Prospective Evaluation of XRCC-1 as a Biopredictor for Survival Outcomes in Patients of Head and Neck Squamous Cell Carcinoma (HNSCC): A Meta-analysis

Background: Epidemiologic studies have demonstrated that X-ray repair cross-complementary group 1 (XRCC1) is one of the susceptibility factors in head and neck squamous cell carcinoma(HNSCC) patients. However, its clinical prognostic impact remains controversial. Thus, a meta-analysis was performed to clarify the survival value of XRCC1 in HNSCC patients. Methods: Following the Preferred Reporting Items or Systematic Reviews Meta Analyses (PRISMA) 2020 guidelines, literature searches were systematically performed by PubMed, EMBASE and Web of Science with a manual retreive to evaluate the prognostic consequence of XRCC1 in HNSCC patients. Hazard ratios (HRs) and 95% condence intervals (CIs) were collected to estimate the correlation between XRCC1 and the survival outcomes of HNSCC patients. Results: Ten studies including 2086 HNSCC patients who satised the inclusion and exclusion criteria were included in this meta-analysis. The meta-analysis showed that high XRCC1 expression and Arg399Gln and Arg194Trp were signicantly correlated with poorer overall survival (OS), with HRs of 1.97 (95% CI, 1.36–2.84, P<0.001), 1.30 (95% CI, 1.12–1.51, P<0.001), and 1.65 (95% CI, 1.18–2.32 P<0.05), respectively. Conclusion: XRCC1 was associated with poorer survival outcomes in HNSCC patients. Hence, XRCC1 is a potential therapeutic target for HNSCC. 2 -based Q test and the I 2 test. The xed-effects model was employed for analysis without obvious statistical heterogeneity between studies (P > 0.10, I 2 < 50%). Otherwise, the random-effects model was applied. Moreover, we performed subgroup analysis to explore the potential source of heterogeneity. Sensitivity analysis was also enrolled to investigate the inuence of each individual study on the overall pooled results. Publication bias was assessed by Begg’s and Egger’s. All of the statistical tests used in this meta-analysis were performed with Stata version 12.0 (StataCorp, College Station, TX, USA). the method of HR extrapolation from the Kaplan-Meier graph may also generate heterogeneity because the HRs of the ve records enrolled were calculated by Parmer's methods and not directly reported. Last, studies with positive results are more likely to be published and thus more likely to be enrolled. Hence, the survival value of XRCC1 in HNSCC may to some extent have been overestimated in this meta-analysis. In summary, our systematic review or meta-analysis was conducted to identify XRCC1 from all primary relevant articles and to evaluate the potential prognostic value of XRCC1 in HNSCC. The results demonstrated that both XRCC1 polymorphisms (Arg194Trp and Arg399Gln) and XRCC1 protein high expression exhibited poor survival in HNSCC patients. The prognostic value of XRCC1 in HNSCC patients is undoubtedly known. However, larger prospective studies should be conducted in the future to further verify the outcomes from this current meta-analysis


