Abstract
This cohort study evaluates whether adult survivors of Hodgkin lymphoma have excess mortality compared with the general population.
Introduction
Treatment advances in Hodgkin lymphoma (HL) have driven 5-year relative survival rates close to 90% in the United States,1 resulting in a growing population of long-term survivors at risk for late effects that contribute substantially to mortality.2,3,4,5 Although previous studies have reported increased all-cause mortality among HL survivors,4 these estimates are largely derived from populations that do not adequately represent contemporary long-term survivors diagnosed during adulthood.3,4,5,6,7,8 Most prior reports have focused on childhood cancer survivors,6 patients evaluated during early follow-up,7 or those treated before the widespread adoption of contemporary strategies such as positron emission tomography–adapted therapy or involved-field radiation.6,7,8 As a result, existing mortality estimates have limited applicability for informing survivorship care among adults. To address this gap, we conducted a population-based analysis to quantify excess all-cause mortality among 5-year HL survivors diagnosed during adulthood compared with the general US population.
Methods
We conducted a retrospective cohort study using data from 16 population-based registries of the Surveillance, Epidemiology, and End Results (SEER) program. Eligible patients were aged 20 years or older at diagnosis with first-primary HL diagnosed during 2000 to 2017, received chemotherapy with or without radiotherapy, and survived at least 5 years after diagnosis. Follow-up was available up to December 2023. Race and ethnicity was obtained from the SEER database variable to identify demographic groups at higher risk of all-cause mortality. This study was deemed exempt from review and the requirement for informed consent due to the use of publicly available deidentified data by University of Texas MD Anderson Cancer Center Institutional Review Board. This research follows the STROBE reporting guideline.
Standardized mortality ratios (SMRs) and excess absolute risks (EARs) per 10 000 patient-years were estimated using expected deaths from US population mortality rates matched by age, sex, race and ethnicity, and calendar period, assuming a Poisson distribution. Follow-up was censored 5 years after the index date to ensure comparable estimates across broad diagnostic periods (2000-2007 vs 2008-2017). SMR heterogeneity across subgroups was assessed using likelihood ratio tests comparing a null model with those adjusted for the variable. Cumulative incidence of mortality was estimated using competing-risks methods. Two-sided P < .05 was considered statistically significant. Analyses were conducted in R version 4.5.2 (R Project for Statistical Computing).
Results
Among 20 057 HL survivors, the median (IQR) age was 36 (27-50) years (53.7% male; 15.0% Hispanic, 10.3% non-Hispanic Black, and 69.2% non-Hispanic White; 63.4% with nodular sclerosis) (Table). Mortality was significantly higher among HL survivors compared with the general US population (2899 observed vs 1480.8 expected deaths; SMR, 1.96; 95% CI, 1.89-2.03), corresponding to an EAR of 81.9 (95% CI, 75.8-88.0) per 10 000 patient-years. Relative mortality risk was higher for all major causes of death, primarily from malignant conditions. However, the largest absolute number of excess deaths was attributable to other noncancer and cardiovascular causes (Table).
Table. Patient Characteristics, Standardized Mortality Ratios, and Absolute Excess Risk Per 10 000 by Patient Factors From 16 SEER Registries, 2000-2017.
