Age-stratified survival and competing mortality in adult acute lymphoblastic leukemia and lymphoblastic lymphoma: a population-based competing-risk and restricted mean survival analysis
- 1Department of Hematology, Guizhou Provincial People’s Hospital, Guiyang, Guizhou, China
- 2The First Clinical Medical College, Zunyi Medical University, Zunyi, Guizhou, China
Abstract
Background: Survival differences by age are well recognized in adult acute lymphoblastic leukemia/lymphoma (ALL/LBL), but their timing, competing-mortality structure, and behavior across recent diagnosis periods remain less clear.
Methods: This retrospective SEER cohort included adults diagnosed during 2010-2023, excluding Burkitt leukemia/lymphoma. We estimated overall survival (OS), 3-year restricted mean survival time (RMST), and cumulative incidence of ALL/LBL-attributable and other/unknown-cause death. Cox, Fine-Gray, spline, and exploratory time-split models assessed age associations.
Results: Among 9,491 patients, 4,618 died: 3,828 deaths were attributed to ALL/LBL and 790 to other or unknown causes. Three-year RMST was 1.89, 2.10, and 2.14 years in 2010-2014, 2015-2019, and 2020-2023, respectively; the recent-period difference was 0.044 years (95% CI -0.016 to 0.104; P = 0.150). Median follow-up in 2020–2023 was 22 months. Compared with age 18–39 years, the adjusted overall-mortality HR for age >=75 years was 6.258 (95% CI 5.627-6.960). Because proportional hazards were violated, this HR should be read as a summary association. The exploratory HR was 8.201 (95% CI 7.224-9.311) during the first year and 3.496 (95% CI 2.885-4.236) thereafter. At 3 years, cumulative incidence in patients aged >=75 years was 70.4% for ALL/LBL-attributable death by registry attribution and 12.8% for other/unknown-cause death. The spline test supported departure from linearity (P<0.001); the age-by-period interaction was not significant (P = 0.182), whereas an exploratory age-by-subtype interaction was nominally significant (P = 0.024, unadjusted for multiplicity).
Conclusion: Older age was associated with higher mortality, especially in the first year after diagnosis. ALL/LBL-attributable death predominated by registry attribution even in the oldest group. The 3-year comparison did not establish a survival difference between 2015–2019 and 2020-2023; limited follow-up and residual case-mix differences preclude concluding that progress has stopped.
Related
- Cumulative-incidence crossover and conditional mortality after definitive surgery for endometrial cancer: a population-based competing-risk analysis
- Development and validation of a competing-risk nomogram for predicting cancer-specific mortality in prostate cancer: a SEER population-based study
- Spatial distribution and age-specific geodemographic patterns of Acute Lymphoblastic Leukemia (ALL) mortality in Ecuador: an ecological analysis across elevation gradients (2011–2022)
- Competing Mortality, Not Breast-Cancer Mortality, Explains the Survival Deficit of Breast Cancer After a Prior Non-Breast Cancer: A SEER Competing-Risks Analysis
- Competing Risk Prediction of COVID-19 Hospital Mortality with Model Diagnostics and Exploratory Evaluation in Mexico
BibTeX
@article{li2026agestratified,
title = {{Age-stratified survival and competing mortality in adult acute lymphoblastic leukemia and lymphoblastic lymphoma: a population-based competing-risk and restricted mean survival analysis}},
author = {Zhe Li and Jianglin Yu},
journal = {Frontiers in Oncology},
year = {2026},
month = oct,
doi = {10.3389/fonc.2026.1913564},
url = {https://doi.org/10.3389/fonc.2026.1913564},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Age-stratified survival and competing mortality in adult acute lymphoblastic leukemia and lymphoblastic lymphoma: a population-based competing-risk and restricted mean survival analysis
AU - Zhe Li
AU - Jianglin Yu
PY - 2026
DA - 2026/10/12
JO - Frontiers in Oncology
DO - 10.3389/fonc.2026.1913564
UR - https://doi.org/10.3389/fonc.2026.1913564
AB - Background: Survival differences by age are well recognized in adult acute lymphoblastic leukemia/lymphoma (ALL/LBL), but their timing, competing-mortality structure, and behavior across recent diagnosis periods remain less clear. Methods: This retrospective SEER cohort included adults diagnosed during 2010-2023, excluding Burkitt leukemia/lymphoma. We estimated overall survival (OS), 3-year restricted mean survival time (RMST), and cumulative incidence of ALL/LBL-attributable and other/unknown-cause death. Cox, Fine-Gray, spline, and exploratory time-split models assessed age associations. Results: Among 9,491 patients, 4,618 died: 3,828 deaths were attributed to ALL/LBL and 790 to other or unknown causes. Three-year RMST was 1.89, 2.10, and 2.14 years in 2010-2014, 2015-2019, and 2020-2023, respectively; the recent-period difference was 0.044 years (95% CI -0.016 to 0.104; P = 0.150). Median follow-up in 2020–2023 was 22 months. Compared with age 18–39 years, the adjusted overall-mortality HR for age >=75 years was 6.258 (95% CI 5.627-6.960). Because proportional hazards were violated, this HR should be read as a summary association. The exploratory HR was 8.201 (95% CI 7.224-9.311) during the first year and 3.496 (95% CI 2.885-4.236) thereafter. At 3 years, cumulative incidence in patients aged >=75 years was 70.4% for ALL/LBL-attributable death by registry attribution and 12.8% for other/unknown-cause death. The spline test supported departure from linearity (P<0.001); the age-by-period interaction was not significant (P = 0.182), whereas an exploratory age-by-subtype interaction was nominally significant (P = 0.024, unadjusted for multiplicity). Conclusion: Older age was associated with higher mortality, especially in the first year after diagnosis. ALL/LBL-attributable death predominated by registry attribution even in the oldest group. The 3-year comparison did not establish a survival difference between 2015–2019 and 2020-2023; limited follow-up and residual case-mix differences preclude concluding that progress has stopped.
SN - 2234-943X
ER -