Homologous recombination repair capacity in peripheral blood lymphocytes and breast cancer risk (2025)

Search for other works by this author on:

Article history

Abstract

Introduction

Genomic translocations, deletions and duplications are frequently observed in breast tumors (1–3). These genomic rearrangements are thought to be caused by the abnormal repair of DNA double-strand breaks via two major pathways: homologous recombination repair (HRR) and non-homologous end joining (4). During HRR-mediated repair of DNA double-strand breaks, the sister chromatid acts as a template to copy the missing information into the broken locus. Because sister chromatids are identical to each other, DNA damage can be accurately repaired with no major genetic consequences. As many breast cancer genes are involved in HRR (e.g. BRCA1, BRCA2, Rad 52, Rad 51, ATM, etc.), defects in HRR have been assumed to lead to cancer (5,6). Although such a hypothesis has been confirmed in familial breast cancer with BRCA1 and BRCA2 mutations (7), it remains unclear whether the same is true for sporadic breast cancer.

Numerous biomarkers of HRR capacity have been evaluated in the past decade, including germline BRCA1 and BRCA2 mutations, somatic HR pathway mutations, genomic instability, gene expression signatures and functional HRR assays (8). Other than germline mutations, the remaining biomarkers are tumor-based. Germline BRCA1 and BRCA2 mutations are rare and are nearly exclusively found in familial breast cancer. Thus, the variation of HRR capacity to the genetic susceptibility of sporadic breast cancer risk cannot be assessed at all. Host cell reactivation (HCR) assay measures the expression level of damaged reporter genes. Compared with assays measuring the efficiency of the multiple steps of excision repair individually, HCR covers the entire pathway and assesses the global effects. Furthermore, compared with other phenotypic assays, HCR is advantageous as it is relatively fast, objective and uses undamaged cells. Using a phenotypic approach to measure DNA-repair capacity in humans has been most frequently carried out using peripheral blood lymphocytes (PBLs) from cases and matched controls. Because the inherited genetic susceptibility factor should be present in all normal somatic cells, PBLs are the most convenient material for testing DNA-repair capacity. Studies have used PBL samples from lung, breast, bladder, liver and skin cancer cases and controls, and have consistently demonstrated elevated cancer risk in subjects with poorer repair capabilities (9–12).

We conducted a two-stage study using a luciferase plasmid-based HCR assay. In the discovery stage, we compared HRR capacity in PBLs obtained from breast cancer cases and matched healthy controls. In the validation stage, we validated the association in a nested breast cancer case–control study using pre-diagnostic PBLs.

Materials and methods

Study population

The study participants in the discovery stage were selected from an ongoing breast cancer case–control study beginning in 2012. Participants were patients at The University of Texas M. D. Anderson Cancer Center (Houston, TX) with newly diagnosed (defined by the presence of malignant breast epithelial cells) and histologically confirmed (by microscopic analysis and molecular subtype) breast cancer. Blood samples were drawn prior to any cancer treatment. Controls were identified largely from female residents of Harris County using random digit dialing. Written informed consent was obtained from each study participant. To assess the relationship between HRR capacity and breast cancer risk, we selected 152 cases from all cases recruited during the first 6 months of 2017 (n = 183). The selection was mainly based on the availability of PBL samples and the completeness of demographic and clinical characteristics at baseline. Equal number of controls were selected from a pool of controls recruited in 2017 (n = 205). The cases and controls were frequently matched on age, gender and ethnicity. Self-reported ethnic background was used to define race and ethnicity.

To confirm the results, we included a validation population with 50 incident breast cancer cases and 50 controls identified from Mano-A-Mano, the Mexican American Cohort study (MAC). Detailed description of breast cancer cases in the MAC study has been described previously (13,14). To the end of 1 December 2017, with a median follow-up time of 8.2 years, a total of 126 newly diagnosed breast cancers were identified. Among them, 109 were validated through the Texas Cancer Registry and had blood samples that were collected at baseline. The case selection was based on the availability of PBL samples in the biorepository and the time of breast cancer diagnosis (at least 1 year after their initial enrollment into the MAC and blood donation). The cases and controls were matched on age at recruitment (±2 years), date of biospecimen collection (±1 year) and gender. The study protocol was approved by the Institutional Review Board at M. D. Anderson Cancer Center.

