Background: Cancer incidence and mortality are rising globally, with the sharpest increases occurring in low- and middle-income countries (LMICs). In Brazil, cancer outcomes are strongly influenced by social, geographic, and healthcare-related disparities within a dual healthcare system composed of the publicly funded Unified Health System (Sistema Único de Saúde – SUS) and a private sector. Although Brazil has national mortality data and a broad network of cancer registries, population-level cancer survival estimates stratified by healthcare coverage remain scarce and fragmented, limiting monitoring of outcomes across tumor types and health systems.

Objectives: To analyze hospital-based cancer registry data from oncology centres accredited by the National Cancer Institute in Brazil’s southernmost state and provide a comprehensive assessment of five-year overall survival for the 19 most common cancers, comparing outcomes between public and private healthcare systems, and evaluating temporal trends and sociodemographic gradients.

Methods: We included all eligible hospital-based cancer registries from Rio Grande do Sul. Individuals aged 15–99 years diagnosed with one of the 19 most common cancers between January 1, 2005, and December 31, 2018 were eligible. Cancer registry records were probabilistically linked to the national Mortality Information System to ascertain vital status through December 31, 2023. Overall survival (OS) was defined as time from diagnosis to death from any cause or censoring. Primary analyses were restricted to cases diagnosed from 2010 onwards to minimize under-ascertainment of deaths in earlier years. Five-year age-standardized OS was estimated using Kaplan–Meier methods. Multivariable Cox proportional hazards models with a gamma frailty term for clustering by centre were fitted to estimate hazard ratios (HRs) for all-cause mortality comparing public versus private healthcare coverage, adjusted for age, sex, education level, and period of diagnosis.

Results: Nineteen hospitals participated, representing approximately 60% of the state’s hospital-based cancer registries. Among 255,295 registered tumors, 85,345 cases diagnosed from 2010 onwards with complete healthcare coverage information were included in the main analyses. Five-year age-standardized OS varied markedly across cancers, ranging from 8.6% for pancreatic cancer to 87.6% for thyroid cancer in the most recent period. OS was consistently lower among patients treated in the public healthcare system. After adjustment, public-sector patients had significantly higher mortality risks for 12 of the 19 cancers studied, with the largest disparities observed for thyroid, breast, prostate, stomach, colorectal cancer, and melanoma. Temporal analyses showed heterogeneous survival trends. Although improvements between 2010–2014 and 2015–2018 were observed for several cancers, including lung cancer and melanoma, gains were substantially greater in the private sector, resulting in persistent—and in some cases widening—survival inequities.

Conclusions/Implications: ?Cancer survival in Brazil remains deeply unequal. Despite overall improvements, patients treated in the public healthcare system experience substantially worse outcomes for most common cancers, reflecting delayed diagnosis, limited access to advanced diagnostics and therapies, and broader socioeconomic inequalities. This study offers a scalable model for hospital-based survival analyses and underscores the need to strengthen routine survival surveillance through systematic linkage of cancer and mortality databases. Targeted investments to ensure timely diagnosis, equitable treatment access, and inclusion in clinical trials—particularly within SUS—are urgently needed.