NM_006231.4:c.6748-18G>A is an intronic variant in POLE (intron 48) that is present at extremely low frequency in population databases (gnomAD v2.1: 8/249,726 alleles, AF=0.0032%; gnomAD v4.1: 13/1,594,924 alleles, AF=0.00082%), meeting PM2 at supporting strength.1 SpliceAI predicts no splice impact (max delta score 0.0), and this variant does not involve canonical splice consensus sequences. No functional studies or computational tools support a deleterious effect.2 This variant has been reported in ClinVar as Likely benign by a single clinical laboratory (Invitae, SCV002332903, VCV001540941). The cited publication (PMID:28492532) is a methodology paper describing the Sherloc classification framework and does not provide variant-specific evidence.3 The custom León-Castillo et al. 2020 POLE framework (PM1, PS4, PP3, BP4) applies only to missense variants in the exonuclease domain; this intronic variant is outside the scope of the custom rules, which rely on specific variant-level evidence from the paper's supplementary tables where this variant is absent.4 Overall, only one criterion is met (PM2_Supporting), which is insufficient to reach a classification under ACMG/AMP 2015 combination rules. The variant remains a Variant of Uncertain Significance (VUS) by default, though the available evidence leans benign (ClinVar Likely benign, SpliceAI no impact).5