To evaluate the safety, toxicity, and maximum tolerated dose (MTD) of IFN beta-1a (Rebif, Serono Laboratories, Inc.) in patients with malignant diseases unresponsive to standard therapies and to assess the pharmacodynamics and pharmacokinetics associated with IFN beta-1a administration, an open-label, single-center phase I study was designed. Thirty-four patients were enrolled and treated with IFN beta-1a. All had measurable solid neoplasms or evaluable hematological malignancies. All patients received a single i.v. bolus dose of IFN-beta-1a on day 1, followed 7 days later by daily s.c. injections for 28 consecutive days. Successive groups of three patients received increasingly higher doses (in geometric progression from 1.5 million international units (MIU)/m2 to 24 MIU/m2) until dose-limiting toxicities were noted. Pharmacokinetic and biological studies, including measurement of the activity of 2',5'-oligoadenylate synthetase (2',5'-OAS) in peripheral blood mononuclear cells and serum levels of soluble Tac (CD 25) and beta-2 microglobulin, were performed on patients who agreed to participate. i.v. and s.c. doses of IFN beta-1a up to 24 MIU/m2 were administered. The most frequent adverse events (AEs) were constitutional symptoms. Grade III AEs during i.v. dosing included fever, elevation of bilirubin, and infection unrelated to therapy. No grade IV events were seen. AEs noted during continuous s.c. therapy included fever, liver transaminase increase, albuminuria, fatigue, nausea, myalgia, and rigors. Dose-limiting toxicities were encountered during s.c. dosing at the 24-MIU/m2 and 18-MIU/m2 dose levels and included gastrointestinal toxicity, elevations of aspartate aminotransferase and alanine aminotransferase, and albuminuria. The s.c. MTD was determined to be 12 MIU/m2, although there was great variability in the individual patient's ability to tolerate IFN beta-1a. 2',5'-OAS activity, thought to be indicative of IFN activity, increased within hours after i.v. and s.c. dosing, with the level remaining persistently elevated during the s.c. daily injections. The highest peak level was attained in the 6-MIU/m2 group. There was no evidence that the increase in 2',5'-OAS activity decayed with repetitive dosing, nor was there evidence of accumulation in this pharmacodynamic marker. Serum beta-2-microglobulin levels showed a modest time- and dose-dependent increase after s.c. administration of IFN beta-1a, with the largest increase seen at the 24-MIU/m2 dose level. There were no clear dose-dependent responses noted in soluble Tac serum levels. IFN beta-1a was well-tolerated when administered by a single i.v. bolus injection at doses up to and including 24 MIU/m2. Daily s.c. injections for at least 28 days were well-tolerated at doses up to and including 12 MIU/m2, with some patients tolerating doses twice as high as this. The MTD for the i.v. route could not be clearly determined according to the guidelines of the protocol. However, i.v. bolus doses up to 24 MIU/m2 were relatively well-tolerated. For the s.c. route, the MTD was determined to be 12 MIU/m2, but there was great interpatient variability, with some patients able to tolerate higher doses.