Women who have had breast cancer in the past are at increased risk of developing a second primary cancer (SPC), including second primary breast cancer (SPBC) or a second primary non-breast cancer (SPNBC). In the Multiethnic Cohort (MEC) Study, we conducted a prospective cohort analysis in 3,223 female breast cancer survivors from five racial/ethnic populations (White, African American, Japanese American, Latino, and Native Hawaiian) to assess the association of rare pathogenic variants (PV) in 37 known cancer predisposition genes with risk of SPC. A total of 719 (22.3%) women developed SPC, of which, 323 (10.0%) were SPBC. Germline PVs in BRCA1 (HR, 2.28; 95% CI, 1.11–4.65) and ERCC2 (HR, 3.51; 95% CI, 1.29–9.54) were significantly enriched in women with SPC. In the subtype analysis for SPBC, a significant association of ERCC2 PVs (HR, 5.09; 95% CI, 1.58–16.4) and a suggestive association of BRCA2 PVs (HR, 2.24; 95% CI, 0.91–5.55) were observed. There was also a higher risk of SPNBC in carriers of BRCA1 PVs (HR, 2.98; 95% CI, 1.21–7.36). These results provide evidence that germline PVs in BRCA1, BRCA2, and ERCC2 contribute to the development of SPC in breast cancer survivors. These findings also suggest that compromised DNA repair mechanisms could be a predisposition factor for SPC in patients with breast cancer, supporting the need for closer monitoring of SPC in women carrying PVs in these genes.

Significance:

This multiethnic study links germline pathogenic variants in BRCA1, BRCA2, and ERCC2 to the development of second primary cancer in breast cancer survivors, providing biological insights and biomarkers to guide patient monitoring.

Women who have had breast cancer in the past are at increased risk of developing a second primary cancer (SPC), with contralateral breast cancer (CBC) being the most common secondary malignancy. Identifying risk factors for SPC is essential for cancer prevention efforts, especially with an increasing population of breast cancer survivors (1). Past epidemiologic studies have identified several factors associated with the development of CBC or second primary breast cancer (SPBC) in breast cancer survivors. The risk of developing SPBC is significantly higher among women with younger age at initial diagnosis (<40 years; refs. 2–6), a positive family history of breast cancer in first-degree relatives (6, 7), and women with hormone receptor (HR) negative tumors (2, 5, 8–10). Women treated with radiotherapy have a 16% to 26% higher risk of developing a subsequent breast tumor than nonirradiated women (2, 5, 11). In contrast, hormonal therapy and chemotherapy were associated with a modest protective effect (2, 7, 12). SPBC is also found to be more common in African American breast cancer survivors compared with European Americans (2, 4–6, 8).

Genetic predisposition, notably pathogenic variants (PVs) in BRCA1 or BRCA2 susceptibility genes, contributes to the risk of SPBC. A meta-analysis of 20 cohort studies reported that the 5-year cumulative risk of SPBC in women carrying BRCA1 and BRCA2 PVs was approximately 5-fold and 3-fold, respectively, as in non-BRCA carriers (13). Furthermore, the increased risk of SPBC was much greater in BRCA1/2 carriers with an initial breast cancer diagnosis before age 40 (14). In addition to BRCA1 and BRCA2, a recent targeted sequencing study including 75,550 women with unilateral breast cancer and 7,728 women with SPBC also reported significant associations with PVs in CHEK2, PALB2, and TP53 in European American and African American women. These associations were nonsignificant or not tested in Latino or Asian Americans, due to the low prevalence of the PVs in these two ethnic groups (15).

In addition to SPBC, breast cancer survivors have been reported to have a significantly elevated risk for subsequent cancers of the thyroid (16–19), uterine corpus (16, 19–21), ovary (16, 19, 20), esophagus, stomach, colon, lung, melanoma of the skin, sarcoma, and acute myeloid leukemia (16). Second primary non-breast cancers (SPNBC) have been associated with older age at initial breast cancer diagnosis (17, 19, 20, 22–24) as well as radiotherapy treatment (24–27). There is also a small increased risk of developing acute myeloid leukemia, melanoma, and uterine cancer after receiving chemotherapy. Hormonal treatment with tamoxifen is known to increase the risk for uterine cancer (24).

Little is known regarding the genetic risk of developing a non-breast SPC among breast cancer survivors. The familial clustering of multiple primary cancers involving breast, ovary, and uterine corpus, the increased risk of thyroid cancer and melanoma after breast cancer, as well as the reciprocally elevated risk of breast cancer after these tumors, all suggest shared genetic etiology. In 3,223 female breast cancer survivors from the Multiethnic Cohort (MEC) Study, we conducted a prospective analysis to test the hypothesis that known cancer predisposition genes may play a role in the development of SPC in breast cancer survivors.

Study population

The MEC is an ongoing prospective cohort study designed to examine the association of lifestyle and genetic factors with the incidence of cancer. The design and establishment of the MEC in 1993 to 1996 has been described elsewhere (28). Incident cases of breast cancer were identified through a linkage of MEC participants to the SEER tumor registries in Hawaii and California. The linkage also provided clinical information on tumor stage, grade, hormone receptor status, and first course of treatment. The University of Hawaii and the University of Southern California Institutional Review Boards approved the study protocol, and all participants provided written informed consent in accordance with the principles outlined in the Declaration of Helsinki.

Previously, a nested case-control study of 3,641 female patients with breast cancer and 3,689 unaffected women from the MEC was included in the CAnceR RIsk Estimates Related to Susceptibility (CARRIERS) Consortium for targeted sequencing (29–31). The CARRIERS study was approved by the institutional review board at the Mayo Clinic. From this targeted sequencing study population, the current analysis included 3,223 women with invasive breast cancer during follow-up (December 31, 2017), excluding women with any in situ cancer (N = 297) or women who had any cancer diagnosed before cohort entry (N = 372). According to the SEER Breast Multiple Primary and Histology Coding Rules (32), 719 (22.3%) women had later developed SPC. On the basis of the first cancer diagnosis following breast cancer, 323 cases (10.0%) were SPBC and 396 (12.3%) were SPNBC.

Targeted sequencing and pathogenic variants

Germline DNA from these 3,223 patients with breast cancer was isolated from peripheral blood (89.4%), mouthwash (4.7%), or saliva (5.9%) samples using QIAGEN DNA extraction kits. Along with the remaining CARRIERS study samples, the MEC DNA samples were analyzed using a custom amplicon-based QIAseq panel (Qiagen) covering all coding regions and consensus splice sites from 37 cancer predisposition genes, as described previously (29, 31). High-quality sequence data (read depth of > 20 times) were obtained for 99.3% of the targeted regions. From the variants with a minor allele frequency less than 0.01 in patients with breast cancer, all loss-of-function variants (nonsense, frameshift, consensus splice sites) or intronic and missense variants identified as “pathogenic” or “likely pathogenic” in the ClinVar database were classified as pathogenic variants (PV; ref. 29). We restricted germline PVs to those with an alternate allele fraction (AAF) between 0.30 and 0.70 to exclude suspected mosaic somatic variants derived from the expansion of clonal populations of blood cells. On the basis of the existing evidence on breast cancer susceptibility, the genes included in the custom panel were categorized as established breast cancer predisposition genes (ATM, BARD1, BRCA1, BRCA2, CDH1, CHEK2, NF1, PALB2, PTEN, RAD51C, RAD51D, and TP53), candidate breast cancer predisposition genes (BLM, BRIP1, CDKN2A, ERCC3, FANCC, FANCM, MLH1, MRE11A, MSH2, MSH6, NCN, RAD50, RECQL, RINT1, SLX4, and XRCC2), and other cancer predisposition genes (APC, EPCAM, ERCC2, KRAS, MEN1, MUTYH, PMS2, PPM1D, and PRSS1).