| Characteristic | All patients, No. (%) | Deaths, No. | Person-years | SMR (95% CI) | P for heterogeneity | EAR (95% CI) | |
|---|---|---|---|---|---|---|---|
| Observed | Expected | ||||||
| Overall | 20 057 (100) | 2899 | 1480.8 | 173 243.1 | 1.96 (1.89-2.03) | NA | 81.9 (75.8-88.0) |
| Age at diagnosis, y | |||||||
| Median (range) | 36 (20-89) | NA | NA | NA | NA | NA | NA |
| 20-39 | 11 497 (57) | 676 | 268.2 | 106 887.3 | 2.52 (2.34-2.72) | <.001 | 38.2 (33.4-42.9) |
| 40-59 | 6005 (30) | 1026 | 521.6 | 51 735.3 | 1.97 (1.85-2.09) | 97.5 (85.4-109.6) | |
| 60-89 | 2555 (13) | 1197 | 691.0 | 14 620.5 | 1.73 (1.64-1.83) | 346.1 (299.7-392.5) | |
| Sex | |||||||
| Female | 9480 (47) | 1072 | 576.6 | 83 831.7 | 1.86 (1.75-1.97) | .03 | 59.1 (51.4-66.7) |
| Male | 10 577 (53) | 1827 | 904.1 | 89 411.4 | 2.02 (1.93-2.12) | 103.2 (93.8-112.6) | |
| Race and ethnicity | |||||||
| Hispanic | 3001 (15) | 384 | 146.8 | 23 444.8 | 2.62 (2.37-2.89) | <.001 | 101.2 (84.8-117.6) |
| Non-Hispanic American Indian or Alaska Native | 87 (<1) | 21 | 6.3 | 697.1 | 3.31 (2.16-5.08) | 210.4 (81.5-339.2) | |
| Non-Hispanic Asian or Pacific Islander | 1029 (5) | 102 | 28.8 | 8273.4 | 3.54 (2.91-4.29) | 88.4 (64.5-112.3) | |
| Non-Hispanic Black | 2061 (10) | 330 | 162.1 | 17 292.6 | 2.04 (1.83-2.27) | 97.1 (76.5-117.7) | |
| Non-Hispanic White | 13 879 (69) | 2062 | 1136.7 | 123 535.2 | 1.81 (1.74-1.89) | 74.9 (67.7-82.1) | |
| Year of diagnosisa | |||||||
| 2000-2007 | 8578 (43) | 683 | 263.7 | 40 707.4 | 2.59 (2.4-2.79) | <.001 | 103.0 (90.4-115.6) |
| 2008-2017 | 11479 (57) | 755 | 392.8 | 45 770.1 | 1.92 (1.79-2.06) | 79.1 (67.4-90.9) | |
| Hodgkin lymphoma subtype | |||||||
| NSHL | 12 711 (63) | 1446 | 761.8 | 116 979.2 | 1.90 (1.80-2.00) | .03 | 58.5 (52.1-64.9) |
| MCHL | 2287 (11) | 564 | 269.6 | 19 624.3 | 2.09 (1.93-2.27) | 150 (126.3-173.7) | |
| LRHL | 553 (3) | 99 | 62.6 | 4670.4 | 1.58 (1.30-1.93) | 77.9 (36.2-119.7) | |
| LDHL | 145 (1) | 38 | 21.4 | 1334.7 | 1.77 (1.29-2.44) | 124.1 (33.6-214.6) | |
| cHL-NOS | 4361 (22) | 752 | 365.3 | 30 634.4 | 2.06 (1.92-2.21) | 126.2 (108.7-143.8) | |
| Cancer stage | |||||||
| I-II | 11 870 (59) | 1400 | 815.2 | 108 308.6 | 1.72 (1.63-1.81) | <.001 | 54 (47.2-60.8) |
| III-IV | 7650 (38) | 1405 | 620.1 | 60 134.7 | 2.27 (2.15-2.39) | 130.5 (118.3-142.8) | |
| Unknown | 537 (3) | 94 | 45.5 | 4799.8 | 2.06 (1.69-2.53) | 101 (61.4-140.6) | |
| Radiotherapy | |||||||
| No | 12 483 (62) | 2028 | 954.8 | 98 572.7 | 2.12 (2.03-2.22) | <.001 | 108.9 (99.9-117.8) |
