OR = 1.25, 95%CI(0.81,1.92), P = 0.331], CT genotype seems to be not related to the response to platinum. The forest plot is shown in Fig 3(B). It should be noted that there was only one study in the colon, nasopharyngeal and cervical cancer groups.
(a) Result of the response to platinum-based chemotherapy in all cancer types. (b) Result of subgroup analysis of the response to platinum-based chemotherapy according to cancer types. (c) The funnel plot of publishing bias. (d) Result of the sensitivity analysis. (e) TSA on the response to platinum. The analysis included 4889 participants, which exceeded the required information size (n = 2703).
The funnel plot was basically symmetrical (Fig 3(C)). Begg’s test and Egger’s test showed that P = 0.675 and 0.110, respectively, indicating that there was no publication bias. The sensitivity analysis showed that the result was reliable (Fig 3(D)). The TSA results showed that the enrolled studies were adequate (Fig 3(E)).
2.1.3 Homozygous model (CC vs TT). Number of studies and people involved in the study as described in 2.1.1. ERCC1 rs11615 CC genotype was related to a good response to platinum-based chemotherapy. Homozygous model (CC vs TT): I2 = 53.6%, random effect model was used, OR = 1.84, 95%CI(1.32,2.58), P<0.001, as shown in Fig 4(A). In the subgroup analysis, esophageal cancer [I2 = 0%, OR = 3.85, 95%CI (1.22,12.13), P = 0.021], ovarian cancer [I2 = 48.0%, OR = 6.15, 95% CI (2.56,14.29), P<0.001] and colon cancer [OR = 3.50, 95% CI (1.04,11.74), P = 0.043] showed that CC predicted a better response to platinum than the TT genotype. In gastric cancer [I2 = 58.3%, OR = 1.18, 95%CI(0.67,2.06), P = 0.568], lung cancer [I2 = 0%, OR = 1.00, 95%CI(0.47,2.16), P = 0.993], nasopharyngeal [OR = 2.20,95%CI(0.11,42.73)], cervical cancer [OR = 0.52,95%CI(0.02,12.12)] and osteosarcoma [I2 = 50.5%, OR = 1.75, 95%CI(1.03,2.97), P = 0.039], CC vs TT seems to be not related to the response to platinum. The subgroup plot is shown in Fig 4(B). It should be noted that there was only one study in the colon, nasopharyngeal and cervical cancer groups.
(a) Result of the response to platinum-based chemotherapy in all cancer species. (b) Result of subgroup analysis of the response to platinum-based chemotherapy according to cancer species. (c) The funnel plot of publishing bias. (d) Result of the sensitivity analysis. (e) TSA on the response to platinum. The analysis included 4889 participants, which exceeded the required information size (n = 2919).
The funnel plot was basically symmetrical (Fig 4(C)). Begg’s test and Egger’s test showed that P = 0.675 and 0.175, respectively, indicating that there was no publication bias. The sensitivity analysis showed that the result was reliable (Fig 4(D)). The TSA results showed that the enrolled studies were adequate (Fig 4(E)).
2.2 The rs2298881 A/C polymorphisms.
2.2.1 Dominant model (CA+AA vs. CC). The rs2298881 polymorphism has a wild-type C allele and variant A allele. A total of 6 studies [16, 17, 19, 20, 37, 38], involving 1678 patients were included. There were 3 studies on gastric cancer, 2 on osteosarcoma and 1 on ovarian cancer. The CA+AA and CC genotypes showed no difference in platinum sensitivity. Dominant model (CA+AA vs CC): I2 = 69.4%, random effect model was used, OR = 0.90, 95% CI (0.55,1.49), P = 0.693, as shown in Fig 5(A). In the subgroup analysis, the CA+AA vs CC genotype was not related to platinum response in the gastric cancer [I2 = 28.5%, OR = 0.74, 95% CI (0.46,1.18), P = 0.204], osteosarcoma [I2 = 90.9%, OR = 1.09, 95% CI (0.13,9.53), P = 0.936] and ovarian cancer [OR = 1.10, 95% CI (0.75,1.62), P = 0.638] groups, as shown in Fig 5(B).
(a) Result of the response to platinum-based chemotherapy in all cancer types. (b) Result of subgroup analysis of the response to platinum-based chemotherapy according to cancer type. (c) The funnel plot of publishing bias. (d) Result of the sensitivity analysis. (e) TSA on the response to platinum. The analysis included 1678 participants, which does not exceed the required information size (n = 1767).
The funnel plot was basically symmetrical (Fig 5(C)). Begg’s test and Egger’s test showed that P = 0.452 and 0.829, respectively, indicating that there was no publication bias. The sensitivity analysis showed that the result was robust (Fig 5(D)). The TSA results showed that the enrolled studies were not adequate and need to be verified by more relevant studies (Fig 5(E)).
