The presence of secondary type or DNMT3A mutations does not affect prognosis in NPM1 mutated AML treated with intensified fludarabine-high dose cytarabine and idarubicine (FLAI) induction regimen. Free

Background: Acute myeloid leukaemia (AML) with mutated Nucleophosmin 1 (NPM1) is recognized as a distinct subtype of AML in the 2022 WHO and ICC classification, characterized by favourable clinical outcome. However, evidence from studies shown that NPM-mut AML shows biological heterogeneity and co-occurring gene mutations could modulate the prognosis.

Despite several studies suggested that other molecular features such as DNMT3A mutations could be associated with worse outcome in NPM-mut AML, only FLT3-ITD is currently considered for further risk stratification in NPM-mut AML.

In 2022, European LeukemiaNet (ELN) classification recognized a panel of mutations associated with secondary AML, including SRSF2, SF3B1, U2AF1, ZRSR2, ASXL1, EZH2, BCOR, and STAG2 (secondary-type mutation, STMs). These mutations, in absence of other driver mutations such as NPM1, have been assigned to the adverse-risk group; however, their prognostic relevance in the specific context of NPM-mut AML is not defined.

Moreover, most data on the prognostic role of gene mutations and co-mutations have been derived from patients treated with the conventional “3+7” induction. It has reported that FLAI (fludarabine, high-dose cytarabine, and idarubicin) regimen may overcome the negative prognostic impact of FLT3-ITD in NPM-mut AML, but there are still no data on the impact of STMs or DNMT3A mutation in this setting.

Aim: The aim of this study was to evaluate the prognostic impact of co-occurring STM and/or DNMT3A mutation in a cohort of patients with NPM-mut AML treated with FLAI regimen.

Methods: We retrospectively analysed 53 adult patients (median age: 55 years; range: 29–60) with NPM-mut AML diagnosed and classified according to the WHO 2022 classification. All patients received induction chemotherapy with FLAI regimen in our centre.

Next-generation sequencing (NGS) was performed on bone marrow samples at diagnosis using the Myeloid Solution panel by SOPHiA Genetics, encompassing 34 critical genes mutations. Sample were processed on Illumina MiSeq platform and analysis was performed with SOPHiA DDM® Software. Molecular MRD was assessed in all patients who achieved complete remission (CR), by reverse transcription polymerase chain reaction (RT-PCR) assays targeting NPM1 mutations. Measurable residual disease MRD negativity was defined according to standardized European LeukemiaNet thresholds.

Results: Of the 53 patients evaluated, 6 (11%) had one or more STM, whereas 9 (17%) had DNMT3A mutations. Three patients (6%) presented with both STMs and DNMT3A mutations. FLT3-ITD was identified in 22 out of 53 tested patients (42%). High-risk cytogenetic abnormalities, specifically complex hyperdiploid karyotype, were detected in 3 patients (6%).

After the first cycle of FLAI, 48 patients (91%) achieved a complete morphological remission (CR), 3 patients (6%) were resistant, and 2 patients (3%) died before treatment response could be evaluated. Among those achieving CR, MRD was assessed after the second cycle and found to be negative in 44 of 53 (83%) patients. Notably, the presence of STMs, DNMT3A mutations, FLT3-ITD, or high-risk cytogenetics did not significantly influence CR rates or the likelihood of achieving MRD negativity.

The three patients who were refractory to FLAI had normal karyotype and did not have STM sor DNMT3A mutation; however, all showed KMT2A (MLL) mutations.

After a median follow-up of 61 months (95%: 43.7–69.4), median overall survival (OS) and median disease-free survival (DFS) were not reached.

The presence of STMs, DNMT3A, FLT3-ITD mutations, or adverse-risk cytogenetics had no statistically significant impact on long-term OS or DFS (p=not significant). However, MRD negativity after the first treatment cycle was strongly associated with improved survival outcomes, including significantly longer OS and DFS (p<0.05), highlighting the prognostic value of early molecular response.

Conclusions: Our findings suggest that the use of the FLAI induction regimen is highly effective in patients with NPM-mut AML, with high rates of MRD-negative remission and durable survival, regardless of concomitant presence of adverse molecular features such as STMs, DNMT3A, or FLT3-ITD mutation.

Very few data are available about prognostic impact of KMT2A mutations. In our cohort, although rare, KMT2A mutations appeared to be associated with resistance to FLAI, suggesting a potential role as a predictive molecular marker for treatment failure.