Adoption of universal next generation sequencing for Lynch syndrome screening in endometrial cancer

Abstract

<h3>Objectives:</h3> To assess implementation of a novel microsatellite instability (MSI) screen using next generation sequencing (NGS) in endometrial cancer (EC) through comparison to prior Lynch syndrome screening strategies, as well as to compare characteristics of patients with microsatellite stable (MSS) and microsatellite unstable endometrial cancer. <h3>Methods:</h3> Retrospective review was performed of patients with newly diagnosed endometrial cancer at a single academic institution during three time periods and using different screening strategies: 1) 01/2010 - 01/2013 with provider-directed mismatch repair immunohistochemistry (MMR IHC) in women ≤60, 2) 01/2013 - 01/2016 with universal MMR IHC in all women ≤60, and 3) 10/2018 - 10/2019 with universal screening for MSI using NGS regardless of age. In cohorts 1 and 2, women with abnormal MMR IHC were referred to a genetic counselor who ordered MLH1 promoter methylation or germline genetic testing as indicated. In cohort 3, reflexive MLH1 promoter methylation was performed for patients with MSI and no MMR germline or somatic mutations, with referral to genetic counseling for suspected germline MMR mutation or unexplained MSI. Statistical analysis was performed using a t-test for continuous variables and Fisher's exact test for categorical variables. ≤60, 2) 01/2013 - 01/2016 with universal MMR IHC in all women ≤60, and 3) 10/2018 - 10/2019 with universal screening for MSI using NGS regardless of age. In cohorts 1 and 2, women with abnormal MMR IHC were referred to a genetic counselor who ordered MLH1 promoter methylation or germline genetic testing as indicated. In cohort 3, reflexive MLH1 promoter methylation was performed for patients with MSI and no MMR germline or somatic mutations, with referral to genetic counseling for suspected germline MMR mutation or unexplained MSI. Statistical analysis was performed using a t-test for continuous variables and Fisher's exact test for categorical variables. <h3>Results:</h3> There were 133 subjects in cohort 1, 135 subjects in cohort 2, and 140 subjects in cohort 3. The proportion of patients screened for Lynch syndrome increased over time (39%, 94%, and 100% of intended screening population). The proportion of positive screens was similar between the 3 cohorts (27%, 21%, and 27%). The proportion with positive screens who underwent confirmatory testing with germline genetic testing or MLH1 promoter methylation was 50% in cohort 1 and 59% in cohort 2. Following implementation of reflexive MLH1 promoter methylation testing when indicated, the proportion increased to 89% (p=0.001). Attendance at genetic counseling appointments following referral was 50%, 59%, and 61% in each cohort. Eleven patients were diagnosed with Lynch syndrome across the 3 cohorts (4%, 3%, and 1% respectively). In cohort 3, subjects with MSI EC had similar clinicopathologic characteristics compared with MSS cancers, and over half of MSI cancers were identified in subjects over the age of 60 (71%). Eight subjects had incidental somatic mutations identified through NGS, including 6 <i>POLE</i> mutations, 1 <i>BRIP1</i> mutation, and a subject with mutations in <i>TP53, AKT1,</i> and <i>FBXW7.</i> At our institution, MSI testing by NGS was billed at $905 versus $1200 for MMR IHC screening. <h3>Conclusions:</h3> Transition to universal screening using NGS with reflexive MLH1 promoter methylation testing increased the number of subjects completing genetic assessment for Lynch syndrome after a positive screen, likely as somatic causes of MSI were ruled out, allowing more focused genetics referrals. Similar characteristics between subjects with MSI and MSS EC supports universal screening regardless of clinical characteristics. In addition to identifying Lynch syndrome, detection of MSI through NGS can identify somatic alterations with therapeutic implications, and can be performed at a lower cost than MMR IHC.