2.2.2 Heterozygous model (CA vs CC). Number of studies and patients involved in the study as described in 2.2.1. The CA and CC genotypes showed no difference in platinum sensitivity. Heterozygous model (CA vs CC): I2 = 45.9%, fixed effect model was used, OR = 0.94, 95% CI(0.72,1.23), P = 0.653, as shown in Fig 6(A). In the subgroup analysis, the CA vs CC genotype was not related to platinum response in the gastric cancer [I2 = 42.2%, OR = 0.75, 95% CI (0.32,1.73), P = 0.495], osteosarcoma [I2 = 81.2%, OR = 1.12, 95% CI (0.57,2.17), P = 0.747] and ovarian cancer [OR = 1.11, 95% CI (0.74,1.67), P = 0.600] groups, as shown in Fig 6(B).
(a) Result of the response to platinum-based chemotherapy in all cancer species. (b) Result of subgroup analysis of the response to platinum-based chemotherapy according to cancer species. (c) The funnel plot of publishing bias. (d) Result of the sensitivity analysis. (e) TSA on the response to platinum. The analysis included 1678 participants, which exceeded the required information size (n = 1083).
The funnel plot was basically symmetrical (Fig 6(C)). Begg’s test and Egger’s test showed that P = 0.462 and 0.860, respectively, indicating that there was no publication bias. The sensitivity analysis showed that the result was robust (Fig 6(D)). The TSA results showed that the enrolled studies were adequate (Fig 6(E)).
2.2.3 Homozygous model (AA vs CC). Number of studies and patients involved in the study as described in 2.2.1. The AA and CC genotypes showed no difference in platinum sensitivity. Homozygous model (AA vs CC): I2 = 76.2%, random effect model was used, OR = 0.90, 95% CI(0.46,1.76), P = 0.758, as shown in Fig 7(A). In the subgroup analysis, the AA vs CC genotype was not related to platinum response in the gastric cancer [I2 = 45.8%, OR = 0.69, 95% CI (0.36,1.30), P = 0.248], osteosarcoma [I2 = 93.3%, OR = 1.31, 95% CI (0.09,19.60), P = 0.843] and ovarian cancer [OR = 1.04, 95% CI (0.61,1.79), P = 0.873] groups, as shown in Fig 7(B).
(a) Result of the response to platinum-based chemotherapy in all cancer species. (b) Result of subgroup analysis of the response to platinum-based chemotherapy according to cancer species. (c) The funnel plot of publishing bias. (d) Result of the sensitivity analysis. (e) TSA on the response to platinum. The analysis included 1678 participants, which exceeded the required information size (n = 1076).
The funnel plot was basically symmetrical (Fig 7(C)). Begg’s test and Egger’s test showed that P = 1.000 and 0.969, respectively, indicating that there was no publication bias. The sensitivity analysis showed that the result was robust (Fig 7(D)). The TSA results showed that the enrolled studies were adequate (Fig 7(E)).
2.3 The rs3212986 A/C polymorphisms in ERCC1.
2.3.1 Dominant model (CA+AA vs. CC). The rs3212986 polymorphism has a wild-type C allele and variant A allele. A total of 9 studies [16, 17, 20, 24, 25, 33–36], involving 2545 patients were included. There are 7 studies on gastric cancer, 1 on lung cancer, 1 on ovarian cancer and 1 on osteosarcoma. The CA+AA vs. CC phenotype showed no difference in response to platinum. Dominant model (CA+AA vs. CC): I2 = 65.8%, random effect model was used, OR = 0.91, 95% CI(0.68,1.22), P = 0.542, as shown in Fig 8(A). A subgroup analysis was subsequently carried out. In the subgroup analysis, the CA+AA vs. CC genotype was not related to platinum response in the gastric cancer [I2 = 56.9%, OR = 1.05, 95% CI (0.77,1.43), P = 0.772], osteosarcoma [OR = 1.10, 95% CI (0.62,1.96), P = 0.742] and ovarian cancer [OR = 0.71, 95% CI (0.50,1.02), P = 0.061] groups. The CA+AA vs. CC genotype was related to the response to platinum in lung cancer [OR = 0.23, 95% CI (0.10, 0.57), P = 0.002]. The results are shown in Fig 8(B). There was only one study in osteosarcoma, lung cancer and ovarian cancer.
(a) Result of the response to platinum-based chemotherapy in all cancer types. (b) Result of subgroup analysis of the response to platinum-based chemotherapy according to cancer type. (c) The funnel plot of publishing bias. (d) Result of the sensitivity analysis. (e) TSA on the response to platinum. The analysis included 2545 participants, which exceeded the required information size (n = 2511).
The funnel plot was basically symmetrical (Fig 8(C)). Begg’s test and Egger’s test showed that P = 0.592 and 0.993, respectively, indicating that there was no publication bias. The sensitivity analysis showed that the results were robust (Fig 8(D)). The TSA results showed that the enrolled studies were adequate (Fig 8(E)).
2.3.2 Heterozygous model (CA vs CC). Number of studies and patients involved in the study as described in 2.3.1. The CA and CC genotypes showed no difference in response to platinum. Heterozygous model (CA vs CC): I2 = 55.8%, random effect model was used, OR = 0.91, 95% CI(0.68,1.23), P = 0.551, as shown in Fig 9(A). A subgroup analysis was subsequently carried out. In the subgroup analysis, the CA vs CC genotype was not related to platinum response in gastric cancer [I2 = 43.6%, OR = 1.04, 95% CI (0.75,1.43), P = 0.821], osteosarcoma [OR = 0.98, 95% CI (0.54,1.79), P = 0.949] or ovarian cancer [OR = 0.87, 95% CI (0.59,1.28), P = 0.489]. The CA vs CC genotype was related to the response to platinum in the lung cancer [OR = 0.20, 95% CI (0.07, 0.57), P = 0.002] groups, as shown in Fig 9(B).
