Acute Myeloid Leukemia (AML) is a heterogeneous hematologic malignancy characterized by multiple genetic abnormalities. The Internal Tandem Duplication (ITD) mutation of the FLT3 gene is one of the most common and prognostically unfavorable molecular alterations.

Although several studies have investigated the effects of FLT3-ITD length and genomic insertion site, the literature reports conflicting evidence regarding their differential impact on disease behavior and patient prognosis. This highlights the need for a more thorough understanding of these variables. In this single-Center observational study, we aim to evaluate the impact of FLT3-ITD length and insertion site on clinical outcomes in patients (pts) with AML.

Over 5 years, we analyzed a cohort of 384 newly diagnosed adult AML patients, prospectively and retrospectively. Of these, 72 (19%) harbored the FLT3-ITD mutation, as detected by NGS or RT-PCR. The length of the ITD, measured in base pairs (bp), was correlated with clinical outcomes and biological variables to identify potential patterns and interactions. For 37 pts who underwent NGS (51%), the precise mutation locus was characterized to provide further insight into its genomic context. To stratify pts and reinforce the clinical relevance of ITD length, an optimal cut-point was determined using maximally selected rank statistics (MAXSTAT).

The median age was 64 years. NPM1 was the most frequently mutated gene by RT-PCR (36/72, 50%), while NGS identified NPM1 (22/37, 59%) and DNMT3A (16/37, 43%) as the most common co-mutations. Six pts exhibited multiple ITDs. The median ITD length was 48 bp (IQR 30-69). Exon 14 was affected in 35 out of 37 (95%) cases, while exon 15 has been observed to be involved in a smaller portion (2/37, 5%). Additionally, 6 out of 37 (16%) of cases had intronic involvement. MAXSTAT analysis resulted in the identification of 63 bp as threshold discriminating two distinct pts' groups. Notably, the two groups did not differ significantly in terms of baseline characteristics, supporting the robustness of the stratification method employed.

FLT3-ITD length was found to be associated with an increased risk of relapse. Indeed, pts who relapsed had a median length of 71 bp (IQR 47-75) strongly higher compared to a median of 42 bp (IQR 27.8-63), of those who did not relapsed (p=0.018, OR 1.03). Moreover, the 5-year Cumulative Incidence of Relapse was significantly higher in pts with ITD above 63 bp (sHR=3.57; 95% CI: 1.21-10.56; p=0.0216), together with a trend toward worse Overall Survival and Event-Free Survival.

In addition, when combining age and ITD length in the analysis, the negative prognostic impact of longer ITDs on survival remained significant even in older pts, who generally have a poorer baseline prognosis (p=0.0003). This suggests that ITD length is an independent predictor of outcome regardless of patient age.

Analysis of the mutation locus revealed that pts with long ITDs were significantly more likely to have intronic involvement (p=0.001).

No differences were observed regarding the number of unique co-mutations and ITD length overall. However, a noteworthy association emerged between ITD length and DNMT3A mutation status: DNMT3A-mutated pts exhibited longer ITD on average, and were approximately five times more likely to have an ITD length greater than 63 bp.

Five-year OS analysis revealed that pts harboring both DNMT3A mutations and an ITD length ≥ 63 bp (10/37, 27%) had a significantly higher risk of mortality compared to other subgroups (p=0.02). A further Kaplan-Meier survival analysis, limited to pts with DNMT3A mutation, showed a significant difference between those with ITD greater than 63 bp and those with ITD lesser than 63 bp (p=0.03).

In conclusion, this study suggests that longer FLT3-ITDs are associated with a higher incidence of relapse and poorer clinical outcomes, supporting their relevance for risk stratification. We report, for the first time, a prognostically significant association between DNMT3A mutation and ITD length. Additionally, the longer ITDs involved intronic regions more than shorter ones, probably leading to distinct protein structure. Further studies on larger cohort are needed to confirm the data and, eventually, integrate ITD length, location, and the co-occurrence of DNMT3A mutations into prognostic models to help refine risk stratification and treatment decisions, including early intensification or allogeneic stem cell transplantation.

This content is only available as a PDF.
2025
Sign in via your Institution