Background: Cancer has emerged as the leading cause of death globally, accounting for approximately 10.4 million lives annually. Driven by population aging and continuous evolution of risk factors, the global burden of cancer is projected to escalate further in the coming decades. Concurrently, the planet is experiencing unprecedented rates of surface warming. Specifically, the global mean surface temperature has risen by 1.2 °C since the 1960s, and is anticipated to exceed 2 °C above pre-industrial levels by the 2050s even under moderate-emission scenarios. A growing body of epidemiological evidence has established robust associations between climate change and elevated mortality risks from cardiovascular and respiratory diseases. However, its potential role in cancer mortality remains largely underexplored. If temperature extremes can accelerate tumour proliferation, impair immune surveillance, or delay timely diagnosis, climate change could impose a substantial yet previously unquantified burden on future cancer incidence and mortality.
Objectives: In this ecological study, we aimed to investigate the association between ambient temperature variability and cancer mortality at a global scale, and further explore the potential effect modifiers.
Methods: We extracted age-standardized mortality rate (ASMR) for cancer, demographic, and socioeconomic data from the Global Burden of Disease (GBD) Results Database, and environmental and geographical statistics from the TerraClimate dataset, covering 204 countries and regions between 1990 and 2019. A generalized additive mixed model was used to evaluate the association between temperature variability and cancer ASMR, with country code as a random effect to account for between-region heterogeneity. Potential confounders included the sociodemographic index (SDI), smoking prevalence, fine particulate matter (PM_2.5) concentration, ozone concentration, year, and other key climate variables. The adjusted excess risk (ER) and its 95% confidence interval (CI) were used to represent the percentage change in cancer ASMR for each 1 °C increase in temperature variability.
Results: From 1990 to 2019, the global median ASMR of cancer was 126.16 per 100,000 population (interquartile range [IQR]: 99.26/10⁵, 151.63/10⁵), and the median temperature variability was 0.53 ? (IQR: 0.23 ?, 0.98 ?). The fully adjusted regression model analysis revealed a positive association between temperature variability and ASMR. For each 1 °C increase in temperature variability, the overall cancer ASMR increased by 3.18% (95% CI: 2.62%, 3.75%) in the year of exposure (lag 0). This association was more pronounced in individuals aged 40 to 59, and living in areas with low SDI or with severe air pollution. Additionally, the lag effect of temperature variability on cancer mortality peaked at lag02 years (ER = 5.09%; 95% CI: 3.59%, 6.62%), indicating a cumulative effect of temperature variability over a 3-year period.
Conclusions: Our study provides evidence that temperature variability may be associated with an increase risk of cancer mortality, with heterogeneous effects across age groups and regions with different socioeconomic and environmental conditions. These findings suggest that targeted climate adaptation strategies tailored to vulnerable populations are crucial to mitigating the potential cancer mortality burden attributable to climate change. Future studies are warranted to validate these findings using individual-level data and explore underlying biological pathways to establish causality.

Global map of temperature variability and cancer mortality across 200 countries and regions from 1990 to 2019.