(a) Result of the response to platinum-based chemotherapy in all cancer species. (b) Result of subgroup analysis of the response to platinum-based chemotherapy according to cancer species. (c) The funnel plot of publishing bias. (d) Results of the sensitivity analysis. (e) TSA on the response to platinum. The analysis included 2545 participants, which exceeded the required information size (n = 1984).
The funnel plot was basically symmetrical (Fig 9(C)). Begg’s test and Egger’s test showed that P = 0.251 and 0.391, respectively, indicating that there was no publication bias. The sensitivity analysis showed that the result was robust (Fig 9(D)). The TSA results showed that the enrolled studies were adequate (Fig 9(E)).
2.3.3 Homozygous model (AA vs CC). Number of studies and number of patients involved in the study as described in 2.3.1. The AA and CC genotypes showed no difference in response to platinum. Homozygous model (AA vs CC): I2 = 66.1%, random effect model was used, OR = 0.96, 95% CI(0.56,1.64), P = 0.883, as shown in Fig 10(A). In the subgroup analysis, the AA vs CC genotype was not related to platinum response in the gastric cancer [I2 = 56.2%, OR = 1.19, 95% CI (0.66,2.13), P = 0.562], osteosarcoma [OR = 1.93, 95% CI (0.64,5.84), P = 0.247] and lung cancer [OR = 0.33, 95% CI (0.08,1.30), P = 0.113] groups. The AA vs CC genotype was related to the response to platinum in the ovarian cancer [OR = 0.40, 95% CI (0.22, 0.73), P = 0.003] group. The above results are shown in Fig 10(B).
(a) Result of the response to platinum-based chemotherapy in all cancer species. (b) Result of subgroup analysis of the response to platinum-based chemotherapy according to cancer species. (c) The funnel plot of publishing bias. (d) Results of the sensitivity analysis. (e) TSA on the response to platinum. The analysis included 2545 participants, which exceeded the required information size (n = 1984).
The funnel plot was basically symmetrical (Fig 10(C)). Begg’s test and Egger’s test showed that P = 0.466 and 0.336, respectively, indicating that there was no publication bias. The sensitivity analysis showed that the result was robust (Fig 10(D)). The TSA results showed that the enrolled studies were adequate (Fig 10(E)).
3. Overall survival in ERCC1 polymorphism
3.1 The rs11615 polymorphism.
3.1.1 Homozygous model (TT vs CC). A total of 8 studies [12–15, 19, 21, 24, 34] involving 1569 patients were included in the homozygous model. There were 3 studies on ovarian cancer, 2 on osteosarcoma, 2 on gastric cancer, and 1 on esophageal cancer. CC predicted longer overall survival than the TT genotype. Homozygous model (TT vs CC): I2 = 25.5%, fixed effect model was used, HR = 1.67, 95% CI(1.27,2.20), P<0.001, as shown in Fig 11(A). In subgroup analysis, TT vs CC genotype was not related to OS in esophageal cancer [HR = 1.67, 95%CI(0.46,6.04), P = 0.434], gastric cancer [I2 = 0%,HR = 1.67,95%CI(0.97,2.88), P = 0.064] and osteosarcoma [I2 = 77.5%,HR = 1.51,95%CI(0.75,3.06), P = 0.246] groups. The CC genotype was related to longer overall survival in ovarian cancer [I2 = 57.7%, HR = 1.71, 95% CI (1.18, 2.49), P<0.001], as shown in Fig 11(B). There was only one study in the esophagus.
(a) Result of the overall survival of the homozygous model in all cancer species. (b) Result of subgroup analysis of the overall survival of the homozygous model according to cancer type. (c) Results of the sensitivity analysis. (d) TSA on the overall survival in homozygous model. (e) Result of the overall survival of the heterozygous model in all cancer type. (f) Result of subgroup analysis of the overall survival of the heterozygous model according to cancer type. (g) Result of the sensitivity analysis. (h) TSA of overall survival in the heterozygous model.
Begg’s test and Egger’s test showed that P = 0.902 and 0.530, respectively, indicating that there was no publication bias. The sensitivity analysis showed that the result was reliable (Fig 11(C)). TSA showed that the AIS was 1569, and the sample size met the requirement (Fig 11(D)).
3.1.2 Heterozygous model (CT vs CC). A total of 7 studies [12, 13, 15, 19, 21, 24, 34] involving 1360 patients were included in the homozygous model. Heterozygous model (CT vs CC): I2 = 40.1%, fixed effect model was used, HR = 1.22, 95% CI (0.95,1.57), P = 0.115, as shown in Fig 11(E). Subgroup analysis was subsequently carried out. In subgroup analysis, CT vs CC genotype was not related to OS in esophageal cancer [HR = 1.11, 95%CI(0.25,4.98), P = 0.892], osteosarcoma [I2 = 0%,HR = 0.94,95%CI(0.56,1.59), P = 0.824], gastric cancer [I2 = 79.0%, HR = 1.30, 95%CI(0.84,2.01), P = 0.247] and ovarian cancer [I2 = 66.4%,HR = 1.35,95%CI(0.92,1.98), P = 0.123] groups. The forest plot is shown in Fig 11(F). There was only one study in the esophageal cancer group.
