This Phase I study was performed to assess the feasibility of administering L-778,123, a peptidomimetic farnesyl protein transferase (FPTase) inhibitor, as a continuous i.v. infusion for 7 days every 3 weeks and to determine the recommended dose for subsequent disease-directed trials. This study also sought to characterize the pharmacological behavior of L-778,123 and to determine whether the desired biological effect, inhibition of protein farnesylation, could be detected and assessed during treatment. Patients with advanced solid malignancies were treated with L-778,123 as a continuous i.v. infusion for 7 days every 3 weeks at doses ranging from 35 to 1120 mg/m(2)/day. On the basis of preclinical studies, toxicity assessments included cardiac telemetry, electrocardiograms, and electroretinograms in addition to more routine safety monitoring laboratory tests. Plasma sampling was performed to characterize the pharmacokinetics of L-778,123, and peripheral blood mononuclear cells (PBMCs) were sampled to detect and monitor the inhibitory effects of L-778,123 on the prenylation of HDJ2, a chaperone protein that undergoes farnesylation. Twenty-five patients received 51 complete courses of L-778,123. An unacceptably high incidence of dose-limiting toxicities, consisting of grade 4 thrombocytopenia, significant prolongation of the QT(c) interval, and profound fatigue, was observed at the 1120 mg/m(2)/day dose level. At the next lower L-778,123 dose level, 560 mg/m(2)/day, seven new patients had no unacceptable toxicity. Instead, myelosuppression was mild to moderate and QT(c) prolongation was negligible. Pharmacokinetics were linear, and L-778,123 plasma concentrations at steady-state (mean, 8.09 +/- 3.11 microM at 560 mg/m(2)/day) exceeded IC(50) values (range, 0.07-5.35 microM) required for growth inhibition and cytotoxicity in preclinical studies. The systemic clearance of L-778,123 averaged 106.4 +/- 45.6 ml/min/m(2), and the terminal half-life of elimination was 2.8 +/- 1.0 h. L-778,123 inhibited HDJ2 prenylation for the duration of the drug infusion in a dose-dependent manner, but seemed to plateau above 560 mg/m(2)/day. At the 560 mg/m(2)/day dose level, the mean percentage of HDJ2 protein in its unprenylated form increased from 1.41% +/- 1.71% (pretreatment) to 28.76% +/- 6.10% (day 4) and 30.86 +/- 4.96 (day 8) and declined to 2.28% +/- 2.11% one week after drug discontinuation (day 16). L-778,123 administered as a continuous 7-day i.v. infusion for 7 days every 21 days is well tolerated at doses of 560 mg/m(2)/day and results in biologically relevant concentrations and consistent inhibition of HDJ2 prenylation in PBMCs. Although the relationship between drug-related inhibition of HDJ2 prenylation in PBMCs and both prenylation of relevant proteins and growth inhibition in tumor cells is unknown, serial analyses of HDJ2 prenylation provide a pharmacodynamic marker of protein prenylation that may be useful in optimizing the development of drugs targeting FPTase.