Burden of cancer in China from 1990 to 2023 and projections to 2050

China bears one of the largest cancer burdens worldwide, yet comprehensive assessments integrating long-term age-sex patterns with uncertainty-aware projections based on the latest Global Burden of Disease (GBD) 2023 estimates remain limited. Using GBD 2023 data, we quantified trends in age-standardized incidence rates (ASIR) and age-standardized mortality rates (ASMR) for all malignant neoplasms in China from 1990 to 2023 and projected future trajectories through 2050. Period-specific changes were assessed using Joinpoint regression, and medium- to long-term projections were generated with a Bayesian age-period-cohort model incorporating uncertainty. Between 1990 and 2023, ASIR remained largely stable, with an average annual percent change (AAPC) of -0.02% per year, whereas ASMR declined substantially, with an AAPC of -1.55% per year. In 2023, ASIR and ASMR reached 2,259.08 and 114.02 per 100,000 population, respectively. Age-specific analyses demonstrated a steep increase in cancer burden with advancing age, characterized by higher incidence among middle-aged women and higher mortality among older men. Mortality trends exhibited distinct patterns, including an accelerated decline after 2008, a transient rebound during 2014-2017, and a modest upturn after 2020. Projections by the Bayesian age-period-cohort (BAPC) model indicate that ASIR is likely to rise modestly after a prolonged plateau-predominantly driven by women-while ASMR may gradually increase again after 2025, with greater uncertainty and rebound risk among men, particularly after 2040. These findings reveal an emerging structural divergence in China's cancer burden, with incidence increasingly concentrated in women and mortality in men. This pattern highlights the need for sex- and age-specific cancer control strategies, including strengthened risk-factor control and multimorbidity management in men, high-quality screening and survivorship care in women, and continuous surveillance to recalibrate policies as observed trends deviate from projected trajectories.