Begg’s test and Egger’s test showed that P = 0.548 and 0.398, respectively, indicating that there was no publication bias. The sensitivity analysis showed that the results were not robust (Fig 11(G)). TSA showed that the AIS was 1360, and the sample size met the requirement (Fig 11(H).
3.2 The rs2298881 A/C polymorphisms.
3.2.1 Homozygous model (AA vs CC). A total of 3 studies [19, 20, 38] were included in the homozygous model, involving 663 patients. There were 3 studies on gastric cancer, 2 on osteosarcoma and 1 on ovarian cancer. Homozygous model (AA vs CC): I2 = 26.4%, fixed effect model was used, HR = 1.80, 95% CI(1.01,3.18), P = 0.044, as shown in Fig 12(A). Because there were too few studies included, the meta-analysis was conducted again after the study was trimmed and completed to verify the sensitivity. HR = 1.06, 95% CI (0.68,1.67), P = 0.800, as shown in Fig 12(B). AA vs CC was not related to overall survival. TSA showed that the AIS was 663, and the sample size met the requirement (Fig 12(C).
(a) Result of the overall survival of the homozygous model. (b) Result of the overall survival of the homozygous model after meta-trim. (c) TSA on the overall survival in homozygous model. (d) Result of the overall survival of the heterozygous model. (e) Result of the overall survival of the heterozygous model after meta-trim. (f) TSA of overall survival in the heterozygous model.
3.2.2 Heterozygous model (AC vs CC). Number of studies and number of patients involved in the study as described in 3.2.1. AC vs CC was not related to overall survival. Heterozygous model (AC vs CC): I2 = 0%, fixed effect model was used, HR = 1.17, 95% CI (0.80,1.73), P = 0.416, as shown in Fig 12(D). The result of the meta-analysis after trim and fill showed that HR = 1.10, 95% CI (0.77,1.56), P = 0.616 (Fig 12(E)). TSA showed that the AIS was 663, and the sample size met the requirement (Fig 12(F).
3.3 The rs3212986 A/C polymorphisms.
3.3.1 Heterozygous model (AC vs CC). A total of 5 studies [17, 20, 34–36] involving 1159 patients were included in the heterozygous model. There were 7 studies on gastric cancer, 1 on osteosarcoma, 1 on lung cancer, and 1 on ovarian cancer. AC vs CC was not related to overall survival. Heterozygous model (AC vs CC): I2 = 0%, fixed effect model was used, HR = 1.24, 95% CI (0.95,1.61), P = 0.108, as shown in Fig 13(A).
(a) Result of the overall survival of the heterozygous model. (b) Result of sensitivity analysis in the heterozygous model. (c) TSA on the overall survival in heterozygous model. (d) Result of the overall survival of the homozygous model. (e) Result of sensitivity analysis in the homozygous model. (f) TSA of overall survival in the homozygous model.
Begg’s test and Egger’s test showed P = 1.000 and 0.580, respectively, indicating that there was no publication bias. The sensitivity analysis showed that the result was reliable (Fig 13(B)). TSA showed that the AIS was 1159, and the sample size met the requirement (Fig 13(C)).
3.3.2 Homozygous model (AA vs CC). Number of studies and number of patients involved in the study as described in 3.3.1. Homozygous model (AA vs CC): I2 = 59.4%, random effect model was used, HR = 1.82, 95% CI(0.87,3.80), P = 0.109, as shown in Fig 13(D).
Begg’s test and Egger’s test showed that P = 0.806 and 0.422, respectively, indicating that there was no publication bias. The sensitivity analysis showed that the result was reliable (Fig 13(E)). TSA showed that the AIS was 1159, and the sample size met the requirement (Fig 13(F)).
Discussion
The study explored the effects of ERCC1 rs11615, rs2298881 and rs3212986 polymorphisms on platinum responsiveness and OS. Platinum responsiveness may reflect short-term effects, whereas the OS reaction can reflect long-term effects. These two results are correlated, but not necessarily consistent, and therefore both need to be evaluated. In ERCC1 rs11615, the CT and CC genotypes predicted a better response to platinum in ovarian cancers than the TT genotypes. Polymorphisms were also associated with OS. The CC genotype was associated with longer overall survival in ovarian cancer patients than in TT patients. In ERCC1 rs2298881 and rs3212986, the A/C polymorphism was not related to the response to platinum and OS.
