Antithymocyte globulin (ATG) is used as prophylaxis against graft-versus-host disease (GVHD). Current dosing regimens for ATG are empiric and weight-based, and do not account for patient-specific factors. Furthermore, the target of ATG, recipient T cells post-cytotoxic chemotherapy, is not a function of recipient weight. We hypothesized the recipient peripheral blood absolute lymphocyte count (ALC) on the day of ATG administration would interact with the dose of ATG administered to predict transplantation outcomes. We retrospectively analyzed 135 patients who received ATG for GVHD prophylaxis for unrelated allogeneic hematopoietic cell transplantation at 3 different doses: 10 mg/kg, 7.5 mg/kg, and 5 mg/kg. There was no difference in 2-year overall survival (OS) among ATG dosing groups; however, deaths from infectious complications were significantly higher with higher doses of ATG (3.7% versus 19% versus 26.7%; P = .02). Severity of chronic GVHD was lower with higher doses of ATG (28% versus 24% versus 4%; P = .03). In multivariate analysis, the median peripheral blood ALC on day of ATG administration and the total amount of ATG interacted to predict OS (hazard ratio, .09; P = .03). For low recipient ALC (10th percentile, or .56 × 10/µL), a higher total ATG dose was associated with a greater risk of death, whereas for high recipient ALC (90th percentile, or 24.96 × 10/µL), a higher ATG dose was associated with a lower risk of death. Our findings suggest that the interaction between ATG and its target, the recipient lymphocyte, could represent a new paradigm for ATG dosing.