McLaren et al.(1) recently used whole-exome sequencing to show that GNPAT p.D519G (rs11558492) is associated with a high-iron phenotype in HFE C282Y homozygous men. Three other studies examined this association in different Caucasian populations with conflicting results. (2)(3)(4) However, the role of GNPAT in the absence of iron overload or in the HFE wild-type population remains unclear.Here, we investigated iron indices before and after iron supplementation and their association with genetic polymorphisms. This study was approved by the ethics committee, and informed consent was obtained from all participants. We screened healthy reproductive-age women and excluded those with renal insufficiency, anemia, gastrointestinal bleeding, or chronic inflammation and those that were pregnant or breastfeeding; thus, 83 healthy female volunteers were eligible for analysis. Participants fasted overnight, and baseline iron parameters were measured at 8 AM. They were then administered sodium ferrous citrate (element iron, 100 mg), and blood was sampled 2 hours later. Data are presented in Table 1.Regarding TMPRSS6 rs855791, iron parameters at baseline or after iron challenge did not significantly differ between genotypes. For GNPAT rs11558492, 64 participants were wild type (A/A) and 19 carried variants (18 A/G and 1 G/G). The G allele frequency was 12% (20 of 166), similar to 14% in the general population in 1000 Genomes (chi-square test: P 5 0.416). Fasting serum iron levels were higher in subjects with GNPAT variants than in wild-type subjects (131.2 vs. 110.5 lg/dL; P 5 0.045), but there was no significant difference in fasting transferrin saturation (TS; 38.2% vs. 33.7%; P 5 0.153). After iron supplementation, serum iron levels and TS were both significantly higher in GNPAT variants (iron, 320.2 vs. 284.5 lg/dL; P 5 0.025; TS, 85.8% vs. 78.9%; P 5 0.040). Furthermore, multivariate analysis showed that the GNPAT genotype was an independent factor associated with baseline iron levels (P 5 0.045; adjusted R 2 5 0.037); the factors analyzed included age,