The prognostic value of patient-reported Health-Related Quality of Life and Geriatric Assessment in predicting early death in 6769 older (≥70 years) patients with different cancer tumors
Chantal Quinten,Cindy Kenis,Lore Decoster,Philip R Debruyne,Inge De Groof,Christian Focan,Frank Cornelis,Vincent Verschaeve,Christian Bachmann,Dominique Bron,Syvlie Luce,Gwenaëlle Debugne,Heidi Van den Bulck,Jean-Charles Goeminne,Dirk Schrijvers,Katrien Geboers,Benedicte Petit,Christine Langenaeken,Ruud Van Rijswijk,Pol Specenier,Guy Jerusalem,Jean-Philippe Praet,Katherine Vandenborre,Michelle Lycke,Johan Flamaing,Koen Milisen,Jean-Pierre Lobelle,Hans Wildiers
DOI: https://doi.org/10.1016/j.jgo.2020.03.017
Abstract:Objectives: We aimed to determine the prognostic value of baseline Health-Related Quality Of Life (HRQOL) and geriatric assessment (GA) to predict three-month mortality in older patients with cancer undergoing treatment. Methods: Logistic regressions analysed HRQOL, as measured with the EORTC Global Health Status (GHS) scale, and geriatric information prognostic for early mortality controlling for oncology variables. The assessment was established with the odds ratio (OR), 95% confidence interval (CI) and level of significance set at p < 0.05. Discriminative power was evaluated with area under the curve (AUC). Results: In total, 6769 patients were included in the study, of whom 1259 (18.60%) died at three months. Our model showed higher odds of early death for patients with lower HRQOL (GHS, OR 0.98, 95% CI 0.98-0.99; p < 0.001), a geriatric risk profile (G8 Screening Tool, 1.94, 1.14-3.29; p = 0.014), cognitive decline (Mini Mental State Examination, 1.41, 1.15-1.72; p = 0.001), being at risk for malnutrition (Mini Nutritional Assessment-Short Form, 1.54, 1.21-1.98; p = 0.001), fatigue (Visual Analogue Scale for Fatigue, 1.45, 1.16-1.82; p = 0.012) and comorbidities (Charlson Comorbidity index, 1.23, 1.02-1.49; p = 0.033). Additionally, older age, poor ECOG PS and being male increased the odds of early death, although the magnitude differed depending on tumor site and stage, and treatment (all p < 0.05). Predictive accuracy increased with 3.7% when including HRQOL and GA in the model. Conclusion: The results suggest that, in addition to traditional clinical measures, HRQOL and GA provide additional prognostic information for early death, but the odds differ by patient and tumor characteristics.