Quantifying evidence for phenotypic specificity (PP4) for syndromic phenotypes: Large-scale integration of rare germline FH variants from diagnostic laboratory testing for HLRCC (Hereditary Leiomyomatosis and Renal Cell Cancer) and renal cancer

Abstract

PurposeHereditary Leiomyomatosis and Renal Cell Cancer (HLRCC) is a rare cancer susceptibility syndrome exclusively attributable to pathogenic variants in FH. This paper quantitatively weights the phenotypic context (PP4/PS4) of such very rare variants in FH. MethodsWe collated clinical diagnostic testing data on germline FH variants from 387 individuals with HLRCC and 1,780 individuals with renal cancer, and compared the frequency of very rare variants in each phenotypic cohort against 562,295 population controls. We generated pan-gene very rare variant likelihood ratios (PG-VRV-LRs), domain-specific likelihood ratios for missense variants (DS-VRMV-LR) using spatial clustering analysis, and log2.08 likelihood ratios (LLRs) as applicable within the updated ACMG/AMP Variant Classification Framework. ResultsFor HLRCC, the PG-VRV-LR was estimated to be 2,669.4 (95% CI: 1,843.4-3,881.2, LLR 10.77) for truncating variants and 214.7 (185.0-246.9, LLR 7.33) for missense variants. For renal cancer, the PG-VRV-LR was 95.5 (48.9-183.0, LLR 6.23) for truncating variants and 5.8 (3.5-9.3, LLR 2.39) for missense variants. Clustering analysis in HLRCC cases revealed three hotspot regions wherein the DS-VRMV-LR increased to 1226.9. ConclusionThese data provide quantitative measures for very rare missense and truncating variants in FH, which reflect the differing phenotypic specificity of HLRCC and renal cancer, and may be applicable in clinical variant classification.