We optimized the modulation of drug resistance by the irreversible augmentation of cytotoxicity of coincubating vinblastine (VNB) with VP-16 and the reversible increase in cytotoxicity of coincubation of verapamil (VPL) with VNB and VP-16. VPL was administered as a loading dose (i.v.) (0.15 mg/kg) and then administered as a constant infusion (0.005 mg/kg) over 6 days. 24 h after verapamil, VNB 2 mg/m2 IVP was administered and followed 1 h later by a 5-day simultaneous continuous infusion of VP-16 (200 mg/m2/day) to seven pediatric patients (11 courses) with refractory malignancies. The mean age at treatment was 7.5 +/- 5.3 years, mean prior anthracycline dose (303 +/- 210 mg/m2) with a range of 0-606 mg/m2. Toxicity was limited to cardiac and hematological. The median nadir of the WBC was 900 at 14.5 +/- 0.5 days and platelet count 32,000 at 15.5 +/- 0.8 days. There were two episodes of bacterial sepsis both of which responded to i.v. antibiotics. Five of 11 courses resulted in first-degree block and one course in second-degree block. At Hour 120 of the VPL infusion the PR interval was 0.18 +/- 0.01 versus 0.13 +/- 0.01 at Hour 0 (P less than 0.0004). The ejection fraction by two-dimensional echocardiogram was not significantly different at Hour 0, 24, or 120 of the infusion (60.6 +/- 2.7 versus 52.7 +/- 5.1 versus 51.8 +/- 5.0%). The cardiac index was also not significantly different at Hour 0, 24, or 120 (4.39 +/- 0.2 versus 4.21 +/- 0.6 versus 3.91 +/- 0.5 liters/min/m2). The 15-min VPL level was 1954.5 +/- 391/ng/ml and steady state levels at Hour 24 and 120 of the infusion were 468.1 +/- 59 and 422.8 +/- 52 ng/ml, respectively. Two of 11 treatment courses resulted in hypotension secondary to inordinately high 24-h levels of VPL (1233 and 1263 ng/ml). These two episodes required inotropic support but did not require the discontinuation of VPL. There were 8 of 11 partial responses, the majority of which consisted of peripheral cytoreduction of leukemic blasts and one M-2A response in AML. The levels of VPL achieved in this study have been shown to augment the in vitro cytotoxicity of vinblastine and VP-16 to resistant cell lines. Further clinical studies are needed to determined the maximal-tolerated dose of VPL in a Phase I study and to examine its efficacy in selected relapsed pediatric patients.