Toxicity of a quinocarmycin analog, DX-52-1, in rats and dogs in relation to clinical outcome

Purpose: Quinocarmycin analog DX-52-1 is a cyanated derivative of quinocarmycin, a compound isolated from cultures of Streptomyces melanovinaceus. DX-52-1 was selected for preclinical development because it showed efficacy against melanoma cell lines in the NCI human tumor cell screen and melanoma xenografts in mice. This report describes studies in rats and dogs to determine the maximum tolerated dose (MTD) and identify dose-limiting toxicities (DLT) in each species in different regimens to establish a safe starting dose and potential target organs of DX-52-1 for phase I clinical trials. Methods: DX-52-1 was administered to Fischer 344 rats using repeated intravenous (i.v.) slow bolus injections following q3hx3 and q3hx3,q7dx3 regimens, and to beagle dogs using a single injection, 6-h continuous i.v. infusion (c.i.v.) and weekly 6-h c.i.v. for 3 weeks. Endpoints evaluated included clinical observations, body weights, hematology, serum clinical chemistry, and microscopic pathology of tissues. Results: The MTD of DX-52-1 was a total dose of 18 mg/m(2) body surface area for q3hx3 administration in rats and 30 mg/m(2) for a single c.i.v. administration in dogs. The total dose MTD for rats on a weekly (q3hx3,q7dx3) regimen was 54 mg/m(2), and for dogs on the weekly x3 (6-h c.i.v.) infusion was 60 mg/m(2). In rats, significant elevations in blood urea nitrogen and creatinine were observed together with acute renal tubular necrosis histologically. Modest increases in liver enzymes were also observed, as were decreases in reticulocytes that were unaccompanied by histologic changes in liver and bone marrow. In dogs, adverse signs included vomiting/retching, diarrhea, and transient hypothermia; also red blood cells, hemoglobin, hematocrit, and lymphocytes were decreased. Histologic evaluation of tissues from dogs revealed necrosis and cellular depletion of the bone marrow, and extensive damage to the entire gastrointestinal tract, including marked cellular necrosis of the mucosa and lymphoid necrosis of the gastrointestinal associated lymphoid tissue. Destruction of the mucosal lining of the intestinal tract was likely responsible for dehydration, toxemia, septicemia, and shock seen in moribund dogs. Conclusions: The MTD values were comparable between rats and dogs given roughly similar dose regimens (single dose or weekly) and both species tolerated a higher total dose with weekly administration. However, the principal target organ responsible for DLT in rats was the kidney, whereas in dogs, the most severe effects were on the gastrointestinal tract and bone marrow. Both renal and gastrointestinal toxicities were reported in patients after 6-h c.i.v. infusions in a limited phase I clinical trial, indicating that neither animal model alone was predictive of DX-52-1-induced toxicity in humans, and that both species were required to define human toxicity.