Clinical Implications of BRCA Mutations in Advanced Biliary Tract Cancer

Biliary tract cancer (BTC) is a rare malignancy that accounts for 3–5% of all cancers worldwide and is classified as intrahepatic cholangiocarcinoma (IH-CCC), extrahepatic cholangiocarcinoma (EH-CCC), and gallbladder (GB) cancer according to its location [1, 2]. As most patients are diagnosed at an advanced stage, fewer than 30% of patients with BTC are candidates for surgical resection [3]. In addition, BTC has low complete resection rates and high recurrence rates owing to its anatomical location and aggressive potential [4]. The current standard for frontline BTC chemotherapy is a combination of gemcitabine and cisplatin, which results in only 11.7 months of median overall survival (OS) and 8.0 months of median progression-free survival (PFS) [5]. Lamarca et al. [6] conducted the phase 3 randomized trial ABC-06, which demonstrated that 5-fluoropyrimidine/oxaliplatin, as a second-line treatment, resulted in a better OS compared to that with active symptom control (6.2 months vs. 5.3 months, p = 0.031). Recently, first-line durvalumab, an immune checkpoint inhibitor, plus chemotherapy improved OS and PFS versus placebo plus chemotherapy in TOPAZ-1 trial, with limited survival benefit (median OS; 12.8 months, median PFS; 7.2 months) [7]. Therefore, the identification and understanding of novel biomarkers and targets are essential for developing a therapeutic strategy for BTC.

Over the last few years, molecular diversity and actionable mutations in various cancers have been identified, and these studies have become more prevalent with the development of next-generation sequencing (NGS). Nakamura et al. [8] reported that nearly 40% of all BTC cases have potentially targetable genetic alterations. Breast cancer gene (BRCA1/2) mutations are detected in up to 5% of all BTC cases, and BRCA2 mutations occur more frequently in GB cancers [8‒10]. BRCA1/2 mutations result in a deficiency in DNA damage repair (DDR) related to homologous recombination. The accumulation of DNA double-strand breaks leads to genomic instability and increases the risk of malignant transformation [11, 12]. BRCA mutations have been used as biomarkers to predict the sensitivity to DNA-damaging therapies, particularly platinum-based alkylating agents [13, 14], as well as to poly-(ADP-ribose)-polymerase inhibitors (PARPis) in ovarian, breast, prostate, and pancreatic cancers [15‒18]. Despite these recent advances, information on BRCA-mutated BTC remains limited. Herein, we sought to identify the clinical significance of genetic profiles and demonstrate the clinical and therapeutic implications of BRCA mutations in advanced BTC.

We screened patients with advanced BTC who participated in targeted high-throughput sequencing at Keimyung University Dongsan Hospital between December 2018 and December 2021. In total, 25 patients with advanced BTC who received palliative first-line platinum-based chemotherapy were identified. All patients were pathologically diagnosed with BTC based on the World Health Organization classification. BTC includes IH-CCC, EH-CCC, and GB cancer but excludes cancers of the duodenum and ampulla of Vater. Data including patient age, sex, tumor location, histological differentiation, sites and numbers of metastases, disease status, history of palliative chemotherapy, and results of NGS were retrospectively reviewed from medical records. The outcomes were collected as survival data and treatment responses. OS was calculated from the date of first-line platinum-based chemotherapy commencement to the date of death. PFS was defined as the difference between the date of first-line platinum-based chemotherapy commencement and the date of disease progression or death. Response Evaluation Criteria in Solid Tumors (RECIST) version 1.1 was used to evaluate treatment responses. NGS was conducted by the K-MASTER project (DSMC 2018-05-019) for 18 patients. This study was approved by the Institutional Review Board of Keimyung University Dongsan Hospital (DSMC 2022-05-100).

All tissue specimens were confirmed to be adequate for sequencing by the pathologists. Sections were prepared from formalin-fixed paraffin-embedded (FFPE) tumor tissue samples. If the tumor tissue sample was not available, blood samples were prepared for cell-free DNA analysis. Genomic DNA from FFPE tissues was extracted using the QIAamp DNA FFPE Tissue kit (Qiagen, Hilden, Germany) according to the manufacturer’s instructions. Cell-free DNA was isolated from plasma using the QIAamp Circulating Nucleic Acid Kit (Qiagen) according to the manufacturer’s instructions. Targeted sequencing was performed using three NGS platforms including the K-MASTER cancer panel (CancerScan [19]), Oncomine Comprehensive Plus Panel (Thermo Fisher Scientific, Waltham, MA, USA), and Axen Cancer Panel 1 (Macrogen, Seoul, Korea) [20]. The tissue-based NGS panels, K-MASTER cancer panel (12 patient samples, 409 genes) and Oncomine Comprehensive Plus Panel (7 patient samples, 496 genes), were used for 19 patient samples. A liquid biopsy panel, Axen Cancer Panel 1 (88 genes), was used for 6 patient samples. These panels can detect mutations, copy number variants, and the fusion of multiple genes.

