PHASE-I AND PHARMACOKINETIC STUDY OF A NEW ANTINEOPLASTIC AGENT - PYRAZINE DIAZOHYDROXIDE (NSC-361456)

Pyrazine diazohydroxide (PZDH) is a novel antineoplastic agent that ap pears to form DNA adducts via the reactive pyrazine diazonium ion and produces substantial antitumor activity in preclinical models. We cond ucted a phase I trial to determine the maximally tolerated dose of PZD H that could be administered as a 5-min i.v. bolus for 5 consecutive d ays repeated every 28 days. Thirty-one patients with advanced cancer r efractory to standard therapy received a total of 65 cycles of therapy at 7 sequential PZDH dose levels: 18, 36, 45, 56, 75, 100, and 133 mg /m2/day. At the maximally tolerated dose (133 mg/m2/day x 5), all 4 pa tients experienced grade 3-4 thrombocytopenia, and 3 of 4 had grade 3- 4 neutropenia. At the recommended phase II dose (100 mg/m2/day x 5), t he median WBC nadir following the first cycle was 2.5 x 10(3)/mul (ran ge, 0.6-7.6) occurring on day 36, and the median platelet nadir was 87 x 10(3)/mul (range, 9-155) occurring on day 26. Nausea and vomiting o ccurred at all dose levels, but were well controlled with ondansetron. No evidence of hepatic, renal, pulmonary, cardiac, venous, dermatolog ical, or neurological toxicity was observed. Pharmacokinetic evaluatio ns were performed on 28 of the 31 patients using an analytical method including derivatization of the parent drug to 2-chloropyrazine. We re port the total 2-chloropyrazine, which represents PZDH converted per m ethod plus PZDH converted in vivo. Although the assay quantitation lim it is 10 ng/ml, PZDH could only be detected at the first dose level fo r 30-90 min after the i.v. bolus. Compartmental modeling of the first 4 dose levels was most consistent with a 2-compartment model. Subseque nt dose levels revealed a third phase to the plasma decay curve. The a rea under the plasma drug concentration-time curve increased proportio nally with dose; there was no evidence for dose-dependent pharmacokine tics. Pharmacokinetic parameters for 12 patients analyzed by the 3-com partment model revealed an alpha-half-life (t1/2alpha) of 2.83 +/- 1.5 7 (mean +/- SD), a t1/2beta of 11.9 +/- 4.42, and a t1/2gamma of 161 /- 47.1 min, with a mean clearance of 1.86 +/- 0.91 liters/min. At the 100- and 133-mg/m2 dose levels, the mean areas under the plasma drug concentration-time curve were 105 and 169 mug min/ml, respectively. Th ere was a moderate correlation between body surface area and clearance (r = 0.45, P = 0.015) but a better correlation between weight and cle arance (r = 0.53, P = 0.004). At the recommended phase II dose, there are insufficient data available to assess whether variability in bone marrow toxicity is due to pharmacokinetic variability or to other fact ors. We identified a correlation (r = -0.68, P = 0.004) between serum chloride concentration and platelet nadir in patients treated at the 1 00- and 133-mg/m2 dose levels. Phase II studies at a PZDH dose of 100 mg/m2/day x 5 every 42 days should be considered, although dose based upon body weight may be more appropriate. Additional phase I studies t o explore and possibly modulate the relationship between PZDH toxicity and serum chloride are also warranted.