1-BETA-ARABINOFURANOSYLCYTOSINE IN THERAPY OF LEUKEMIA - PRECLINICAL AND CLINICAL OVERVIEW

Although the mechanisms of therapeutic efficacy of cytosine arabinoside (Ara-C) are multifactorial, the pharmacodynamic basis for its cytotoxicity and therapeutic efficacy lies in its intracellular metabolism and the retention of the active metabolite, Ara-C triphosphate (Ara-CTP), which is a competitive inhibitor of DNA polymerase. Additional determinants of tumor cell sensitivity include Ara-CMP incorporation into cellular DNA, the size of the competing normal metabolite, deoxycytidine/5'-triphosphate pool, and the heterogeneity in growth kinetics of tumor cells, S-phase vs cells in other phases of the cell cycle. With high-dose Ara-C, substantial amounts of Ara-CTP are formed in phases of the cell cycle. The presence of high intracellular concentration with prolonged retention of Ara-CTP could lead to the inhibition of cell growth of the cells entering S-phase as a consequence of inhibition of DNA-polymerase and/or incorporation into cellular DNA, resulting in a chain termination. Pharmacokinetically, Ara-C is rapidly eliminated from plasma. In mice, pharmacokinetic parameters of Ara-C are not sufficient predictors for the observed differences in their in vivo antitumor activity. Although these mice were bearing different tumor types (L1210 Ara-C sensitive or P-388 relatively more resistant), the observed differences in tumor response were achieved under identical plasma Ara-C concentrations and area under the concentration time curve. The observed antitumor activity in L1210 cells is primarily associated with higher Ara-CTP pools and retention (T1/2 > 4 hr) in tumor cells as compared with normal bone marrow cells. In the least responsive tumor (P-388), although Ara-CTP pools were sufficiently high, retention of the drug in tumor cells and in normal cells is poor with a T1/2 < 2 hr. Thus, unlike mice bearing leukemia L1210 cells, alteration of the mode and dose of administration of Ara-C in mice bearing P-388 could only result in increased host toxicity with no therapeutic gain. Similarly in patients with acute nonlymphocyte leukemia (ANLL), there is no significant correlation between plasma Ara-C concentration and the intracellular concentrations or retentions of Ara-CTP. In some patients the highest Ara-CTP pools in leukemic myeloblast cells are achieved at a lower level of plasma Ara-C and decrease further with the increase of plasma Ara-C. Thus, in the in vivo model system and in ANLL patients with no prior chemotherapy, Ara-CTP retention is a critical factor associated with response to this agent, in particular its direct association with duration of complete response. In contrast, in patients with relapsed disease, Ara-CTP retention is no longer a predictive parameter for remission duration after relapse. In model systems, in vitro and in vivo resistance to Ara-C has been associated predominantly with a single mechanism, i.e. a low level of deoxycytidine kinase. While this mechanism of resistance can be easily achieved in model systems, its expression in ANLL remains a rare event, representing less than 5% of cases evaluated, regardless of whether the cells were obtained from patients with no prior chemotherapy or in patients who have been heavily pretreated with a regimen containing Ara-C. Thus, unlike model systems, expression of low levels of deoxycytidine kinase in tumor cells following repetitive treatment with Ara-C has not been easily observed clinically.