1. Introduction
Radiation therapy (RT) is the primary care for locally advanced non-small cell lung cancer (NSCLC). However, while resulting in tumor shrinkage, RT also leads to damage of surrounding normal tissues and caused normal tissue toxicities [
1]. Commonly occurred RT-induced toxicities have limited the planed dose and therefore compromised the efficiency of local and/or regional control and negatively influence patients’ prognosis. Accurate prediction before treatment will therefore enable personalized dosimetric design, and maximize treatment effect while minimizing potential toxicities. Currently, the risk of RT-induced toxicities were assessed based on clinical-based factors, including physical characteristics of the radiation beam; treatment dose, fractionation and time; volume and type of normal tissue receiving radiation; radiosensitizer usage; and comorbidities [
2]. However, large variation exists in patients with similar clinical parameters. It suggests a potential role of host genetic factors in determining an individual’s response to RT and therefore susceptibility of developing RT-induced toxicities [
3].
Generally, DNA double-strand breaks (DSBs) are the major genotoxic lesions induced by ionization radiation. There are two distinct and complementary pathways for DSB repair, namely, homologous recombination (HR) and non-homologous end joining (NHEJ). DSBs repairing play a key role in maintaining genomic stability and integrity after RT [
4]. It has been reported that deficiencies in DSB repair genes were associated with high incidence of chromosome aberrations and increased tumor radiation sensitivity [
5,
6,
7,
8]. Therefore, it is reasonable to presume that variants in DSB repair pathway genes could modify the radiosensitivity as well as clinical outcomes after RT.
Several studies have reported the association between SNPs in several DSB repair pathway genes and RT-induced esophagitis and pneumonitis, such as
BRCA1 [
9],
ATM [
10,
11],
P53 [
10],
RAD51 [
4] and
LIG4 [
12]. However, these studies mainly used single gene-based approach without validation of their findings. We have previously used pathway-based approaches to identify genetic variations in inflammation pathway genes as predictors of radiation-induced toxicities in NSCLC patients [
13], which provided more coverage compared to single-gene-based approaches. In this study, to the best of our knowledge, we, for the first time, utilized a pathway-based approach to investigate genetic variations within DSB pathway genes in a relatively large, well-characterized population and analyzed their role in developing esophagitis or pneumonitis following definitive RT with a validation step. Our goal is to identify potential DSB-related biomarkers which will be used to facilitate personalized dosage design.
4. Discussion
RT destroys tumor cells mainly by damaging their DNA, especially double-strand DNA. However, such an effect also occurs in normal cells. An individual’s capacity for repairing DNA damage may determine the presence and extent of radiation toxicities. It has been known for years that genetic variations in DNA repair genes might regulate patients’ sensitivity to RT [
1]. In this study, we systematically investigated genetic variation within DSB repair pathway as potential predictive biomarkers for radiation-induced esophagitis and pneumonitis in NSCLC patients treated with definitive radiation or concurrent chemoradiation. To control for false discovery, a validation step was also included. We found that an intronic SNP
BLM:rs7165790 was significantly associated with a decreased risk of esophagitis in both the discovery and validation populations. A strong cumulative effect was observed for these SNPs on radiation toxicities.
The most significant SNP for esophagitis was an intronic variant
BLM:rs7165790. Bloom syndrome (BS) gene product,
BLM, encodes the protein RecQL3 helicase, an enzyme that restores malfunctioning replication forks during DNA replication. The RecQ helicase BLM has been also shown to be involved in single-strand DNA resection at the initial stages of homologous recombination [
27,
28]. Grabarz
et al. found that
BLM is an essential factor involved in DSB repair initiation and is essential for the maintenance of genome stability [
29]. Mutations in
BLM could lead to Bloom syndrome, which is an autosomal-recessive genetic disorder that is associated with increased levels of spontaneous sister-chromatid exchanges (SCEs), genome instability, as well as elevated cancer susceptibility [
30]. Broberg
et al. performed a case-control study to indicate that a variant allele of rs2532105 in BLM showed increased risk for breast cancer [
31]. In our study,
BLM:rs7165790 was the only SNP validated to be associated with decreased risk of developing esophagitis. Gene-based test also showed that
BLM was among the most significant genes associated with risk of esophagitis. All these results support that this gene plays an important role in radiation-induced esophagitis. Although located in the intron region where its function was not obvious, rs7165790 was identified as borderline significant in expression quantitative trait locus (eQTL) relationship with
BLM (
p = 0.0644). It is possible that this SNP, or other SNPs tagged by it, could contribute to the alteration of
BLM expression, which warrants further investigation.
