Phase II and pharmacodynamic studies of pyrazine diazohydroxide (NSC 361456) in patients with advanced renal and colorectal cancer

Pyrazine diazohydroxide (PZDH) is a novel antitumor agent that forms DNA adducts via the reactive pyrazine diazonium ion. In a recent Phase I study of PZDH, we identified a recommended Phase II dose of 100 mg/m²/day x 5, given as a 5-min i.v. bolus with the cycles repeated every 42 days (N.J. Vogelzang, et aL, Cancer Res., 54: 114-119, 1994). There was a moderate negative correlation between serum chloride concentration and logarithm platelet nadir, suggesting the hypothesis that PZDH is activated in an acidic environment, leading to more toxicity in acidotic patients. Therefore, the University of Chicago Phase II cooperative network conducted two Phase II studies of PZDH in renal cancer (15 patients, 2 with liver metastases) and in 5-fiuorouracil-refractory colorectal cancer (14 patients, 13 with liver metastases) to determine efficacy in each disease and to correlate safety and tolerance of the drug with PZDH pharmacokinetics/pharmacodynamics and with arterial blood gas measurements. There were no responses seen in either tumor type. The primary toxicity of PZDH was myelosuppression with neutropenia (absolute neutrophil count, < 1000/μ1) and thrombocytopenia (<50,000 cells/μl), seen in 41 and 24% of all cycles, respectively. Other grade 3 and 4 toxicities were rare. Pharmacodynamic analysis revealed no significant correlation between plasma levels at 5, 60, and 120 min; WBCs; absolute neutrophil and platelet count nadirs; and initial serum chloride or blood pH levels. The colorectal patients experienced significantly more thrombocytopenia than did the renal cancer patients (median platelet nadir after cycle 1 was 151 x 10³/μl for renal patients versus 76 x 10³/μl for colon patients; P = 0.04), suggesting either that prior 5-fluorouracil and leucovorin reduced bone marrow reserve or that colorectal patients with liver metastases experienced more PZDH toxicity. Regression analyses revealed a possible relationship (P = 0.06) between serum pH and thrombocytopenia (i.e., for each increase of 0.03 in pH, there was a 34% increase in the platelet nadir), but there was no relationship between serum chloride and thrombocytopenia. Curiously, an increase in alkaline phosphatase was associated with an increase in the platelet nadir (P = 0.02). If PZDH continues to be developed as an antineoplastic agent, further studies of these relationships are suggested.