Author Response: D313Y Variant in Fabry Disease: A Systematic Review and Meta-analysis

We would like to thank Bauer et al.¹ for their interest and constructive criticism on our work.

1. Regarding the p.D313Y frequency, we repeated the analysis using the R software version 3.5.0 (package: meta),² and identical results were documented. Let us again remind the readers that meta-analysis of proportions was performed after the implementation of the variance-stabilizing double-arcsine transformation, using the random-effects model.³

2. We further performed a sensitivity analysis, removing the studies indicated by Bauer et al.; however, subgroup differences between patients with Fabry disease (FD) suspicion vs the general population remained significant (p < 0.01). In addition, leave-one-out meta-analysis to investigate for potential exaggerated effect sizes did not reveal any significantly influential study.

3. We have performed a thorough quality assessment of the included studies, and the results are provided in a supplementary file together with the main manuscript. Indeed, the included studies are of moderate quality, mostly suffering by the lack of control group. Therefore, further epidemiologic, basic research studies and well-designed registries of D313Y variation carriers are warranted, as we discuss in our manuscript.

4. According to the prespecified protocol for our systematic review, online databases presenting GLA mutations in the general population were excluded, thus failing to include the frequency reported in the genome aggregation database. Yet, even if this reported frequency is included, the overall results and the subgroup differences do not differ compared with our primary analysis.

5. In the subgroup of general population, Colon et al. screened patients with reduced α-galactosidase activity presenting a theoretical risk of false negatives⁴; however, in the study of Koulousios et al. healthy patients were evaluated irrespective of α-galactosidase activity.⁵ As mentioned earlier, the leave-one-out meta-analysis did not disclose any significantly influential study. After further comparison between the frequencies among patients with suspected FD vs the frequency reported in the genome aggregation database, we detected again significant subgroup differences (p < 0.01).

In conclusion, following the results of this meta-analysis, which can only provide prevalence estimates and cannot be compared with results of well-designed epidemiologic studies that are warranted, we propose not to a priory exonerate the presence of D313Y variation but rather thoroughly monitor the D313Y carriers for FD manifestations.