The ERCC1 protein is an excision nuclease that acts within the NER pathway and is involved in platinum metabolism. There are many mechanisms of platinum resistance, and NER is the main mechanism leading to moderate resistance [39]. From the above results, it is easy to see that the effect of ERCC1 polymorphism on platinum responsiveness is related to cancer type. This may be related to the influence of the NER pathway on the occurrence and development of different cancers. NER is a key pathway for dysregulated DNA repair in esophageal adenocarcinoma [40], ovarian cancer [41] and gastric cancer [8]. ERCC1 rs11615, rs229881 and rs3212986 are the most intensively studied SNPs. ERCC1 is a key rate-limiting enzyme in the multistep NER process. ERCC1, in collaboration with the XPF protein, is involved in DNA lesion recognition. Therefore, functional ERCC1 SNPs may contribute directly to phenotypes of drug sensitivity by modifying the function of related genes and reflecting platinum sensitivity as an innate trait [8].
The TT subtype of rs11615 was associated with platinum resistance, whereas CT and TT containing the T allele were associated with shorter OS. This result can be explained by biological significance. The rs11615 T allele of the ERCC1 polymorphism was found to be associated with high mRNA expression of the corresponding gene [42]. Functional studies confirmed the substantial effect of the ERCC1 rs11615C>T SNPs on the phenotype of NER capacity [42–44], which leads to enhanced DNA repair capacity and even to the development of platinum resistance.
The reasons for this significant ethnic difference are unclear which may be due to significant race-specific differences in linkage disequilibrium (LD) between rs11615 SNP and other nearby alleles that may cause platinum resistance [45]. The difference results in a possible lower prognostic and predictive value of the rs11615 SNP in African Americans and Africans than in Caucasians and Asians. In the Asian (CHB+JPT) population, at least five upstream polymorphisms within or directly adjacent to the ERCC1 gene were associated with rs11615. In addition to altering ERCC1 expression, phenotypic alterations in rs11615 may also lead to alterations in adjacent genes [46].
Our study not only analyzed the platinum sensitivity and prognosis of different SNPs but also compared different cancers horizontally. It can be seen that different cancers and different SNPs have varying effects on platinum sensitivity and prognosis. Ovarian cancer identified in this study was closely related to ERCC1 gene polymorphisms. Ovarian cancer patients do benefit from DNA damage repair pathway inhibitors, such as Rucaparib [47–49], Olaparib [50–52] and Niraparib [53–56]. These drugs target DNA damage repair (DDR) pathways, such as BRCA1/2 mutation and homologous recombination deficiency. In addition, HRR deficiency can also cause hereditary ovarian cancer [57]. All the above evidence suggests that ovarian cancer is a genomically unstable tumor, which may be the reason why the NER pathway represented by ERCC1 affects it more significantly. The lack of a specific DDR pathway may lead to mutation or chromosomal rearrangement and other consequences, promoting genomic instability and tumorigenesis [58]. Other similar studies have examined the relationship between only one SNP and one type of cancer. Our study focused on different cancer types and races. The limitation of this study is that some results may have boundary effects. But all of the results are reliable through sensitivity analysis. We did not search for documents published in other languages except English and Chinese, which may lead to bias due to language restrictions. In clinical practice, clinicians may prefer to select chemotherapy schemes other than platinum containing chemotherapy for patients with ERCC1 polymorphism rs11615 TT genotype, such as early use of PARP inhibitors and anti-angiogenesis, especially in ovarian cancer.
Conclusion
The ERCC1 rs11615 CT and CC genotypes predicted a better response to platinum than the TT genotype in esophageal and ovarian. The CC genotype was related to longer overall survival than TT in ovarian cancer. The ERCC1 rs2298881 and rs3212986 A/C polymorphisms were not related to the response to platinum and OS.
References
- 1. Faivre S, Chan D, Salinas R, Woynarowska B, Woynarowski JM. DNA strand breaks and apoptosis induced by oxaliplatin in cancer cells. Biochem Pharmacol. 2003;66(2):225–37. pmid:12826265
