CAR-T Cell Therapy Versus Salvage Chemotherapy in Relapsed/Refractory DLBCL: A Systematic Review and Meta-Analysis Integrating Radiologic, Laboratory, and Histopathologic Correlates.

Background: Relapsed or refractory diffuse large B-cell lymphoma (R/R DLBCL) remains a major therapeutic challenge, particularly among patients with high-risk molecular features or primary refractory disease. Chimeric antigen receptor T-cell (CAR-T) therapy has emerged as a promising treatment strategy; however, its comparative effectiveness versus salvage chemotherapy requires comprehensive evaluation. Methods: This systematic review and meta-analysis were conducted in accordance with PRISMA 2020 and MOOSE guidelines. Randomized controlled trials and high-quality comparative cohort studies evaluating CAR-T therapy versus salvage chemotherapy in adult patients with R/R DLBCL were included. Primary outcomes were overall survival (OS) and progression-free survival (PFS). Secondary outcomes included complete metabolic response (CMR), toxicity profiles, and multidisciplinary correlates (radiologic, laboratory, and histopathologic). Random-effects models, meta-regression analyses, and GRADE assessment were applied. Results: Eight studies (n = 2,150) were included. CAR-T therapy significantly improved PFS (HR 0.55, 95% CI 0.42-0.71; p < 0.001) and OS (HR 0.68, 95% CI 0.54-0.86; p = 0.001). CMR rates were higher in the CAR-T group (56.0%) compared with salvage chemotherapy (24.5%) (RR 2.28, 95% CI 1.82-2.86; p < 0.001). Elevated inflammatory markers and tumor burden were associated with increased risk of immune-related toxicities. Subgroup analyses demonstrated greater benefit in primary refractory and double-hit lymphoma. CAR-T therapy was associated with cytokine release syndrome (9% grade ≥3) and ICANS (12%), whereas salvage chemotherapy demonstrated higher rates of hematologic toxicity. Conclusions: CAR-T therapy provides superior survival and response outcomes compared with salvage chemotherapy in R/R DLBCL, particularly in high-risk populations. Integration of radiologic, laboratory, and molecular predictors may enhance patient selection and optimize toxicity management.