| Yes | 7574 (38) | 871 | 526.1 | 74 670.4 | 1.66 (1.55-1.77) | 46.2 (38.5-54.0) | |
| Radiotherapy in cancer stage I-II | |||||||
| No | 5771 (49) | 750 | 387.1 | 47 980.8 | 1.94 (1.8-2.08) | <.001 | 75.6 (64.4-86.8) |
| Yes | 6099 (51) | 650 | 428.1 | 60 327.8 | 1.52 (1.4-1.64) | 36.8 (28.5-45.1) | |
| Radiotherapy in cancer stage III-IV | |||||||
| No | 6313 (83) | 1214 | 535.3 | 47 242.6 | 2.27 (2.14-2.4) | .95 | 143.7 (129.2-158.1) |
| Yes | 1337 (17) | 191 | 84.7 | 12 892 | 2.26 (1.95-2.59) | 82.5 (61.5-103.5) | |
| Cause of deathb | |||||||
| Solid malignant neeoplasm | NA | 391 | 273.4 | NA | 1.43 (1.29-1.58) | NA | 6.8 (4.6-9.0) |
| Digestive system | NA | 112 | 91.5 | NA | 1.22 (1.01-1.46) | NA | 1.2 (0-2.4) |
| Respiratory system | NA | 154 | 80.7 | NA | 1.91 (1.62-2.23) | NA | 4.2 (2.8-5.6) |
| Skin excluding basal and squamous | NA | 27 | 7.4 | NA | 3.63 (2.43-5.18) | NA | 1.1 (0.5-1.7) |
| Lymphoma | NA | 653 | 11.2 | NA | 58.15 (53.8-62.73) | NA | 37.0 (34.2-39.9) |
| Hodgkin lymphoma | NA | 417 | 0.7 | NA | 638.65 (579.3-701.93) | NA | 24.0 (21.7-26.3) |
| Non-Hodgkin lymphoma | NA | 236 | 10.6 | NA | 22.31 (19.59-25.28) | NA | 13.0 (11.3-14.8) |
| Other hematologic malignant neoplasms | NA | 47 | 18.2 | NA | 2.59 (1.92-3.4) | NA | 1.7 (0.9-2.4) |
| Unspecified cancer | NA | 92 | 32.0 | NA | 2.87 (2.33-3.5) | NA | 3.5 (2.4-4.5) |
| Cardiovascular | NA | 670 | 410.8 | NA | 1.63 (1.51-1.76) | NA | 15.0 (12.0-17.9) |
| Diseases of heart | NA | 554 | 316.7 | NA | 1.75 (1.61-1.9) | NA | 13.7 (11.0-16.4) |
| Other cardiovascular diseases | NA | 116 | 94.1 | NA | 1.23 (1.02-1.47) | NA | 1.3 (0-2.5) |
| Other noncancer | NA | 1046 | 735.1 | NA | 1.42 (1.34-1.51) | NA | 17.9 (14.3-21.6) |
| Infections | NA | 91 | 48.3 | NA | 1.88 (1.52-2.3) | NA | 2.5 (1.4-3.5) |
| Other chronic conditions | NA | 298 | 221.5 | NA | 1.35 (1.2-1.5) | NA | 4.4 (2.5-6.4) |
| Other or ill-defined | NA | 657 | 465.3 | NA | 1.41 (1.31-1.52) | NA | 11.1 (8.2-14.0) |
Abbreviations: cHL-NOS, classic Hodgkin lymphoma, not otherwise specified; EAR, excess absolute risk; LDHL, lymphocyte-depleted Hodgkin lymphoma; LRHL, lymphocyte-rich Hodgkin lymphoma; MCHL, mixed cellularity Hodgkin lymphoma; NA, not applicable; NSHL, nodular sclerosis Hodgkin lymphoma; SEER, Surveillance, Epidemiology, and End Results; SMR, standardized mortality ratio.
Median (IQR) follow-up after index date was 174 (151-200) months for the 2000 to 2007 period and 69 (39-100) months for the 2008 to 2017 period.
