Novel scoring tool predicts 1-year mortality in elderly cancer patients

MetaGENUA, a novel six-variable scoring scheme, can reliably predict mortality in elderly cancer patients, a recent study has shown.

Researchers conducted a single-centre, prospective analysis of 162 elderly cancer patients (mean age, 78±5.5 years; 57 percent female) who underwent a comprehensive geriatric assessment (CGA) before receiving treatment. The outcome of interest was 1-year mortality.

More than half of the participants had impairments in at least three clinical domains of the CGA and were identified as frail. This was mostly driven by multimorbidity, a heightened risk of malnutrition and an initial decline in instrumental activities of daily living (IADL).

Twenty-two patients died within the first year of surgery, yielding an overall 1-year mortality rate of 13.5 percent. Seventy-seven percent of the deaths were related to cancer.

Six variables made up the final prediction model: age, IADL, cancer stage, malnutrition risk, geriatric depression scale scores and cumulative illness rating scale scores. Taken together, the model had an overall accuracy of 0.65 in the training cohort. There was also relative ease of use, with an estimated mean time of MetaGENUA score generation of 20 minutes.

Upon validation, the model yielded a sensitivity of 0.59 and a specificity of 0.86. The resulting area under the receiver operator characteristic curve was 0.72 (95 percent confidence interval, 0.55–0.87).

“We anticipate that the routine use of MetaGENUA will allow a more rational administration of cancer treatments and translate into improved clinical outcomes both in terms of oncological and geriatric endpoints,” the researchers said.