Impactof NOTCH1, FBXW7 and SETD2 Mutations On Early Treatment Response In Childhood T-Cell Acute Lymphoblastic Leukemia：A China Children’s Leukemia Group Study

T-cell lymphoblastic leukemia (T-ALL) is an aggressive malignancy accounting for about 15% of newly diagnosed acute lymphoblastic leukemia (ALL) in children. Although the prognosis has been improved by intensified therapies, the outcome of high risk patients is still not optimistic. Activating NOTCH1 and/or FBXW7 mutations have been reported to be the leading genetic abnormality in T-ALL with good prognosis but few studies about the role of NOTCH1 and FBXW7 in T-ALL were conducted in China and their mutation distributions and clinical prognosis are still unclear. On the other hand, loss of function mutations of SETD2, a histone methyltransferase specific for trimethylation of histone H3 on Lys36 (H3K36me3), has been found in renal cell carcinoma, gliomas and early T cell precursor ALL (ETP-ALL). SETD2 is a potential tumor suppressor gene and may play a deleterious role in human cancer. We attempted to investigate the role of these mutations in Chinese T-ALL children. Thirty-seven bone marrow samples from childhood patients with newly diagnosed T-ALL (26 boys and 11 girls; age＜14 years) enrolled into the China Children’s Leukemia Group (CCLG) ALL-2008 trail from September 2010 to December 2012 were analyzed for SETD2, NOTCH1, FBXW7 mutations. All exons of SETD2, exons 25-28, 34 of NOTCH1 and exons 5, 7-12 of FBXW7, including intron-exon boundaries, were sequenced by Sanger method.

NOTCH1 mutations were observed in 43.2% (n=16) of the T-ALL patients. Thirteen missense mutations were found in HD-N domain, HD-C domain and TAD domain of NOTCH1 while only frameshift insertions and deletions were identified in PEST domain. Besides missense mutations, five inframe insertions were also observed in HD-N domain of NOTCH1. We only found one FBXW7 synonymous mutation in our patients, which was extremely low compared with that in other studies. SETD2 somatic mutations were found in 8.1% (n=3, one female and two males) of 37 T-ALL patients. Of the three patients, two patients had truncated SETD2 protein because of nonsense mutation (c.6229 C＞T) and frameshift insertion (c.7516_7517insTTATA) respectively. Both SETD2 and NOTCH1 mutations were found in one patient. In our study, no significant relationships between NOTCH1 status and age, sex, mediastinal or Central Nervous System (CNS) involvement, immunophenotype and cytogenetics were observed. However, White Blood Cell (WBC) count at diagnosis in patients without NOTCH1 mutation were much higher (221×10⁹/L vs 76.95×10⁹/L, P=0.015). Patients with NOTCH1 mutations had a better early treatment response: higher prednisone good response rate (81.2% vs 47.6%, p=0.048), higher incidence of bone marrow M1 status (blast cell＜5%) on day 15 (73.3% vs 33.3%, p=0.04), and higher rate of favorable minimal residual disease (MRD) level on day 88 (100% vs 59.9%, p=0.012). In contrast, patients without NOTCH1 mutations were more likely to fall into the high risk group (71.4% vs 21.4%, p= 0.006) according to our treatment protocol. After a median follow-up of 11 months, 1 patient did not reach complete remission (CR), 2 patients died during induction and 2 patients relapsed in 6 months. Patients who suffered relapse and induction failure are all in the NOTCH1 wild type group. SETD2 mutations seemed to have no relationship to sex, age, WBC, cytogenetics, CNS and mediastinal involvement. In contrast to previous study that found SETD2 mutations only in EPT ALL patients, none of our SETD2 mutated patients were ETP ALL, two were cortical-T ALL and one was pre-T ALL. All of the patients with SETD2 mutations were alive without the disease at the last follow-up. However, all of the patients were in the high risk group. The patient who had both NOTCH1 and SETD2 mutations responded poorly to prednisone. Although the other two patients without NOTCH1 mutations had good response to prednisone, favorable BM status on day 15 and low MRD on day 33, they had very high MRD levels (0.21% and 0.44% respectively) on day 88.

NOTCH1 activating mutations were found in 43.2% of pediatric T-ALL patients enrolled in the CCLG ALL-2008 study with good treatment response while FBXW7 mutation rate was much lower. SETD2 mutation was a recurrent event in pediatric T ALL not only in ETP ALL. It seems that patients with SETD2 mutations have high risk of poor response to chemotherapy regardless of NOTCH1 status. Further studies are needed to find out the prognostic significance of SETD2 mutation in pediatric T ALL.