Longitudinal follow-up after identification of variants of uncertain significance in highly penetrant cancer predisposition genes.

10607 Background: ASCO and the NCCN recommend that patients with variants of uncertain significance (VUS) in hereditary cancer predisposition genes undergo periodic variant reassessment, since reclassification of VUS as pathogenic or likely pathogenic (P/LP) may alter clinical management. However, the extent to which such longitudinal follow-up occurs in practice has not been studied. Methods: Data were retrospectively reviewed from patients who underwent germline genetic testing for cancer-related indications from 2015-2025 at an adult genetics clinic in an academic-affiliated, safety-net healthcare system. Analyses were restricted to patients found to have VUS (without P/LP variants) in any of 23 highly-penetrant cancer predisposition (HPCP) genes based on the NCCN guidelines for genetic/familial high-risk assessment: APC , ATM , BMPR1A , BRCA1 , BRCA2 , BRIP1 , CDH1 , CDKN2A , CHEK2 , EPCAM , HOXB13 , MLH1 , MSH2 , MSH6 , MUTYH (biallelic), PALB2 , PMS2 , PTEN , RAD51C , RAD51D , SMAD4 , STK11 , and TP53 . The primary outcome of interest was the percent of patients who adhered to clinician recommendations for follow-up after VUS disclosure, excluding those for whom insufficient time had elapsed to assess follow-up status. Chi-square tests were used to assess if follow-up rates differed among patients with VUS that were considered “clinically concordant” (personal history of a tumor consistent with the gene’s neoplastic spectrum), “possibly concordant” (history of a relevant neoplasm in a 1st/2nd degree relative but not the patient), or “likely incidental” (no relevant personal or family cancer history). Results: Of 2028 tested patients, 25% (n = 501) were found to have one or more cancer-related VUS (without P/LP variants), including 308 with VUS in HPCP genes. At the time of initial result disclosure, 11% of these patients were discharged from further follow-up, while the remaining 89% were advised to return for variant reassessment, typically within 2 years. Adherence to these recommendations was low overall, with only 38% returning for any post-disclosure follow-up. Rates of follow-up were similar among patients with VUS classified as clinically concordant (40%), possibly concordant (34%), or likely incidental (45%); p = 0.47. Conclusions: Despite clear clinician guidance to return for variant reassessment, most patients with VUS in highly-penetrant genes did not engage in recommended follow-up, even in the presence of a clinically compatible personal and/or family cancer history. Alternative approaches to longitudinal VUS management - such as automated recontact and/or direct-to-patient notification of variant reclassifications - should be explored to mitigate the risk of loss to follow-up in this population.