Introduction
Head and neck squamous cell carcinomas (HNSCC) is the fth most common cancer in the world and accounts for approximately 3% of new cancer cases annually. It is a malignant disease of the upper aerodigestive tract, including oral, pharyngeal and laryngeal regions. Current treatment strategies for patients with HNSCC involve surgery, chemotherapy, immunotherapy, radiotherapy and so on. However, with such treatments, the survival results of HNSCC patients were still not satisfactory. The survival of HNSCC patients remain unsatisfactory.
Speci cally, the ve-year survival rate of HNSCC patients is estimated to be near 50% due to high mortality and early disease recurrence. Therefore, there is an urgent need to identify new predictive biomarkers to improve the overall survival (OS) of HNSCC patients. Epidemiologic data showed that tobacco and alcohol consumption are the main etiological factors in the carcinogenesis of HNSCC. These risk factors, such as benzo(a)pyrene from tobacco, can induce an increase in reactive oxygen species levels that leads to oxidative DNA damage in the epithelial cells of the head and neck region, which stimulates HNSCC occurrence. However, not all individuals with these risk factors develop HNSCC, suggesting that individual susceptibility might also play a role in HNSCC carcinogenesis.
The DNA repair pathway can generally correct base lesions. In contrast, defects in the DNA repair system can also result in genetic instability and carcinogenesis. Single-nucleotide polymorphisms (SNPs) are genetic variants that are related to cancer susceptibility [1]. Hence, common genetic polymorphisms in genes involved in DNA repair capacity may play key roles in individual susceptibility and prognosis in HNSCC.
The primary DNA repair pathway that corrects base lesions is the base excision repair. X-ray repair cross-complementary group 1 (XRCC1) is a master DNA repair gene involved in base excision repair [2]. The capacity to repair damaged cells with XRCC1 is encoded by polymorphic genes that may modify the expression of encoded proteins and proteins. Consequently, the DNA repair ability is modi ed. Recently, XRCC1 gene SNPs within conserved sequences have been identi ed. Arg399Gln (rs25487) and Arg194Trp (rs1799782) are the main variant alleles [3,4].
A number of studies have investigated the association between XRCC1 polymorphisms and HNSCC risk. However, the association between XRCC1 SNPs or expression and the survival outcomes in HNSCC patients is still largely controversial or unknown. Thus, a meta-analysis including a total of 10 studies with 2086 patients was performed by systematically searching for relevant publications from the PubMed, EMBASE, and Web of Science databases to provide a systematic evaluation of the diagnostic and prognostic role of XRCC1 polymorphisms and expression in HNSCC patients.

Search strategy
Comprehensively searches in PubMed, Web of Science and EMBASE were performed using the following search terms: (XRCC1 or X-ray repair cross-complementing 1) and (prognosis OR outcome OR mortality OR survival OR progression OR recurrence) and (head and neck or laryngeal or tonsil or oropharyngeal or oral or oropharynx or nasopharyngeal) and (squamous cell cancer or carcinoma).Related papers published from September 2006 to March 2021 were selected.

Selection criteria
The inclusion criteria of the meta-analysis were as follows: 1) all patients diagnosed with HNSCC; 2) the study revealed the association between XRCC1 and survival of HNSCC; 3) selected the English language of publications only; 4) including hazard ratios (HR) with prognostic endpoints and the 95% con dence interval(CI) directly, or other data such as Kaplan-Meier survival curves that can be estimate for OS. The criteria of exclusion included: 1) studies without exsiting HR and 95% CI data for meta-analysis, or other outcomes for OS such as Kaplan-Meier survival curves. 2) Other text formats: abstracts, reviews, letters and case reports; and 3) The data of the article has nothing to do with HNSCC or XRCC1. In case of the same cohort reported by several publications, Latest literature gained the priority .

Data extraction
Following the Preferred Reporting Items or Systematic Reviews Meta Analyses (PRISMA) 2021 guidelines [5], electronic searches for relevant studies were performed in PubMed, EMBASE and Web of Science databases till March 1, 2021. Data were carefully extracted and tabulated by two independent researchers (LQ Zhou and Fan Yang) to minimize variation. In order to achieve a consensus, disagreements were solved by adjustments of the third researcher(YJ Wang). The following information was extracted from the studies: author, year, country, ethnicity, cancer type, sample size, age, follow-up, method, and HR value. The Newcastle-Ottawa Scale (NOS) was included to assess the quality of the included publications, and a star system (maximum is nine stars) was adopted to evaluate a study in three domains: comparability of study groups, selection of participants and ascertainment of outcomes of interest. Scores of NOS ≥ 6 indicated high-quality studies[6]. Reporting recommendations for tumor marker prognostic studies (REMARK) were also applied to evaluate study quality in cancer-related metaanalyses [7].

Statistical analysis
The value of survival outcome between XRCC1 polymorphisms or expression and HNSCC was measured by HR and 95% CI, which were obtained directly or estimated from the primary publications or estimated by p values and other published data following Parmer's methods [8].
Heterogeneity was assessed by the χ 2 -based Q test and the I 2 test. The xed-effects model was employed for analysis without obvious statistical heterogeneity between studies (P > 0.10, I 2 < 50%). Otherwise, the random-effects model was applied. Moreover, we performed subgroup analysis to explore the potential source of heterogeneity. Sensitivity analysis was also enrolled to investigate the in uence of each individual study on the overall pooled results. Publication bias was assessed by Begg's and Egger's. All of the statistical tests used in this metaanalysis were performed with Stata version 12.0 (StataCorp, College Station, TX, USA).