The distribution of BRCA mutations and clinical features was assessed using the χ2 test or Fisher’s exact test for categorical variables. The Kaplan-Meier method was used to evaluate OS and PFS from the time of first-line platinum-based chemotherapy, and survival outcomes between the groups were compared using a log-rank test. Univariate and multivariate Cox proportional hazards regression analyses of OS and PFS were used to evaluate the predictive and prognostic value of clinical factors and genetic mutations, including BRCA mutations. The results are presented as hazard ratios (HRs) and corresponding 95% confidence intervals (CIs); p < 0.05 was considered significant. Statistical analyses were performed using IBM SPSS Statistics for Windows (version 25.0; IBM Corp., Armonk, NY, USA).

The characteristics of the 25 patients are shown in Table 1. Sixteen patients (64.0%) were younger than 65 years of age, and 16 patients (64.0%) were male. BTC cases consisted of IH-CCC (9, 36.0%), EH-CCC (5, 20.0%), and GB cancer (11, 44.0%). The histology of the tumors was classified as moderately differentiated (13, 52.0%), poorly differentiated (7, 28.0%), or unclassified (5, 20.0%). Sixteen patients (64.0%) experienced postoperative recurrence. Fourteen patients (56.0%) had liver metastasis at presentation, and 19 (76.0%) had more than one metastatic site. Thirteen patients (52.0%) received fewer than six cycles of first-line chemotherapy, and 17 patients (68.0%) received second-line chemotherapy. Death had occurred in 21 patients at the cutoff time. The median OS and PFS of all patients were 11.9 months (95% CI: 9.2–14.6) and 5.6 months (95% CI: 3.8–7.3), respectively. Comparisons between the BRCA-mutated and -intact groups are shown in Table 2. Of the 25 patients, nine (36.0%) had BRCA mutations. No significant differences in the clinicopathological characteristics were observed between the groups.

We analyzed the genomic landscape of the study population using the NGS data. In Figure 1, mutations and clinical data are combined and illustrated for each patient. The most common mutation was in TP53 (52.0%). Genomic alterations in ATM (32.0%), ERBB2 (24.0%), NOTCH1 (20.0%), and FGFR3 (20.0%) were also reported at high frequencies. BRCA mutations were observed in 9 patients (36.0%), including those in BRCA1 (E23fs, V271M) and BRCA2 (K2729N, R2494*, G2044V, H3056Y, S3364C, V208G). Univariate analysis of OS according to the genetic mutations was then performed (Fig. 2). Among the genes with mutation frequencies greater than 10%, TP53 and ATM mutations were associated with OS (TP53: HR 2.719, 95% CI: 1.074–6.881, p = 0.035; ATM: HR 2.780, 95% CI: 1.091–7.082, p = 0.032). There was no significant association between OS and BRCA mutations (HR 0.538, 95% CI: 0.183–1.576, p = 0.258).

All patients received first-line platinum-based chemotherapy for advanced BTC, and their response was evaluated using computed tomography scans. The PFS and best response to chemotherapy are presented in Figure 1. Patients with BRCA mutations had a slightly improved PFS compared to those with intact BRCA (6.7 months [range, 2.7–10.7 months] vs. 5.3 months [range, 3.6–7.0 months], p = 0.090; Figure 3a). However, there was no significant difference in OS between groups (BRCA mutant vs. intact; 10.6 months [range, 3.6–17.6 months] vs. 11.9 months [range, 7.5–16.3 months], p = 0.252; Figure 3b). Metastatic status at presentation was significantly associated with PFS in both univariate (HR 2.959, 95% CI: 1.196–7.324, p = 0.019) and multivariate (HR 5.363, 95% CI: 1.599–17.989, p = 0.007) analyses (Table 3). BRCA mutations were significantly associated with PFS in the multivariate analysis (HR 0.150, 95% CI: 0.034–0.655, p = 0.012).

In the present study, we evaluated the clinical significance of BRCA mutations in patients with advanced BTC treated with first-line platinum-based chemotherapy and revealed that BRCA mutations are associated with PFS by performing multivariate analysis. Although the OS of the BRCA mutant group was slightly shorter than that of the intact BRCA group, the difference was not statistically significant. Furthermore, we identified the frequency and HR of high-risk genetic mutations and discovered potential targets, using NGS, in patients with BTC.

