No Evidence That PTEN Abnormalities Impact on Outcome in Pediatric Patients with T-Cell Acute Lymphoblastic Leukemia (T-ALL) Treated on the MRC UKALL2003 Trial

Although outcome has improved for pediatric patients with T-ALL, ≈25% of cases relapse and prognosis post-relapse remains poor. Molecular characterisation at diagnosis can provide additional information for risk-stratification. We previously reported that patients with double NOTCH1 or NOTCH1+FBXW7 mutations (NOTCH1±FBXW7^Double) have a very good outcome and should not be considered for more intensive therapy in first remission, even if slow responders or MRD-positive after induction therapy. However, recent studies have suggested that this may be modulated by the presence of coincident abnormalities such as in the PTEN gene. Truncating mutations and genomic loss of this gene have been described in T-ALL, but their prognostic impact in patients is unclear, with reports of either no effect or reduced event-free survival (EFS). Furthermore, subgroup analysis has shown that the adverse impact of a PTEN mutation is either not seen in the presence of a NOTCH1 mutation or, conversely, that it ablates the benefit of a NOTCH1 mutation. In order to determine whether these abnormalities impact on the good outcome seen in NOTCH1±FBXW7^Double cases in our cohort, and whether they can refine stratification of cases with single NOTCH1 mutations (NOTCH1^Single) or wild-type NOTCH1 (NOTCH1^WT), we investigated PTEN genotype in samples from 145 patients treated on the MRC UKALL2003 trial and correlated this with outcome in the different subgroups.

The entire coding region (exons 1-9) of the PTEN gene was screened for mutations using heteroduplex analysis. Samples with abnormal chromatograms were further investigated. Mutations were detected in 21 patients (14%); 17 (81%) had exon 7 mutations, 2 exon 6+7, and 2 exon 5 mutations. All were small insertions, deletions or indels; 89% were predicted to lead to C-terminal truncation and loss of protein function, 11% were in-frame size changes. Exon 7 mutant levels were quantified by size analysis in 19 patients; median total mutant level was 48% of all PTEN alleles (range, 10%-96%). Of note, in the 21 mutated cases, only 7 (33%) had a single mutation; 8 had 2, 3 had 3, and 3 had 4 mutations. Based on total mutant level, 11 cases were considered to have monoallelic (heterozygous) mutations and 10 cases biallelic (homozygous/hemizygous or compound heterozygous) mutations. To investigate loss of genomic material, Illumina CytoSNP-850k SNP array analysis was performed on all samples. Partial or complete loss of the PTEN gene was detected in 15 patients (10%), 12 with heterozygous and 3 homozygous loss. This data was consistent with quantitative analysis of the relative allele levels of two common intronic SNPs (rs1903858 and rs555895) studied in 76 informative patients, which indicated that in 2 of 6 informative cases with heterozygous loss, only the 3’ end of the gene was deleted. Putting together the mutation and SNP data, 32 patients (22%) had abnormalities in the PTEN gene (PTEN^ABN), 19 (59%) scored as monoallelic (PTEN^MONO), and 13 (41%) biallelic (PTEN^BI).

There was no significant difference according to overall PTEN genotype in either early response to therapy (P>.99) or MRD status at day 29 of induction therapy (P=.28). Long-term outcome also did not significantly differ, EFS at 5 years was 78% in PTEN^ABN and 85% in PTEN^WT patients (P=.37), overall survival (OS) 81% and 91% respectively (P=.1). These results did not change if grouped according to PTEN type, EFS 74% in PTEN^MONO and 85% in PTEN^BI patients (P=.46), although the number of patients in these groups was very small. The incidence of PTEN abnormalities did not differ according to NOTCH1/FBXW7 genotype, 59% PTEN^ABN patients had a NOTCH1/FBXW7 mutation compared to 67% PTEN^WT patients (P=.7). There was no evidence that PTEN genotype impacted on the favorable outcome of the NOTCH1±FBXW7^Double group, none of the 5 PTEN^ABNNOTCH1±FBXW7^Double patients relapsed and all remain alive. Similarly, no significant difference was observed in the NOTCH1^Single and NOTCH1^WT patients.

In conclusion, although we found that loss of PTEN through either gene mutation or genomic deletion was relatively common in pediatric T-ALL patients, including total loss of PTEN in nearly one-half of mutated patients, this appeared to have no effect on either response to treatment or long-term outcome with current therapies, and therefore screening of PTEN is not warranted in pediatric T-ALL for potential use in risk-adapted therapy.