When splicing is not all or none: Implications for variant classification

Variants with intermediate functional effects—those that neither abolish nor fully preserve gene function—constitute an underappreciated source of genetic complexity and a grey zone for classification within the current ACMG/AMP binary pathogenic-or-benign framework. Here, we address this issue using GT>GC (+2T>C) 5′ splice site variants, approximately 15% of which retain variable amounts of wild-type transcript, as a tractable model. Drawing on disease-associated GT>GC variants and those analyzed by full-length gene splicing assays, we show that residual wild-type transcript levels form a continuum from barely detectable to near normal. Notably, GT>GC variants retaining substantial wild-type transcript (e.g., SPINK1 c.194+2T>C, HBB c.315+2T>C, and BRCA2 c.8331+2T>C) frequently present classification challenges. Our findings also indicate that in classical Mendelian disorders caused by monoallelic or biallelic complete loss-of-function alleles, GT>GC variants are not invariably pathogenic (e.g., DOCK8 c.3234+2T>C). Minigene analyses of selected CFTR and BAP1 variants further illustrate the inherent difficulty of predicting and experimentally quantifying residual wild-type transcript levels (e.g., CFTR c.2490+2T>C and CFTR c.4242+2T>C). Because GT>GC variants that generate wild-type transcript exemplify the degeneracy and context dependence of splice-site recognition, insights from their study extend to other splice-altering variants outside the invariant +1, −1, and −2 positions. Overall, our analysis highlights that grey-zone variants with intermediate functional effects represent a biological reality that complicates the genotype– phenotype relationship and underscores the need for classification frameworks that accommodate the full spectrum of variant effects and the true complexity of human genetic disease.