Dynamic prognostic stratification and survival prediction in elderly patients with diffuse large B-cell lymphoma: a real-world evidence study

Elderly patients with diffuse large B-cell lymphoma (DLBCL) show substantial heterogeneity in disease characteristics, treatment tolerance, and survival outcomes. This study aimed to identify prognostic factors and evaluate their time-dependent impact on survival in elderly DLBCL patients. This retrospective study included 120 elderly DLBCL patients diagnosed between January 2020 and February 2026. Clinical, pathological, and treatment-related characteristics were collected. Patients were classified according to first-line treatment strategy as R-CHOP, reduced-dose/palliative chemotherapy, or non-chemotherapy. Kaplan–Meier analysis, log-rank tests, and multivariable Cox regression models were performed to evaluate factors associated with progression-free survival (PFS) and overall survival (OS). The 1-, 2-, and 3-year PFS rates were 76.67%, 58.33%, and 47.50%, respectively, while corresponding OS rates were 81.67%, 67.50%, and 59.17%. Higher IPI scores were independently associated with inferior PFS (HR = 25.28, P < 0.001) and OS (HR = 55.77, P < 0.001). Component-based analysis identified advanced age, Ann Arbor stage III–IV, multiple extranodal involvement, and elevated LDH as independent predictors of poorer survival. The prognostic relevance of individual factors varied across follow-up periods. Patients receiving R-CHOP had better 3-year OS than those receiving reduced-dose/palliative chemotherapy or no chemotherapy (75.00% vs. 46.15% vs. 15.38%, P < 0.001). In elderly DLBCL patients, IPI remains a robust prognostic tool, while individual clinical factors demonstrate differential prognostic relevance over time. Tumor burden, disease extent, and host vulnerability jointly contribute to long-term outcomes. Treatment strategies should be individualized considering disease risk and patient fitness.