Cancer patients were treated with recombinant interleukin-2 (IL-2) in a Phase I clinical trial. Patients were given four repetitive weekly cycles of four days of continuous i.v. IL-2 infusions followed by 3 days of observation. A transient 80% decrease in the number of circulating peripheral blood lymphocytes (PBLs) was noted 24 h after initiation of the IL-2 infusion. The in vitro IL-2 induced proliferative response, and natural killer (NK) activity of the PBLs recovered at this time was only 10-20% of that by the same number of PBLs obtained prior to IL-2 therapy. This effect was transient and rebound increases in circulating lymphocytes expressing high NK and lymphokine-activated killer functions were demonstrated at the end of a 4 day IL-2 infusion. To study further whether the drop in lymphocyte activity observed at initiation of IL-2 therapy was due to activation of a suppressor mechanism, patient PBLs isolated 24 h into the IL-2 infusion were mixed with their pre-IL-2 therapy PBLs at different ratios and assayed in NK and proliferation assays. These mixing experiments did not prove suppression to be the mechanism for the decreased response; rather, the PBLs obtained 24 h into IL-2 therapy merely diluted out the in vitro response of PBLs obtained prior to therapy. Although there was a considerable drop in circulating PBLs 24 h after initiation of each of three subsequent weekly IL-2 treatment cycles, these remaining lymphocytes in the peripheral blood were functional in both NK and IL-2 proliferative assays. These PBLs obtained 24 h into each of the subsequent three cycles showed a progressive increase in their in vitro NK and IL-2-induced proliferative activity, reaching levels two to three times higher than that of pretherapy PBLs by the fourth cycle. Thus, IL-2 caused a transient disappearance of lymphocytes from the circulation at the initiation of each cycle, but lymphocyte function was impaired only in the first IL-2 cycle. These data suggest that resting IL-2 responsive cells initially leave the circulation upon exposure to IL-2, but that such cells become activated and some remain detectable in the circulation when subsequent weekly cycles of IL-2 are given.