For the radiation-induced pneumonitis, the top SNP was synonymous with SNP rs1051772 in
TOPBP1, which was significantly associated with decreased risk of radiation-induced pneumonitis and showed the same trend of effects between discovery and validation populations and reached statistical significance in the combined meta-analysis. Increasing evidence has indicated that
TOPBP1 participated in DNA replication checkpoint control and played important roles in maintaining genomic stability [
32]. SNPs in
TOPBP1 were also associated with some cancer risks [
10,
11]. The mechanisms underlying radiation-induced toxicities need to be further explored.
MUS81 was the top gene that was significantly associated with radiation-induced pneumonitis by gene-based analysis. 3 SNPs (rs13817, rs558114 and rs635375) in
MUS81 were shown to be significantly related to radiation-induced pneumonitis in the discovery group only, and the exact function is still unclear. The N-terminus was also proposed to contain a BLM-interacting domain [
33], and whether these two genes interact with each other in mediating radiation toxicities needs to be further studied.
Previous studies have reported the association of
ATM polymorphisms with radiation-induced esophagitis [
34,
35,
36]. In our study, we did not find a similar association, although 10 genetic variants tagging the
ATM gene were included in the analysis. Since most of the prior reports are candidate gene studies genotyping a limited number of SNPs in the Asian population while our study is pathway-based in the non-Hispanic white population, differences in study design and demographic group may affect the final results.
Interestingly, we did not find overlapping SNPs between esophagitis and pneumonitis, suggesting that different biological mechanisms or responses to chemoradiation-induced damage may play a role in the development of these toxicities. This is supported by our finding that compared to radiation-only treatment, concurrent chemoradiation treatment did not seem to affect the distribution of patients with pneumonitis but significantly increased the incidence of esophagitis (
Supplementary Table S6). However, since this study only focused on genetic variants in the DSB pathway, we could not rule out that other pathways, such as inflammation, might share common susceptibility factors for esophagitis and pneumonitis. Future whole-genome profiles of genetic variation associated with these adverse events are necessary to find the most significant and/or common genetic factors for clinical application.
Our study has several advantages. First, other than single genes, we systematically investigated the effects of genetic variations (440 SNPs) within major genes in DNA-DSB repair pathway (44 genes). Second, we performed the first comprehensive dosimetric and clinical data collection to enable this pathway-based analysis. Moreover, to reduce the potential false positive findings, we adopted a two-phase screening and validation approach in the analysis.
There were, however, some limitations to this study. First, the sample size was relatively small due to the fact that radiation therapy is mostly used in late-stage patients in our population, while the majority of patients were treated initially by surgery or by a combination of surgery and chemotherapy. However, the sample size is relatively large compared to other studies of this kind [
37]. Second, although we included an internal validation group, the possibility of false positives still exists. Future external independent validation should be included to further validate our findings.
In summary, this is the first pathway-based study for association between single nucleotide polymorphisms in DNA DSB repair pathway genes and risk of radiation-induced pneumonitis and esophagitis for NSCLC Patients. Our results provide strong support for the claim that DSB pathway-related genetic variations serve as a potential biomarker to predict radiation toxicities and further guide the RT for NSCLC patients. With further investigations, we can test the predictive value by combining these significant SNPs with some commonly used variables in clinics that may be also related to radiation toxicities.