Statistical analysis

In this prospective cohort analysis, we assessed the association of known cancer predisposition genes with the incidence of SPC (N = 719), and subsequently with SPBC (N = 323) or SPNBC (N = 396). The subtype analysis in SPNBC excluded women with SPBC and vice versa. In addition, 33 women who had undergone a bilateral mastectomy were further excluded from the analysis of SPBC. Women who harbored any PV in each of the sequenced genes were considered carriers. The single gene association test was performed comparing carriers to noncarriers. Our ability to estimate the relative risks for individual genes was limited to 11 genes with at least five carriers observed in all women and at least one carrier in women with and without SPC (ATM, BRCA1, BRCA2, CHEK2, PALB2, BLM, BRIP1, ERCC3, FANCC, ERCC2, and MUTYH). In all analyses, women contributed person-time at risk from the diagnosis date of the initial breast tumor until the diagnosis date of the second primary cancer of interest, death, or end of follow-up (December 31, 2017). The HR and 95% confidence intervals (CI) were estimated using Cox proportional hazard models with age as the time metric. The proportionality assumption was tested by Schoenfeld residuals and found to be met. To control for potential confounders, all regression models were adjusted for age at first breast tumor diagnosis, race/ethnicity (White, African American, Japanese American, Latino, and Native Hawaiian), family history of breast cancer in first-degree relatives (positive vs. negative), SEER summary stage of the first breast tumor (localized, regional, and distant), estrogen receptor (ER) status (positive vs. negative) and progesterone receptor (PR) status (positive vs. negative) of the first breast tumor. First course of radiotherapy (administered vs. not administered) was also included in the final model on SPBC. These variables were included because they were potential confounders (P < 0.05 in univariate analysis) or because they are established or suspected risk factors for the specific SPC phenotype. Missing values were included as an “unknown” category so that all observations could be included in the analyses. We also performed sensitivity analyses that (i) estimated the P values from the Cox proportional hazard models using a saddlepoint approximation to account for the small number of carriers (33), (ii) imputed the missing values of covariates with the ethnic-specific expected values (imputation method), and (iii) used multistate models to model a third primary cancer (N = 96) as a subsequent event following the diagnosis of a SPC while accounting for competing risk of death. All statistical analyses were performed with R v.3.6 (34). P values less than 0.05 were considered statistically significant. Tests of statistical significance were two-sided.

We performed gene set analysis to assess whether being a carrier of any gene in the gene set was associated with the risk of SPC. Gene sets were determined either by prior evidence of these genes with breast cancer (12 established breast cancer genes, 16 candidate breast cancer genes, and 9 other cancer predisposition genes) or by DNA repair pathways. We tested six DNA repair pathways including 25 DNA repair genes (DRG) involved in base excision repair (BER; MUTYH), nucleotide excision repair (NER; ERCC3, ERCC2), double-strand break repair via homologous recombination (HR; ATM, BLM, BARD1, BRCA1, BRCA2, BRIP1, MRE11A, PALB2, RAD50, RAD51C, RAD51D, RECQL, SLX4, XRCC2), mismatch repair (MMR; MLH1, MSH2, MSH6, PMS2), Fanconi Anemia (FA; BRCA1, BRCA2, BRIP1, FANCC, FANCM, PALB2, RAD51C, SLX4), and other related genes (CHEK2, TP53; refs. 35, 36). We performed additional analysis on DNA repair pathways, excluding genes that were associated with SPC in our single-gene association analysis (P < 0.10). All variables adjusted in the single gene regression models were also included in the gene set tests.

Data availability

The data analyzed in this study are publicly available in Database of Genotypes and Phenotypes (dbGaP) at phs002820.v1.

Patient demographics and clinical characteristics

The overall study cohort included 3,223 females with incident breast cancer, of which, 719 SPC cases were identified during an average follow-up time of 11.2 years (up to 24.7 years), including 323 SPBC cases and 396 SPNBC cases (Table 1; Supplementary Fig. S1). On average, women with SPC, SPBC, or SPNBC were first diagnosed with breast cancer at age of 66.5, 65.5, and 67.3, respectively, in comparison to 68.1 for women with primary breast cancer (PBC) only, but these differences were not statistically significant. Among the 323 women with SPBC, 253 (78.3%) had two breast cancer diagnoses more than 6 months apart, 60 (18.6%) had synchronous bilateral breast cancer, and the remaining 10 SPBC cases had two breast tumors of different histological subtypes diagnosed within 6 months. With the adjustment for the other covariates, women with a positive family history of breast cancer in first-degree relatives were more likely to be diagnosed with SPC (HR, 1.30; 95% CI, 1.07–1.56; P = 0.007; Supplementary Table S1) with the association being slightly stronger for SPBC (HR, 1.42; 95% CI, 1.08–1.88; P = 0.01) than for SPNBC (HR, 1.28; 95% CI, 0.99–1.64; P = 0.06). Compared with White women, Latinas had a substantially lower risk of SPC, SPBC, or SPNBC (HR of 0.59–0.75, P of 0.002–0.06), and Japanese American women were less likely to be diagnosed with SPC and SPNBC (HR of 0.65–0.76, P of 0.002–0.006). The risks of SPC, SPBC, or SPNBC were not statistically different between African American women and White women. Although we observed no significant relationship between hormone receptor status and SPC, positive PR status of the first breast cancer was marginally associated with a lower risk of SPNBC (HR, 0.74; 95% CI, 0.55–1.00; P = 0.05; Supplementary Table S1). We also found a significant inverse association of first-course radiotherapy with SPBC (HR, 0.78; 95% CI, 0.62–0.99; P = 0.04).

Table 1.

Demographic and clinical characteristics of study population (N = 3,223).

