Validation of the Index for Facial Angiofibromas: Data analysis from a randomized controlled trial of sirolimus gel treatment in patients with tuberous sclerosis complex

Abstract

AbstractThe Index for Facial Angiofibromas (IFA), a novel scoring system for angiofibromas, has been validated in patients with tuberous sclerosis complex (TSC). The objective of this analysis was to further validate the IFA using data from a clinical trial of topical sirolimus in patients with TSC. This was an analysis of photographs from a Phase III trial conducted in Japan (NCT02635789). Patients (n = 62) were randomized 1:1 to receive sirolimus or placebo gel for 12 weeks. Changes in angiofibromas were independently assessed using the primary composite endpoint, the Facial Angiofibroma Severity Index (FASI), and the IFA. Thresholds for a clinically meaningful change in IFA score were evaluated using receiver operating characteristic (ROC) analysis. The IFA scores had good‐to‐excellent inter‐assessor reliability, very high intra‐assessor reliability, and could be used to evaluate the distribution of disease severity at baseline. High correlations were observed between the categorized change from baseline in IFA scores and the primary composite endpoint (Kendall's coefficient of concordance, W = 0.8655, p < 0.0001), and between the change from baseline in IFA and FASI scores (Kendall's coefficient of concordance, W = 0.745, p < 0.0001). By ROC analysis, an optimal IFA cut‐off point of 1.667 was determined to distinguish patients with markedly improved or improved angiofibromas from those with slightly improved or unchanged angiofibromas (area under the curve 0.937) as determined by the primary composite endpoint. The IFA score is potentially clinically useful because of its high validity and reliability. A decrease in score from baseline of ≥1.667 may be considered clinically meaningful.