Survival Outcomes and Prognostic Factors in Elderly Patients with Hodgkin Lymphoma: A Retrospective Cohort Study

Background: Hodgkin lymphoma (HL) has a bimodal age distribution, and outcomes in patients aged ≥60 years remain inferior to those in younger cohorts because of a higher comorbidity burden, reduced treatment tolerance, and disease biology that is incompletely characterized in real-world elderly cohorts. Methods: We retrospectively analyzed 42 consecutive patients aged ≥60 years with Hodgkin lymphoma. Demographic, clinical, laboratory, and treatment data were extracted from medical records. Overall survival (OS) and progression-free survival (PFS) were estimated using the Kaplan–Meier method and compared with the log-rank test; univariable Cox proportional-hazards regression was used to identify factors associated with OS and PFS. Because complete response and refractory disease are determined only after treatment, and refractory disease is itself a component of the composite PFS endpoint, these two variables were excluded from the Cox and Kaplan–Meier prognostic models and are instead reported descriptively. Results: Median age at diagnosis was 68 years (range, 60–94); 71.4% of patients had a high or very-high Charlson Comorbidity Index, and 95.2% presented with B symptoms. Twenty patients (47.6%) died during follow-up; the Kaplan–Meier median overall survival (OS) was 88 months (95% CI, 25–not reached). Progression-free survival (PFS), defined as the composite of relapse, refractory disease, or death from any cause, whichever occurred first, was reached as an event in 21 patients (50.0%); the Kaplan–Meier median PFS was 72 months (95% CI, 8–not reached). On univariable Cox regression, older age (HR 1.09/year, p = 0.002), ECOG performance status with >50% of the day spent in bed (HR 22.4 for OS, p = 0.003; HR 11.1 for PFS, p = 0.002), high International Prognostic Score (IPS) (HR 11.1 for OS, p = 0.007; HR 17.7 for PFS, p = 0.003), stage IV disease (HR 6.24, p = 0.036), splenic involvement (HR 4.33, p = 0.002), extranodal involvement (HR 6.55, p = 0.001), lower hemoglobin (HR 0.80, p = 0.024), lower albumin (HR 0.32, p = 0.004), lower Prognostic Nutritional Index (HR 0.93, p = 0.006), and higher neutrophil-to-lymphocyte ratio (HR 1.07, p = 0.031) were associated with inferior OS. Complete response was achieved in 28 patients (66.7%) and primary refractory disease occurred in 6 patients (14.3%); both showed a marked descriptive association with outcome, but because these variables are assessed after treatment—and refractory disease is itself part of the PFS event definition—they were excluded from the Cox and Kaplan–Meier prognostic models to avoid guarantee-time bias and circularity, and are reported descriptively only. When PFS was defined using this composite endpoint rather than relapse alone, ECOG performance status, IPS risk group, and Ann Arbor stage were each significantly associated with both OS and PFS on Kaplan–Meier/log-rank testing (all p < 0.05)—indicating that performance status is not dissociated from progression risk, as a relapse-only PFS definition had previously suggested, but instead reflects both survival endpoints. Conclusions: In this real-world elderly HL cohort, performance status, IPS risk group, extranodal/splenic disease involvement, and inflammatory-nutritional biomarkers identified patients at risk of poor survival. The cohort also had a high comorbidity burden (71.4% high-risk and 19.0% very-high-risk Charlson Comorbidity Index), but sparse strata and unstable hazard-ratio estimates meant our data support describing the cohort as highly comorbid without demonstrating a statistically robust CCI–survival association in this dataset. Failure to achieve complete response and primary refractory disease were each associated with markedly worse outcomes descriptively, but were intentionally excluded from the formal prognostic models given their post-treatment, time-dependent nature and circular relationship with the PFS endpoint. These hypothesis-generating findings support incorporating geriatric and biomarker-based risk assessment alongside conventional staging when planning treatment for older patients with Hodgkin lymphoma, and warrant validation in larger prospective cohorts.