OverallaPBCSPCaSPBCSPNBC
CharacteristicsN = 3,223N = 2,504N = 719N = 323N = 396
Race/ethnicity 
 White 755 (23.4) 563 (22.5) 192 (26.7) 79 (24.5) 113 (28.5) 
 African American 600 (18.6) 466 (18.6) 134 (18.6) 60 (18.6) 74 (18.7) 
 Native Hawaiian 268 (8.3) 197 (7.9) 71 (9.9) 38 (11.8) 33 (8.3) 
 Japanese American 992 (30.8) 786 (31.4) 206 (28.7) 102 (31.6) 104 (26.3) 
 Latino 608 (18.9) 492 (19.6) 116 (16.1) 44 (13.6) 72 (18.2) 
Age at cohort entry, mean (SD) 59.4 (8.2) 59.3 (8.3) 59.8 (8.0) 58.8 (8.0) 60.7 (7.8) 
Age at first breast cancer diagnosis 
 Mean (SD) 67.7 (9.0) 68.1 (9.1) 66.5 (8.7) 65.5 (9.1) 67.3 (8.2) 
 45–60 690 (21.4) 513 (20.5) 177 (24.6) 100 (31.0) 77 (19.4) 
 60–70 1,184 (36.7) 910 (36.3) 274 (38.1) 110 (34.1) 164 (41.4) 
 70–80 1,077 (33.4) 842 (33.6) 235 (32.7) 100 (31.0) 135 (34.1) 
 80+ 272 (8.4) 239 (9.5) 33 (4.6) 13 (4.0) 20 (5.1) 
First-degree family history of breast cancer 
 Negative 2571 (79.8) 2023 (80.8) 548 (76.2) 240 (74.3) 308 (77.8) 
 Positive 505 (15.7) 366 (14.6) 139 (19.3) 63 (19.5) 76 (19.2) 
 Missing 147 (4.6) 115 (4.6) 32 (4.5) 20 (6.2) 12 (3.0) 
Tumor stage of first breast cancer 
 Localized 2,318 (71.9) 1,784 (71.2) 534 (74.3) 242 (74.9) 292 (73.7) 
 Regional 786 (24.4) 615 (24.6) 171 (23.8) 74 (22.9) 97 (24.5) 
 Distant 55 (1.7) 47 (1.9) 8 (1.1) 5 (1.5) 3 (0.8) 
 Missing 64 (2.0) 58 (2.3) 6 (0.8) 2 (0.6) 4 (1.0) 
Estrogen receptor (ER) status of first breast cancer 
 Negative 538 (16.7) 420 (16.8) 118 (16.4) 55 (17.0) 63 (15.9) 
 Positive 2,333 (72.4) 1,831 (73.1) 502 (69.8) 222 (68.7) 280 (70.7) 
 Missing 352 (10.9) 253 (10.1) 99 (13.8) 46 (14.3) 53 (13.4) 
Progesterone receptor (PR) status of first breast cancer 
 Negative 801 (24.9) 621 (24.8) 180 (25.0) 77 (23.8) 103 (26.0) 
 Positive 1,935 (60.0) 1526 (60.9) 409 (56.9) 186 (57.6) 223 (56.3) 
 Missing 487 (15.1) 357 (14.3) 130 (18.1) 60 (18.6) 70 (17.7) 
Radiation treatment of first breast cancerb 
 Not administered 1,523 (47.3) 1,185 (47.3) 338 (47.0) 163 (50.5) 175 (44.2) 
 Administered 1,655 (51.3) 1,281 (51.2) 374 (52.0) 156 (48.3) 218 (55.1) 
 Missing 45 (1.4) 38 (1.5) 7 (1.0) 4 (1.2) 3 (0.8) 
Chemotherapy of first breast cancerb 
 Not administered 2,249 (69.8) 1,751 (69.9) 498 (69.3) 218 (67.5) 280 (70.7) 
 Administered 899 (27.9) 692 (27.6) 207 (28.8) 100 (31.0) 107 (27.0) 
 Missing 75 (2.3) 61 (2.4) 14 (1.9) 5 (1.5) 9 (2.3) 
Hormonal therapy of first breast cancerb 
 Not administered 1,675 (52.0) 1,291 (51.6) 384 (53.4) 181 (56.0) 203 (51.3) 
 Administered 1,452 (45.1) 1,138 (45.4) 314 (43.7) 133 (41.2) 181 (45.7) 
 Missing 96 (3.0) 75 (3.0) 21 (2.9) 9 (2.8) 12 (3.0) 
OverallaPBCSPCaSPBCSPNBC
CharacteristicsN = 3,223N = 2,504N = 719N = 323N = 396
Race/ethnicity 
 White 755 (23.4) 563 (22.5) 192 (26.7) 79 (24.5) 113 (28.5) 
 African American 600 (18.6) 466 (18.6) 134 (18.6) 60 (18.6) 74 (18.7) 
 Native Hawaiian 268 (8.3) 197 (7.9) 71 (9.9) 38 (11.8) 33 (8.3) 
 Japanese American 992 (30.8) 786 (31.4) 206 (28.7) 102 (31.6) 104 (26.3) 
 Latino 608 (18.9) 492 (19.6) 116 (16.1) 44 (13.6) 72 (18.2) 
Age at cohort entry, mean (SD) 59.4 (8.2) 59.3 (8.3) 59.8 (8.0) 58.8 (8.0) 60.7 (7.8) 
Age at first breast cancer diagnosis 
 Mean (SD) 67.7 (9.0) 68.1 (9.1) 66.5 (8.7) 65.5 (9.1) 67.3 (8.2) 
 45–60 690 (21.4) 513 (20.5) 177 (24.6) 100 (31.0) 77 (19.4) 
 60–70 1,184 (36.7) 910 (36.3) 274 (38.1) 110 (34.1) 164 (41.4) 
 70–80 1,077 (33.4) 842 (33.6) 235 (32.7) 100 (31.0) 135 (34.1) 
 80+ 272 (8.4) 239 (9.5) 33 (4.6) 13 (4.0) 20 (5.1) 
First-degree family history of breast cancer 
 Negative 2571 (79.8) 2023 (80.8) 548 (76.2) 240 (74.3) 308 (77.8) 
 Positive 505 (15.7) 366 (14.6) 139 (19.3) 63 (19.5) 76 (19.2) 
 Missing 147 (4.6) 115 (4.6) 32 (4.5) 20 (6.2) 12 (3.0) 
Tumor stage of first breast cancer 
 Localized 2,318 (71.9) 1,784 (71.2) 534 (74.3) 242 (74.9) 292 (73.7) 
 Regional 786 (24.4) 615 (24.6) 171 (23.8) 74 (22.9) 97 (24.5) 
 Distant 55 (1.7) 47 (1.9) 8 (1.1) 5 (1.5) 3 (0.8) 
 Missing 64 (2.0) 58 (2.3) 6 (0.8) 2 (0.6) 4 (1.0) 
Estrogen receptor (ER) status of first breast cancer 
 Negative 538 (16.7) 420 (16.8) 118 (16.4) 55 (17.0) 63 (15.9) 
 Positive 2,333 (72.4) 1,831 (73.1) 502 (69.8) 222 (68.7) 280 (70.7) 
 Missing 352 (10.9) 253 (10.1) 99 (13.8) 46 (14.3) 53 (13.4) 
Progesterone receptor (PR) status of first breast cancer 
 Negative 801 (24.9) 621 (24.8) 180 (25.0) 77 (23.8) 103 (26.0) 
 Positive 1,935 (60.0) 1526 (60.9) 409 (56.9) 186 (57.6) 223 (56.3) 
 Missing 487 (15.1) 357 (14.3) 130 (18.1) 60 (18.6) 70 (17.7) 
Radiation treatment of first breast cancerb 
 Not administered 1,523 (47.3) 1,185 (47.3) 338 (47.0) 163 (50.5) 175 (44.2) 
 Administered 1,655 (51.3) 1,281 (51.2) 374 (52.0) 156 (48.3) 218 (55.1) 
 Missing 45 (1.4) 38 (1.5) 7 (1.0) 4 (1.2) 3 (0.8) 
Chemotherapy of first breast cancerb 
 Not administered 2,249 (69.8) 1,751 (69.9) 498 (69.3) 218 (67.5) 280 (70.7) 
 Administered 899 (27.9) 692 (27.6) 207 (28.8) 100 (31.0) 107 (27.0) 
 Missing 75 (2.3) 61 (2.4) 14 (1.9) 5 (1.5) 9 (2.3) 
Hormonal therapy of first breast cancerb 
 Not administered 1,675 (52.0) 1,291 (51.6) 384 (53.4) 181 (56.0) 203 (51.3) 
 Administered 1,452 (45.1) 1,138 (45.4) 314 (43.7) 133 (41.2) 181 (45.7) 
 Missing 96 (3.0) 75 (3.0) 21 (2.9) 9 (2.8) 12 (3.0) 

aThe first primary breast cancer was diagnosed between 1993 and 2016 and the second primary cancer was diagnosed between 1994 and 2017.

bFirst-course treatment from SEER cancer registry.

Association of PVs in cancer predisposition genes with SPC

We were able to test for association with 11 genes that had at least five carriers in all women and at least one carrier in women with and without SPC (Table 2). BRCA2 was the most frequently affected gene, with 20 of 3,233 (0.62%) breast cancer patients harboring a germline PV, followed by PALB2 (16 carriers, 0.50%), BRCA1 (14 carriers, 0.43%), and ATM (14 carriers, 0.43%; Supplementary Fig. S2; Supplementary Table S2). Two genes were significantly associated with SPC: BRCA1 (HR, 2.28; 95% CI, 1.11–4.65; P = 0.024) and ERCC2 (HR, 3.51; 95% CI, 1.29–9.54; P = 0.014), after controlling for potential confounders. We observed a significant association of ERCC2 (HR, 5.09; 95% CI, 1.58–16.4; P = 0.007) and a suggestive association of BRCA2 (HR, 2.24; 95% CI, 0.91–5.55; P = 0.08) with SPBC. The association of BRCA1 remained significant in the analysis of SPNBC (HR, 2.98; 95% CI, 1.21–7.36; P = 0.02), and a similar suggestive association was also found for SPBC (HR, 2.64; 95% CI, 0.82–8.43; P = 0.10). There were no appreciable differences in the results from the sensitivity analyses (Supplementary Table S3).

Table 2.

Association of germline PVs in cancer predisposition genes with SPC, SPBC, or SPNBC.

