The ClinVar Variant Classification History tool just picked up its tenth monthly snapshot. Between September and October, 2,020 variant classifications changed across the ACMG SF v3.3 panel, and one gene stood out from the rest: GLA, the gene behind Fabry disease.
GLA had 206 classification changes this month, a rate of 95.1 per 1,000 tracked variants. The next closest gene on the panel came in at 16.9. Nothing else this month came close.
Most of those 206 changes were the usual noise, shuffles between adjacent compound labels like Pathogenic and Pathogenic/Likely pathogenic that don't reflect any real change in evidence. Filtering those out leaves a cleaner signal: a cluster of 19 GLA variants that moved from Likely pathogenic down to Uncertain significance in a single month.
That direction is worth sitting with. Most of the reclassification activity this tool has tracked so far moves the other way, variants gaining confidence as evidence accumulates. A downgrade cluster this size, in one gene, in one month, usually means something specific happened rather than gradual drift. Seventeen of the nineteen variants were single-submitter, one-star records, the kind of classification that reflects one lab's assessment rather than a reviewed consensus. That pattern is consistent with a single submitter re-evaluating a batch of calls rather than the field disagreeing.
It's a useful reminder for anyone using ClinVar data in a clinical pipeline: a one-star Likely pathogenic call carries real uncertainty, and that uncertainty can resolve in either direction. The classification history matters as much as the current snapshot.