In recent years, identification of the genomic landscape of BTC has led to the search for novel potential therapeutic targets, providing opportunities to overcome the limitations of poor survival outcomes [21, 22]. In 2015, an analysis of 260 cases of BTC revealed the spectra of genomic alterations, including targetable growth factor-mediated signaling pathways (FGFR, PKA, EGFR, and ERBB3) and epigenetic regulators (ARID, IDH1/2, BAP1, and MLL2/3) [8]. Over the last 2 years, based on phase 2 trials, the US Food and Drug Administration granted accelerated approval for the use of FGFR inhibitors (pemigatinib and infigratinib) to treat patients with previously treated metastatic cholangiocarcinoma with FGFR2 gene fusions/rearrangements [23, 24]. In addition, the efficacy of the IDH inhibitor ivosidenib was studied in patients with advanced IDH1-mutated cholangiocarcinoma in a phase 3 trial (ClarIDHy); risk was reduced by 51% (p < 0.0001) compared with that in the placebo group in the adjusted analysis [25]. In our study, an IDH1 mutation was detected in 8% of the cases, and there were no FGFR2 fusion genes. The problem with targeted therapy for BTC is the vast genetic diversity. Therefore, the genomic landscape needs to be evaluated to identify new therapeutic targets and increase the number of therapeutic candidates.

BRCA1/2 is the most well-known DDR gene and has been well-studied in various cancers. Although BRCA mutations are generally correlated with poor prognosis, previous studies have shown successful responses to platinum-based chemotherapy and PARPis with BRCA-mutated BTCs. Golan et al. [26] conducted a multicenter retrospective study of BRCA-associated cholangiocarcinoma. Among the 18 patients, 13 received platinum-based treatment and four received PARPis. They reported that the median OS was 25 months (95% CI: 15.2–40.6 months) for 11 patients with stage III/IV disease, which was a much better survival outcome when compared with previous OS data for BTC. In the present study, we compared the response of 25 patients with recurrent/metastatic BTC in association with BRCA mutations to platinum-based chemotherapy. Although patients with BRCA mutations had slightly improved PFS, this did not increase the OS, which was unlike the results of the study that included PARPis. In addition, knowledge from past research on pancreatic ductal adenocarcinoma has formed the basis of trials to explore the effectiveness of PARPis in the treatment of BTCs, owing to their histological and anatomical similarity. Therefore, the clinical importance of PARPis has been emphasized, and many clinical trials are evaluating their role in the treatment of advanced BTC.

The BRCAness phenotype is defined as a homologous recombination repair defect that exists without a BRCA mutation and is related to a cluster of genes in the DDR pathway [27]. Although the development of sequencing technology and the clinical application of the NGS panel has made it easy to detect numerous DDR genes, there is a lack of consensus on how many and which gene variants should be included in BRCAness and DDR deficiency in association with BTC [10, 28]. Heeke et al. [29] revealed that the most common gene mutations associated with homologous recombination deficiency status were in ARID1A (14.3%), BAP1 (7.6%), ATM (4.1%), BRCA2 (2.3%), CHEK2 (2.3%), and PALB2 (1.2%) based on 870 patients with BTC. Additionally, Chae et al. [30] classified germline and somatic mutations in ATM, ATR, BAP1, BARD1, BRCA1, BRCA2, BRIP1, CHEK2, FAM175A, GEN1, MLH1, MSH2, MSH6, MRE11A, NBN, PALB2, PMS2, RAD50, RAD51, RAD51C, RAD51D, and XRCC2 as DDR genes. They reported DDR gene mutations in 55 (63.5%) patients and found that these mutations were significantly associated with improved OS (21.0 vs. 13.3 months, p = 0.009) and PFS (6.9 vs. 5.7 months, p = 0.013) in 88 patients with BTC who received platinum-based chemotherapy. In the present study, although NGS included diverse DDR genes, we focused on BRCA because the definition of BRCAness and DDR genes in BTC is unclear.

This study has several limitations, including its retrospective nature, small sample size, and heterogeneity of the cohort with respect to characteristics such as tumor location and metastatic status. These factors could result in selection bias and might have affected the outcomes. In addition, we used two different specimens (tissue/blood) and three different NGS platforms for genetic analysis, and NGS could not differentiate between the somatic and germline variants. Overall, our results should be validated with a prospective large-scale study. Nevertheless, this study contributes important information to this field and suggests directions for further research on advanced BTC.

In conclusion, we identified a heterogeneous genetic profile for BTC, and some genetic mutations might be associated with survival outcomes. This study demonstrated that BRCA mutations might have a role as predictive biomarkers for palliative first-line platinum-based chemotherapy in patients with advanced BTC. These data provide a rationale for screening genetic profiles and testing targeted therapies, including PARPis, to overcome the limited OS.

The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki. The study was approved by the Institutional Review Board of Keimyung University Dongsan Hospital, and the requirement for written informed consent was waived due to the retrospective nature of the study (DSMC 2022-05-100).

The authors have no conflicts of interest to declare.

No funding was used for this research.

Conception and design of the study: Keon Uk Park and Hyera Kim. Acquisition of data: Hyera Kim, Jin Young Kim, and Keon Uk Park. Analysis and interpretation of data: Hyera Kim. Drafting the manuscript: Hyera Kim. Critical revision of the manuscript for important intellectual content: Keon Uk Park. All authors have read and approved the final version of the manuscript.

All data generated or analyzed during this study are included in this article. Further inquiries can be directed to the corresponding author.