Characteristics and Predictors of Early Deaths in Newly Diagnosed Classical Hodgkin Lymphoma

Background：Despite the growing availability and utilization of new therapies for classical Hodgkin lymphoma (cHL), early death from the disease continues to be a significant challenge, especially among older patients. Previous studies have rarely focused on early mortality in patients with cHL.

Method: A retrospective analysis was conducted on the clinicopathological data of primary classical Hodgkin lymphoma (cHL) patients from 17 registries in the Surveillance, Epidemiology and End Results (SEER) database between 2000 and 2020. The patients were randomly divided into training and validation cohorts in a 7:3 ratio. In the training cohort, multivariate logistic regression analysis was used to identify risk factors for early mortality in cHL patients. A nomogram prediction model was then constructed. The model was evaluated using receiver operating characteristic (ROC) curves, calibration curves, and decision curve analysis (DCA). External validation was performed using clinical information from 297 newly diagnosed cHL patients at Tianjin Medical University Cancer Institute and Hospital (TMUCIH) between 2010 and 2022. Early death (ED) is defined as death within three months of diagnosis.

Results：In the SEER database from 2000 to 2020, spanning 17 registries, there were 23,921 patients with classical Hodgkin lymphoma. Among these, 6,792 patients died, with 794 (11.7% of all deaths) occurring within three months of diagnosis. Of these early deaths, 638 (79.6%) were cancer-specific. Notably, 71.79% (570 cases) of the 794 early death patients were aged 60 or older. In the TMUCIH cohort, there were 13 cases of early deaths, of which 9 were elderly patients. Since information on B symptoms was included in the SEER database starting from 2010, we selected patients from the 2010-2020 SEER database to construct the logistic regression model. Patients were divided into two cohorts: an elderly cohort (≥60 years) and a younger cohort (<60 years). At baseline, the elderly cohort had a higher proportion of Black patients, a higher incidence of mixed cellularity subtype, and more advanced staging compared to the younger cohort. Comparing the clinical characteristics of the younger cohort in the SEER database with the younger cohort in TMUCIH revealed no significant differences. The same was true for the older cohorts. Logistic regression analysis showed that in the younger cohort, mixed cellularity, unclassified subtypes, B symptoms, and advanced stage were closely associated with all-cause early mortality. Male gender, unclassified subtypes, and advanced stage were independent risk factors for cancer-specific early mortality. In the older cohort, Asian and American Indian race, presence of B symptoms, mixed cellularity, and unclassified subtypes increased the risk of cancer-specific early mortality. Importantly, the use of chemotherapy was associated with reduced risks of both all-cause early mortality and cancer-specific early mortality in both cohorts. The nomogram demonstrated high predictive accuracy for early mortality in cHL patients, as confirmed by internal and external validation.

Conclusions: This study demonstrates from both the SEER database and real-world data that advanced age is a significant high-risk factor for early mortality in patients with classical Hodgkin lymphoma. The models predicting early mortality differ between elderly and younger patients. This study highlights the critical importance of supportive interventions for both young and elderly patients with classical Hodgkin lymphoma.

Keywords: Classical Hodgkin lymphoma, older adults, early death, predictors, real-world data, SEER