Lymphoma Diagnosis and TreatmentRadiomics and Machine Learning in Medical ImagingCNS Lymphoma Diagnosis and Treatment

Guido Faggian, Roberto Faggian, Michela Salzano, C. Stavolo, Teresa Argenziano, A. Diglio, Angela Faggian

2026.2.4Italian Journal of Medicine

DOI: 10.4081/itjm.2026.2406

tlooto Summary

This systematic review synthesizes recent evidence from PubMed literature, highlighting how an optimal multimodal approach reduces mortality and improves progression-free survival, incorporating radiomics and artificial intelligence for more accurate predictions.

Abstract

The classification of lymphomas represents a crucial element in the clinical management of these heterogeneous tumors, profoundly influencing prognosis, follow-up, and therapeutic strategies through the integration of clinical, imaging, and molecular aspects. The 5th Edition of the World Health Organization Classification of Hematolymphoid Tumors (WHO-HAEM5) of 2022 introduced significant updates, recognizing new entities based on genomic and molecular data, such as high-grade B-cell lymphomas with MYC/BCL2 rearrangements and subtypes with non-germinal center B-cell-like phenotype. Fluorodeoxyglucose positron emission tomography/computed tomography imaging remains the essential tool for staging, therapeutic response assessment, and relapse detection, with sensitivity exceeding 90% in aggressive lymphomas such as diffuse large B-cell lymphoma (DLBCL) and Hodgkin lymphoma, surpassing the limitations of conventional computed tomography and predicting outcomes using metrics like SUVmax and Deauville score. Molecular aspects, including MYD88/CD79B mutations in DLBCL and markers such as circulating tumor DNA, refine risk stratification, guiding targeted therapies such as Bruton's tyrosine kinase inhibitors, venetoclax, or CAR-T cells. This systematic review synthesizes recent evidence (2020-2025) from PubMed literature, highlighting how an optimal multimodal approach reduces mortality and improves progression-free survival, incorporating radiomics and artificial intelligence for more accurate predictions. Challenges persist in access to advanced technologies and prospective validation, but the integration of these developments promises personalized management, reducing overtreatment and improving quality of life.

Citation format

FAGGIAN, Guido, et al. Integrating WHO-HAEM5 classification, 18F-FDG PET/CT imaging, and molecular profiling in lymphoma management: Implications for prognosis, follow-up, and personalized therapy (2020-2025 systematic review). Italian Journal of Medicine, 2026, 20(1).