OLFML2A expression in acute myeloid leukemia, with exploratory findings in acute lymphoblastic leukemia: Prognostic significance and re‑analysis of public single‑cell transcriptomic data

This retrospective study analyzed peripheral blood mononuclear cell samples from 34 patients with acute leukemia [26 with acute myeloid leukemia (AML) and 8 with acute lymphoblastic leukemia (ALL)] and 15 healthy controls to investigate olfactomedin‑like 2A (OLFML2A) expression and its potential clinical significance. OLFML2A mRNA expression was significantly higher in the patient cohort than in the control group, and expression was higher in AML than in ALL. In the institutional cohort, elevated OLFML2A expression in AML was associated with fusion‑gene positivity, chromosomal abnormalities and adverse‑risk disease. Findings in ALL were based on a small sample (n=8) and should therefore be regarded as exploratory. Although a possible subtype‑dependent difference between AML and ALL was considered, the ALL subgroup was too small to support a definitive directional conclusion. This observation should therefore be regarded as hypothesis‑generating and may further support disease‑specific OLFML2A biology. Receiver operating characteristic analysis in the local cohort suggested preliminary diagnostic discrimination, particularly for AML, although no independent external diagnostic validation cohort was available. In the combined local cohort, higher OLFML2A expression was associated with shorter progression‑free survival, whereas overall survival did not differ significantly. Because the TCGA validation cohort consisted of AML cases only, all public‑database survival analyses were restricted to AML and supported OLFML2A as an adverse prognostic factor in that disease. Re‑analysis of public single‑cell transcriptomic datasets detected OLFML2A expression in natural killer cells, monocytes/macrophages, endothelial cells, hematopoietic progenitor‑related populations and malignant cells, providing supportive transcriptomic context rather than direct mechanistic validation. Overall, these findings support OLFML2A as a candidate biomarker in AML, whereas the observations in ALL remain preliminary and require validation in larger cohorts.