Abstract BACKGROUND: CDKN2A germline pathogenic/likely pathogenic variants (PVs) cause familial atypical multiple mole melanoma syndrome, which increases melanoma (MEL) and pancreatic cancer (PC) risk. The interpretation of genetic test results can be complicated by discordant variant classifications. CDKN2A c.146T>C (ClinVar ID:127523) is such a variant and is classified as likely pathogenic at some laboratories and as uncertain at others. Furthermore, the variant frequency in Hispanic/Latinos (H/Ls) is higher than typically expected for a PV in CDKN2A and some functional studies support that it is a PV. METHODS: Genetic counseling databases at various institutions were queried for demographic, personal and family cancer history data for individuals with known PVs in CDKN2A, with c.146T>C, and c.–2G>A (ClinVar ID:141245), a benign CDKN2A variant (BV) seen in H/Ls to serve as a comparator. The association between personal and/or family history of PC and/or MEL and demographic/clinical characteristics was tested using Fischer exact test, Chi-square test, and post-hoc test with ANOVA. Odds ratios were calculated with multinomial logistic regression. RESULTS: Overall, 219 individuals were identified; 109 with c.146T>C, 60 with other CDKN2A PVs, and 50 with the BV; 82% were female, 76% H/L, and 70% had a personal cancer history – 18% (n=27) had MEL and 10% (n=16) had PC. By variant group, 99% (c.146T>C) and 96% (BV) were H/L; 85% and 84% of Mexican ancestry, respectively. MEL was reported in 38% (n=22) of other PVs, 5% (n=5) of c.146T>C, and not at all in the BV group. Those with c.146T>C were 92% less likely to have MEL than other CDKN2A PVs (p<0.001), with no difference in mean age at diagnosis, 32y vs. 40y, respectively (p=0.71). Seven (12%) of other PVs had PC compared to 6% (n=7) of c.146T>C and 4% (n=2) of the BV. While this difference was not significant (p=0.27), the mean age at diagnosis in c.146T>C was significantly younger than other PVs (32y vs. 71y; p=0.02). When including family history, those with c.146T>C were 80% less likely to have a personal AND family history of MEL than other PVs (p<0.001), but the odds of having both a personal AND family history of PC was not significantly different (p=0.52). CONCLUSION: In the largest CDKN2A c.146T>C study (n=109) to date, MEL was significantly less frequent in individuals with this variant compared to other PVs CDKN2A. A similar, non-significant, trend was seen for PC. Notably, individuals with c.146T>C who developed PC presented at a significantly younger age than individuals with other CDKN2A PVs. The presence of c.146T>C in H/Ls of mostly Mexican ancestry suggests it may be a founder variant. While more research is needed to further refine cancer penetrance and associations, the conservative approach is to manage c.146T>C as per CDKN2A PV guidelines. Given the frequency of this variant in genomic databases, clarifying clinical impact is vital for risk- appropriate management and to reduce existing cancer disparities faced by many H/Ls. Citation Format: Daisy Hernandez, Ashlie Browning, Amber Gemmell, Karlena Lara-Otero, Mariana Niell-Swiller, Kathyrn A. Mraz, Amie Blanco, Yvonne Cardona, Emmeline Chang, Jacob Comeaux, Julie O. Culver, James M. Ford, Natalia Gutierrez, Daniella Kamara, Beth Y. Karlan, Ming Li, Jacqueline Mersch, Qi Nie, Maria L. Wei, Charité N. Ricker. Exploring the cancer phenotype of a potential CDKN2A founder variant, c.146T>C (p.Ile49Thr), in Hispanic/Latinos of predominately Mexican ancestry [abstract]. In: Proceedings of the 17th AACR Conference on the Science of Cancer Health Disparities in Racial/Ethnic Minorities and the Medically Underserved; 2024 Sep 21-24; Los Angeles, CA. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2024;33(9 Suppl):Abstract nr C123.
Read full abstract