According to the GLOBOCAN 2022 database, lung cancer remains the leading cancer worldwide in regard to both incidence and mortality [1]. Recent advances in imaging technology and histopathology have resulted in increasing numbers of patients diagnosed with multiple lung cancers (MLC), defined by the presence of two or more independent malignant lesions in the lungs, occurring either synchronously or metachronously [2,3,4]. Accurate distinction between multiple primary lung cancers (MPLC) and intrapulmonary metastasis (IM) is crucial, as it directly affects disease staging and treatment strategies [5,6,7,8]. For comparing genomic profiles across tumor loci, several methods based on Next-Generation Sequencing (NGS) technology have recently been developed [9,10,11]. Among those, whole-genome sequencing (WGS) and whole-exome sequencing (WES) provide comprehensive genomic information for better discrimination of MPLC from IM [12,13], however, their use for determining the status of every tumor examined in routine clinical settings is impractical. Integration of targeted sequencing panels with histopathological evaluation has therefore become a promising approach, and its clinical utility warrants further investigation.

In addition to tumor classification, elucidation of the contribution of germline pathogenic variants (GPVs) to lung adenocarcinoma carcinogenesis is an important issue. In a study of primary lung cancers in 7788 patients, germline genetic testing identified enrichment of GPVs in BRCA2 DNA repair associated (BRCA2), ATM serine/threonine kinase (ATM), checkpoint kinase 2 (CHEK2), BRCA1 DNA repair associated (BRCA1), and mismatch repair genes as compared with an unaffected control group [14]. Among high- or moderate-penetrance DNA damage repair genes, BRCA2 exhibited the highest detection prevalence in lung cancer cases, especially those with lung adenocarcinoma (LUAD) [15]. Additionally, a recent study found that GPVs in BRCA2 and TP53 were associated with early-onset LUAD in an Asian population [16].

A large population-based genetic association study found that rare pathogenic variants in cancer predisposition genes were more strongly associated with multiple primary cancer diagnoses (2.6-fold greater) as compared with single primary cancer diagnoses (1.8-fold greater) [17]. Furthermore, enriched GPVs in BRCA2, TP53, and RAD51D have consistently been found in Asian patients with dual primary breast and lung malignancies, thus highlighting the potential contribution of genetic predisposition in multiple cancer etiology [18]. Although genetic risk models have substantially advanced our understanding of GPVs in several malignancies, including breast, ovarian, prostate, and pancreatic cancers [19,20,21,22], their clinical significance in lung cancer remains unclear.

In this study, we aimed to analyze tumors from patients with multiple lung cancer adenocarcinoma by integrating histopathological and molecular classification, and to identify putative germline pathogenic variants in cancer susceptibility genes that may contribute to lung cancer etiology, with particular focus on BRCA2.

Targeted panel sequencing for 26 tumors obtained from 11 patients diagnosed with multiple lung adenocarcinomas was performed for the present study. Tumors in nine of the patients exhibited distinct molecular profiles and were classified as MPLC, while those in one were classified as IM, and the final patient exhibited both IM and MPLC (Fig. 2). These results are in line with the International Association for the Study of Lung Cancer guidelines and other reports, which caution against classification of multiple lesions as IM or MPLC based solely on morphological analysis, and strongly recommend integrating molecular testing [27,28,29].

In addition to providing improved diagnostic precision, the present findings also highlight the potential contribution of shared predisposition VUS among patient lesions, which may underlie the development of multiple independent tumors. Several germline variants were identified across multiple tumor foci, including alterations in the homologous recombination gene BRCA2, tumor suppressor genes TP53, KMT2D, PIK3R1, and NRG1, and oncogenes CARD11, PDGFRB, AR, and KIT (Table 2). Currently, the contribution of VUS to multiple lung cancer development and their clinical relevance remain unclear due to insufficient data to assess pathogenicity. However, it is possible that these germline variants cumulatively contribute to tumor development. Additional functional studies and computational predictions are required to determine their clinical significance.

