Survival outcomes in stage I-III colorectal cancer patients over 65 years of age: results from the National Cancer Database

Authors: Hardikar S, Weil CR, Lloyd S, Cohan JN, Supiano MA, Ose J, Peoples AR, Gupta S, Pelletier K, Extermann M, Siegel EM, Shibata D, Ulrich CM

Category: Survivorship & Health Outcomes/Comparative Effectiveness Research
Conference Year: 2022

Abstract Body:
PURPOSE. Older patients (>65 years) are frequently under-represented in clinical trials that determine cancer treatment guidelines, and consequently may experience worse survival outcomes, even after controlling for advanced age. METHODS. The National Cancer Database (NCDB) was queried to evaluate overall survival in stage I-III colorectal cancer patients (2004-2017) over 65 years of age. Patients with metastatic disease and non-adenocarcinoma histology were excluded. Standard of care (SOC) therapy was defined as any recommended treatment option listed within site- and stage-specific National Comprehensive Cancer Network guidelines. The Kaplan-Meier product limit method was used to compare the 2-year overall survival (OS) in stage- and site-specific subgroups among those who did vs. did not receive SOC treatment.RESULTS. A total of 498,285 patients met inclusion criteria; the median age of the patients was 76 years. The majority were non-Hispanic White (88%), female (52%), Medicare insured (86%), colon cancer patients (76%) with a Charlson comorbidity index (CCI) of 0 (63%). Overall, OS was better among those who received SOC treatments for all stage- and site-specific subgroups. For example, among stage I colon cancer patients, 2-year OS for those receiving SOC therapy was 86% compared to 49% among those receiving non-SOC treatments (p<0.05). Results comparing 2-year OS in SOC vs. non-SOC remained significantly different in subgroup analysis for stage I rectal (87% vs. 47%, respectively), stage II-III colon (80% vs. 53%, respectively) and stage II-III rectal cancer patients (85% vs. 67%, respectively). We are currently working on implementing multivariable Cox proportional hazards modeling accounting for age and other comorbidities that are important predictors of survival. CONCLUSIONS. Compared to patients receiving SOC therapy, overall survival is poorer among stage I-III colorectal cancer patients who receive non-SOC treatments. Multivariable Cox proportional hazard analyses adjusted for relative survival and expected remaining life years based on patients' age are underway to investigate in more detail the associations of specific treatments with overall survival in older colorectal cancer patients.

Keywords: colorectal cancer, overall survival, age, geriatric