HRR assay in PBLs

The HRR assay is a modified HCR assay which specifically assesses HRR capacity. In brief, we constructed two plasmids, pGL-luc-del1 and pGL-luc-del2, by deleting a portion of the Luc report gene in the PGL-3-luc vector. A ~400 bp homolog sequence remained between two deletions. Due to the deletions, none of the plasmids could produce normal luciferase protein. Both plasmids were co-transfected into cells and cultured for 48 h. If the cells had normal HRR ability, a normal pGL-3-luc plasmid that could synthesize luciferase would be regenerated. Therefore, luciferase activity provided an index for HRR ability. Approximately 0.5 × 10⁶ PBLs were cultured in RPMI 1640 supplemented with 20% FCS and PHA at 37°C and 5% CO₂ and incubated for 72 h. The DEAE-dextran (Sigma–Aldrich) method was used to transfect aliquots with pGL-luc-del1 and pGL-luc-del2 or pGL-luc alone as the positive control. The cultures were then incubated for 40 h before harvesting. Afterwards, the cell pellets were suspended in reporter lysis buffer (Promega), frozen, thawed once in ethanol-dry ice and 37°C water baths, and then centrifuged. For each LUC assay, 20 µl of cell extract supernatant was mixed with 100 µl of Luciferase Assay Substrate (Promega) in a tube at room temperature. LUC activity in arbitrary light-intensity units was measured with a luminometer (Turner). LUC light-intensity units were recorded for the cells with undamaged pGL-luc alone (control), and for the cells with damaged, co-transfected pGL-luc-del1 and pGL-luc-del2. DNA repair capacity (%) was calculated as the product of 100% and the ratio of the damaged plasmid values to the undamaged plasmid values. Duplicate procedures were performed for each sample. The average coefficient of variation was 8.06% (range 3.34–11.29%), indicating high reproducibility of the assay.

Statistical analysis

We used the statistical software package SAS version 9.4 (SAS, Cary, NC, USA) for all analyses. First, we evaluated whether HRR capacity and selected demographic characteristics differed between breast cancer patients and healthy controls. The Student t-test was used for two-level dichotomous variables, and analysis of variance was used for variables with more than two levels. Next, we used the Student t-test to evaluate whether mean HRR capacity differed across categories in each of the selected demographic variables of the cases and controls, and tumor characteristics of the cases. We also compared case–control difference in HRR capacity in each category of each selected demographic variable. For the association between HRR capacity and breast cancer risk, we used unconditional multivariate logistic regression to estimate odds ratios (ORs) and 95% CIs. The analysis was adjusted for potential confounders. HRR capacity was treated as a continuous variable or as a categorical variable in dichotomous and tertile analyses. In dichotomized analysis, HRR capacity was designated as ‘high’ or ‘low’ using the controls’ 75% levels of HRR capacity as cutoffs. In tertile analysis, HRR capacity was designated as ‘high’, ‘medium’ or ‘low using the controls’ tertile levels of HRR capacity as cutoffs. In the validation analysis, HRR capacity was treated as a continuous variable. We applied similar multivariate logistic regression analysis to assess relationships between HRR capacity and breast cancer risk.

Results

A total of 152 breast cancer cases and 152 healthy controls were included in the analysis (Table 1). In general, the case and control groups were well-matched on age group and race. No significant differences were observed for other demographic variables, including BMI category, family history (one first degree or second degree relative diagnosed with breast or ovarian cancer at any age), smoking status, alcohol status and physical activity. In terms of tumor characteristics, 68.42% of cases were ER+/PR+, 31.58% were triple negative breast cancer (TNBC) and the majority of the cases (76.97%) were diagnosed at an early stage (I and II). Overall, the cases had statistically significantly lower HRR capacity than the controls (7.86 % versus 10.95%, P < 0.0001).

