Epstein-barr virus-encoded small RNA incorporated prognostic model enables precision risk stratification and guides treatment decisions in peripheral t-cell lymphoma: a multicenter retrospective cohort

Epstein-Barr virus-encoded small RNA (EBER), a hallmark of EBV infection, is a poor prognostic factor in peripheral T-cell lymphoma (PTCL). Current clinical-based prognostic scores inadequately identify high-risk EBER-positive patients or guide therapy, underscoring the need for improved risk-stratification and treatment strategies. This multicenter cohort study, encompassing 167 PTCL patients, systematically analyzed the impact of EBER status on patient survival and treatment response. Utilizing LASSO-penalized Cox regression, a novel prognostic risk scoring system was developed incorporating EBER status and clinical indicators. With a median follow-up of 22.1 months, 63 patients (38%) died. The objective response rate was 57%. EBER-positive status was associated with older age, hypoalbuminemia, high IPI scores, shorter median overall survival (mOS), higher positivity rates for CD30, CD4, BCL6, PD-1, and poorer response to first-line chemotherapy. Multivariate analysis identified independent adverse prognostic factors (p < 0.05): Albumin < 40, Platelet-to-Monocyte Ratio ≤ 300, Lactate Dehydrogenase > 250, Age > 70, and EBER-positivity. A prognostic model incorporating these factors stratified patients into three significantly distinct risk groups (p < 0.001): Low-risk (n = 45; 3-year OS 87.6%, mOS not reached), Intermediate-risk (n = 60; 3-year OS 49.7%, mOS 32.8 months), High-risk (n = 62; 3-year OS 25.1%, mOS 14.3 months). The model outperformed existing models and demonstrated excellent discrimination, stability, clinical utility across PTCL subgroups. This novel prognostic score, integrating subtype-specific marker and clinical features, provides a refined framework for precise PTCL risk stratification and treatment guidance.