PII-078 - EVALUATION OF G6PD ACTIVITY IN ALL OF US

N. Powell¹, R. Geck², T. Shugg¹, T. Skaar¹, M. Dunham²; ¹Indiana University School of Medicine, Division of Clinical Pharmacology, ²University of Washington, Department of Genome Sciences.

Background: The WHO recently convened a panel of advisors in January 2022 to review G6PD variant classifications, noting a need for more studies of G6PD deficiency associations with G6PD variants. To this end, we aimed to characterize G6PD enzymatic activity in people with G6PD variants using real-world evidence (RWE) in the v7 release of the All of Us data. We expect this will aid in the interpretation and implementation of the updated G6PD CPIC guidelines to avoid toxicity from certain drugs.

Methods: G6PD coding variants were extracted from All of Us WGS exonic Plink files. G6PD percent activity was calculated as the measured activity value divided by the middle of the available or inferred reference range (detailed methodology to be in poster). Each participant’s lowest value was used. An exception was granted by All of Us to present findings of small sample sizes.

Results: There were 31 G6PD haplotype/genotype groups in the All of Us data where G6PD activity data was available and interpretable (n=1236). Figure 1 shows activity values align with the literature, with a few new insights. The All of Us results confirm G6PD activities for A- (c.376A>G;968T>C), Mediterranean (c.563C>T), and Santa Maria (c.376A>G;542A>T) variants; the median G6PD activity for compound heterozygotes (assumed based on population haplotype frequencies) or hemizygotes for these variants is < 10%. Hemi- or homozygosity for the Asahi (c.202G>A) and A- (c.376A>G;202G>A) was associated with a median activity < 25%, aligning with previous reports. We identified a heterozygote for the Dagua missense variant (c.595A>G) with 25% G6PD activity; this variant has not previously been reported with associated G6PD activity results.

Conclusion: The replication of expected G6PD activities across G6PD genotype groups demonstrates the feasibility of using RWE in All of Us to characterize novel or understudied G6PD variants such as the Dagua variant. The prevalence of low G6PD activity beyond c.202G>A genotype groups should help inform choice of G6PD genotyping panels for the US and global populations as well as the need for continued G6PD activity testing. The presence of a few normal activity values in the A- (c.376A>G;202G>A) group may reflect outliers caused by analytical or hematological reasons, demonstrating the added value of G6PD genotyping in addition to activity assays.