Introduction of a Variant Classification System for Analysis of Genotype-Phenotype Relationships in Heritable Retinoblastoma

SIMPLE SUMMARY: Heritable retinoblastoma is a genetic disease that predisposes to develop multiple retinoblastomas in childhood and other extraocular tumors later in life. It is caused by genetic variants in the RB1 gene. Here we present a new classification for genetic variants in the RB1 gene (REC) that focuses on the variant’s effect. The different classes, REC-I to -V, correlate with different risks of tumor predisposition. REC correlated with different clinical courses when applied in our study cohort. REC aims to facilitate risk estimation for physicians, patients and their families, and researchers and to improve the definition of the necessity of screening examination. ABSTRACT: Constitutional haploinsufficiency of the RB1 gene causes heritable retinoblastoma, a tumor predisposition syndrome. Patients with heritable retinoblastoma develop multiple retinoblastomas early in childhood and other extraocular tumors later in life. Constitutional pathogenic variants in RB1 are heterogeneous, and a few genotype-phenotype correlations have been described. To identify further genotype-phenotype relationships, we developed the retinoblastoma variant effect classification (REC), which considers each variant’s predicted effects on the common causal mediator, RB1 protein pRB. For validation, the RB1 variants of 287 patients were grouped according to REC. Multiple aspects of phenotypic expression were analyzed, known genotype-phenotype associations were revised, and new relationships were explored. Phenotypic expression of patients with REC-I, -II, and -III was distinct. Remarkably, the phenotype of patients with variants causing residual amounts of truncated pRB (REC-I) was more severe than patients with complete loss of RB1 (REC-II). The age of diagnosis of REC-I variants appeared to be distinct depending on truncation’s localization relative to pRB structure domains. REC classes identify genotype-phenotype relationships and, therefore, this classification framework may serve as a tool to develop tailored tumor screening programs depending on the type of RB1 variant.