Phenotypically Deficient Urinary Elimination of Carboxyphosphamide after Cyclophosphamide Administration to Cancer Patients1

The 0-24-h urinary metabolic profile of cyclophosphamide was inves tigated in a series of 14 patients with various malignancies receiving combination chemotherapy including i.v. cyclophosphamide. This was accomplished using combined thin-layer chromatography-photographydensitometry, which can quantitate cyclophosphamide and its four prin cipal urinary metabolites (4-ketocyclophosphamide, nor-nitrogen mus tard, Carboxyphosphamide, and phosphoramide mustard). Recovery of drug-related metabolites was 36.5 ±17.8% (SD) dose, the most abundant metabolites being phosphoramide mustard (18.5 ±16.1% dose) and unchanged cyclophosphamide (12.7 ±9.3% dose). The most variable metabolite was Carboxyphosphamide, with five patients excreting 0.3% dose or less. These patients were termed low carboxylators (LC) and could be distinguished from high carboxylators (HC) by a carboxylation index (relative percentage as Carboxyphosphamide multiplied by 10). Mean carboxylation indices for the LC and HC phenotypes were 3.4 ± 2.6 and 151 ±115, respectively. There were no associations between patient age, sex, body weight, tumor type, or concomitant drug therapy and carboxylation phenotype. Neither 4-ketocyclophosphamide nor nornitrogen mustard excretion differed between LC and HC phenotypes; however, HC patients had a greater excretion of cyclophosphamide (46.4 ±15.5 relative percentage) than LC patients (19.4 ±12.6%). The DNA cross-linking cytotoxic metabolite phosphoramide mustard was elevated more than 2-fold in the LC (76.5 ±13.9%) compared with the HC (33.0 ±12.2%) phenotype. It is concluded that these data represent the first evidence of a defect in cyclophosphamide metabolism, and it is proposed that this arises from a hitherto unrecognized aldehyde dehydrogenase genotype.