A randomized, double-blind trial compared treatment with the immune modulator WF10 (ten patients) and placebo (nine patients) administered in cycles over 3 months among individuals with advanced AIDS. There were no notable clinical adverse events; changes in hematologic and chemistry values were comparable in the two groups. In both groups, median HIV-RNA PCR values remained stable. Immunologic variables showed a consistent tendency to increase in the WF10 group and to decrease in the control group, with significant differences between groups for median WBC, lymphocyte, CD19, and CD35 values. Ten infections occurred in the control group, four of which were Pneumocystis carinii pneumonia (PCP), and three in the WF10 group none of which was PCP. Five patients in the control group were hospitalized during the trial for a total of 53 days; no patients in the WF10 group were hospitalized. Over a subsequent 9-months follow-up, six patients from the control group and one from the WF10 group died. These results indicate that WF10 administration appears safe, may enhance immunologic function, and unlike other macrophage-activating cytokines does not increase HIV expression in this patient population. Further studies of WF10 in larger patient populations are warranted.