Study selection and characteristics
As shown in Fig. 1, 231 published articles were identi ed by searching the electronic databases. Out of 231 studies, duplicates (n = 131), reviews, abstracts and letters (n = 16) and the studies not related to the topics (n = 44 )were excluded. The retrieved publications(n = 40) were further evaluated by reading their full texts, and 30 studies were discarded due to the lack of speci c data regarding HNSCC or XRCC1. Finally, 10 studies with 2086 HNSCC patients were selected.
The characteristics of the enrolled studies were summarized in Table 1. Six for HNSCC, one for laryngeal squamous cell carcinoma (LSCC), one for oral squamous cell carcinomas (OSCC),one for nasopharyngeal carcinoma (NPC) and one for oropharyngeal squamous cell carcinoma (OPSCC) were included. Sample sizes of the publications ranged from 75 to 531. Two publications enrolled > 500 patients. XRCC1 SNPs were explored by polymerase chain reaction restriction fragment length polymorphism (PCR-RFLP) and immunohistochemical (IHC) staining of XRCC1 protein technology were applied for XRCC1 protein expression. In the original literature, half of the studies reported the HR and 95% CI directly, and the remaining studies only provided Kaplan-Meier survival curves. All of the publications' NOS scores were above 6 and the REMARK scores were between 13 and 16. Table 1 Characteristics of the studies examined in the meta-analysis Association between XRCC1 and survival in HNSCC patients Ten articles with 2086 patients in the meta-analysis evluated the survival value of XRCC1 for patients with HNSCC. The pooled HR for high XRCC1 expression was 1.97 (95%CI, 1.36-2.84, P < 0.001), for the XRCC1 Arg399Gln was 1.30 (95% CI, 1.12-1.51, P < 0.001) and for the XRCC1 Arg194Trp was 1.65 (95% CI, 1.18-2.32, P < 0.005). The heterogeneity between high XRCC1 expression and OS (I 2 = 16.1%, P heterogeneity = 0.304) was slight, and moderate heterogeneity was noted between XRCC1Arg194Trp and OS. (I 2 = 53.0%, P heterogeneity =0.119), while the heterogeneity between XRCC1 Arg399Gln and OS was low (I 2 = 46.2%, P heterogeneity = 0.115) (Fig. 2).

Author
Year

Sensitivity analysis
During the process of comprehensive comparisons, we conducted sensitivity analyses to evaluate the effect of each single data point against the aggregate. The recalculated outcomes (Fig. 3)were not substantially in uenced, suggesting that the combined effect size of the metaanalysis results was stable and reliable.

Publication bias
Begg's funnel plots and Egger's test were applied to validate potential bias from searched publications. The result indicates bias existed (p = 0.014). Therefore, "trim and ll" analysis was further utilized, and the pooled HR of 1.253 (95% CI: 1.115-1.408) remained statistically signi cant ( Figure. 4),overall, the results are relatively reliable in spite of the signi cant publication bias.

