Background: Colorectal cancer (CRC) is prevalent in the older population. Geriatric assessment (GA) has previously been found to predict treatment tolerance and postoperative complications in older cancer patients. The aim of this study was to explore whether GA also predicts 1-year and 5-year survival after CRC surgery in older patients and to compare the predictive power of GA with that of established prognostic factors such as TNM classification of malignant tumors (TNM) stage and age.
Materials and methods: A cohort of 178 CRC patients aged 70 and older were followed prospectively. All patients went through elective surgery, and GA was performed presurgery. The GA resulted in patients being divided into two groups: frail or nonfrail. All patients were followed for 5 years or until death. Data were analyzed by Kaplan-Meier plots and the Cox proportional hazards model.
Results: Seventy-six patients (43%) were frail, and one hundred and two (57%) were nonfrail. Twenty-three patients (13%) died during the first year after surgery. One-year survival was 80% in the frail group and 92% in the nonfrail group. Five-year survival was significantly lower in frail (24%) than nonfrail patients (66%), and this difference was apparent both within the stratums of TNM stages 0-II and TNM stage III. In multivariable analysis adjusting for TNM stage, age, and sex, frailty was an independent prognostic factor for survival.
Conclusion: A GA-based frailty assessment predicts 1-year and 5-year survival in older patients after surgery for CRC. In localized and regional disease, the impact of frailty upon 5-year survival is comparable with that of TNM stage.
Keywords: Colorectal neoplasms; Frail elderly; Geriatric assessment; Mortality.