In an adjuvant clinical trial for high-risk patients with malignant melanoma by using recombinant interleukin-2 (rIL-2) and recombinant interferon-alpha 2b (rIFN-alpha 2b), we monitored the development of antibodies against rIFN-alpha and various immunoparameters as biologic markers for IFN activity in vivo. Thirty-one patients (22 men, nine women) with high-risk malignant melanoma received eight 6-week cycles of rIL-2 and rIFN-alpha. Serum samples of all patients were screened for the presence of antibodies against IFN-alpha by a solid enzyme immunoassay (EIA). Specimens testing positive in the EIA were assessed for their ability to neutralize the antiviral effects of IFN-alpha in vitro in an antibody-neutralizing bioassay (ANB). Furthermore, serum levels of neopterin, beta 2-microglobulin, soluble IL-2 receptor (sIL-2R), anticardiolipin and antithyroglobulin were evaluated. Of 31 patients, 11 (36%) developed binding antibodies; three (27%) of them had antibodies with neutralizing capacities (range, 350-28,000 INU/ml). Of male patients, 8 (36%) of 22 versus 1 (11%) of nine female patients developed antibodies. Statistical analysis (unpaired t test) revealed that all patients with antibody titers showed significant (p < 0.04) lower serum levels of beta 2-microglobulin and reproducible decreases in sIL-2R levels, whereby those with neutralizing antibodies showed significantly (p < 0.0001) lower values than did those with binding antibodies. Elevations of anticardiolipin (17 of 31) and antithyroglobulin (one of 31) were not correlated to the presence of IFN antibodies. Our results show the in vivo significance of antibodies against rIFN-alpha, especially of those with neutralizing capacities. Monitoring of antibody formation as well as immunoparameters like beta 2-microglobulin in clinical trials can contribute to identifying patients who, if necessary, might benefit from alternative IFN treatment, for instance, by using natural IFNs.