· JAMA neurology 2026 · Cross-sectional population-based modeling study · n=183 data sources

Burden of Central Nervous System Cancer in the United States, 1990-2021.

Cited 1 times in the scientific literature.

Level 4 - case-series / case-control

Cross-sectional epidemiological modeling study using secondary Global Burden of Disease data.

PubMed 41182787 · doi:10.1001/jamaneurol.2025.4286 · record verified 2026-08-27

What was done

Researchers conducted a cross-sectional analysis using the Global Burden of Disease (GBD) Study 2021 dataset, incorporating data from 183 sources to estimate central nervous system (CNS) cancer metrics in the United States from 1990 to 2021. Outcomes included overall and age-standardized incidence, prevalence, mortality, disability-adjusted life-years (DALYs), years of life lost, and years lived with disability per 100,000 population, stratified by state/division, sex, age group, and Sociodemographic Index (SDI).

What was found

In 2021, there were 31,780 incident CNS cancer cases (95% UI, 29,971.1 to 32,843.9) across the US. Age-standardized rates per 100,000 were 6.91 for incidence (95% UI, 6.58 to 7.12), 4.1 for mortality (95% UI, 3.87 to 4.22), and 134.38 for DALYs (95% UI, 129.83 to 137.95). From 1990 to 2021, overall incidence did not change significantly, whereas DALY rates decreased by 15.77% (95% UI, -17.75% to -13.68%) and mortality rates fell by 8.41% (95% UI, -11.09% to -6.22%). Disease burden was consistently higher in males. Age-specific incidence followed a bimodal distribution, decreasing in children under 5 years (-34.42% to -11.56%) but increasing in adults over 70 years. Persistently elevated burdens were observed in Mississippi, Alabama, Kentucky, Kansas, and West Virginia. SDI was negatively correlated with DALYs (rho = -0.6860, P < .001) and mortality rates (rho = -0.6391, P < .001).

Why it matters

This study establishes multi-decade, state-level benchmarks showing that while US CNS cancer survival and disability have improved modestly, persistent geographic, socioeconomic, and age-related disparities remain.

Limits

The findings rely on secondary GBD modeling rather than direct registry follow-up, which is susceptible to reporting heterogeneity across jurisdictions and time. The abstract does not provide data disaggregated by specific tumor histologies or assess the impact of particular clinical treatments.

Cited by