Julia De Schrevel ^1,2* Clémentine Faucon ^1,2* François-Xavier Sibille ^3,4 Laura Dumont ^2,5* Francois R. HERRMANN ⁶ Héloïse Rouvière ^1,2* Sandra De Breucker ^1,2,6*

• ¹ Hôpital Erasme, Université libre de Bruxelles, Brussels, Belgium
• ² Université libre de Bruxelles, Brussels, Belgium
• ³ Université Catholique de Louvain, Louvain-la-Neuve, Walloon Brabant, Belgium
• ⁴ CHU UCL Namur Site Godinne, Namur, Belgium
• ⁵ Département d'oncologie médicale, Institut Jules Bordet, Université libre de Bruxelles, Anderlecht, Brussels, Belgium
• ⁶ Hôpitaux universitaires de Genève (HUG), Genève, Geneva, Switzerland

The Comprehensive Geriatric Assessment (CGA) is recommended for assessing frailty in older cancer patients but is time-consuming. The G8 screening tool was developed to select frail patients requiring CGA to optimize resources. The Edmonton Frail Scale (EFS) is another frailty scale validated for preoperative frailty screening, but scarcely studied in the field of oncogeriatrics. In this study, we examined the added value of the EFS in older cancer patients already considered as frail by the G8, by analyzing the association of EFS with CGA adjusted for age, gender, metastatic stage and comorbidity. We also analyzed the association of EFS with the one-year mortality rate after adjusting for cancer type and metastatic stage.Methodology: This retrospective study included patients aged over 70 years old with a new diagnosis of cancer, considered as potentially frail according to the G8 and who had had a CGA (N=380).The EFS identified 329 (86.58%) patients as frail and having a statistically significant predicted number of pathological components on the CGA (r =0.64, p<0.001).When adjusted for age, sex, comorbidity, and metastatic stage, the EFS was independently associated with the CGA (p<0001), as well as with comorbidity (p=0.004). The patients who died in the first year (43%) had a significantly higher mean EFS score (8/17) than living patients (6/17) (p<0.0001). After adjustment for cancer type and stage, EFS independently predicted one-year mortality (OR 1.17; IC95% 1.08-1.28; pseudoR 2 = 0.228, p<0.001).The EFS is a reliable tool for predicting frailty identified by the CGA in an older cancer population pre-selected as frail by the G8. EFS is an independent predictor of one-year mortality after adjustment for confounding factors. Validation of the EFS as a screening tool for frailty in cancer requires further studies to assess its performance in patients with normal G8 scores.