PBCSPCSPBCbSPNBC
GeneaCarriers (%)Carriers (%)HR (95% CI)cPCarriers (%)HR (95% CI)cPCarriers (%)HR (95% CI)cP
Established breast cancer predisposition gene 
ATM 10 (0.40) 4 (0.56) 1.48 (0.55–3.98) 0.44 1 (0.31) 0.85 (0.12–6.29) 0.88 3 (0.76) 1.91 (0.61–6.00) 0.27 
BARD1 2 (0.08) 0 (0.00) — — 0 (0.00) — — 0 (0.00) — — 
BRCA1 6 (0.24) 8 (1.11) 2.28 (1.11–4.65) 0.024 3 (0.93) 2.64 (0.82–8.43) 0.10 5 (1.26) 2.98 (1.21–7.36) 0.018 
BRCA2 15 (0.60) 5 (0.70) 1.18 (0.49–2.88) 0.71 5 (1.55) 2.24 (0.91–5.55) 0.08 0 (0.00) — — 
CDH1 0 (0.00) 1 (0.14) — — 0 (0.00) — — 1 (0.25) — — 
CHEK2 3 (0.12) 3 (0.42) 2.03 (0.64–6.46) 0.23 1 (0.31) 1.81 (0.24–13.55) 0.56 2 (0.51) 2.32 (0.56–9.60) 0.25 
NF1 2 (0.08) 0 (0.00) — — 0 (0.00) — — 0 (0.00) — — 
PALB2 12 (0.48) 4 (0.56) 0.89 (0.33–2.40) 0.82 2 (0.62) 0.96 (0.24–3.94) 0.96 2 (0.51) 0.95 (0.23–3.87) 0.95 
PTEN 0 (0.00) 0 (0.00) — — 0 (0.00) — — 0 (0.00) — — 
RAD51C 5 (0.20) 0 (0.00) — — 0 (0.00) — — 0 (0.00) — — 
RAD51D 2 (0.08) 0 (0.00) — — 0 (0.00) — — 0 (0.00) — — 
TP53 0 (0.00) 1 (0.14) — — 0 (0.00) — — 1 (0.25) — — 
Candidate breast cancer predisposition gene 
BLM 4 (0.16) 2 (0.28) 2.50 (0.61–10.18) 0.20 1 (0.31) 4.12 (0.55–30.92) 0.17 1 (0.25) 2.74 (0.38–19.79) 0.32 
BRIP1 4 (0.16) 1 (0.14) 1.02 (0.14–7.34) 0.98 1 (0.31) 1.99 (0.27–14.40) 0.50 0 (0.00) — — 
CDKN2A 1 (0.04) 0 (0.00) — — 0 (0.00) — — 0 (0.00) — — 
ERCC3 4 (0.16) 1 (0.14) 1.24 (0.17–8.90) 0.83 0 (0.00) — — 1 (0.25) 2.21 (0.31–16.07) 0.43 
FANCC 5 (0.20) 2 (0.28) 1.28 (0.32–5.16) 0.73 1 (0.31) 1.32 (0.18–9.52) 0.78 1 (0.25) 1.07 (0.15–7.66) 0.95 
FANCM 2 (0.08) 1 (0.14) — — 1 (0.31) — — 0 (0.00) — — 
MLH1 0 (0.00) 1 (0.14) — — 0 (0.00) — — 1 (0.25) — — 
MRE11A 1 (0.04) 0 (0.00) — — 0 (0.00) — — 0 (0.00) — — 
MSH2 0 (0.00) 0 (0.00) — — 0 (0.00) — — 0 (0.00) — — 
MSH6 0 (0.00) 1 (0.14) — — 1 (0.31) — — 0 (0.00) — — 
NBN 3 (0.12) 0 (0.00) — — 0 (0.00) — — 0 (0.00) — — 
RAD50 3 (0.12) 1 (0.14) — — 1 (0.31) — — 0 (0.00) — — 
RECQL 3 (0.12) 1 (0.14) — — 0 (0.00) — — 1 (0.25) — — 
RINT1 3 (0.12) 1 (0.14) — — 1 (0.31) — — 0 (0.00) — — 
SLX4 2 (0.08) 0 (0.00) — — 0 (0.00) — — 0 (0.00) — — 
XRCC2 5 (0.20) 0 (0.00) — — 0 (0.00) — — 0 (0.00) — — 
Other cancer predisposition gene 
APC 0 (0.00) 0 (0.00) — — 0 (0.00) — — 0 (0.00) — — 
EPCAM 0 (0.00) 1 (0.14) — — 0 (0.00) — — 1 (0.25) — — 
ERCC2 2 (0.08) 4 (0.56) 3.51 (1.29–9.54) 0.014 3 (0.93) 5.09 (1.58–16.41) 0.007 1 (0.25) 2.00 (0.27–14.59) 0.49 
KRAS 0 (0.00) 0 (0.00) — — 0 (0.00) — — 0 (0.00) — — 
MEN1 0 (0.00) 1 (0.14) — — 0 (0.00) — — 1 (0.25) — — 
MUTYH 5 (0.20) 1 (0.14) 0.51 (0.07–3.65) 0.50 0 (0.00) — — 1 (0.25) 0.88 (0.12 - 6.30) 0.90 
PMS2 3 (0.12) 1 (0.14) — — 0 (0.00) — — 1 (0.25) — — 
PPM1D 1 (0.04) 2 (0.28) — — 1 (0.31) — — 1 (0.25) — — 
PRSS1 2 (0.08) 1 (0.14) — — 0 (0.00) — — 1 (0.25) — — 
PBCSPCSPBCbSPNBC
GeneaCarriers (%)Carriers (%)HR (95% CI)cPCarriers (%)HR (95% CI)cPCarriers (%)HR (95% CI)cP
Established breast cancer predisposition gene 
ATM 10 (0.40) 4 (0.56) 1.48 (0.55–3.98) 0.44 1 (0.31) 0.85 (0.12–6.29) 0.88 3 (0.76) 1.91 (0.61–6.00) 0.27 
BARD1 2 (0.08) 0 (0.00) — — 0 (0.00) — — 0 (0.00) — — 
BRCA1 6 (0.24) 8 (1.11) 2.28 (1.11–4.65) 0.024 3 (0.93) 2.64 (0.82–8.43) 0.10 5 (1.26) 2.98 (1.21–7.36) 0.018 
BRCA2 15 (0.60) 5 (0.70) 1.18 (0.49–2.88) 0.71 5 (1.55) 2.24 (0.91–5.55) 0.08 0 (0.00) — — 
CDH1 0 (0.00) 1 (0.14) — — 0 (0.00) — — 1 (0.25) — — 
CHEK2 3 (0.12) 3 (0.42) 2.03 (0.64–6.46) 0.23 1 (0.31) 1.81 (0.24–13.55) 0.56 2 (0.51) 2.32 (0.56–9.60) 0.25 
NF1 2 (0.08) 0 (0.00) — — 0 (0.00) — — 0 (0.00) — — 
PALB2 12 (0.48) 4 (0.56) 0.89 (0.33–2.40) 0.82 2 (0.62) 0.96 (0.24–3.94) 0.96 2 (0.51) 0.95 (0.23–3.87) 0.95 
PTEN 0 (0.00) 0 (0.00) — — 0 (0.00) — — 0 (0.00) — — 
RAD51C 5 (0.20) 0 (0.00) — — 0 (0.00) — — 0 (0.00) — — 
RAD51D 2 (0.08) 0 (0.00) — — 0 (0.00) — — 0 (0.00) — — 
TP53 0 (0.00) 1 (0.14) — — 0 (0.00) — — 1 (0.25) — — 
Candidate breast cancer predisposition gene 
BLM 4 (0.16) 2 (0.28) 2.50 (0.61–10.18) 0.20 1 (0.31) 4.12 (0.55–30.92) 0.17 1 (0.25) 2.74 (0.38–19.79) 0.32 
BRIP1 4 (0.16) 1 (0.14) 1.02 (0.14–7.34) 0.98 1 (0.31) 1.99 (0.27–14.40) 0.50 0 (0.00) — — 
CDKN2A 1 (0.04) 0 (0.00) — — 0 (0.00) — — 0 (0.00) — — 
ERCC3 4 (0.16) 1 (0.14) 1.24 (0.17–8.90) 0.83 0 (0.00) — — 1 (0.25) 2.21 (0.31–16.07) 0.43 
FANCC 5 (0.20) 2 (0.28) 1.28 (0.32–5.16) 0.73 1 (0.31) 1.32 (0.18–9.52) 0.78 1 (0.25) 1.07 (0.15–7.66) 0.95 
FANCM 2 (0.08) 1 (0.14) — — 1 (0.31) — — 0 (0.00) — — 
MLH1 0 (0.00) 1 (0.14) — — 0 (0.00) — — 1 (0.25) — — 
MRE11A 1 (0.04) 0 (0.00) — — 0 (0.00) — — 0 (0.00) — — 
MSH2 0 (0.00) 0 (0.00) — — 0 (0.00) — — 0 (0.00) — — 
MSH6 0 (0.00) 1 (0.14) — — 1 (0.31) — — 0 (0.00) — — 
NBN 3 (0.12) 0 (0.00) — — 0 (0.00) — — 0 (0.00) — — 
RAD50 3 (0.12) 1 (0.14) — — 1 (0.31) — — 0 (0.00) — — 
RECQL 3 (0.12) 1 (0.14) — — 0 (0.00) — — 1 (0.25) — — 
RINT1 3 (0.12) 1 (0.14) — — 1 (0.31) — — 0 (0.00) — — 
SLX4 2 (0.08) 0 (0.00) — — 0 (0.00) — — 0 (0.00) — — 
XRCC2 5 (0.20) 0 (0.00) — — 0 (0.00) — — 0 (0.00) — — 
Other cancer predisposition gene 
APC 0 (0.00) 0 (0.00) — — 0 (0.00) — — 0 (0.00) — — 
EPCAM 0 (0.00) 1 (0.14) — — 0 (0.00) — — 1 (0.25) — — 
ERCC2 2 (0.08) 4 (0.56) 3.51 (1.29–9.54) 0.014 3 (0.93) 5.09 (1.58–16.41) 0.007 1 (0.25) 2.00 (0.27–14.59) 0.49 
KRAS 0 (0.00) 0 (0.00) — — 0 (0.00) — — 0 (0.00) — — 
MEN1 0 (0.00) 1 (0.14) — — 0 (0.00) — — 1 (0.25) — — 
MUTYH 5 (0.20) 1 (0.14) 0.51 (0.07–3.65) 0.50 0 (0.00) — — 1 (0.25) 0.88 (0.12 - 6.30) 0.90 
PMS2 3 (0.12) 1 (0.14) — — 0 (0.00) — — 1 (0.25) — — 
PPM1D 1 (0.04) 2 (0.28) — — 1 (0.31) — — 1 (0.25) — — 
PRSS1 2 (0.08) 1 (0.14) — — 0 (0.00) — — 1 (0.25) — — 

aThe association analysis was limited to 11 genes with at least five carriers in all women and at least one carrier in women with and without the event of interest.

bA total of 33 women with bilateral mastectomy were further excluded in the analysis of SPBC.

cHRs and 95% CIs were calculated from Cox proportional hazard models adjusting for age at initial breast cancer diagnosis, race/ethnicity, first-degree family history of breast cancer, SEER summary stage, ER status, and PR stats of first breast tumor. Further adjustment on radiotherapy was included in the analysis of SPBC.

