MODY probability calculator utility in individuals' selection for genetic testing: Its accuracy and performance

Abstract

Introduction: MODY probability calculator (MPC) represents an easy-to-use tool developed by Exeter University to help clinicians prioritize which individuals should be oriented to genetic testing. We aimed to assess the utility of MPC in a Portuguese cohort with early-onset monogenic diabetes. Methods: This single-centre retrospective study enrolled 132 participants submitted to genetic testing between 2015 and 2020. Automatic sequencing and, in case of initial negative results, generation sequencing were performed. MODY probability was calculated using the probability calculator available online. Positive and negative predictive values (PPV and NPV, respectively), accuracy, sensitivity and specificity of the calculator were determined for this cohort. Results: Seventy-three individuals were included according to inclusion criteria: 20 glucokinase (GCK-MODY); 16 hepatocyte nuclear factor 1A (HNF1A-MODY); 2 hepatocyte nuclear factor 4A (HNF4A-MODY) and 35 DM individuals with no monogenic mutations found. The median probability score of MODY was significantly higher in monogenic diabetes-positive subgroup (75.5% vs. 24.2%, p <.001). The discriminative accuracy of the calculator, as expressed by area under the curve, was 75% (95% CI: 64%-85%). In our cohort, the best cut-off value for the MODY calculator was found to be 36%, with a PPV of 74.4%, NPV of 73.5% and corresponding sensitivity and specificity of 76.2% and 71.4%, respectively. Conclusions: In a highly pre-selected group of probands qualified for genetic testing, the Exeter MODY probability calculator provided a useful tool in individuals' selection for genetic testing, with good discrimination ability under an optimal probability cut-off of 36%. Further geographical and population adjustments are warranted for general use.