Prognostic Impact of PTPN11 mutations in Acute Myeloid Leukemia

Background: The classification of acute myeloid leukemia (AML) is being refined as knowledge of associated genetic alterations increases. The role of the protein tyrosine phosphatase non-receptor type 11 (PTPN11) oncogene is well-established in Noonan syndrome and juvenile myelomonocytic leukemia, although less is known about its implications in adult AML. PTPN11 is involved in RAS/MAPK signaling pathway regulation. In this study, we sought to characterize the co-mutational landscape and outcomes of PTPN11 mutations in patients with AML.

Methods: We retrospectively analyzed a database of 645 patients aged > 18 years with AML treated at our institution from 2014 to 2024. The PTPN11^wild-type cohort was selected to match age, ECOG score, and Charlson Comorbidity Index (CCI) score median and range of the PTPN11^mut cohort. Data was collected on molecular and cytogenetic profiles, European Leukemia Net (ELN) risk, response after induction therapy, and survival. Composite complete response (CCR) was defined as complete response, CR with incomplete hematologic recovery, or CR with partial hematologic recovery. Median relapse-free survival (mRFS) and overall survival (mOS) were compared using the Kaplan Meier method. RFS included patients who had primary refractory disease. Categorical variables were compared using Fisher's exact test. Continuous variables were compared using the unpaired t test.

Results:PTPN11 mutations were detected in 35 patients with newly diagnosed AML. Those who received intensive induction had a median age of 62 (range, 24-74), median ECOG score of 1 (range, 0-2), and median CCI score of 4 (range, 2-11). The PTPN11^mut non-intensive cohort had a median age of 78 (range, 70-85), ECOG of 2 (range, 1-3), and CCI of 7 (range, 5-12). Within a matched cohort of 216 patients with PTPN11^wild-type, both the intensive and non-intensive groups had similar characteristics to their PTPN11^mut counterpart . Co-occurring mutations in NPM1 were significantly enriched in the PTPN11^mut cohort compared to PTPN11^wild-type (46% v. 25%, p=0.015). There were no significant differences in cytogenetic profiles. The frequency of patients in each ELN genetic risk category was similar between the two cohorts. The mOS was 11.2 months in the PTPN11^mut cohort and 15.3 months in the PTPN11^wild-type cohort (HR=1.31, p=0.236). There were no significant survival differences between PTPN11^mut and PTPN11^wild-type when stratified by induction treatment intensity. The mOS of the PTPN11^mut cohort was significantly shorter than that of the PTPN11^wild-type favorable risk group (mOS=not reached [NR], HR=2.42, p=0.006) and PTPN11^wild-type intermediate risk group (mOS=23.8 months, HR=1.90, p=0.033) but similar to mOS of the PTPN11^wild-type adverse risk group (mOS=12.7 months, HR 0.94, p=0.80). Patients with PTPN11^mut favorable risk AML had significantly worse mRFS (6.5 v. 22.8 months, HR 2.24, p=0.041) and shorter mOS than those with PTPN11^wild-type favorable risk disease (21 months v. NR, HR 2.11, p=0.166), although the latter did not reach significance. Of note, all patients in the PTPN11^mut favorable risk group had NPM1 co-mutations. PTPN11^mut intermediate risk patients also had shorter mOS than PTPN11^wild-type intermediate risk patients (12.3 v. 23.8 months, HR 2.35, p=0.065), trending toward significance. There was no difference in mOS between PTPN11^mut adverse risk and PTPN11^wild-type adverse risk (9.5 v. 10.5 months, HR 1.17, p=0.594). PTPN11^mut with NPM1^wild-type (not stratified by ELN risk) had a mOS of 11.1 months, similar to that of PTPN11^mut with NPM1^mut (12.3 months, HR 0.64, p=0.336).

Conclusions: While PTPN11 mutations frequently co-occur with NPM1 mutations, they are also associated with higher risk mutations. PTPN11^mut is associated with shorter OS and earlier relapse in ELN favorable risk AML compared to favorable risk with PTPN11^wild-type, despite co-occurring NPM1 mutations. These patients should be considered for stem cell transplant in first remission. Co-occurring NPM1^mut alone does not improve the OS of PTPN11^mut. The OS of PTPN11^mut AML more closely resembles that of ELN adverse risk AML. Overall, PTPN11 mutations are associated with poor prognosis. Larger studies are warranted to support the incorporation of PTPN11 into the ELN criteria for higher risk AML.

No relevant conflicts of interest to declare.