Associations of PVs in gene sets with SPC

In the gene set analysis, we observed a higher risk of SPC among women carrying a PV in any of the 37 genes (HR, 1.44; 95% CI, 1.07–1.94; P = 0.02) as well as in carriers of the 9 other cancer predisposition genes (HR, 2.08; 95% CI, 1.14–3.79; P = 0.02; Table 3). Women carrying a PV in any DRG had a significantly higher risk of SPC (HR, 1.40; 95% CI, 1.02–1.92; P = 0.04), especially among carriers of genes of the MMR pathway (HR, 4.03; 95% CI, 1.23–13.20; P = 0.02) and NER pathway (HR, 2.57; 95% CI, 1.06–6.27; P = 0.04). Carriers of other DNA repair related genes (i.e., CHEK2 and TP53) were suggested to have a 2.73-fold risk of SPC (95% CI, 0.99–7.50; P = 0.05).

Table 3.

Gene set associations with SPC, SPBC, or SPNBC.

PBCSPC
Gene set (no. genes)No. carriers (%)No. carriers (%)HR (95% CI)P
SPC 
All cancer genes (37) 110 (4.39) 49 (6.82) 1.44 (1.07–1.94) 0.01 
BC susceptibility genes (28)a 97 (3.87) 38 (5.29) 1.31 (0.94–1.82) 0.11 
 Established breast cancer genes (12) 57 (2.28) 26 (3.62) 1.40 (0.93–2.09) 0.10 
 Candidate breast cancer genes (16) 40 (1.60) 12 (1.67) 1.14 (0.64–2.02) 0.66 
Other cancer genes (9)a 13 (0.52) 11 (1.53) 2.08 (1.14–3.79) 0.02 
DNA repair genes (25)b 98 (3.91) 42 (5.84) 1.40 (1.02–1.92) 0.04 
 HR pathway (14) 74 (2.96) 26 (3.62) 1.18 (0.79–1.76) 0.42 
 FA pathway (8) 51 (2.04) 21 (2.92) 1.26 (0.81–1.97) 0.30 
 MMR pathway (4) 3 (0.12) 3 (0.42) 4.03 (1.23–13.20) 0.02 
 NER pathway (2) 6 (0.24) 5 (0.70) 2.57 (1.06–6.27) 0.04 
 BER pathway (1) 5 (0.20) 1 (0.14) 0.51 (0.07–3.65) 0.50 
 Other related genes (2) 3 (0.12) 4 (0.56) 2.73 (0.99–7.50) 0.05 
SPBCc 
All cancer genes (37) 108 (4.37) 23 (7.12) 1.56 (1.01–2.40) 0.045 
BC susceptibility genes (28)a 95 (3.84) 19 (5.88) 1.47 (0.92–2.37) 0.11 
 Established breast cancer genes (12) 56 (2.27) 12 (3.72) 1.46 (0.81–2.65) 0.21 
 Candidate breast cancer genes (16) 39 (1.58) 7 (2.17) 1.46 (0.69–3.11) 0.33 
Other cancer genes (9)a 13 (0.53) 4 (1.24) 1.95 (0.72–5.28) 0.19 
DNA repair genes (25)b 96 (3.89) 21 (6.50) 1.60 (1.02–2.51) 0.04 
 HR pathway (14) 72 (2.91) 14 (4.33) 1.45 (0.84–2.50) 0.19 
 FA pathway (8) 50 (2.02) 13 (4.02) 1.76 (1.00–3.12) 0.05 
 MMR pathway (4) 3 (0.12) 1 (0.31) 3.43 (0.43–27.20) 0.24 
 NER pathway (2) 6 (0.24) 3 (0.93) 3.04 (0.96–9.69) 0.06 
 BER pathway (1) 5 (0.20) 0 (0.00) – – 
 Other related genes (2) 3 (0.12) 1 (0.31) 1.81 (0.24–13.55) 0.56 
SPNBC 
All cancer genes (37) 110 (4.39) 26 (6.57) 1.44 (0.96–2.15) 0.08 
BC susceptibility genes (28)a 97 (3.87) 19 (4.80) 1.23 (0.77 – 1.97) 0.38 
 Established breast cancer genes (12) 57 (2.28) 14 (3.54) 1.44 (0.84–2.49) 0.19 
 Candidate breast cancer genes (16) 40 (1.60) 5 (1.26) 0.88 (0.36–2.13) 0.77 
Other cancer genes (9)a 13 (0.52) 7 (1.77) 2.43 (1.14–5.17) 0.02 
DNA repair genes (25)b 98 (3.91) 21 (5.30) 1.32 (0.85–2.07) 0.22 
 HR pathway (14) 74 (2.96) 12 (3.03) 1.06 (0.59–1.91) 0.83 
 FA pathway (8) 51 (2.04) 8 (2.02) 0.93 (0.46–1.89) 0.84 
 MMR pathway (4) 3 (0.12) 2 (0.51) 4.99 (1.17–21.38) 0.03 
 NER pathway (2) 6 (0.24) 2 (0.51) 2.11 (0.52–8.58) 0.30 
 BER pathway (1) 5 (0.20) 1 (0.25) 0.88 (0.12–6.30) 0.90 
 Other related genes (2) 3 (0.12) 3 (0.76) 3.55 (1.10–11.46) 0.03 
PBCSPC
Gene set (no. genes)No. carriers (%)No. carriers (%)HR (95% CI)P
SPC 
All cancer genes (37) 110 (4.39) 49 (6.82) 1.44 (1.07–1.94) 0.01 
BC susceptibility genes (28)a 97 (3.87) 38 (5.29) 1.31 (0.94–1.82) 0.11 
 Established breast cancer genes (12) 57 (2.28) 26 (3.62) 1.40 (0.93–2.09) 0.10 
 Candidate breast cancer genes (16) 40 (1.60) 12 (1.67) 1.14 (0.64–2.02) 0.66 
Other cancer genes (9)a 13 (0.52) 11 (1.53) 2.08 (1.14–3.79) 0.02 
DNA repair genes (25)b 98 (3.91) 42 (5.84) 1.40 (1.02–1.92) 0.04 
 HR pathway (14) 74 (2.96) 26 (3.62) 1.18 (0.79–1.76) 0.42 
 FA pathway (8) 51 (2.04) 21 (2.92) 1.26 (0.81–1.97) 0.30 
 MMR pathway (4) 3 (0.12) 3 (0.42) 4.03 (1.23–13.20) 0.02 
 NER pathway (2) 6 (0.24) 5 (0.70) 2.57 (1.06–6.27) 0.04 
 BER pathway (1) 5 (0.20) 1 (0.14) 0.51 (0.07–3.65) 0.50 
 Other related genes (2) 3 (0.12) 4 (0.56) 2.73 (0.99–7.50) 0.05 
SPBCc 
All cancer genes (37) 108 (4.37) 23 (7.12) 1.56 (1.01–2.40) 0.045 
BC susceptibility genes (28)a 95 (3.84) 19 (5.88) 1.47 (0.92–2.37) 0.11 
 Established breast cancer genes (12) 56 (2.27) 12 (3.72) 1.46 (0.81–2.65) 0.21 
 Candidate breast cancer genes (16) 39 (1.58) 7 (2.17) 1.46 (0.69–3.11) 0.33 
Other cancer genes (9)a 13 (0.53) 4 (1.24) 1.95 (0.72–5.28) 0.19 
DNA repair genes (25)b 96 (3.89) 21 (6.50) 1.60 (1.02–2.51) 0.04 
 HR pathway (14) 72 (2.91) 14 (4.33) 1.45 (0.84–2.50) 0.19 
 FA pathway (8) 50 (2.02) 13 (4.02) 1.76 (1.00–3.12) 0.05 
 MMR pathway (4) 3 (0.12) 1 (0.31) 3.43 (0.43–27.20) 0.24 
 NER pathway (2) 6 (0.24) 3 (0.93) 3.04 (0.96–9.69) 0.06 
 BER pathway (1) 5 (0.20) 0 (0.00) – – 
 Other related genes (2) 3 (0.12) 1 (0.31) 1.81 (0.24–13.55) 0.56 
SPNBC 
All cancer genes (37) 110 (4.39) 26 (6.57) 1.44 (0.96–2.15) 0.08 
BC susceptibility genes (28)a 97 (3.87) 19 (4.80) 1.23 (0.77 – 1.97) 0.38 
 Established breast cancer genes (12) 57 (2.28) 14 (3.54) 1.44 (0.84–2.49) 0.19 
 Candidate breast cancer genes (16) 40 (1.60) 5 (1.26) 0.88 (0.36–2.13) 0.77 
Other cancer genes (9)a 13 (0.52) 7 (1.77) 2.43 (1.14–5.17) 0.02 
DNA repair genes (25)b 98 (3.91) 21 (5.30) 1.32 (0.85–2.07) 0.22 
 HR pathway (14) 74 (2.96) 12 (3.03) 1.06 (0.59–1.91) 0.83 
 FA pathway (8) 51 (2.04) 8 (2.02) 0.93 (0.46–1.89) 0.84 
 MMR pathway (4) 3 (0.12) 2 (0.51) 4.99 (1.17–21.38) 0.03 
 NER pathway (2) 6 (0.24) 2 (0.51) 2.11 (0.52–8.58) 0.30 
 BER pathway (1) 5 (0.20) 1 (0.25) 0.88 (0.12–6.30) 0.90 
 Other related genes (2) 3 (0.12) 3 (0.76) 3.55 (1.10–11.46) 0.03 

