Introduction: Predictors for mortality and toxicity in older patients with cancer are mainly studied in cohorts with various cancers at different stages. This study aims to identify predictive geriatric factors (PGFs) for early death and severe chemotherapy related adverse events (CRAEs) in patients aged ≥70 years with metastatic non-small-cell lung cancer (mNSCLC).
Material and methods: This is a secondary analysis of the multicenter, randomized, phase 3 ESOGIA trial that compared, for patients ≥70 years with mNSCLC, a treatment algorithm based on performance status and age to another algorithm based on geriatric assessment. To identify PGFs of three-month mortality and grade 3, 4, or 5 CRAEs, multivariate Cox models and logistic models, adjusted for treatment group and center, and stratified by randomization arm, were constructed.
Results: Among 494 included patients, 145 (29.4%) had died at three months and 344 (69.6%) had severe chemotherapy toxicity. For three-month mortality, multivariate analyses retained mobility (Test Get up and Go), instrumental activity of daily living (IADL) dependence and weight loss as PGFs. The combined effect of IADL ≤2/4 and weight loss ≥3 kg was strongly associated with three-month mortality (adjusted hazard ratio: 5.71 [95% confidence interval [CI]: 2.64-12.32]). For chemotherapy toxicity, Charlson Comorbidity Index ≥2 was independently associated with grade3, 4, or 5 CRAEs (adjusted odds ratio [95% CI]: 1.94 [1.06-3.56]).
Discussion: Mobility, IADL dependence, and weight loss were predictive of three-month mortality in a population aged ≥70 years treated for mNSCLC, while comorbidities were independently associated with severe chemotherapy toxicity.
Keywords: Frail dimension; Geriatric assessment; Lung cancer; Mortality; Toxicity.
Copyright © 2023. Published by Elsevier Ltd.