While certain antibodies directed against the human immunodeficiency virus (HIV) envelope have the potential to suppress virus replication in vitro, the impact of neutralizing antibodies in vivo remains unclear. In a recent proof-of-concept study, the broadly neutralizing monoclonal antibodies 2G12, 4E10, and 2F5 exhibited inhibitory activities in vivo, as exemplified by a delay of the viral rebound following the interruption of antiretroviral therapy. Unexpectedly, the antiviral effect seen was most prominently due to 2G12 activity. To further investigate whether differential HIV-inhibitory activity was due to different pharmacokinetic properties of the antibodies, we performed a formal pharmacokinetic analysis with 14 patients. Repeated infusions at high dose levels were well tolerated by the patients and did not elicit an endogenous immune response against the monoclonal antibodies. The pharmacokinetic parameters of all three antibodies correlated with each other. Mean estimates were 0.047, 0.035, and 0.044 liter/kg for the central volume of distribution of 2G12, 4E10, and 2F5, respectively, and 0.0018, 0.0058, and 0.0077 liter/kg . day for the systemic clearance of 2G12, 4E10, and 2F5, respectively. Monoclonal antibody 2G12 had a significantly longer elimination half-life (21.8 +/- 7.2 days [P < 0.0001]) than monoclonal antibodies 4E10 (5.5 +/- 2.2 days) and 2F5 (4.3 +/- 1.1 days). The comprehensive pharmacokinetic data from this long-term multiple-dose phase II study were coherent with those from previous short-term phase I studies, as assessed by compartmental and noncompartmental techniques. The anti-HIV type 1 antibodies studied showed distribution and elimination kinetics similar to those seen for other human-like antibodies. Further studies examining tissue concentrations to explain the differential in vivo activity of the anti-gp120 antibody compared with those of the two anti-gp41 antibodies are warranted.