Longitudinal trajectories of triglyceride glucose index and colorectal cancer mortality in Chinese older adults: a large prospective cohort study

<h2>Abstract</h2><h3>Background</h3> Colorectal cancer (CRC) is the second leading cause of cancer-related deaths worldwide in 2022. Triglyceride glucose (TyG) index, as a more convenient and reliable predictor of insulin resistance, is associated with cardiometabolic diseases, but its relationship with CRC remains unclear. Although evidence from previous prospective cohort studies indicated that people with higher baseline TyG index were associated with onset of CRC, most of these studies have focused on elevated baseline TyG index levels rather than its dynamic changes over time. This study aimed to evaluate the long-term trajectories of the TyG index and the mortality due to CRC. <h3>Methods</h3> We used data from a Multi-center Elderly Health Management Cohort Study in Luzhou and Zunyi, China. Participants aged ≥65 years were enrolled from January 2017 and followed up until December 2021. Baseline data came from annual health examinations from the health management programs for the elderly in basic public health services including standardized questionnaires and laboratory tests. Demographic information, lifestyles, and health conditions were collected. TyG index was calculated by fasting blood glucose and triglycerides. Deaths from all-cause and CRC were obtained from linkage with China's national death registration system. CRC death as underlying cause of death was defined as International Classification of Diseases, 10th revision codes C18-C20. The latent class trajectory modeling method was used to analyze the 5-year TyG index trajectories from registration and the Cox proportional hazards regression model was utilized to examine the relationship between baseline TyG index level, the trajectory of TyG index, and CRC mortality. <h3>Findings</h3> Among 767,483 older adults, 52.9% were females; the mean age at baseline was 70.4 years. A total of 39,738 deaths (including 700 from CRC) were recorded in a median follow-up of 3.8 years [IQR 2.4-5.2]). Three longitudinal patterns based on the 5-year TyG index were identified: "medium-stable trajectory" (n=374,345, 48.8%), "low-stable trajectory" (n=334,645, 43.6%), and "high-stable trajectory" (n=58,493, 7.6%). Compared with the lowest quartile TyG index in the baseline, there was a borderline significant higher risk of CRC mortality for the highest quartile (hazard ratio 1.23, 95% confidence interval 0.98-1.54, p=0.07) after adjustment for age, sex, years of schooling, alcohol consumption, smoking, exercise frequency, diabetes, hypertension, cardiovascular and cerebrovascular diseases. Compared with medium-stable TyG index pattern, individuals with a high-stable TyG index had lower CRC mortality, with a crude hazard ratio (95% CI) of 0.71 (0.50-0.99). In the fully adjusted model, neither low-stable trajectory group nor high-stable trajectory group showed any significant association with CRC mortality (low-stable trajectory hazard ratio: 0.96 (95% CI: 0.82-1.13); high-stable trajectory hazard ratio: 0.78 (95% CI: 0.55-1.11)). <h3>Interpretation</h3> Although a borderline significant association was found between highest quartile TyG index at baseline and CRC mortality, there was no significant association with CRC mortality for trajectory pattern of TyG index. Such finding might be due to the slightly shorter follow-up period but to some extent, it also suggests that strict adherence to normal values for metabolic indicators may not be necessary in older Chinese adults, although more prospective evidence with longer follow-up is still needed.