Pathogenic germline variants among Thai patients with colorectal cancer: A study in Genomics Thailand Project.

Abstract

e22534 Background: Germline variant testing in colorectal cancer (CRC) patients (pts) has been applied in many countries. Limited data is known regarding germline CRC prevalence in Thailand and other low to middle-income countries in Southeast Asia. Identifying the variants in different ethnicities may show the variety in frequency of the pathogenic germline variants (PGVs) and their relationship with clinical characteristics. Our study reports on the distribution of PGVs and their clinical findings in Thai CRC pts. Methods: We use the germline testing results from the Genomics Thailand Project during 2020-2023, which performed next-generation sequencing on 50,000 pts in 5 disease groups. CRC pts in our institution with a potential risk of carrying PGVs were sent to a geneticist to obtain a family history prior to undergoing germline testing with 25-96 cancer-predisposing genes. Pts with complete pedigree data and germline testing results were enrolled in our retrospective cohort to analyze the variants and clinical findings from chart reviews. Descriptive statistics, the Cox proportion hazards model, and the Kaplan-Meier method were used to evaluate the results. Results: We included 137 CRC pts with a mean age of 47 years at diagnosis, and 58.4% were male. Of these, 38/137 (27.8%) pts were identified as pathogenic/likely pathogenic(P/LP). 26/38 (68.4%) were pathogenic Lynch syndrome (LS), 16 (42.1%) MLH1, 6 (15.8%) MSH2, 2 (5.3%) MSH6, and 2 (5.3%) PMS2. Among pathogenic non-LS, the most frequent variant was 4 (10.5%) APC, followed by 3 (7.9%) ATM, 2 (5.3%) BRCA2, 1 (2.6%) PTEN, 1 (2.6%) BARD1, and 1 (2.6%) monoallelic MUTYH. The LS group was significantly associated with a family history of CRC (p < 0.001), endometrial cancer (p = 0.001), and a synchronous/metachronous CRC (p = 0.003) compared to the non-PGVs group. Likewise, the presence of CRC in a first-degree relative (p = 0.025), more than one CRC pts in the family (p = 0.027), or early onset CRC (p = 0.012) in a first-degree relative was also significantly higher in the pathogenic non-LS group compared to the non-PGVs group. CRC pts who had more than one primary cancer in a lifetime were not significant in the LS or non-LS group compared to the non-PGVs group. At a median follow-up time of 60 (range 7-453) months, Pts in the PGVs group tended to have a shorter median time from the first CRC diagnosis to a second event defined as any of metachronous/synchronous CRC, recurrent CRC, or second cancer in a different organ compared to the non-PGVs group (135 vs. 228 months, p = 0.11). Conclusions: 27.8% pts in our cohort carried P/LP germline variants. The PGVs group was associated with a shorter time to the second event and a higher rate of familial cancer. Our study supports the benefits of genetic cancer screening. [Table: see text]