- 2. Reed E. ERCC1 and clinical resistance to platinum-based therapy. Clin Cancer Res. 2005;11(17):6100–2. pmid:16144907
- 3. Mu D, Hsu DS, Sancar A. Reaction mechanism of human DNA repair excision nuclease. J Biol Chem. 1996;271(14):8285–94. pmid:8626523
- 4. Metzger R, Leichman CG, Danenberg KD, Danenberg PV, Lenz HJ, Hayashi K, et al. ERCC1 mRNA levels complement thymidylate synthase mRNA levels in predicting response and survival for gastric cancer patients receiving combination cisplatin and fluorouracil chemotherapy. J Clin Oncol. 1998;16(1):309–16. pmid:9440758
- 5. Wood RD. Nucleotide excision repair in mammalian cells. J Biol Chem. 1997;272(38):23465–8. pmid:9295277
- 6. Ludovini V, Floriani I, Pistola L, Minotti V, Meacci M, Chiari R, et al. Association of cytidine deaminase and xeroderma pigmentosum group D polymorphisms with response, toxicity, and survival in cisplatin/gemcitabine-treated advanced non-small cell lung cancer patients. J Thorac Oncol. 2011;6(12):2018–26. pmid:22052224
- 7. Tiseo M, Bordi P, Bortesi B, Boni L, Boni C, Baldini E, et al. ERCC1/BRCA1 expression and gene polymorphisms as prognostic and predictive factors in advanced NSCLC treated with or without cisplatin. Br J Cancer. 2013;108(8):1695–703. pmid:23549037
- 8. Yin M, Yan J, Martinez-Balibrea E, Graziano F, Lenz HJ, Kim HJ, et al. ERCC1 and ERCC2 polymorphisms predict clinical outcomes of oxaliplatin-based chemotherapies in gastric and colorectal cancer: a systemic review and meta-analysis. Clin Cancer Res. 2011;17(6):1632–40. pmid:21278243
- 9. Liu X, Zhang Z, Deng C, Tian Y, Ma X. Meta-analysis showing that ERCC1 polymorphism is predictive of osteosarcoma prognosis. Oncotarget. 2017;8(37):62769–79. pmid:28977987
- 10. Tang N, Lyu D, Zhang Y, Liu H. Association between the ERCC1 polymorphism and platinum-based chemotherapy effectiveness in ovarian cancer: a meta-analysis. BMC Womens Health. 2017;17(1):43. pmid:28623887
- 11. Zhang Y, Cao S, Zhuang C, Chen J, Chen X, Sun H, et al. ERCC1 rs11615 polymorphism and chemosensitivity to platinum drugs in patients with ovarian cancer: a systematic review and meta-analysis. J Ovarian Res. 2021;14(1):80. pmid:34148553
- 12. Huang X, Liu C, Cui Y, Zhang H, Liu Y, Zhou X, et al. Association between XRCC1 and ERCC1 single-nucleotide polymorphisms and the efficacy of concurrent radiochemotherapy in patients with esophageal squamous cell carcinoma. Oncol Lett. 2017;13(2):704–14. pmid:28356949
- 13. Qi BL, Li Y, Wang N, Zhou RM, Hu P, Kang S. [Polymorphisms of ERCC1 gene and outcomes in epithelial ovarian cancer patients with platinum-based chemotherapy]. Zhonghua Fu Chan Ke Za Zhi. 2013;48(11):847–52 pmid:24444563
- 14. SY Y, Y L, S K. Effect of ERCC1 gene polymorphism on response to platinum-based chemotherapy and prognosis in patients with epithelial ovarian cancer. Journal of modern integrative medicine. 2017;26(11):1156–9+70
- 15. XY H, J X, F W, J W, ZF H. Association of ERCC1 gene polymorphism with TP chemotherapy sensitivity and prognosis in patients with epithelial ovarian cancer. Anti-tumor Pharmacy. 2017;7(06):692–6+717
- 16. Bao Y, Yang B, Zhao J, Shen S, Gao J. Role of common ERCC1 polymorphisms in cisplatin-resistant epithelial ovarian cancer patients: A study in Chinese cohort. Int J Immunogenet. 2020;47(5):443–53. pmid:32173978
- 17. Cao ZH, Yin HP, Jiang N, Yu B. Association between ERCC1 and ERCC2 gene polymorphisms and chemotherapy response and overall survival in osteosarcoma. Genet Mol Res. 2015;14(3):10145–51. pmid:26345951
- 18. Ji WP, He NB. Investigation on the DNA repaired gene polymorphisms and response to chemotherapy and overall survival of osteosarcoma. Int J Clin Exp Pathol. 2015;8(1):894–9 pmid:25755792
- 19. Sun Y, Wu Y, Li W, Kong Z, Zou X. Genetic polymorphisms in nucleotide excision repair pathway influences response to chemotherapy and overall survival in osteosarcoma. Int J Clin Exp Pathol. 2015;8(7):7905–12 pmid:26339355
- 20. Li J, Zuo X, Lv X, Kong F, Xu W, Yang S. Association of DNA repair gene polymorphisms with response to chemotherapy and prognosis of gastric cancer in a Chinese population. Tumour Biol. 2014;35(8):7569–74. pmid:24793015
- 21. Yang LM, Li XH, Bao CF. Glutathione S-transferase P1 and DNA polymorphisms influence response to chemotherapy and prognosis of bone tumors. Asian Pac J Cancer Prev. 2012;13(11):5883–6. pmid:23317281
