NEXT-GENERATION DIAGNOSTICS IN ACUTE LYMPHOBLASTIC LEUKEMIA: FROM ULTRA-SENSITIVE MEASURABLE RESIDUAL DISEASE TO ARTIFICIAL INTELLIGENCE (2020–2026)

Abstract

Objectives: Acute lymphoblastic leukemia (ALL) diagnostics have changed rapidly in recent years, moving beyond conventional morphology, flow cytometry, and targeted molecular testing toward ultra-sensitive measurable residual disease (MRD) assays, genome-wide profiling, and artificial intelligence (AI). This review examines the major diagnostic advances introduced between 2020 and 2026 and assesses how close they are to routine clinical implementation. Methods: A structured narrative review of PubMed, Embase, Scopus, and Web of Science was performed, focusing primarily on studies published between January 2020 and August 2026, with earlier landmark studies included where necessary. Results: NGS-based MRD testing identifies residual leukemia beyond the sensitivity of conventional flow cytometry, including up to 46% of otherwise MRD-negative cases. Genome-wide approaches, particularly optical genome mapping and RNA sequencing, increase the detection of clinically relevant genomic abnormalities to approximately 90–95%. AI-based tools can substantially shorten flow-cytometric MRD analysis and support leukemia classification using routine laboratory data. However, their clinical readiness remains uneven, particularly for AI and liquid-biopsy approaches. Conclusions: Diagnostic technologies in ALL are advancing faster than the evidence guiding their clinical use. The next challenge is therefore not simply detecting more diseases but determining which findings should change treatment. Standardization, prospective validation, accessibility, and integration into clinical pathways will be essential for translating technological progress into meaningful patient benefits.