In-Vitro Efficacy of the Deoxyguanoside Analogs Forodesine (BCX-1777) and ARA-G in Pediatric Acute Lymphoblastic Leukemia.

Abstract 2038

Poster Board II-15

Purine nucleoside phosphorylase (PNP) deficiency in humans is associated with elevated deoxyguanosine (dGuo) plasma levels. DGuo is converted into dGTP inducing apoptosis in T-cells and this provides the rationale for the development of deoxyguanosine analogues as a potential treatment option for T-cell malignancies. Forodesine (BCX-1777; BioCryst-Mundipharma) is an efficient blocker of PNP activity, thereby boosting the conversion of dGuo into dGTP and raising intracellular dGTP levels. AraG (9-b-D-arabinofuranosyl-guanine) is a compound that is resistant to PNP-mediated degradation that is efficiently converted into AraGTP. AraGTP becomes incorporated in the DNA, blocking DNA synthesis and promoting apoptosis. In a phase II clinical trial, the AraG prodrug Nelarabine enforced a complete remission rate of 55% for pediatric T-ALL patients at 1^st relapse. (Berg, JCO 2005). Clinical data of Forodesine treatment in pediatric ALL patients are not yet available.

As tested on primary pediatric acute lymphoblastic leukemia (ALL) patient samples (4 T-ALL, 2 BCP-ALL), 1μM of Forodesine is sufficient to completely block PNP and abolish rapid dGuo degradation resulting in a median 7.9 (range 0.5-378) fold raise of intracellular dGTP levels. Accumulation of dGTP is comparable for T-ALL (n=31) and BCP-ALL (n=11) patient samples. This reflects equal intrinsic ability of salvage nucleotide synthesis for both T-ALL and BCP-ALL cells. Cytotoxic effect of Forodesine was tested on primary leukemia cells from newly diagnosed pediatric ALL patients in-vitro by incubating cells with Forodesine (1μM) in the presence of increasing concentrations of dGuo (0.001-50μM). In accordance with selective T-cell toxicity, T-ALL cells were more sensitive to Forodesine/dGuo treatment (median T-ALL LC50 value: 1.1μM dGuo/1μM Forodesine, n=27, p=0.001) compared to BCP-ALL cells, which had a median LC50 value of 8.8μM dGuo/1μM Forodesine (n=30). All patients that responded demonstrated dGTP accumulation (1.5-222.1 fold), although the raise of dGTP levels did not correlate with Forodesine/dGuo toxicity (r²= 0.10, p=0.22).

Studying in-vitro responsiveness to AraG, T-ALL cells were more sensitive compared to BCP-ALL cells (p=0.0002) with a median AraG LC50 value of 20.5μM for T-ALL samples (n=24) versus 48.3μM for BCP-ALL samples (n=20). Remarkably, TELAML1 positive BCP-ALL cases were insensitive to AraG treatment (median LC50 value >50μM, n=9). No correlation was identified between in-vitro Forodesine/dGuo and AraG cytotoxicities (r²=0.05, p=0.29). Most patient samples that displayed AraG resistance still responded to Forodesine/dGuo treatment. This may be explained by the fact that the uptake of both drugs may be facilitated by different transporters. Using RQ-PCR we could demonstrate that AraG toxicity, in contrast to Forodesine, was significantly associated with ENT1 (equilibrative nucleoside transporter 1) expression levels (p=0.008), which was previously identified as strong predictor for AraC cytotoxicity in pediatric ALL (Stam RW. et al., Blood 2003). AraG cytotoxicity strongly correlated with AraC cytotoxicity (r²=0.71, p<0.0001). We found no significant correlation between Forodesine sensitivity and the expression levels of other nucleoside transporters (CNT1, CNT2, CNT3, ENT2), kinases (dCK, dGK), nucleotidases (NT5C1A, NT5C2, PNI) or other enzymes that are involved in dGuo metabolism (PNP, RRM1, RRM2).

In conclusion, T-ALL cells are more sensitive to Forodesine/dGuo treatment in-vitro than BCP-ALL cells that have nearly 8 fold higher dGuo LC50 values. Resistance to AraG treatment does not preclude responsiveness to Forodesine treatment and vice versa, indicating that Forodesine and AraG rely on different cellular mechanisms for cytotoxicity, possibly involving differences in dependence on the nucleoside transporter ENT1.