The Favorable Effect of Activating NOTCH1 Receptor Mutations On Long Term Outcome in T-ALL Is Treatment Related and Can Be Separated From NOTCH Pathway Activation by FBXW7 Loss of Function.

Abstract 908

Precursor T-cell acute lymphoblastic leukemia (T-ALL) remains one of the challenges in pediatric oncology. While overall treatment results have considerably improved over recent years, T-ALL relapses are still frequently fatal. Therefore, it is vital to identify molecular risk factors that allow early and effective treatment stratification. We have recently shown that activating NOTCH 1 receptor mutations signify a favorable prognosis in patients treated on protocols of the ALL-BFM study group, although this effect has not been found with other treatment protocols (see abstract submitted by Zuurbier et al on behalf of the DOC and COALL study groups). We have now tested the hypothesis that activation of the Notch pathway at different steps may have similar clinical effects and may act in a synergistic fashion. An analysis of an extended data set of 301 children with T-ALL, who were enrolled in the ALL-BFM 2000 protocol confirms the overall favorable effect of activating NOTCH 1 mutations (pEFS 0.87 SE 0.03 vs. 0.74 SE 0.04 p=0.0091; relapse incidence at 5 years 0.07 SE 0.02 vs. 0.17 SE 0.03, p= 0.006). Interestingly, subgroup analysis now reveals that the NOTCH 1 effect can only be observed in the cohort of 168 patients with a rapid early treatment response (pEFS 0.95 SE 0.03 in the Notch 1 mutated vs. pEFS 0.80 SE 0.05 in the non-mutated group, p=0.01, incidence of CNS relapse 0.01 vs. 0.03), who were stratified into the ALL-BFM 2000 standard and medium risk groups. By contrast, there is no significant effect of NOTCH 1 mutations on long term outcome in the cohort of 133 patients, who had initially responded more slowly and were stratified into the BFM high-risk group (pEFS 0.70 both, with and without NOTCH 1 mutations). In striking contrast to the BFM patients, EORTC high-risk patients (who do not receive prophylactic cranial radiotherapy) with NOTCH 1 mutations have a dismal prognosis and tend to develop CNS relapses (see abstract submitted by Clappier et al for the EORTC study group), whereas CNS relapses are rare in BFM-treated patients in general but also in the high-risk group, who received prophylactic cranial radiotherapy. A possible mechanistic link between NOTCH 1 activation and T-ALL CNS tropism has recently been suggested by the finding that the chemokine receptor CCR7 is controlled by NOTCH 1 and required for targeting leukemic cells to the CNS (Buonamici et al., Nature, 2009). We thus suggest that the influence of NOTCH 1 receptor mutations on long term outcome is likely treatment dependent.

We next analyzed the effect of inactivating mutations of the E3 ubiquitin ligase FBXW7 in 294 patients of whom sufficient DNA was available. Mutations were found in 42 (14%) patients. Interestingly, 88% (37/42) of the patients with FBXW7 mutations showed a prednisone good response (PGR), whereas only 12% (5/42) showed a prednisone poor response (PPR). By contrast, in the group without FBXW7 mutations 55% (138/252) showed a PGR and 45% (114/252) a PPR (p = 0.00003). With regard to prednisone response FBXW7 and NOTCH 1 mutations (both present in 30 patients) clearly synergize (PGR in 93%, 28/30; PPR in 7%, 2/30). The MRD kinetics were available in 39/42 patients with FBXW7 mutations. A favorable minimal residual disease (MRD) response on days 33 and 78 was observed in 16/39 and in 28/39 patients, respectively, and an unfavorable MRD response in 23/39 and 11/39 patients, respectively. On day 33, the difference between the FBXW7-mutated and non-mutated groups was significant (p=0.03), whereas FBXW7 mutations lost their significant predictive value for a favorable early treatment response on day 78, although there was still a trend (p=0.07). Inactivation of FBXW7 is thus associated with a rapid early treatment response in a univariate analysis and also significantly synergizes with activation of the NOTCH 1 receptor in this regard. Surprisingly, however, FBXW7 inactivation does not synergize with NOTCH 1 mutations to predict a significantly more favorable long term outcome (pEFS 0.86 in both, NOTCH 1 only and in NOTCH 1/FBXW7 double mutations). In conclusion, the favorable long term effect of NOTCH 1 receptor gain of function is specified to operate only in those patients with rapid early treatment response. This effect is separable from that of FBXW7 inactivation, which predicts excellent early treatment response but does not add to the favorable long term treatment outcome that is specified by NOTCH 1 receptor mutations alone.