Multilineage Dysplasia as Assessed by Immunophenotype in Acute Myeloid Leukemia: A Prognostic Tool in a Genetically Undefined Category

• Published in: Cancers 2020-10, Vol.12 (11), p.3196
• Main Authors: Mannelli, Francesco, Bencini, Sara, Piccini, Matteo, Gianfaldoni, Giacomo, Bonetti, Maria Ida, Peruzzi, Benedetta, Caporale, Roberto, Scappini, Barbara, Pancani, Fabiana, Ponziani, Vanessa, Signori, Leonardo, Zizza, Michela, Annunziato, Francesco, Bosi, Alberto
• Format: Article
• Language: English
• ISSN: 2072-6694; 2072-6694
• DOI: 10.3390/cancers12113196

Acute myeloid leukemia (AML) “with myelodysplasia-related changes (MRC)” is considered a separate entity by the World Health Organization (WHO) classification of myeloid neoplasms. While anamnestic and cytogenetic criteria provide objective attribution to this subset, with clear unfavorable prognostic significance, the actual role of multi-lineage dysplasia (MLD) as assessed by morphology is debated. The aim of our work was to study MLD by a technique alternative to morphology, which is multiparameter flow cytometry (MFC), in a large series of 302 AML patients intensively treated at our Center. The correlation with morphology we observed in the unselected analysis reiterated the capability of the MFC-based approach at highlighting dysplasia. MLD data, estimated through an immune-phenotypic score (IPS), provided no insight into prognosis when considered overall nor within well-defined genetic categories. Of interest, IPS-related dysplasia conveyed significant prognostic information when we focused on genetically undefined patients, triple-negative for NPM1, FLT3 and CEBPA (TN-AML). In this context, the lack of dysplastic features (IPS_0) correlated with a significantly higher CR rate and longer survival compared to patients showing dysplasia in one or both (neutrophil and erythroid) cell lineages. The impact of IPS category maintained its validity after censoring at allogeneic HSCT and in a multivariate analysis including baseline and treatment-related covariates. In a subgroup featured by the lack of genetic determinants, our data could help address the relative unmet needs in terms of risk assessment and treatment strategy, and provide insight into prediction of response in the rapidly evolving therapeutic scenario of AML.