Cause of death grouping is based on the variable “SEER Cause of Death Recode 1969+ (03/01/2018)” which is based on the International Statistical Classification of Diseases and Related Health Problems, Tenth Revision codes for solid malignant neoplasms (C15-C22.4, C22.7, C22.9, C23-C25, C26.0, C26.8-C26.9, C45.1, C48.0-C48.2, C48.8 [digestive system], C30-C34, C38.1-C38.4, C38.8, C39, C45.0 [respiratory system], C43-C44, C46 [skin excluding basal and squamous], C00-C14, C37-C38.0, C40-C41, C45.2, C47, C49-C58, C60-C75); lymphoma (C81 [Hodgkin lymphoma], C82-86 and C96.3 [non-Hodgkin lymphoma]); other hematologic malignant neoplasms (C90.0, C90.2, C90.3 [myeloma], C90.1, C91.0-C92.9, C93.0-C93.3, C93.7, C93.9, C94.0-C94.5, C94.7, C95.0-C95.2, C95.7, C95.9 [leukemia]); unspecified cancer (C26.1, C45.7+, C45.9+, C76-C80, C88, C94.6, C96.0-C96.2, C96.4-C97, D00-D48); cardiovascular (I00-I09, I11, I13, I20-I51 [diseases of the heart], I10, I12, I60-I69, I70, I71, I72-I78 [other cardiovascular diseases]); other noncancer (A15-A19, A40-A41, A50-A53, J09-J18 [infections], A34, E10-E14, G30, J40-J47, K25-K28, K70, K73-K74, N00-N07, N17-N19, N25-N27, O00-O95, O98-O99, P00-P96, Q00-Q99 [other chronic conditions]; R00-R99, U01-U03, V01-Y09, Y35, Y85-Y87.1, Y89.0 [other or ill-defined], remaining codes per SEER documentation).
Excess mortality varied across subgroups. Hispanic patients, non-Hispanic American Indian or Alaska Native patients, non-Hispanic Asian or Pacific Islander patients, and those diagnosed at ages 20 to 39 years had SMRs greater than 2.50. However, the highest EARs were observed among adults aged 60 to 89 years, non-Hispanic American Indian or Alaska Native patients, and those diagnosed with mixed cellularity HL. SMRs and EARs were lower in 2008 to 2017 than in 2000 to 2007. Radiotherapy was associated with lower SMR and EAR only among patients with stage I or II disease (Table).
With a median (IQR) follow-up of 173 (120-226) months since diagnosis, the overall 10-year cumulative mortality was 8.0% (95% CI, 7.6%-8.4%), increasing to 15.0% (95% CI, 15.0%-16.0%) at 15 years. At 10 years, mortality form noncancer causes was highest at 2.5% (95% CI, 2.3%-2.8%), followed by other neoplasms at 2.1% (95% CI, 1.9%-2.3%), cardiovascular conditions at 1.6% (95% CI, 1.5%-1.8%), and HL at 1.6% (95% CI, 1.5%-1.8%). HL-related deaths plateaued over time, while mortality from other causes continued to increase (Figure).
Figure. Line Graph of Cumulative Mortality Among 5-Year Hodgkin Lymphoma Survivors From 16 Surveillance, Epidemiology, and End Results Registries, 2000-2017.
NHL indicates non-Hodgkin lymphoma.
Discussion
In this retrospective population-based cohort study, 5-year HL survivors diagnosed during adulthood experienced nearly 2-fold higher all-cause mortality than the general US population, with varying estimates across demographic and clinical subgroups. Previous studies among childhood survivors have reported higher SMRs of 4 to 8,4,6,7 possibly reflecting differences in age at diagnosis, treatment protocols, background comorbidity, and lack of conditioning on 5-year survival. Notably, the long-term predominance of noncancer and cardiovascular mortality supports integrated survivorship care beyond lymphoma recurrence surveillance. Our estimates quantify population-based disparities and excess mortality among adults diagnosed in the era of risk-adapted therapy, suggesting the need for continued improvement in survivorship care. The SEER registries lack detailed clinical and socioeconomic information, limiting adjustment for potential confounding; therefore, associations with radiotherapy should not be interpreted as causal. Furthermore, follow-up may be insufficient to detect late radiation-associated outcomes, including second primary malignant neoplasms and cardiovascular disease, often emerging decades after treatment.
