Phase I clinical investigation of 9,10-anthracenedicarboxaldehyde bis[(4,5-dihydro-1H-imidazol-2-yl)hydrazone] dihydrochloride with correlative in vitro human tumor clonogenic assay.

9,10-Anthracenedicarboxaldehyde bis[(4,5-dihydro-1H-imidazol-2-yl)hydrazone] dihydrochloride (bisantrene) is a new anthracene bishydrazone derivative which was entered into a Phase I clinical trial (one dose weekly for 3 weeks) because it showed significant antitumor activity in a number of animal tumor models and in vitro in the human tumor stem cell assay. When possible, patients were entered into the phase I study if their tumors showed in vitro sensitivity to bisantrene and resistance to standard agents, using a human tumor stem cell assay. Thirty-one patients were treated with bisantrene over a 10-month period, starting at a dose of 70 mg/sq m/week. The appearance of leukopenia determined the dose-limiting toxicity of bisantrene. The maximally tolerated dose appeared to be 200 mg/sq m in that three of five patients tolerated these weekly-for-3-weeks doses while experiencing only mild or moderate leukopenia. In contrast, the 220-mg/sq m dose caused moderate to life-threatening leukopenia after just two weekly doses in four of five patients. Local bisantrene toxicity included mild to severe arm swelling, phlebitis, pain, urticaria, and erythema in 68% of the patients. In general, these toxicities were well tolerated and rapidly reversible, but two patients had severe local swelling for up to 6 months. In this Phase I trial, bisantrene showed clinical antitumor activity against both hematological cancer (i.e., lymphoma and myeloma) and solid tumors (i.e., bladder, lung, and renal cancer and melanoma). Of importance, four of the six responses occurred in patients whose therapy was selected on the basis of in vitro sensitivity to bisantrene using the human tumor stem cell assay. One patient with disseminated melanoma had complete disappearance of an axillary node metastasis (for more than 6 months) while developing a brain metastasis, suggesting that bisantrene does not concentrate in the central nervous system.