Discussion
Head and neck cancers are a heterogeneous group of secondary cancers [9], with a high rate of early recurrence and mortality. However, radiotherapy and chemotherapy have been widely used in the clinic and have recently been shown to be effective. The survival rates of advanced HNSCC are approximately 50% and have not changed signi cantly in recent years. This may be due to the individual's treatment response controlled by intrinsic genetics. In a previous study, oncogene genes in HNSCC were identi ed through extensive DNA sequencing and genetic analysis [10]. Therefore, it is reasonable to quest potential biomarkers of treatment by analyzing genomic features.
DNA repair genes have been considered driver genes of HNSCC due to their frequent mutation. Defects in DNA repair promote genomic instability and carcinogenesis [11]. DNA repair systems play an indispensable role in protecting cells against carcinogenic agents from internal and external stimuli [12]. Statistical analyses indicated that the DNA repair status of a tumor is associated with poor prognosis in cancer [13,14]. XRCC1 is a major DNA repair gene involved in base excision repair (BER) for small base lesions resulting from oxidation and alkylation damage [15,16]. More than 300 validated SNPs in the XRCC1 gene were reported in the dbSNP database (http://www.ncbi.nlm.nih.gov/SNP). Among them, the XRCC1 genes Arg194Trp and Arg399Gln, which occur within conserved sequences, are the most frequently mutated genes and the most extensively studied.
Studies have investigated the association between XRCC1 Arg399Gln/XRCC1 Arg194Trp polymorphisms and HNSCC risk [17][18][19][20][21][22][23]. However, these results were contradictory. Wang [24] conducted a meta-analysis on the association of XRCC1 Arg399Gln polymorphisms with HNSCC risk. Nevertheless, they did not observe any precise estimation of this relationship based on 18 published studies. Sturgis [25] reported a reduced risk between XRCC1 Arg194Trp and HNSCC while Andrew[26] found a weak elevation between HNSCC risk and the XRCC1 Arg194Trp polymorphism. XRCC1 protein is involved in BER [27,28], and its protein expression alters the sensitivity of cells to radiation and chemotherapeutic agents [29]. It was reported that high protein expression levels of XRCC1 may be a risk factor for HNSCC [30,31]. In addition, XRCC1 protein expression is common in HNSCC, and high XRCC1 protein expression may confer poorer survival, regardless of the primary tumor site or stage [31,32]. In summary, these results either contrast with each other or are not accurate conclusions.
In addition, this systematic review or meta-analysis prospectively evaluated XRCC1 as a biopredictor of survival outcomes in patients with HNSCC. There is no doubt that our study is the rst meta-analysis including 10 published studies with 2086 patients to comprehensively evaluate the value of XRCC1 (SNPs and protein expression) and offer useful information for clinical decision-making in HNSCC.
After analyzing and summarizing all of the selected data, the results indicate that high expression and SNPs (Arg194Trp and Arg399Gln) of XRCC1 signi cantly predicted poorer OS in HNSCC patients with HRs of 1.97, 1.65, and 1.30, respectively. These ndings con rmed that XRCC1 could be widely applied as a diagnostic marker and therapeutic target in HNSCC patients. As a genetic association literature, the Hardy-Weinberg (HWE) principle is also considered to avoid methodological weaknesses, such as biased selection of subjects or genotyping errors.
The enrolled studies were all in agreement with HWE.
This meta-analysis should be interpreted within the context of its limitations. First, studies containing only English languages and only published articles were included, which may contribute to additional bias. Therefore, publication bias is very likely to occur. In light of Begg's funnel plots and Egger's test, no signi cant publication bias was present in this meta-analysis. Second, the number of articles was limited, and the sample size was relatively small in the present meta-analysis. False-positive or false-negative ndings may have occurred in small sample sizes. Therefore, our results of the survival value in analyses might have occurred by chance, and more large-scale and comprehensive studies are needed to achieve a more persuasive conclusion. Third, the method of HR extrapolation from the Kaplan-Meier graph may also generate heterogeneity because the HRs of the ve records enrolled were calculated by Parmer's methods and not directly reported. Last, studies with positive results are more likely to be published and thus more likely to be enrolled. Hence, the survival value of XRCC1 in HNSCC may to some extent have been overestimated in this meta-analysis.
In summary, our systematic review or meta-analysis was conducted to identify XRCC1 from all primary relevant articles and to evaluate the potential prognostic value of XRCC1 in HNSCC. The results demonstrated that both XRCC1 polymorphisms (Arg194Trp and Arg399Gln) and XRCC1 protein high expression exhibited poor survival in HNSCC patients. The prognostic value of XRCC1 in HNSCC patients is undoubtedly known. However, larger prospective studies should be conducted in the future to further verify the outcomes from this current meta-analysis Declarations Figure 2 Forest plot indicating the association between XRCC1 polymorphisms (Arg194Trp and Arg399Gln) / high XRCC1 protein expression and OS in HNSCC Figure 3