Development of a predictor of one-year mortality in older patients with cancer by geriatric and oncologic parameters

Objectives: More than 60% of the new cancer diagnoses are currently made in older adults, a highly heterogeneous population. Reliable and time-saving tools to define older adults' prognosis are needed to inform the oncologist's decisions in routine clinical practice. We sought to define a multi-domain classification tool for the prediction of all-cause one-year mortality in a cohort of older adults with solid tumors. Materials and Methods: We conducted a single-centre, prospective study of patients with solid cancer aged 65 or older and with G8 score ≤ 14. All patients underwent a comprehensive geriatric assessment (CGA) before starting their surgical or medical treatment. One-year mortality was recorded. A CGA-based prediction tool of one-year mortality was developed and subsequently validated in two independent training and testing cohorts with a 70/30 split, respectively. Results: 162 patients were enrolled. Mean patient age was 78 ± 5.5 years. Forty-three percent of the patients were men. Colorectal and breast cancer were the most common diagnoses. The clinical variables selected for the development of the new classifier (MetaGENUA®) were: mini-nutritional assessment (MNA), instrumental day life activities (IADL), Cumulative Illness Rating Scale (CIRS), geriatric depression scale (GDS), age, and cancer stage. In our independent validation cohort, MetaGENUA® showed high specificity (0.86) and AUC = 0.71 (95% CI = 0.55–0.87). Conclusions: MetaGENUA® predicts one-year mortality in older patients with cancer with high specificity. As such, MetaGENUA® is predicted to reveal as a useful tool to guide the oncologist's decisions in clinical practice.