Trends in pancreatic cancer incidence, prevalence, and survival outcomes by histological subtypes: a retrospective cohort study.

Background: Pancreatic cancer (PC) is a heterogeneous disease with various histological and molecular subtypes. This study aimed to provide updated epidemiological estimates, survival outcomes, and treatment information for PC based on histological subtypes in the USA.

Methods: Data from the US Cancer Statistics and Surveillance, Epidemiology, and End Results (SEER)-17 databases (2000-2020) were used, including adults aged ≥20 years who were diagnosed with PC. The trends of incidence and prevalence by histological types were calculated by using the Joinpoint Regression model. Survival by histological type was analysed by using Kaplan-Meier curves and log-rank tests for group comparisons.

Results: Overall, the age-adjusted PC incidence per 100,000 increased from 9.54 to 12.05 in SEER-17 and from 9.75 to 12.19 in the US Cancer Statistics between 2001 and 2019. A further SEER-17 study comprised 113,681 PC cases that were sorted by histologic type between 2000 and 2020. The incidence per 100,000 of invasive intraductal papillary mucinous neoplasm (IPMN) and invasive mucinous cystic neoplasm (MCN) decreased (IPMN from 0.67 to 0.20 and MCN from 0.05 to 0.01) whereas that of other histological subtypes increased. Survival analysis indicated the best outcomes for solid pseudopapillary tumors and the poorest for squamous cell carcinoma. At the localized stage, the proportion of surgery in the treatment modalities varied depending on the biological behavior; the proportion of surgery for pancreatic neuroendocrine tumor was the highest and that for pancreatic ductal adenocarcinoma (PDAC) was the lowest. At the distant metastasis stage, a chemotherapy-based regimen remained the primary treatment of PDAC, pancreatic neuroendocrine tumor, and IPMN.

Conclusions: PC incidence and prevalence have been increasing. The incidence of IPMN and MCN decreased whereas that of other subtypes increased. Treatment distribution varies among subtypes and stages.