Development of a Geriatric Prognostic Scoring System for Predicting Survival After Surgery for Elderly Patients With Gastrointestinal Cancer

Background The number of elderly patients with gastrointestinal cancer is rising as the population ages. This study aimed to assess the impact of a preoperative geriatric assessment on postoperative survival and to develop a geriatric prognostic scoring system (GPSS) for elderly patients. Methods Patients ( n = 544) age 75 years or older who had undergone radical surgery for gastrointestinal cancer were recruited for this observational study. Geriatric assessments (GAs) using the Barthel Index, the Mini-Mental State Examination, Instrumental Activities of Daily Living, the Vitality Index, and the Geriatric Depression Score were administered before surgery. Multivariable analysis was performed using a Cox proportional hazard regression model to identify significant prognostic factors. The GPSS was developed using regression coefficients of the multivariable regression to predict overall survival (OS). Thereafter, 165 consecutive patients were prospectively validated to test the authors’ model. Results The independent predictors of OS appeared to be GA as well as age, type of cancer, clinical stage, performance status, and body mass index. The patients were classified into high- and low-risk groups according to the GPSS. The overall 3-year survival was 79% in the low-risk group and 26% in the high-risk group (hazard ratio [HR], 5.69; 95% confidence interval [CI] 4.35–7.42; p < 0.0001). Furthermore, when GPSS was applied to independent cohorts, the patients in the high-risk group showed significantly poorer prognoses than those in the low-risk group (HR, 4.49; 95% CI 2.65–7.60; p < 0.0001). Conclusions Geriatric assessments were closely associated with postoperative OS. The GPSS is useful in predicting postoperative prognosis and may help determine treatment strategies for elderly patients with gastrointestinal cancer.