effect of AZA use in HTX recipients who are heterozygotes for TPMT variants on clinical outcomes is unknown. Aim: To investigate the relationship between genetic variation in TPMT, TPMT activity, active metabolite levels, and AZA efficacy and safety in HTX recipients. Methods: Genotyping for TPMT variants (TPMT *2 G238C, *3B G460A, and *3C A719G) was performed on 101 patients treated with AZA who underwent orthotopic HTX. Erythrocyte TPMT enzyme activity was measured in 94 of these patients and was considered normal (15.1-26.4 U/mL), intermediate (6.3-15.0 U/mL) and reduced (!6.3 U/mL). AZA metabolite level (6TGN) was measured in a subset of 23 recipients. All clinical and demographic data was abstracted by retrospective chart review. Results: There were 10 patients who were heterozygous for TPMT variants and 91 who were wild-type (WT). Heterozygotes have lower TPMT activity level as compared to WT (14.1 6 4.35 vs. 21 6 4.47, p!0.001). There was a moderate inverse correlation between TPMT activity and 6-TGN levels (p50.07). The mean AZA dose at baseline was similar in the two groups (1406 53 mg vs. 166 6 45 mg, p50.074) but 6 months after HTX, AZA dose was lower in heterozygotes as compared to WT (35 6 66 mg vs. 99 6 82 mg, p50.025). The reason for lower mean AZA dose in heterozygotes was due to a high discontinuation rate as compared to WT (8/10 vs. 31/91, p!0.05). The most common reason for discontinuation in both groups was rejection. There was no significant difference in incidence and time to onset of leukopenia between the two groups. Conclusions: TPMT variant heterozygotes have lower TPMT activity levels compared to WT. Carriers of TPMT variants do not have increased risk for leukopenia with AZA therapy, but are more likely to discontinue the drug due to development of rejection. Based on our results, HTX recipients with TPMT heterozygosity do not need AZA dose adjustment to prevent leukopenia. These patients should be monitored closely for rejection, similar to WT HTX recipients.