CEP701 and PKC412 Predictably and Reliably Inhibit FLT3 Phosphorylation in Primary AML Blasts but Their Induction of a Cytotoxic Response Appears to Be Much More Variable.

Abstract Activating mutations of the receptor tyrosine kinase FLT3 are present in approximately one-third of AML cases and are associated with an adverse prognosis. FLT3 is expressed in over 90% of cases of AML and many non-mutants show evidence of FLT3 activation, which may play a significant signalling role in leukaemogenesis, making FLT3 an attractive therapeutic target. CEP701 (Cephalon) and PKC412 (Novartis) are orally-bioavailable indolocarbazole derivatives that potently inhibit FLT3 phosphorylation. We studied the relationship between in vitro inhibition of FLT3 phosphorylation and induction of cytotoxicity in primary AML blasts from 12 patients. 7 of the cases were FLT3 mutants (6 ITDs and 1 D835 point mutant), the amount of mutant RNA varying between 7% and 84% of total FLT3 RNA expressed. The blasts were exposed for 1 hour to a range of concentrations of CEP701 and PKC412, lysed and immunoprecipitated with an anti-FLT3 antibody. After sequential immunoblotting with anti-phosphotyrosine and anti-FLT3 antibodies, inhibition of FLT3 phosphorylation was measured by densitometry. Both drugs inhibited FLT3 phosphorylation in all samples with lower concentrations required in FLT3 mutants. CEP701 inhibited FLT3 phosphorylation with median IC50s of 3.7nM and 11.9nM in mutant and wild type (WT) cases respectively (p=0.0006). IC50s for PKC412 were 7.7nM and 59.8nM in mutant and WT cases (p=0.0268). Induction of cytotoxicity was assessed by MTS assay following 72-hour exposure of blasts to a range of concentrations of CEP701 and PKC412. Cytotoxic responses to both drugs were greater in FLT3 mutants than WT cases at each dose studied and in terms of IC50 dose (median IC50s in mutant and WT cases: 95nM and 231nM with CEP701, 1.24 μM and 1.61μM with PKC412) although these differences did not reach statistical significance. Annexin V binding apoptosis assay produced similar dose response curves. Both agents showed greater inter-case variability in cytotoxic response than in sensitivity to inhibition of FLT3 phosphorylation. A lack of cytotoxic response to FLT3 inhibition with CEP701 was seen in the ITD mutant with the lowest ratio of mutant to WT FLT3 RNA (0.08) and several WT samples displayed resistance to in vitro induction of cytotoxicity despite almost complete inhibition of FLT3. Induction of cytotoxicity with PKC412 in both mutant and WT cases generally required doses well in excess of those required to fully inhibit FLT3 phosphorylation. Cases were further stratified by flow cytometric measurement of surface FLT3 expression, and by immunoblotting to measure STAT5 dephosphorylation in response to both drugs. No significant difference in overall FLT3 expression was seen between mutant and WT cases. Interestingly the highest FLT3 expression level was seen in a wild type case that was highly sensitive to CEP701. Inhibition of STAT5 phosphorylation appeared closely linked to FLT3 inhibition, although in some cases a good cytotoxic response was achieved despite failure to inhibit STAT5, suggesting involvment of other signalling pathways. In summary, although both CEP701 and PKC412 predictably and reliably inhibit FLT3 phosphorylation in primary AML blasts, their induction of cytotoxicity appears to be much more variable. A number of factors may influence this including variations in level of dependency on FLT3 signalling for blast survival, mutant to WT allele ratio and overall FLT3 expression level. Effects on targets other than FLT3 also need to be considered.