- 22. Hong W, Wang K, Zhang YP, Kou JY, Hong D, Su D, et al. Methylenetetrahydrofolate reductase C677T polymorphism predicts response and time to progression to gemcitabine-based chemotherapy for advanced non-small cell lung cancer in a Chinese Han population. J Zhejiang Univ Sci B. 2013;14(3):207–15. pmid:23463763
- 23. Cheng J, Ha M, Wang Y, Sun J, Chen J, Wang Y, et al. A C118T polymorphism of ERCC1 and response to cisplatin chemotherapy in patients with late-stage non-small cell lung cancer. J Cancer Res Clin Oncol. 2012;138(2):231–8. pmid:22102173
- 24. Park SR, Kong SY, Nam BH, Choi IJ, Kim CG, Lee JY, et al. CYP2A6 and ERCC1 polymorphisms correlate with efficacy of S-1 plus cisplatin in metastatic gastric cancer patients. Br J Cancer. 2011;104(7):1126–34. pmid:21364592
- 25. Li F, Sun X, Sun N, Qin S, Cheng H, Feng J, et al. Association between polymorphisms of ERCC1 and XPD and clinical response to platinum-based chemotherapy in advanced non-small cell lung cancer. Am J Clin Oncol. 2010;33(5):489–94. pmid:20351547
- 26. Chen S, Huo X, Lin Y, Ban H, Lin Y, Li W, et al. Association of MDR1 and ERCC1 polymorphisms with response and toxicity to cisplatin-based chemotherapy in non-small-cell lung cancer patients. Int J Hyg Environ Health. 2010;213(2):140–5. pmid:20189873
- 27. Liang J, Jiang T, Yao RY, Liu ZM, Lv HY, Qi WW. The combination of ERCC1 and XRCC1 gene polymorphisms better predicts clinical outcome to oxaliplatin-based chemotherapy in metastatic colorectal cancer. Cancer Chemother Pharmacol. 2010;66(3):493–500. pmid:19960344
- 28. Ma BB, Hui EP, Wong SC, Tung SY, Yuen KK, King A, et al. Multicenter phase II study of gemcitabine and oxaliplatin in advanced nasopharyngeal carcinoma—correlation with excision repair cross-complementing-1 polymorphisms. Ann Oncol. 2009;20(11):1854–9. pmid:19549713
- 29. Chung HH, Kim MK, Kim JW, Park NH, Song YS, Kang SB, et al. XRCC1 R399Q polymorphism is associated with response to platinum-based neoadjuvant chemotherapy in bulky cervical cancer. Gynecol Oncol. 2006;103(3):1031–7. pmid:16875718
- 30. Ryu JS, Hong YC, Han HS, Lee JE, Kim S, Park YM, et al. 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(3):311–6. pmid:15140544
- 31. Su D, Ma S, Liu P, Jiang Z, Lv W, Zhang Y, et al. Genetic polymorphisms and treatment response in advanced non-small cell lung cancer. Lung Cancer. 2007;56(2):281–8. pmid:17222938
- 32. Zhang Q, Lv LY, Li BJ, Zhang J, Wei F. Investigation of ERCC1 and ERCC2 gene polymorphisms and response to chemotherapy and overall survival in osteosarcoma. Genet Mol Res. 2015;14(3):11235–41. pmid:26400354
- 33. Yu X, Xiao H, Zhao B, Zhang X, Wang G. DNA repair gene ERCC1 C118T polymorphism predicts sensitivity of recurrent esophageal cancer to radiochemotherapy in a Chinese population. Thorac Cancer. 2015;6(6):741–8. pmid:26557912
- 34. Zhong G, Li HK, Shan T, Zhang N. Genetic variability of DNA repair mechanisms in chemotherapy treatment outcome of gastric cancer patients. Genet Mol Res. 2015;14(4):17228–34. pmid:26681216
- 35. Zheng DL, Tang GD, Chen YN, Zhang T, Qin MB. Genetic variability of ERCC1 and ERCC2 genes involved in the nucleotide excision repair pathway influences the treatment outcome of gastric cancer. Genet Mol Res. 2016;15(2). pmid:27173253
- 36. Mo J, Luo M, Cui J, Zhou S. Prognostic value of ERCC1 and ERCC2 gene polymorphisms in patients with gastric cancer receiving platinum-based chemotherapy. Int J Clin Exp Pathol. 2015;8(11):15065–71 pmid:26823845
- 37. Bai Y, Wang L, Li G, Fang X, Li Y, Yang S. Genetic variability of ERCC1 genes in NER pathway influences the treatment outcome of gastric cancer. Int J Clin Exp Pathol. 2015;8(10):13367–73 pmid:26722542
- 38. Chen ZH, Wang L, Luo LP. Association of DNA repair gene polymorphisms with response to chemotherapy and prognosis of gastric cancer. Genet Mol Res. 2014;13(3):7484–91. pmid:25222248
- 39. Shellard SA, Fichtinger-Schepman AM, Lazo JS, Hill BT. Evidence of differential cisplatin-DNA adduct formation, removal and tolerance of DNA damage in three human lung carcinoma cell lines. Anticancer Drugs. 1993;4(4):491–500. pmid:8400352
- 40. Dewalt RI, Kesler KA, Hammoud ZT, Baldridge L, Hattab EM, Jalal SI. Gastroesophageal junction adenocarcinoma displays abnormalities in homologous recombination and nucleotide excision repair. Lung Cancer (Auckl). 2014;5:11–20. pmid:28210138
- 41. Stefanou DT, Souliotis VL, Zakopoulou R, Liontos M, Bamias A. DNA Damage Repair: Predictor of Platinum Efficacy in Ovarian Cancer? Biomedicines. 2021;10(1). pmid:35052761
