Retinoids promote lung alveolarization in animal models and were administered to patients as part of the Feasibility of Retinoid Therapy for Emphysema (FORTE) study. This FORTE substudy investigated the pharmacokinetic profiles of 2 retinoic acid isomers-all-trans-retinoic acid (ATRA) and 13-cis-retinoic acid (13-cRA)-in subjects with emphysema, evaluated strategies to overcome self-induced ATRA catabolism, and identified pharmacodynamic relationships. Comprehensive and limited pharmacokinetics were obtained at multiple visits in emphysema subjects treated with placebo (n = 30), intermittent dosing (4 days/week) with low-dose ATRA (1 mg/kg/day, n = 21), or high-dose ATRA (2 mg/kg/day, n = 25) or daily administration of 13-cRA (1 mg/kg/day, n = 40). High-dose ATRA produced the highest peak plasma ATRA Cmax. However, at follow-up, plasma ATRA C(max) was significantly decreased from baseline in subjects whose day 1 levels exceeded 100 ng/mL (P < .0001). In contrast, administration of 13-cRA produced lower plasma ATRA C(max) (<100 ng/mL), but the levels were significantly higher at follow-up than those on day 1 (P < .001). Plasma ATRA levels as determined on day 1 correlated with changes in pulmonary diffusing capacity at 6 months, consistent with concentration-dependent biologic effects (r2 = -0.25). The authors conclude that intermittent therapy with high-dose ATRA produced the greatest ATRA exposure, but alternative approaches for limiting self-induced ATRA catabolism should be sought.