Next, we assessed the correlation between HRR capacity and selected demographic variables among controls (Table 2). Compared with older women (≥51 years), younger women (<51 years) had higher HRR capacity (11.33% versus 10.57%), but the difference did not reach statistical significance (P = 0.216). Compared with White women, Black women had statistically significantly lower HRR (9.58% versus 11.85%, P = 0.015). Compared with those with no family history of cancer, those with family history of cancer had lower HRR capacity (9.80% versus 11.09%), although the difference did not reach statistical significance (P = 0.201). No statistical significance was observed between Hispanic and White women. Furthermore, no significant difference was observed across BMI category, smoking status, alcohol status and physical activity. The same analysis was also applied to the cases. Similarly, Black breast cancer patients had statistically significantly lower HRR than their White counterparts (6.32% versus 8.54%, P = 0.012). We also assessed the relationship between tumor characteristics and HRR capacity among cases. No significant association was observed between early and late stage tumors. However, TNBC cases had statistically significant lower HRR capacity than cases with ER+/PR+ tumors (6.63% versus 8.43%, P = 0.006). Then, we assessed whether HRR capacity differed between cases and controls in each category of selected characteristics. As expected, the controls had statistically significantly higher HRR capacity than the cases in each category, except with family history of cancer (P = 0.097) possibly due to the small sample size.

We then examined the association between HRR capacity and breast cancer risk (Table 3). If treated as a continuous variable, decreased HRR capacity was associated with 1.42-fold increased risk of breast cancer after adjusting age, race, BMI category, smoking status, alcohol status and physical activity (OR = 1.42, 95% CI: 1.21, 2.53, P < 0.001). In dichotomized analysis, using the 75% levels of HRR capacity in controls as the cutoff point (10.42%), those with lower HRR capacity had 1.97-fold increased risk of breast cancer (OR = 1.97, 95% CI: 1.18, 3.34, P = 0.001). In tertile analysis, the risk association between decreased HRR capacity and breast cancer risk was further validated. Compared with those who had the highest levels of HRR capacity, those with lowest levels of HRR capacity had 1.89-fold increased risk of breast cancer (OR = 1.89, 95% CI: 1.07, 4.23, P = 0.012). In addition, a significant trend of increasing risk of breast cancer was observed when HRR capacity decreased (P < 0.001) (Table 3).

Finally, we attempted to confirm the observed significant association between HRR capacity and breast cancer risk in pre-diagnostic samples (Table 4). As shown in Supplemental Table S1, available at Carcinogenesis Online, the cases and controls were well-matched on age, parity, education level, birthplace, language acculturation, BMI category, smoking status, alcohol drinking and physical activity. Compared with healthy controls (n = 50), incident breast cancer cases (n = 50) had statistically significant lower levels of HRR capacity (9.08% versus 11.25%, P = 0.014). In the univariate analysis, lower HRR capacity was associated with 1.35-fold increased risk of breast cancer (OR = 1.35, 95% CI: 1.10, 4.92, P = 0.019). In the multivariate analysis, lower HRR capacity was associated with 1.21-fold increased risk of breast cancer (OR = 1.21, 95% CI: 1.04, 4.58, P = 0.036) after adjusting age, BMI category, smoking status, alcohol status and physical activity.

Discussion

To date, no study has assessed the relationship between HRR capacity and breast cancer risk. Furthermore, the role of DNA repair capacity in breast cancer risk has not been examined prospectively using pre-diagnostic samples. In the discovery stage, we found that decreased HRR capacity was associated with 1.42-fold increased risk of breast cancer (OR = 1.42, 95% CI: 1.21, 2.53, P < 0.001). The significant association was further confirmed in the validation stage using pre-diagnostic samples, as decreased HRR capacity was associated with 1.21-fold increased risk of breast cancer (OR = 1.21, 95% CI: 1.04, 4.58, P < 0.001). In addition, we found that HRR capacity in PBLs differed by race and tumor subtype. Compared with their counterparts, Black and TNBC patients had statistically lower HRR capacity (P = 0.015 and 0.006, respectively).

The observed positive association between decreased HRR capacity and breast cancer risk is not surprising, as the molecular pathway of HRR has been studied extensively. Many genes that play key roles in the HRR pathway are known breast cancer-related genes, including, for example, BRCA1, BRCA2, Rad 52, Rad 51, ATM, etc. (5,6). The contribution of these genes to breast carcinogenesis is evident for both familial and sporadic cancers (15,16). In a recent pan-cancer study, HRR capacity in tumors was found to be correlated with the total number of pathogenic variants in HRR genes, suggesting the existence of genetic inheritability in HRR capacity (17). In the future, we will assess the extent to which the variation in HRR capacity in PBLs can be explained by genetic variants in HRR genes.

