Aims: Antibody (Ab)-positive pure red-cell aplasia (PRCA) is a very rare but serious adverse event associated with recombinant human erythropoietin treatment (4.1 reports per 100,000 patient-years) in which patients produce antibodies to recombinant and endogenous erythropoietin, halting red blood cell production. In a previous case series, four Thai subjects with chronic kidney disease and Ab-positive PRCA were reported to have the HLA-DRB1*9 allele. To confirm a possible association of HLA-DRB1*9 and Ab-positive PRCA, we performed a pharmacogenomic analysis using subjects from an earlier case–control study of risk factors associated with Ab-positive PRCA, which had been performed using subjects from Europe or Canada. The primary goal of the analysis was to test the association between HLA-DRB1*9 and Ab-positive PRCA. A secondary goal was to perform an exploratory analysis in order to identify additional HLA alleles potentially associated with Ab-positive PRCA. Patients & Methods: Subjects were taken from a case–control study of Ab-positive PRCA in chronic kidney disease patients treated in Europe or Canada. Ab-positive PRCA cases (n = 24) were matched to controls (n = 81) by timing of treatment exposure and, when possible, by location. Results: The allele frequency of HLA-DRB1*9 was 12.5% in cases vs 1.2% in controls (p = 0.002). The frequency of the HLA-DRB1*9/other genotype was 25.0% in cases vs 2.5% in controls (p = 0.004; OR: 10.8 [95% CI: 2.2–53.7]). Within the exploratory analysis, six additional HLA alleles (HLA-A*25, HLA-B*53, HLA-C*12, HLA-DQB1*3, HLA-DQB1*6 and HLA-DRB1*4) were also found to be associated with Ab-positive PRCA. Conclusion: This study confirmed that HLA-DRB1*9 occurs at a significantly higher frequency in Ab-positive PRCA cases than in controls; however, within this sample set, carrying the *9 allele was neither necessary nor sufficient to cause Ab-positive PRCA.