Recently, a Chinese genomewide association study (GWAS) identified four autosomal single-nucleotide polymorphism (SNP) loci as being significantly associated with risk factors for nonobstructive azoospermia (NOA; P < 5 × 10⁻⁸). In the present study, we performed a replication study on two Japanese cohorts from different institutions in order to evaluate whether SNP loci are associated with NOA. The four SNPs (rs12097821, rs2477686, rs10842262, and rs6080550) reported in the Chinese GWAS were genotyped in 490 NOA patients and 1167 controls. To assess the significance of the associations between each of the four SNPs and NOA in the Japanese population, the association results for the two cohorts were combined by meta-analysis. In the meta-analysis, the combined per-allele odds ratios (ORs) for the four SNPs and their respective 95% confidence intervals (CIs) were as follows: rs12097821, OR = 1.10 (CI = 0.89–1.37); rs2477686, OR = 1.11 (CI = 0.87–1.43); rs10842262, OR = 1.11 (CI = 0.94–1.32); and rs6080550, OR = 0.96 (CI = 0.76–1.21). None of the SNPs was significantly associated with NOA (P > 0.05). However, three of four SNPs (rs12097821, rs2477686, and rs10842262) showed associations in the same direction in Japanese men as those reported in the Chinese GWAS. To determine whether the four SNPs are genetic risk factors for NOA, the effect sizes of NOA risk factors require further investigation using larger independent sets of case-control samples of populations, including Japanese and Chinese populations.