PCN337 Survival Outcomes in Patients with Colorectal Cancer in Sweden - a Population-Based Retrospective Registry Study

Abstract

Colorectal cancer (CRC) is the second most common cause of cancer death in the United States. There is a need to better understand treatment outcomes in this patient population in clinical practice, and we therefore used national Swedish registries to conduct a longitudinal, population-based study on survival outcomes. The aim of this study was to describe survival outcomes in CRC patients. All patients diagnosed with CRC in Sweden between 2005 and 2018 (n=104,840) were included and followed until June 2020. Kaplan-Meier curves and logrank tests were used to calculate overall survival (OS). Cox proportional-hazard models adjusted for calendar period of diagnosis (2005-2009, 2010-2014, 2015-2018), age at diagnosis (<60, 60-69, 70-79, 80+), CRC subtype (ICD-10 C18, C19, C20), T stage (T1, T2, T3, T4) and M stage (M0, M1) at diagnosis, were used to assess risk of death. Median age at diagnosis was 72 years. At CRC diagnosis, 79,979 (76.3%) patients had no metastases and 15,811 (15.1%) patients had metastases. Colon cancer was the most common CRC subtype (66.1%), followed by rectal cancer (33%) and rectosigmoid junction cancer (0.9%). Patients with colon cancer had the worst median OS (7.19 years; 95% CI: 7.05–7.32) compared to the rectosigmoid junction cancer (10.21 years; 95% CI: 8.86–12.03) and rectal cancer (8.07 years; 95% CI: 7.85–8.31). We observed a significantly lower risk of death in patients diagnosed with CRC in the later versus earlier calendar periods (2005–2009 vs 2010–2014 [HR: 0.91; 95% CI: 0.89–0.93] and 2005–2009 vs 2015–2018 [HR: 0.80; 95% CI: 0.78–0.82]). Patients with colon cancer had the worst survival among all patients with CRC. CRC diagnosis in the later calendar period was associated with superior survival compared to earlier calendar periods, which may be explained by more effective treatment options.