Blood-Based Biomarkers for the Early Diagnosis of Pediatric Acute Lymphoblastic Leukemia: A Literature Review

Abstract

As the most common childhood malignancy, pediatric acute lymphoblastic leukemia (ALL) demands timely and accurate diagnosis to achieve favorable clinical outcomes. Because current diagnostic methods rely heavily on invasive bone marrow aspiration, reliable and minimally invasive alternatives are urgently needed. This review aims to synthesize and critically evaluate the current evidence on blood-based biomarkers, including circulating RNAs, proteins, metabolites, autoantibodies, and artificial intelligence-assisted diagnostic models, for the early diagnosis of pediatric ALL. Circulating RNA biomarkers and multi-marker diagnostic approaches appear to hold the greatest potential, although the existing evidence remains limited by small, single-center studies, methodological heterogeneity, and insufficient external validation. Several individual markers, including microRNA-146a and a multi-index artificial intelligence model built on routine hematological data, achieved area-under-the-curve values exceeding 0.95, underscoring the diagnostic promise of both molecular and computational approaches. Overall, blood-based biomarkers represent a promising complement to current diagnostic strategies, but standardized multicenter validation and robust multi-marker approaches are required before routine clinical implementation. Continued research in this field could improve early detection while reducing the need for invasive diagnostic procedures in children.