Qiuyu Zhang, Wei Fang, Yanyan Liu
Peripheral T-cell lymphoma (PTCL) is a heterogeneous group of aggressive malignancies with poor prognosis and limited response to standard chemotherapy. Although recent advances in molecular and biological characterization have refined PTCL classification, profound molecular heterogeneity continues to result in highly variable treatment outcomes. Emerging targeted and immunotherapeutic agents show activity in specific subtypes, yet responses remain difficult to predict amid their biological complexity. This review synthesizes current molecular insights, evaluates the limitations of existing therapies, and explores the potential of multi-omics profiling integrated with artificial intelligence (AI) to decode complex signatures for improved subtype classification, treatment response prediction, and survival prognostication. By addressing the "black-box" issue through explainable AI, this approach offers a promising framework to advance PTCL management from empirical treatment toward biomarker-driven precision therapy.