Apparent interaction of dimethyl sulfoxide with cisplatin released from polymer delivery devices injected subcutaneously in dogs.

Local tissue toxicity, systemic toxicity and platinum pharmacokinetics were evaluated in 6 normal healthy beagle dogs injected subcutaneously with two formulations of a polylactide biodegradable polymer (Atrigel) system containing cisplatin. Dogs were injected 4 times at 30 day intervals at platinum dosages of 70, 105 and 157.5 mg/m2 (dose escalation). Once pharmacokinetics were established, 29 dogs with spontaneous stage IIb appendicular osteosarcoma were treated with 4 injections of the same polymer system containing cisplatin at 70 mg/m2 (20 dogs) and 100 mg/m2 (9 dogs) to establish efficacy against micrometastatic disease. Local tissue toxicity was variable. Systemic toxicity, as judged by clinicopathologic evaluation was not noted at any dose level or injection number. Interim (6 month) survival analysis revealed a median disease-free interval of 180 days. Consistent platinum release characteristics were found, however, the lack of toxicity and decreased disease-free-interval raised concerns over the biologic activity of the cisplatin. Prior to completion of the study, it was discovered that dimethyl sulfoxide, the solvent used in the co-polymer system, may be responsible for biologic inactivation of cisplatin. This was subsequently demonstrated in tissue culture assays. The clinical trial was suspended and dogs were treated with traditional chemotherapy.