aGenes were grouped by their prior evidence on breast cancer susceptibility. Established breast cancer genes included ATM, BARD1, BRCA1, BRCA2, CDH1, CHEK2, NF1, PALB2, PTEN, RAD51C, RAD51D, and TP51. Candidate breast cancer genes included BLM, BRIP1, CDKN2A, ERCC3, FANCC, FANCM, MLH1, MRE11A, MSH2, MSH6, NCN, RAD50, RECQL, RINT1, SLX4, and XRCC2. Other cancer predisposition genes included APC, EPCAM, ERCC2, KRAS, MEN1, MUTYH, RMS2, PPM1D, and PRSS1.

bDRGs were grouped by DNA repair pathways. HR pathway included ATM, BLM, BARD1, BRCA1, BRCA2, BRIP1, MRE11A, PALB2, RAD50, RAD51C, RAD51D, RECQL, SLX4, and XRCC2. Fanconi Anemia (FA) pathway included BRCA1, BRCA2, BRIP1, FANCC, FANCM, PALB2, RAD51C, and SLX4. Mismatch repair (MMR) pathway included MLH1, MSH2, MSH6, and PMS2. NER pathway included ERCC3 and ERCC2. BER pathway included MUTYH. Other related genes included CHEK2 and TP53. Note that some genes were included in more than one pathway.

cA total of 33 women with bilateral mastectomy were further excluded in the analysis of SPBC.

Carrying a PV in any of the 37 cancer predisposition genes was significantly associated with a higher risk of SPBC (HR, 1.56; 95% CI, 1.01–2.40; P = 0.045). The risk of SPBC was not statistically different between carriers and noncarriers of PVs in established or candidate breast cancer genes, or PVs in other cancer predisposition genes (Table 3). Women carrying a PV in any DRG had a 1.60-fold risk of SPBC (95% CI, 1.02–2.51; P = 0.04). A positive association with SPBC was also suggested for carriers of genes in the FA pathway (HR, 1.76; 95% CI, 1.00–3.12; P = 0.05) and NER pathway (HR, 3.04; 95% CI, 0.96–9.69; P = 0.06). These associations appeared to be largely driven by BRCA1, BRCA2, or ERCC2 gene (Supplementary Table S4).

Among the 396 women with SPNBC, the top commonly diagnosed cancer sites included uterine corpus (16.4%), lung (14.9%), colon/rectum (14.8%), leukemia (8.1%), pancreas (6.8%), and melanoma (6.8%). There was a suggestive association with SPNBC for carrying a PV in any of the 37 genes (HR, 1.44; 95% CI, 0.96–2.15; P = 0.08). The risk of SPNBC was similar between noncarriers and carriers of the established or candidate breast cancer genes (Table 3). Carriers of PVs in other cancer predisposition genes had a significantly elevated risk of SPNBC (HR, 2.43; 95% CI, 1.14–5.17; P = 0.02) than noncarriers. Although the association of PVs in all DRGs was not statistically significant, carrying a PV in genes involved in the MMR pathway (HR, 4.99; 95% CI, 1.17–21.38; P = 0.03) and in other related genes (HR, 3.55; 95% CI, 1.10–11.46; P = 0.03) was associated with a higher risk of SPNBC.

In this multiethnic cohort of patients with breast cancer, we identified three genes, BRCA1, BRCA2, and ERCC2, in which germline PVs were associated with an increased risk of SPC in breast cancer survivors. In the analysis by SPC subtypes, BRCA2 and ERCC2 were the most associated genes with risk of SPBC whereas for SPNBC, the strongest association was observed with BRCA1. Our gene set analyses also found that the MMR genes and other DNA repair related genes were strongly associated with risk of SPNBC. In general, women carrying a PV in any of these 37 cancer predisposition genes were 44% to 56% more likely to be diagnosed with SPC.

Both BRCA1 and BRCA2 are known to contribute to the risk of subsequent breast malignancies among breast cancer survivors. In a nested case–control study of 705 patients with contralateral breast cancer and 1,398 patients with unilateral breast cancer, all of whom had their first breast cancer diagnosed before age 55, the risk of contralateral breast cancer for BRCA1 and BRCA2 mutation carriers was 4.5- and 3.4-fold, respectively, in comparison to non-BRCA carriers (37). In a large multiethnic targeted sequencing study comprised of 75,550 women with PBC and 7,728 with SPBC, PVs in BRCA1 and BRCA2 were found to be significantly associated with SPBC in White, African American, and Hispanic women with the ORs estimated to range from 1.3 to 2.2 (15). Although in our study the association of BRCA1 and BRCA2 with SPBC was suggestive due to the small number of patients with SPBC, they were similar in magnitude to these published studies.

We found strong and statistically significant associations of germline PVs in the ERCC2 gene with SPC and SPBC. Like BRCA1 and BRCA2 genes, ERCC2 is also involved in DNA damage repair, though via NER processes rather than HR. Deleterious autosomal recessive variants in the ERCC2 gene cause a cancer-prone syndrome, xeroderma pigmentosum (XP) complementation group D (XPD). Most individuals with XP develop multiple skin cancers during their lifetime (38). Although the association between heterozygous germline PVs in XP-related genes and cancer risk is less clear, genetic studies suggest that the two common ERCC2 polymorphisms (Asp312Asn and Lys751Gln) might contribute to the susceptibility of bladder cancer (39), lung cancer (40, 41), and gastric cancer (41, 42), and studies on the association with breast cancer risk have reported conflicting findings (41, 43). These ERCC2 polymorphisms have been previously implicated in the development of SPC. The ERCC2 Asp312Asn variant was found to be associated with the risk of developing second primary esophageal carcinoma in long-term cancer survivors who received radio-chemotherapy for a prior lymphoma or breast cancer (44). In a cohort study of 481 patients with nonmelanoma skin cancer, carriers of the ERCC2 Lys751Gln variant were at an increased risk of SPC, with breast cancer being the third most common SPC observed (45). Because the ERCC2 gene was generally not considered a breast cancer predisposition gene, rare germline PVs in ERCC2 had not been previously examined for association with SPC in breast cancer survivors. Consistent with results from the overall CARRIERS study, in an analysis of breast cancer cases and cancer-free controls from the MEC, germline ERCC2 PVs were not associated with primary breast cancer risk (Supplementary Table S5). Results from this cohort analysis on SPC phenotypes provide initial evidence that women carrying heterozygous germline PVs in ERCC2 genes have a higher risk of developing a second malignancy, especially an SPBC, after their initial breast cancer diagnosis.

Inherited alterations in genes involved in DNA damage recognition and repair have been linked to a variety of cancer predisposition syndromes, where affected individuals are at risk of developing multiple primary tumors over time (46). Previously, PVs in DRGs were found to be associated with subsequent neoplasms in 4,402 survivors of childhood cancer (35). Specifically, PVs in HR or FA genes were significantly associated with an increased risk of subsequent female breast cancer and sarcoma, whereas PVs in NER genes were positively associated with subsequent thyroid cancer. These observed associations appeared to be stronger among patients who received a higher cumulative dosage of chemotherapy or body region-specific radiotherapy. Consistent with these findings, our results further support the contribution of DRGs in the development of multiple primary cancers.

