A reappraisal of APOE genetic effects on Alzheimer’s disease risk in the Japanese population: a meta-analysis

Figures

• 

  Fig. 1. Meta-analysis of APOE-e4*4 vs. APOE-e3*3 in Japanese AD. (A) Forest plots showing ORs with 95% CIs for e4*4 vs. e3*3 (reference) in EOAD, LOAD, and AD, estimated using fixed-effects and random-effects meta-analysis models. Crude ORs with corresponding 95% CIs (Supplementary Table 3) are also shown for reference in each forest plot. (B) For comparison with (A), ORs with 95% CIs from two recent large-scale Japanese datasets [11, 12] are presented separately. As described in Supplementary Fig. 1, these datasets were excluded from the primary meta-analysis in (A) because their substantially larger sample sizes relative to earlier reports could introduce bias. For the dataset by Asanomi et al. [11]., results from fixed- and random-effects meta-analyses, as well as a crude OR, are presented. For the dataset by Miyashita et al. [12]., only the crude OR is shown, as individual-level APOE genotype data were not available [12], precluding meta-analysis. The forest plots for the Asanomi dataset (2019) and the Miyashita dataset (2025) were constructed using numerical data reported in Supplementary Table 6 and Table 3 of Miyashita et al. [12]., respectively