UPDF AI

Population-Level Survival Trends in Hematologic Malignancies: A Surveillance, Epidemiology, and End Results (SEER) Analysis of 494,244 Patients From 2000 to 2020

K. Woodson,J. Alvarenga

2026 · DOI: 10.7759/cureus.111845
Cureus · 0 Citations

TLDR

Survival has improved substantially but unequally across all four malignancies over two decades, and age and race/ethnicity remain independent correlates of mortality after full adjustment, warranting targeted interventions.

Abstract

Population-level survival gains from transformative advances in hematologic malignancy treatment may not be equitably distributed. We examined over two decades of survival trends for acute myeloid leukemia (AML), acute lymphoblastic leukemia (ALL), non-Hodgkin lymphoma (NHL), and multiple myeloma using data from the National Cancer Institute (NCI) Surveillance, Epidemiology, and End Results (SEER) 17 Registries (2000-2020). Kaplan-Meier analysis estimated overall survival (OS) for 423,982 patients aged ≥15 years, stratified by diagnosis year, race/ethnicity, age, sex, and stage; observed OS reflects all-cause mortality. Multivariate Cox proportional hazards regression was performed on 494,244 patients to identify independent correlates of all-cause mortality. Five-year OS improved substantially from 2000 to 2017: myeloma (29.0%→53.7%, +85% relative improvement), ALL (30.7%→53.0%, +73%), AML (15.2%→27.9%, +84%), and NHL (54.5%→65.6%, +20%). Age ≥70 years was the strongest correlate of mortality (hazard ratio (HR): 7.20; 95% CI: 7.01-7.39 vs. <40 years). Non-Hispanic Black and Hispanic patients had significantly higher mortality than non-Hispanic White patients (HR: 1.22 and 1.11, respectively; both: p < 0.001). Each successive diagnosis period was independently associated with lower mortality (HR: 0.72 for 2015-2020 vs. 2000-2004). Survival has improved substantially but unequally across all four malignancies. Age and race/ethnicity remain independent correlates of mortality after full adjustment, warranting targeted interventions.