Global trends in early-onset intrahepatic cholangiocarcinoma incidence from 1993 to 2017: Associations with environmental quality and socioeconomic development.

BACKGROUND: Early-onset intrahepatic cholangiocarcinoma (EOICC), defined as ICC diagnosed before age 50, represents a clinically distinct entity from traditional late-onset ICC (LOICC). However, its global epidemiological patterns and potential links with environmental quality and socioeconomic development remain poorly understood. This study aimed to investigate global EOICC incidence trends from 1993 to 2017 and explore its associations with environmental and socioeconomic factors. METHODS: We extracted cancer incidence data from the World Health Organization-International Agency for Research on Cancer (WHO-IARC) and Cancer Incidence in Five Continents Plus (CI5plus) database (volumes VIII-XII), covering five periods (1993-1997 to 2013-2017) across 40 countries. Incidence rates were calculated per 100 000 population and age-standardized rate (ASR) to the WHO 2000-2025 world standard population. We analyzed temporal trends by calculating annual percentage changes (APCs) using joinpoint regression on period-specific mean ASRs. Human development index (HDI) data were obtained from the United Nations Development Programme (UNDP), and environmental performance index (EPI) data from Yale University. To build a predictive model, we first used least absolute shrinkage and selection operator (LASSO) regression to identify key predictors from HDI and 10 relevant EPI sub-indicators. A support vector machine (SVM) model was then trained on these features to predict EOICC ASRs for 125 countries. RESULTS: Globally, EOICC constitutes 2%-40% of all ICC cases (median 7.57%), with substantial regional variation. While LOICC ASRs remained stable or declined in most countries, EOICC ASRs had been increased in over half of the analyzed nations during the study period. Pearson correlation revealed a significant positive association between EOICC ASR and HDI (R = 0.375, P = 0.017), but not with the mean EPI score (R = 0.207, P = 0.199). Notably, ten specific EPI sub-indicators-including air pollution, heavy metal exposure, and wastewater treatment-showed significant correlations with EOICC ASRs. Linear regression confirmed HDI as an independent contributor to EOICC ASR variability (R2 = 0.140, P = 0.017). The SVM prediction model, built using these key predictors, performed well [R2 = 0.439, the area under the curve (AUC) = 0.897] and was used to estimate EOICC ASRs for 125 countries. CONCLUSIONS: EOICC incidence is rising notably worldwide, particularly in high-socioeconomic countries. This trend may be linked to a combination of modern lifestyle risks and legacy effects of historical environmental exposures. This study provides preliminary macro-level evidence supporting an environmental-socioeconomic etiology for EOICC, though further mechanistic research is needed for validation. Future studies should focus on identifying specific EOICC risk factors and elucidating their pathogenic mechanisms to inform evidence-based public health strategies.