Our study has several strengths and limitations. The MEC is a population-based cohort. The breast cancer cases included in this study represent approximately 70% of all incident invasive breast cancer cases in the MEC identified through December 31, 2017, with similar demographic and clinical characteristics, suggesting that our findings may be broadly generalizable (47). Our study is one of the first multiethnic studies of SPC phenotypes in breast cancer survivors. However, the relatively small numbers in each race/ethnicity group limited our statistical power to perform ethnic-specific analysis. Because of the high number of missing values of hormone receptor status, we were unable to adjust for specific tumor subtypes such as triple-negative or HER2-positive tumors. Finally, although we included first-course chemo/radiation/hormonal treatment from the SEER registry in our analysis, the lack of detailed information on subsequent breast cancer treatments prevented us from fully assessing treatment effects and their potential interactions with PVs on SPC phenotypes.

Results from our analysis provide further evidence that germline PVs in BRCA1 and BRCA2 genes contribute to the development of SPC in breast cancer survivors, and add ERCC2 to the list of non-BRCA1/2 genes associated with SPC. These results also suggest that compromised DNA repair mechanisms could be a predisposition factor for the risk of second malignancies in breast cancer survivors, or more broadly in populations of any cancer survivors. Our findings provide further support for closer monitoring of SPC for women carrying PVs in these genes. The suggestive associations that we observed with SPC phenotypes for other genes (e.g., CHEK2, BLM, PRSS1) and DNA repair pathways (e.g., MMR pathway) require investigation in larger studies.

A.J. de Smith reports grants from Concern Foundation Marni Levine Memorial Breast Cancer Research Fund during the conduct of the study. No disclosures were reported by the other authors.

F. Chen: Formal analysis, investigation, visualization, methodology, writing–original draft, writing–review and editing. S.L. Park: Conceptualization, methodology, writing–review and editing. L.R. Wilkens: Resources, funding acquisition, writing–review and editing. P. Wan: Resources, data curation. S.N. Hart: Resources, data curation. C. Hu: Resources, data curation. S. Yadav: Resources, data curation. F.J. Couch: Resources, data curation, funding acquisition, writing–review and editing. D.V. Conti: Methodology, writing–review and editing. A.J. de Smith: Conceptualization, supervision, funding acquisition, writing–review and editing. C.A. Haiman: Resources, supervision, funding acquisition, writing–review and editing.

This work was supported by the NCI at the NIH (grant nos. T32CA229110 to F. Chen, U01CA164973 to C.A. Haiman and L.R. Wilkens, and R01CA192393 and R35CA253187 to F.J. Couch), the Concern Foundation Marni Levine Memorial Breast Cancer Research Fund (to A.J. de Smith), and the Breast Cancer Research Foundation (to F.J. Couch).

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Note: Supplementary data for this article are available at Cancer Research Online (http://cancerres.aacrjournals.org/).