Two novel truncating BRCA2 variants were detected in the present study. The first variant (p.N900IfsTer4) was found across three tumor lesions from an MPLC case. Screening of the second LUAD cohort identified one patient without any apparent driver mutations, though harboring a distinct truncating BRCA2 variant (p.Q1429FfsTer20). Protein domain mapping revealed the location of p.N900IfsTer4 upstream of the BRC repeat region (amino acid 1009-2083), while p.Q1429FfsTer20 was found within BRC repeat 3, a critical domain essential for RAD51 binding (Fig. 5). The p.N900IfsTer4 variant truncates BRCA2 upstream of all eight BRC repeats and the C-terminal DNA-binding domain, indicating a complete loss-of-function effect. In contrast, the p.Q1429FfsTer20 variant, though retaining the BRC repeats 1 and 2, results in loss of BRC repeats 4–8 and downstream functional domains, potentially leading to homologous recombination deficiency. According to the ACMG/AMP guidelines, these frameshift variants fulfill the PVS1 criterion for a null variant in a well-established tumor suppressor gene. Additionally, both variants satisfy the PM2 criterion, as they are absent from control population databases, including the Exome Aggregation Consortium and gnomAD v4.1, thereby supporting their classification as likely pathogenic variants [26]. Such BRCA2 germline alterations likely result in a haploinsufficient phenotype and may be associated with an increased risk of oncogenesis.

The effects of pathogenic BRCA2 mutations, which impair the efficacy of DNA damage response (DDR), on lung cancer development remain controversial, though some studies have reported findings showing an association of BRCA2 GPVs with a high risk of lung cancer [15,16,30]. A large case-control study that included 63,828 patients with 14 common cancer types and 37,086 controls found that the odds ratio for lung cancer was 1.7-fold greater, though it did not reach a level of statistical significance [31]. These findings highlight the need for further investigation of the contribution of BRCA2-mediated DDR deficiency, not only to the risk of MPLC development but also to the pathogenesis of primary LUAD.

In addition to the carcinogenic contribution of BRCA2 GPVs to LUAD, an increasing clinical concern is whether these tumors exhibit functional homologous recombination deficiency (HRD), which may predict sensitivity to PARP inhibitors. While our approach identified likely germline BRCA2-mediated HRD, further investigation, such as analysis of mutational signatures or functional analyses, is essential to confirm HRD. Thus patients carrying these variants may have opportunities to benefit from PARP inhibitor therapy [32,33], platinum-based chemotherapy [34], nanomedicine-based delivery strategies [35,36,37], and other molecular reprogramming approaches [38].

Recently, precision oncology frameworks that integrate somatic and germline mutation data with genetic counselling have significantly improved diagnostic workflows and personalized care, particularly for LUAD patients with multiple primary tumors or lacking driver mutations [39]. In terms of diagnostic application, NGS analysis using both tissue and liquid biopsy samples may increase detection of somatic and germline variants, thereby enabling comprehensive genomic profiling, enhancing diagnostic precision, and providing additional information for clinical management of NSCLC patients [40]. From a therapeutic perspective, germline genomic profiling can inform treatment selection and classify patients for clinical trials aimed at developing new personalized therapeutic approaches [41].

Nevertheless, the present study has several limitations. First, use of tumor-adjacent regions as normal controls is generally not recommended for confirming germline variants because of various cancer-related phenomena, including possible field effects [42]. As in other retrospective analyses involving deceased patients, the lack of peripheral blood or other normal tissue samples limited our ability to definitively confirm germline status. Furthermore, the present cohorts were relatively small, that have only 11 MPLC and 123 LUAD patients. This may explain why the observed BRCA2 variant frequency in the LUAD cohort (0.81%) was slightly higher than that reported for the general population (0.1-0.4%) [43]. The present study also lacked functional assays to demonstrate HRD in patients with BRCA2 GPVs.

This study identified putative PGVs, including two novel truncating BRCA2 variants, which provide insight into molecular mechanisms involved in lung adenocarcinoma development. Moreover, it is considered that these likely germline variants could serve as novel targets for therapeutic strategies in primary and multiple primary LUAD cases, as well as for early detection in high-risk healthy individuals.