Performance characteristics of brief family history questionnaire to screen for Lynch syndrome in women with newly diagnosed ovarian cancers

Abstract

<h3>Objectives:</h3> Ovarian cancer (OC) is the third most common Lynch syndrome (LS)-associated cancer in women but there is no established screening strategy to identify LS in this population. In centers without access to universal tumor immunohistochemistry (IHC) testing for mismatch repair (MMR) deficiency, family history may identify patients with OC suspected of LS, prompting a referral to genetic counseling. We have previously developed and shown that patient-administered 4-item brief Family History Questionnaire (bFHQ) can effectively identify women with confirmed LS in those with newly diagnosed endometrial cancers. The objective of this study was to assess whether bFHQ can be used as a screening tool to identify women with OC at high risk of LS. <h3>Methods:</h3> Prospective cohort study recruited women with newly diagnosed non-serous/non-mucinous OC from three cancer centers in Ontario, Canada. Participants completed bHFQ, extended Family History Questionnaire (eFHQ; encompassing Amsterdam II criteria, Society of Gynecologic Oncology 20-25% criteria and Ontario Ministry of Health criteria), IHC for MMR proteins and universal germline testing (reference standard) for LS. The sensitivity, specificity, positive and negative predictive values (PPV, NPV) were compared between the family history screening strategies and MMR IHC to identify LS. <h3>Results:</h3> Of 215 women, 169 (79%) were evaluable with both bFHQ and germline mutation status; 12 of these 169 were confirmed to have LS (7%). Median age was 53 years (range 21-71), and for all 12 patients with LS, OC was their sentinel malignancy (100%). Nine of 12 patients (75%) with LS were correctly identified by bFHQ, compared to 6 of 11 (55%) by eFHQ and 11 of 13 (85%) by IHC. The sensitivity, specificity, PPV and NPV values of bFHQ were 75%, 66%, 15% and 98%, compared to 55%, 92%, 35% and 96% for eFHQ and 85%, 90%, 39% and 99% for IHC respectively (Table 1). IHC was the most sensitive and specific approach. The 4-item bFHQ was more sensitive than eFHQ and took less than 10 minutes for each patient to complete. <h3>Conclusions:</h3> Patient-administered bFHQ may serve as an adequate screening tool to triage women with OC for further genetic assessment for LS, especially in centers without access to universal tumor testing for IHC for MMR or microsatellite instability.