Aesun Shin

Background Clinical diabetic traits have been reported to be associated with increased colorectal cancer (CRC) risk in observational studies. Using the Mendelian randomization (MR) analysis method, we examined the causal association between glycemic traits, such as fasting glucose (FG), fasting insulin (FI), and glycosylated hemoglobin A1c (HbA1c), and survival in a cohort of CRC patients. Methods We conducted a two‐sample MR analysis among a cohort of patients with locally advanced CRC at Seoul National University Hospital. Single‐nucleotide polymorphisms robustly associated ( p < 5 × 10 −8 ) with the three glycemic traits were obtained from the Meta‐Analyses of Glucose and Insulin‐related traits Consortium, Asian Genetic Epidemiology Network, and Korea Biobank Array. Three‐year and 5‐year overall survival (OS) and progression‐free survival (PFS) were used as outcomes. Survival analysis was conducted using subgroup analysis by cancer stage and subsite in a multivariate Cox proportional hazards model adjusted for age and sex to examine whether glycemic traits affected survival. Results A total of 509 patients were included in our final analysis. MR analysis showed that HbA1c levels were associated with poor 3‐year OS ( β = 4.20, p = 0.02). Sensitivity analyses did not show evidence of any violations of the MR assumptions. In the cancer subgroup analysis of the Cox proportional hazards model, pooled hazard ratios for FG were significantly associated with poor 3‐year OS and PFS regardless of cancer stage. FI was not significantly associated with any 3‐year survival endpoints. Among Stage III patients, three glycemic traits were significantly associated with both 5‐year OS and PFS. Location‐specific subgroup analysis showed a significant association between three glycemic traits and 5‐year PFS in patients with left‐sided colon cancer. FG was associated with poor 3‐year survival for colon cancer but not rectal cancer. Conclusions Our results suggest that FG and HbA1c could be used to predict prognosis in CRC patients. Lifestyle and/or pharmacological interventions targeting glycemic traits could help improve survival for CRC patients.

Background Physical activity (PA) is recommended to improve the survival of cancer patients. However, the prognostic impact of specific PAs is not well understood. Therefore, we investigated the associations of the duration, type, intensity, and number of PAs one participates in pre‐ and post‐diagnosis with mortality in Korean patients with cancer. Methods Among the participants aged 40–69 years recruited from the Health Examines study, those diagnosed with cancer after baseline ( n = 7749) and within 10 years before baseline ( n = 3008) were included in the analyses for pre‐ and post‐diagnosis PA, respectively. Duration, intensity, type, and number of leisure‐time physical activities participated in were assessed using questionnaires. The Cox proportional hazard model was used to characterize the association between PA and cancer‐specific mortality, adjusting for demographics, behaviors, comorbidities, and cancer stage based on the Surveillance, Epidemiology, and End Results program. Results Pre‐diagnosis, patients participating in vigorous‐intensity activities (hazard ratio [HR]: 0.70, 95% confidence interval [CI]: 0.61–0.82), walking (HR: 0.85, 95% CI: 0.74–0.97), climbing (HR: 0.65, 95% CI: 0.55–0.77), sports (HR: 0.39, 95% CI: 0.25–0.61), and more than two activities (HR: 0.73, 95% CI: 0.63–0.86) had significantly lower all‐cause mortality. Importantly, these associations were only found in patients with colorectal cancer participating in vigorous‐intensity activities (HR: 0.40, 95% CI: 0.23–0.70). Post‐diagnosis, only patients who performed more than two activities (HR: 0.65, 95% CI: 0.44–0.95) had significantly lower all‐cause mortality. Similar associations were found for cancer mortality, both pre‐ and post‐diagnosis. Conclusion Specific characteristics of PA pre‐ and post‐diagnosis may influence the survival of cancer patients.