These findings may inform patient counseling and support the development of risk-based survivorship approaches, as previously suggested,9 focusing on strengthening screening and management practices for cardiovascular disease, subsequent neoplasms, and noncancer conditions. Future studies are needed to identify modifiable clinical factors and social determinants of health associated with overall and subgroup-specific excess mortality to support their integration into survivorship care. Continued population-based surveillance is warranted to assess how novel immunotherapies, including brentuximab-vedotin and nivolumab,10 may shape long-term mortality patterns.
Data Sharing Statement
References
- 1.Teras LR, DeSantis CE, Cerhan JR, Morton LM, Jemal A, Flowers CR. 2016 US lymphoid malignancy statistics by World Health Organization subtypes. CA Cancer J Clin. 2016;66(6):443-459. doi: 10.3322/caac.21357 [DOI] [PubMed] [Google Scholar]
- 2.Lo AC, Chen B, Samuel V, Savage KJ, Freeman C, Goddard K. Late effects in survivors treated for lymphoma as adolescents and young adults: a population-based analysis. J Cancer Surviv. 2021;15(6):837-846. doi: 10.1007/s11764-020-00976-7 [DOI] [PubMed] [Google Scholar]
- 3.Vandtved JH, Øvlisen AK, Baech J, et al. Pulmonary diseases in patients with classical Hodgkin lymphoma relative to a matched background population: a Danish national cohort study. Br J Haematol. 2024;205(2):542-551. doi: 10.1111/bjh.19475 [DOI] [PubMed] [Google Scholar]
- 4.Castellino SM, Geiger AM, Mertens AC, et al. Morbidity and mortality in long-term survivors of Hodgkin lymphoma: a report from the Childhood Cancer Survivor Study. Blood. 2011;117(6):1806-1816. doi: 10.1182/blood-2010-04-278796 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Esbenshade AJ, Lu L, Friedman DL, et al. Accumulation of chronic disease among survivors of childhood cancer predicts early mortality. J Clin Oncol. 2023;41(20):3629-3641. doi: 10.1200/JCO.22.02240 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Rossetti S, Juul SJ, Eriksson F, et al. Long-term cause-specific mortality in adolescent and young adult Hodgkin lymphoma patients treated with contemporary regimens—a nationwide Danish cohort study. Br J Haematol. 2024;205(4):1374-1382. doi: 10.1111/bjh.19586 [DOI] [PubMed] [Google Scholar]
- 7.Núñez-García B, Clemente MB, Sánchez JC, et al. Long-term outcomes in Hodgkin lymphoma survivors: temporary trends and comparison with general population. Hematol Oncol. 2023;41(3):407-414. doi: 10.1002/hon.3131 [DOI] [PubMed] [Google Scholar]
- 8.Favier O, Heutte N, Stamatoullas-Bastard A, et al. ; European Organization for Research and Treatment of Cancer (EORTC) Lymphoma Group and the Groupe d’Etudes des Lymphomes de l’Adulte (GELA) . Survival after Hodgkin lymphoma: causes of death and excess mortality in patients treated in 8 consecutive trials. Cancer. 2009;115(8):1680-1691. doi: 10.1002/cncr.24178 [DOI] [PubMed] [Google Scholar]
- 9.Valcarcel B, Savage KJ, Link BK, et al. Comparison of survivorship care guidelines for patients with lymphoma: recommendations for harmonization and future research agenda. JCO Oncol Pract. 2024;21(6): 801-812. doi: 10.1200/OP.24.00364 [DOI] [PMC free article] [PubMed]
- 10.Ansell SM. Hodgkin lymphoma: 2025 update on diagnosis, risk-stratification, and management. Am J Hematol. 2024;99(12):2367-2378. doi: 10.1002/ajh.27470 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Sharing Statement
