5-Fluorouracil concentrations in human plasma following R,S-1-(tetrahydro-2-furanyl)-5-fluorouracil (ftorafur) administration.

Following administration of i.v. doses of fl,S-1 -(tetrahydro-2furanyl)-5-fluorouracil in cancer patients, discrepant results have been obtained by several investigators when assaying for 5-fluorouracil (FUra) plasma concentrations, ranging from 0.1 to 5 fig/ml at peak concentrations. The newly isolated fl,S-1(tetrahydro-2-furanyl)-5-fluorouracil metabolites with altered tetrahydrofuran moiety convert to FUra in vitro and may con tribute to the high FUra plasma concentrations reported previ ously. We have observed FUra plasma concentrations consid erably below those resulting from therapeutic slow FUra infu sions (generally <0.1 jug/ml). Exceedingly low FUra concen trations after therapeutic doses of the prodrug R,S-1-(tetrahydro-2-furanyl)-5-fluorouracil suggest that metabolically gener ated FUra is further metabolized before redistribution to the systemic circulation. The pyrimidine antimetabolites FT3 and FUra have similar antitumor activities but different selective tissue toxicities (12). FT is considered to be a slow metabolic release form of FUra which has been detected in low concentrations in human plasma after FT administration. However, there are large dis crepancies between the reported peak FUra plasma concen trations resulting from FT administration, ranging from >5 \ig/ ml (2) to 21 /xg/ml (7) and even below 100 ng/ml (1). Significant amounts of FUra metabolites have been isolated in rats (3), dogs, monkeys (5), and cancer patients (2) following [2-14C]FT administration. This supports the hypothesis that FT is metabolized to FUra to a large extent. Moreover, the myelotoxicity observed with FT clinically is comparable to that of FUra when FUra is administered by slow i.v. infusion (1 2). One would, therefore, expect FUra plasma concentrations after FT doses to be similar to those found after equivalent doses of FUra given by slow i.v. infusion, r4owever, this prediction is based on the assumption that metabolically generated FUra freely distributes into the systemic circulation. It is also con ceivable that the intracellularly formed FUra may be rapidly metabolized to further products without reaching the circula tion. The result would be relatively low FUra serum concentra tions, in spite of significant FT conversion to FUra. We report ' This investigation was supported by USPHS Grant GM-16496 from the National Institute of General Medical Sciences, the Earl C. Anthony Fund from the University of California, San Francisco, and Training Grant GM 00728-15 from the NIH. 1 To whom requests for reprints should be addressed. 3 The abbreviations used are: FT, fl,S-1-(tetrahydro-2-furanyl)-5-fluorouracil (ftorafur); FUra, 5-fluorouracil; 4--OH-FT. 4'-hydroxyftorafur; 3'-OH-FT, 3'-hydroxyftorafur; dehydro-FT. dehydroftorafur. Received March 5. 1979; accepted July 10. 1979. here that FUra plasma concentrations in cancer patients after FT administration are considerably below those observed after an equivalent FUra infusion. High FUra concentrations after FT administration measured by other groups may represent an artifact arising mainly from a labile circulating FT metabolite, which we have recently isolated in our laboratory (1, 10). Benvenuto ef al. (2) reported FUra plasma concentrations of 1.7 fig/ml maintained over the observation period of 96 hr in cancer patients given a single FT dose of 4 to 5 g/sq m (total dose, 7 to 8 g). The theoretical FUra dose needed to achieve the plasma concentrations reported by Benvenuto et al. (2) in a 70-kg adult is approximately 24 g over 96 hr4 (37 g FT), assuming a plasma clearance of 2.1 liters/kg/hr for FUra (4). Furthermore, therapeutic FUra plasma concentrations during a 5-day infusion of FUra at the maximally tolerated dose of 30 mg/kg/day range only between 100 and 400 ng/ml (8), and concentrations above 1 fig/ml are normally expected to cause severe toxicities (11). Spurious FUra determination as a metabolite of FT can arise from several sources, (a) The FT dose is usually contaminated with a small amount (0.15%) of FUra (1) which causes relatively high FUra plasma concentrations in the first 30 min after the FT infusion. In our recent study of FT disposition in cancer patients (1), we have observed initial FUra plasma concentra tions of 200 ng/ml which rapidly declined with a half-life of about 10 min. Moreover, Benvenuto ef al. reported FUra con centrations declined exponentially up to 2 hr after FT admin istration, when FT plasma concentrations were the highest. These observations speak against inhibition of FUra metabo lism by FT as a possible cause of unusually high FUra plasma concentrations. (t>)FUra may be formed from FT in vitro during storage and analytical work-up. Even a minor decomposition of FT to FUra may result in serious errors due to the large FT: FUra ratio in plasma (100 to 1000:1 ) (1, 7). (c) FT metabolites may convert to FUra in vitro. We have recently identified 2 hydroxylated FT metabolites, i.e.. 3'-OH-FT and 4'-OH-FT, in rats, rabbits, and cancer patients (1, 10, 13). The 4'-OH-FT isolated from rabbit and human urine was partially converted to FUra by thymidine phosphorylase in vitro (1, 10, 13). This enzyme has been identified in human serum and plasma, and cancer patients were found to have higher enzyme activity than did normal subjects (9). Benvenuto ef al. (2) subsequently found that a mixture of these 2 hydroxylated FT metabolites were partially converted to FUra when incubated with plasma and, to a lesser extent, with aqueous buffer. Furthermore, we have isolated a lipophilic metabolite of FT with an unsaturated * Amount of FUra infused = area under the plasma concentration-time curve of FUra x FUra plasma clearance (6).