- 42. Yu JJ, Lee KB, Mu C, Li Q, Abernathy TV, Bostick-Bruton F, et al. Comparison of two human ovarian carcinoma cell lines (A2780/CP70 and MCAS) that are equally resistant to platinum, but differ at codon 118 of the ERCC1 gene. Int J Oncol. 2000;16(3):555–60. pmid:10675489
- 43. Shi Q, Wang LE, Bondy ML, Brewster A, Singletary SE, Wei Q. Reduced DNA repair of benzo[a]pyrene diol epoxide-induced adducts and common XPD polymorphisms in breast cancer patients. Carcinogenesis. 2004;25(9):1695–700. pmid:15090466
- 44. Qiao Y, Spitz MR, Shen H, Guo Z, Shete S, Hedayati M, et al. Modulation of repair of ultraviolet damage in the host-cell reactivation assay by polymorphic XPC and XPD/ERCC2 genotypes. Carcinogenesis. 2002;23(2):295–9. pmid:11872635
- 45. Consortium. TIH. The International HapMap Project. Nature. 2003;426(6968):789–96. pmid:14685227
- 46. Gao R, Reece KM, Sissung T, Fu SH, Venzon DJ, Reed E, et al. Are race-specific ERCC1 haplotypes in melanoma cases versus controls related to the predictive and prognostic value of ERCC1 N118N? BMJ Open. 2013;3(1). pmid:23293248
- 47. Kristeleit R, Lisyanskaya A, Fedenko A, Dvorkin M, de Melo AC, Shparyk Y, et al. Rucaparib versus standard-of-care chemotherapy in patients with relapsed ovarian cancer and a deleterious BRCA1 or BRCA2 mutation (ARIEL4): an international, open-label, randomised, phase 3 trial. Lancet Oncol. 2022;23(4):465–78. pmid:35298906
- 48. Tomao F, Vici P, Tomao S. Expanding use of rucaparib as maintenance therapy in recurrent ovarian cancer: updates from the ARIEL3 trial. Lancet Oncol. 2020;21(5):616–7. pmid:32359486
- 49. Oza AM, Lorusso D, Aghajanian C, Oaknin A, Dean A, Colombo N, et al. Patient-Centered Outcomes in ARIEL3, a Phase III, Randomized, Placebo-Controlled Trial of Rucaparib Maintenance Treatment in Patients With Recurrent Ovarian Carcinoma. J Clin Oncol. 2020;38(30):3494–505. pmid:32840418
- 50. Penson RT, Valencia RV, Cibula D, Colombo N, Leath CA, Bidziński M 3rd, et al. Olaparib Versus Nonplatinum Chemotherapy in Patients With Platinum-Sensitive Relapsed Ovarian Cancer and a Germline BRCA1/2 Mutation (SOLO3): A Randomized Phase III Trial. J Clin Oncol. 2020;38(11):1164–74. pmid:32073956
- 51. Ray-Coquard I, Pautier P, Pignata S, Pérol D, González-Martín A, Berger R, et al. Olaparib plus Bevacizumab as First-Line Maintenance in Ovarian Cancer. N Engl J Med. 2019;381(25):2416–28. pmid:31851799
- 52. DiSilvestro P, Colombo N, Scambia G, Kim BG, Oaknin A, Friedlander M, et al. Efficacy of Maintenance Olaparib for Patients With Newly Diagnosed Advanced Ovarian Cancer With a BRCA Mutation: Subgroup Analysis Findings From the SOLO1 Trial. J Clin Oncol. 2020;38(30):3528–37. pmid:32749942
- 53. Mirza MR, Åvall Lundqvist E, Birrer MJ, dePont Christensen R, Nyvang GB, Malander S, et al. Niraparib plus bevacizumab versus niraparib alone for platinum-sensitive recurrent ovarian cancer (NSGO-AVANOVA2/ENGOT-ov24): a randomised, phase 2, superiority trial. Lancet Oncol. 2019;20(10):1409–19. pmid:31474354
- 54. Del Campo JM, Matulonis UA, Malander S, Provencher D, Mahner S, Follana P, et al. Niraparib Maintenance Therapy in Patients With Recurrent Ovarian Cancer After a Partial Response to the Last Platinum-Based Chemotherapy in the ENGOT-OV16/NOVA Trial. J Clin Oncol. 2019;37(32):2968–73. pmid:31173551
- 55. Wu XH, Zhu JQ, Yin RT, Yang JX, Liu JH, Wang J, et al. Niraparib maintenance therapy in patients with platinum-sensitive recurrent ovarian cancer using an individualized starting dose (NORA): a randomized, double-blind, placebo-controlled phase III trial(☆). Ann Oncol. 2021;32(4):512–21. pmid:33453391
- 56. Konstantinopoulos PA, Vinayak S. BRCA Mutations and Homologous Recombination Repair Deficiency in Treatment With Niraparib Combined With Pembrolizumab-Reply. JAMA Oncol. 2020;6(3):441. pmid:31876893
- 57. Li SKH, Martin A. Mismatch Repair and Colon Cancer: Mechanisms and Therapies Explored. Trends Mol Med. 2016;22(4):274–89. pmid:26970951
- 58. Chene G, Tchirkov A, Penault-Llorca F. Early telomere shortening and genomic instability in tubo-ovarian preneoplastic lesions—response. Clin Cancer Res. 2013;19(18):5255. pmid:24009148
- 59. Chung CH, Lee JW, Slebos RJ, Howard JD, Perez J, Kang H, et al. A 3’-UTR KRAS-variant is associated with cisplatin resistance in patients with recurrent and/or metastatic head and neck squamous cell carcinoma. Ann Oncol. 2014;25(11):2230–6. pmid:25081901