Comparative Metabolism of 2-(Bis(2-chloroethyl)amino]tetrahydro .2H-1 ,3,2-oxazaphosphorine-2-oxide (Cyclophosphamide) and Its Enantiomers in Humans1

As examples of this phenomenon may be cited the greaten central nervous system stimulant activity of (+ )-amphetamine compared with that of its (—)-isomer(23), the superior antico agulant activity of S-(— )-phenpnocoumon compared with the R-(+)-isomer (10), the inactivity of (+)-methadone in the main tenance of opiate-dependent patients (21), and the superior analgesic activity of (+)-propoxyphene compared with the (—)-antipode(4). In the area of cancer chemotherapy, mel phalan (3-[4-bis(2-chloroethyl)aminophenyl]-L-alanine) was se lected for clinical use in preference to the Disomer (medphalan) on the basis of the superior potency of the L isomer against the Walker 256 carcinoma in rats (2). The L-mannitol analog of mannitol myleran (1,6-dimethanesulfonoxy-1 ,6-dideoxy-Dmannitol) has been reported to have very low antitumon activity compared with the D antipode (3). These examples concern molecules which are dissymmetnical by virtue of a chiral carbon atom. As an example of a biological difference based on chinality at phosphorus may be cited the greater enzyme inhi bition by the S-(—)-isomensof some anticholinestenases, the alkyl S-alkylmethylphosphonothioates compared with their R (+)-antipodes (9). Recent syntheses of the enantiomers of cyclophosphamide (14) and of isotopically labeled (deuterated) variants thereof (5) have enabled their antitumor activity and metabolism to be separately evaluated. Therapeutic differences have been re ported in tests against 2 munine tumors. Against the ADJ/ PC6A plasma cell tumor, (—)-cyclophosphamidehad approxi mately twice the therapeutic index (50% lethal dose/90% inhibitory dose) of the (+)-antipode (6). Smaller differences were observed in tests against the Li 210 leukemia; male mice survived longest when treated with (—)-cyclophosphamide, whereas female mice showed the best response with the (+ )isomer (24). The metabolism of the enantiomens has been compared in several species (mouse, rabbit, and rat), both in vitro, using liver microsomes, and in vivo (5, 7). Particular attention has been paid to the influence of chinality upon the rate of the initial 4-hydroxylation step (Chart 1) and upon the yields of the enzymatically produced detoxification products 4ketocyclophosphamide and carboxyphosphamide. It has been suggested that the relatively selective toxicity of cyclophospha mide toward tumor tissue is related to the more efficient detox ification thereto of 4-hydroxycyclophosphamide and aldophos phamide by normal host tissues (8). If this hypothesis is valid, then differences between the enantiomers regarding the rela tive yields of these products could result in different therapeutic efficacies, as could differences between the rates of metabo lism of the enantiomens in the initial 4-hydroxylation step. The main objective of this study in humans has therefore