The difference in DNA repair capacity between breast cancer cases and healthy controls has been reported previously in case–control studies (18–20). However, these reports focused on nucleotide excision repair (NER) rather than HRR. For example, using a HCR assay, Vergne et al. reported that NER capacity was statistically significantly lower among breast cancer patients than healthy controls in Puerto Rican Hispanic women (19). In another early study, elevated levels of BPDE-DNA adducts, a marker of deficiency in NER, was found in breast cancer patients compared with healthy controls (18). Though these studies have focused on NER and are unable infer causal relationships due to study design, they have illuminated the importance of DNA repair capacity in breast cancer etiology and have further suggested that multiple DNA repair pathways may be involved in breast carcinogenesis.

Our finding that Black controls had lower HRR capacity than White controls is interesting and consistent with a recent report (17). Analyzing data from The Cancer Genome Atlas (TCGA), Sinha et al. discovered that tumors from Blacks had higher genomic instability and lower HRR capacity than those from Whites, and the difference was consistent across many cancer types. They also found that the total number of pathogenic variants in HRR genes was higher in tumors from Blacks than in those from Whites, suggesting that the difference in HRR capacity in tumors has genetic ancestral determinants. In our study, we observed racial differences in HRR capacity in breast cancer cases in addition to controls, as Black cases had significantly lower HRR capacity than White cases (P = 0.012). Thus, our data support the notion that the racial difference in HRR capacity is at least partially due to genetic differences between Blacks and Whites. To further clarify this association, in the future, we will correlate HRR capacity in PBLs with genetic ancestry.

Another interesting finding is that HRR capacity was lower in TNBC patients than in ER+/PR+ patients. HRR deficiency has been implicated in TNBC due to BRCA mutations (8). In fact, Poly (ADP-Ribose) Polymerase inhibitors (PARPi), which target the HRR process, have been approved by the FDA for 10–15% of TNBC patients with germline mutations in BRCA1/2 (21). The double targeting of PARP1/2 and BRCA1/2 mutations which cause HRR deficiency consequently leads to the death of TNBC tumor cells. Whether PARPi have treatment effects in BRCA wildtype TNBC patients is still unclear. Given the mechanism of the drugs, it has been hypothesized that TNBC patients who have no BRCA mutations but have suboptimal HRR capacity (e.g. BRCAness) may respond to PARPi treatment. These patients must be assessed for their HRR capacity, which may be achieved via the HCR assay described in this study.

In brief, we have demonstrated that reduced HRR capacity in PBLs is associated with increased risk of breast cancer. Due to the modest sample size in both stages, our results need to be further validated in large prospective cohort studies. In addition, we must further examine the genetic contribution to HRR capacity in PBLs and possible racial differences. Lastly, HRR capacity was measured in PBLs, which may not reflect HRR capacity in normal breast tissues. Future correlative study is needed to assess the extent to which HRR capacity in PBLs is correlated with HRR capacity in normal breast tissues.

Ethics approval

All procedures performed in this study were approved by the Institutional Review Board at M D Anderson Cancer Center and in accordance with the ethical standards of 1964 Helsinki declaration and its later amendments or comparable ethical standards. Written informed consent was obtained from all participants.

Abbreviations

Abbreviations

• DRC

  DNA-repair capacity

• DSBs

  DNA double-strand breaks

• HCR

  Host cell reactivation

• HRR

  homologous recombination repair

• NER

  nucleotide excision repair

• NHEJ

  non-homologous end joining

• PBLs

  peripheral blood lymphocytes

• TNBC

  triple negative breast cancer

Funding

The study was supported by U01 CA179655 from National Cancer Institutes/National Institutes of Health.

Conflict of Interest Statement: None declared.

Data availability

The data that support the findings of this study are available from the corresponding author upon reasonable request.

References

© The Author(s) 2020. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oup.com.

Issue Section:

CANCER BIOMARKERS AND MOLECULAR EPIDEMIOLOGY

Email alerts

Citing articles via

More from Oxford Academic