Analysis of Time to Treatment and Survival Among Adults Younger Than 50 Years of Age With Colorectal Cancer in Canada

Abstract

ImportanceColorectal cancer (CRC) is uncommon in adults younger than 50 years of age, so this population may experience delays to treatment that contribute to advanced stage and poor survival.ObjectiveTo investigate whether there is an association between time from presentation to treatment and survival in younger adults with CRC.Design, Setting, and ParticipantsThis retrospective cohort study used linked population-based data in Ontario, Canada. Participants included patients with CRC aged younger than 50 years who were diagnosed in Ontario between 2007 and 2018. Analysis was performed between December 2019 and December 2022.ExposureAdministrative and billing codes were used to identify the number of days between the date of first presentation and treatment initiation (overall interval).Main Outcomes and MeasuresThe associations between increasing overall interval, overall survival (OS), and cause-specific survival (CSS) were explored with restricted cubic spline regression. Multivariable Cox proportional hazards models were also fit for OS and CSS, adjusted for confounders. Analyses were repeated in a subset of patients with lower urgency, defined as those who did not present emergently, did not have metastatic disease, did not have cross-sectional imaging or endoscopy within 14 days of first presentation, and had an overall interval of at least 28 days duration.ResultsAmong 5026 patients included, the median (IQR) age was 44.0 years (40.0-47.0 years); 2412 (48.0%) were female; 1266 (25.2%) had metastatic disease and 1570 (31.2%) had rectal cancer. The lower-urgency subset consisted of 2548 patients. The median (IQR) overall interval was 108 days (55-214 days) (15.4 weeks [7.9-30.6 weeks]). Patients with metastatic CRC had shorter median (IQR) overall intervals (83 days [39-183 days]) compared with those with less advanced disease. Five-year overall survival was 69.8% (95% CI, 68.4%-71.1%). Spline regression showed younger patients with shorter overall intervals (<108 days) had worse OS and CSS with no significant adverse outcomes of longer overall intervals. In adjusted Cox models, overall intervals longer than 18 weeks were not associated with significantly worse OS or CSS compared with those waiting 12 to 18 weeks (OS: HR, 0.83 [95% CI, 0.67-1.03]; CSS: HR, 0.90 [95% CI, 0.69-1.18]). Results were similar in the subset of lower-urgency patients, and when stratified by stage.Conclusions and relevanceIn this cohort study of 5026 patients with CRC aged younger than 50 years of age in Ontario, time from presentation to treatment was not associated with advanced disease or poor survival. These results suggest that targeting postpresentation intervals may not translate to improved outcomes on a population level.