Cancer Chemother Pharmacol (1986) 18: 145-152 ancer hemotherapy and harmacology © Springer-Verlag 1986 9- -D-Arabinofuranosyl-2-fluoroadenine 5'-monophosphate pharmacokinetics in plasma and tumor cells of patients with relapsed leukemia and lymphoma* Lynn Danhauser l, William Plunkett 1, Michael Keating 2, and Fernando Cabanillas 2 Departments of 1Chemotherapy Research and 2Hematology, The University of Texas M.D. Anderson Hospital and Tumor Institute at Houston, 6723 Bertner Avenue, Houston TX 77030, USA Summary. The pharmacokinetics of 9-13-D-arabinofurano- syl-2-fluoroadenine (F-ara-A) in plasma and its biological- ly active 5'-triphosphate (F-ara-ATP) in leukemic cells ob- tained from the peripheral blood and bone marrow was evaluated in patients with hematologic malignancies sub- sequent to the first dose of 20-125 mg/m 2 per day for 5 days of F-ara-A 5'-monophosphate (F-ara-AMP) adminis- tered as an IV bolus over 30 min. The terminal half-lives of elimination of both F-ara-A (8 h) in plasma and intracellu- lar F-ara-ATP (15 h) were not dependent upon the dose of F-ara-AMP. The area under the concentration x time curves for F-ara-A and F-ara-ATP, on the other hand, were increased in proportion to the prodrug dose. There was a high correlation between F-ara-ATP levels in circu- lating leukemic cells and those in bone marrow cells aspi- rated at the same time. DNA-synthetic capacity of leukem- ic cells was inversely related to the associated F-ara-ATP concentration. A linear trend was noted when F-ara-ATP levels in pretreatment peripheral blood leukemic cells in- cubated with F-ara-A in vitro were compared with the amount of F-ara-A that was incorporated into nucleic ac- ids. Finally, F-ara-ATP concentrations were three times higher in bone marrow cells from patients with lymphoma- tous bone marrow involvement than from those without evidence of marrow disease. Introduction The striking antileukemic activity of the pyrimidine ana- ~logue 1-[~-D-arabinofuranosylcytosine (ara-C) prompted the search for purine nucleoside derivatives that might be similarly effective agents. 9-[~-D-Arabinofuranosyl-ade- nine (ara-A) proved to be an active antiviral agent [28], but had limited antitumor activity because of its rapid deami- nation by adenosine deaminase [2, 7, 12]. The antitumor efficacy of ara-A could be augmented by combining this agent with a nontoxic dose of the adenosine deaminase in- hibitor 2'-deoxycoformycin [5, 13, 20]. Such combination chemotherapy, however, which elevates the natural nucle- otide dATP, may produce a cellular milieu that is antago- nistic to the inhibitory action of the active moiety, ara- * Supported in part by grants CA28153 and CA32839 from the National Cancer Institute, United States Department of Health and Human Services Offprint requests to: L. Danhauser ATP, on DNA synthesis [24]. The observation that a fluo- rine atom substituted for a hydrogen at the 2-position of the purine component of adenosine could render this com- pound relatively resistant to deamination led to the synthe- sis of 9@D-arabinofuranosyl-2-fluoroadenine (F-ara-A) [17, 181. In the absence of adenosine deaminase inhibitors, F-ara-A produced cytotoxicity against cultured human lymphoblastoid cells that was comparable to that of ara-A combined with 2'-deoxycoformycin [221. Fludarabine phosphate (F-ara-AMP), the soluble 5'- monophosphate derivative of F-ara-A, is currently under- going Phase I-II clinical trials [9, 10, 16, 311 (L. Danhau- ser, W. Plunkett, J. Liliemark, V. Gandhi, S. Iacoboni, M. Keating, 1985, submitted for publication). F-ara-AMP is rapidly dephosphorylated to F-ara-A in both animals and man [6, 15, 19] (L. Danhauser et al., 1985,submitted for publication]. F-ara-A is initially phosphorylated by deoxy- cytidine kinase [4, 81 and then is converted to its biologi- cally active 5'-triphosphate, F-ara-ATP [3, 221 (L. Dan- hauser et al., 1985, submitted for publication). F-ara-A irreversibly inhibits S-adenosylhomocysteine hydrolase [32], whereas F-ara-ATP inhibits ribonucleotide reductase and the DNA polymerases with subsequent inhibition of DNA synthesis [22, 30, 32]. In addition, F-ara-ATP be- comes incorporated into RNA as well as into DNA [20]. The aim of this investigation was to evaluate the phar- macokinetics of F-ara-A in plasma and of F-ara-ATP in leukemic cells obtained from the peripheral blood and bone marrow of patients with hematologic malignancies who had relapsed or failed to respond to prior chemother- apy. We hoped to identify parameters that would prove useful to predict the efficacy of this agent. Materials and methods F-ara-A and erythro-9-(2-hydroxy-3-nonyl)adenine were the kind gifts of Ven L. Narayanan, PhD, Drug Synthesis and Chemistry Branch, National Cancer Institute, Bethes- da, Md. F-ara-AMP for clinical use was supplied by the National Cancer Institute as a sterile, lyophilized powder (200 mg/vial) free of antibacterial preservatives; it was reconstituted with 10ml sterile water. [8-3H]F-ara-A (13-18 Ci/mmol), obtained from Moravek Biochemicals (Brea, Calif), was routinely purified by high-pressure liq- uid chromatography (HPLC) on ~tBondapak C18 (Waters Associates, Inc., Milford, Mass) using previously de- scribed methods [1]. The radioactive fractions eluting in