1.
Miller
KD
,
Nogueira
L
,
Mariotto
AB
,
Rowland
JH
,
Yabroff
KR
,
Alfano
CM
, et al
.
Cancer treatment and survivorship statistics, 2019
.
CA Cancer J Clin
2019
;
69
:
363
85
.
2.
Watt
GP
,
John
EM
,
Bandera
EV
,
Malone
KE
,
Lynch
CF
,
Palmer
JR
, et al
.
Race, ethnicity and risk of second primary contralateral breast cancer in the United States
.
Int J Cancer
2021
;
148
:
2748
58
.
3.
Ramin
C
,
Withrow
DR
,
Davis Lynn
BC
,
Gierach
GL
,
de González
AB
.
Risk of contralateral breast cancer according to first breast cancer characteristics among women in the USA, 1992–2016
.
Breast Cancer Res
2021
;
23
:
24
.
4.
Kurian
AW
,
Canchola
AJ
,
Ma
CS
,
Clarke
CA
,
Gomez
SL
.
Magnitude of reduction in risk of second contralateral breast cancer with bilateral mastectomy in patients with breast cancer: Data from California, 1998 through 2015
.
Cancer
2020
;
126
:
958
70
.
5.
Xiong
Z
,
Yang
L
,
Deng
G
,
Huang
X
,
Li
X
,
Xie
X
, et al
.
Patterns of occurrence and outcomes of contralateral breast cancer: analysis of SEER data
.
J Clin Med
2018
;
7
:
E133
.
6.
Nsouli-Maktabi
HH
,
Henson
DE
,
Younes
N
,
Young
HA
,
Cleary
SD
.
Second primary breast, endometrial, and ovarian cancers in Black and White breast cancer survivors over a 35-year time span: effect of age
.
Breast Cancer Res Treat
2011
;
129
:
963
9
.
7.
Langballe
R
,
Mellemkjær
L
,
Malone
KE
,
Lynch
CF
,
John
EM
,
Knight
JA
, et al
.
Systemic therapy for breast cancer and risk of subsequent contralateral breast cancer in the WECARE Study
.
Breast Cancer Res BCR
2016
;
18
:
65
.
8.
Bessonova
L
,
Taylor
TH
,
Mehta
RS
,
Zell
JA
,
Anton-Culver
H
.
Risk of a second breast cancer associated with hormone-receptor and HER2/neu status of the first breast cancer
.
Cancer Epidemiol Biomark Prev
2011
;
20
:
389
96
.
9.
Kurian
AW
,
McClure
LA
,
John
EM
,
Horn-Ross
PL
,
Ford
JM
,
Clarke
CA
.
Second primary breast cancer occurrence according to hormone receptor status
.
J Natl Cancer Inst
2009
;
101
:
1058
65
.
10.
Reiner
AS
,
Lynch
CF
,
Sisti
JS
,
John
EM
,
Brooks
JD
,
Bernstein
L
, et al
.
Hormone receptor status of a first primary breast cancer predicts contralateral breast cancer risk in the WECARE study population
.
Breast Cancer Res
2017
;
19
:
83
.
11.
Clarke
M
,
Collins
R
,
Darby
S
,
Davies
C
,
Elphinstone
P
,
Evans
V
, et al
.
Effects of radiotherapy and of differences in the extent of surgery for early breast cancer on local recurrence and 15-year survival: an overview of the randomised trials
.
Lancet Lond Engl
2005
;
366
:
2087
106
.
12.
Bertelsen
L
,
Bernstein
L
,
Olsen
JH
,
Mellemkjær
L
,
Haile
RW
,
Lynch
CF
, et al
.
Effect of systemic adjuvant treatment on risk for contralateral breast cancer in the women's environment, cancer and radiation epidemiology study
.
J Natl Cancer Inst
2008
;
100
:
32
40
.
13.
Molina-Montes
E
,
Pérez-Nevot
B
,
Pollán
M
,
Sánchez-Cantalejo
E
,
Espín
J
,
Sánchez
M-J
.
Cumulative risk of second primary contralateral breast cancer in BRCA1/BRCA2 mutation carriers with a first breast cancer: a systematic review and meta-analysis
.
The Breast
2014
;
23
:
721
42
.
14.
Menes
TS
,
Terry
MB
,
Goldgar
D
,
Andrulis
IL
,
Knight
JA
,
John
EM
, et al
.
Second primary breast cancer in BRCA1 and BRCA2 mutation carriers: 10-year cumulative incidence in the Breast Cancer Family Registry
.
Breast Cancer Res Treat
2015
;
151
:
653
60
.
15.
Yao
K
,
KA
CJ
,
Li
S
,
LaDuca
H
,
Hulick
P
,
Gutierrez
S
, et al
.
Prevalence of germline pathogenic and likely pathogenic variants in patients with second breast cancers
.
JNCI Cancer Spectr
2020
;
4
:
pkaa094
.
16.
Supramaniam
R
.
New malignancies among cancer survivors: SEER cancer registries, 1973–2000
.
J Epidemiol Community Health
2008
;
62
:
375
6
.
17.
Li
S
,
Yang
J
,
Shen
Y
,
Zhao
X
,
Zhang
L
,
Wang
B
, et al
.
Clinicopathological features, survival and risk in breast cancer survivors with thyroid cancer: an analysis of the SEER database
.
BMC Public Health
2019
;
19
:
1592
.
18.
Schonfeld
SJ
,
Morton
LM
,
Berrington de González
A
,
Curtis
RE
,
Kitahara
CM
.
Risk of second primary papillary thyroid cancer among adult cancer survivors in the United States, 2000–2015
.
Cancer Epidemiol
2020
;
64
:
101664
.
19.
Corso
G
,
Veronesi
P
,
Santomauro
GI
,
Maisonneuve
P
,
Morigi
C
,
Peruzzotti
G
, et al
.
Multiple primary non-breast tumors in breast cancer survivors
.
J Cancer Res Clin Oncol
2018
;
144
:
979
86
.
20.
Li
Z
,
Wu
Q
,
Song
J
,
Zhang
Y
,
Zhu
S
,
Sun
S
.
Risk of second primary female genital malignancies in women with breast cancer: a SEER analysis
.
Horm Cancer
2018
;
9
:
197
204
.
21.
Liu
J
,
Jiang
W
,
Mao
K
,
An
Y
,
Su
F
,
Kim
BYS
, et al
.
Elevated risks of subsequent endometrial cancer development among breast cancer survivors with different hormone receptor status: a SEER analysis
.
Breast Cancer Res Treat
2015
;
150
:
439
45
.
22.
Qian
X
,
Jia
H
,
Zhang
Y
,
Ma
B
,
Qin
G
,
Wu
Z
.
Risk factors and prediction of second primary cancer in primary female non-metastatic breast cancer survivors
.
Aging
2020
;
12
:
19628
40
.
23.
Hung
M-H
,
Liu
C-J
,
Teng
C-J
,
Hu
Y-W
,
Yeh
C-M
,
Chen
S-C
, et al
.
Risk of second non-breast primary cancer in male and female breast cancer patients: a population-based cohort study
.
PLoS One
2016
;
11
:
e0148597
.
24.
Schaapveld
M
,
Visser
O
,
Louwman
MJ
,
de Vries
EGE
,
Willemse
PHB
,
Otter
R
, et al
.
Risk of new primary nonbreast cancers after breast cancer treatment: a Dutch Population-based study
.
J Clin Oncol
2008
;
26
:
1239
46
.
25.
Grantzau
T
,
Overgaard
J
.
Risk of second non-breast cancer after radiotherapy for breast cancer: a systematic review and meta-analysis of 762,468 patients
.
Radiother Oncol
2015
;
114
:
56
65
.
26.
Grantzau
T
,
Overgaard
J
.
Risk of second non-breast cancer among patients treated with and without postoperative radiotherapy for primary breast cancer: a systematic review and meta-analysis of population-based studies including 522,739 patients
.
Radiother Oncol
2016
;
121
:
402
13
.
27.
Morton
LM
,
Gilbert
ES
,
Hall
P
,
Andersson
M
,
Joensuu
H
,
Vaalavirta
L
, et al
.
Risk of treatment-related esophageal cancer among breast cancer survivors
.
Ann Oncol
2012
;
23
:
3081
91
.
28.
Kolonel
LN
,
Henderson
BE
,
Hankin
JH
,
Nomura
AM
,
Wilkens
LR
,
Pike
MC
, et al
.
A multiethnic cohort in Hawaii and Los Angeles: baseline characteristics
.
Am J Epidemiol
2000
;
151
:
346
57
.
29.
Hu
C
,
Hart
SN
,
Gnanaolivu
R
,
Huang
H
,
Lee
KY
,
Na
J
, et al
.
A population-based study of genes previously implicated in breast cancer
.
N Engl J Med
2021
;
384
:
440
51
.
30.
Domchek
SM
,
Yao
S
,
Chen
F
,
Hu
C
,
Hart
SN
,
Goldgar
DE
, et al
.
Comparison of the prevalence of pathogenic variants in cancer susceptibility genes in black women and non-hispanic white women with breast cancer in the United States
.
JAMA Oncol
2021
;
7
:
1045
50
.
31.
Palmer
JR
,
Polley
EC
,
Hu
C
,
John
EM
,
Haiman
C
,
Hart
SN
, et al
.
Contribution of germline predisposition gene mutations to breast cancer risk in African American Women
.
J Natl Cancer Inst
2020
;
112
:
1213
21
.
32.
Johnson
CH
,
Peace
S
,
Adamo
P
,
Fritz
A
,
Percy-Laurry
A
,
Edwards
BK
.
The 2007 multiple primary and histology coding rules
.
National Cancer of Institute The Surveillance and Epidemiology, and End Results Program
;
2007
.
33.
Bi
W
,
Fritsche
LG
,
Mukherjee
B
,
Kim
S
,
Lee
S
.
A Fast and accurate method for genome-wide time-to-event data analysis and its application to UK Biobank
.
Am J Hum Genet
2020
;
107
:
222
33
.
34.
R: A language and environment for statistical computing. R Found Stat Comput Vienna Aust [Internet]
.
2017
.
Available from
: http://www.R-project.org/.
35.
Qin
N
,
Wang
Z
,
Liu
Q
,
Song
N
,
Wilson
CL
,
Ehrhardt
MJ
, et al
.
Pathogenic Germline mutations in DNA repair genes in combination with cancer treatment exposures and risk of subsequent neoplasms among long-term survivors of childhood cancer
.
J Clin Oncol
2020
;
38
:
2728
40
.
36.
Han
J
,
Haiman
C
,
Niu
T
,
Guo
Q
,
Cox
DG
,
Willett
WC
, et al
.
Genetic variation in DNA repair pathway genes and premenopausal breast cancer risk
.
Breast Cancer Res Treat
2009
;
115
:
613
22
.
37.
Malone
KE
,
Begg
CB
,
Haile
RW
,
Borg
A
,
Concannon
P
,
Tellhed
L
, et al
.
Population-based study of the risk of second primary contralateral breast cancer associated with carrying a mutation in BRCA1 or BRCA2
.
J Clin Oncol
2010
;
28
:
2404
10
.
38.
Cleaver
JE
,
Lam
ET
,
Revet
I
.
Disorders of nucleotide excision repair: the genetic and molecular basis of heterogeneity
.
Nat Rev Genet
2009
;
10
:
756
68
.
39.
Li
S-X
,
Dai
Q-S
,
Chen
S-X
,
Zhang
S-D
,
Liao
X-Y
,
Deng
X
, et al
.
Xeroderma pigmentosum complementation group D (XPD) gene polymorphisms contribute to bladder cancer risk: a meta-analysis
.
Tumour Biol
2014
;
35
:
3905
15
.
40.
Lawania
S
,
Singh
N
,
Behera
D
,
Sharma
S
.
Xeroderma pigmentosum complementation group D polymorphism toward lung cancer susceptibility survival and response in patients treated with platinum chemotherapy
.
Future Oncol
2017
;
13
:
2645
65
.
41.
Xiao
F
,
Pu
J
,
Wen
Q
,
Huang
Q
,
Zhang
Q
,
Huang
B
, et al
.
Association between the ERCC2 Asp312Asn polymorphism and risk of cancer
.
Oncotarget
2017
;
8
:
48488
506
.
42.
Ji
H-X
,
Chang
W-S
,
Tsai
C-W
,
Wang
J-Y
,
Huang
N-K
,
Lee
A-S
, et al
.
Contribution of DNA repair xeroderma pigmentosum group D genotype to gastric cancer risk in Taiwan
.
Anticancer Res
2015
;
35
:
4975
81
.
43.
Tian
Y
,
Lin
X
,
Yang
F
,
Zhao
J
,
Yao
K
,
Bian
C
.
Contribution of xeroderma pigmentosum complementation group D gene polymorphisms in breast and ovarian cancer susceptibility
.
Medicine
2020
;
99
:
e20299
.
44.
Boldrin
E
,
Rumiato
E
,
Fassan
M
,
Rugge
M
,
Cagol
M
,
Marino
D
, et al
.
Genetic risk of subsequent esophageal cancer in lymphoma and breast cancer long-term survival patients: a pilot study
.
Pharmacogenomics J
2016
;
16
:
266
71
.
45.
Brewster
AM
,
Alberg
AJ
,
Strickland
PT
,
Hoffman
SC
.,
Helzlsouer K. XPD polymorphism and risk of subsequent cancer in individuals with nonmelanoma skin cancer
.
Cancer Epidemiol Biomark Prev
2004
;
13
:
1271
5
.
46.
Spry
M
,
Scott
T
,
Pierce
H
,
D'Orazio
JA
.
DNA repair pathways and hereditary cancer susceptibility syndromes
.
Front Biosci
2007
;
12
:
4191
207
.
47.
Sarink
D
,
White
KK
,
Loo
LWM
,
Wu
AH
,
Wilkens
LR
,
Le Marchand
L
, et al
.
Racial/ethnic differences in postmenopausal breast cancer risk by hormone receptor status: The multiethnic cohort study
.
Int J Cancer
2022
;
150
:
221
31
.

Supplementary data