Authors M. DE MAN (1), S. LAURENT (1), K. GEBOES (1) / [1] UZ Gent, Gent, Belgium, Digestive Oncology and Gastroenterology Introduction Cancer is a disease that mainly affects older people and age is considered as a risk factor for chemotherapy toxicity. G8 is a validated comprehensive geriatric assessment (CGA) screening and has a proven prognostic value for functional decline and survival. The Hurria score (H score) is a validated predictive risk stratification schema for chemotherapy toxicity in older patients. Aim The aim of the study is to evaluate the incidence of side effects of chemotherapy treatment in older patients with gastrointestinal tumours and to evaluate G8 and Hurria score as predictive tools. Methods This retrospective study based on prospectively collected data evaluates patient ≥70 years diagnosed with gastrointestinal cancer and who started a new treatment line with chemotherapy (adjuvant, first or further line) between May 2015 and November 2016. Drug dosing was at the discretion of the treating physician. Data on patient, tumour and treatment were collected at baseline, including age, Eastern Cooperative Oncology Group (ECOG) score, weight, length, haematocrit, renal function, activities of daily live (ADL), instrumental activities of daily live (IADL), tumour location and chemotherapy dose are collected at baseline. G8 and H score were performed. Chemotherapy related grade 3/4 side effects (CTCAE 4.03), dose modifications and unexpected hospital admission were recorded during the first treatment period defined as the time between start and first evaluation. Incidence of side effects, number of patients (pts) and events (ev), were compared in patients with high risk (Hh) versus low (Hl) or intermediate risk (Hi) according to H score and in patients with G8 <14 versus patients with G8 ≥14. Chi square test and Student t-test were used to compare respectively the number of patients and events in the risk groups. Results 64 pts (M42/F22) were included, median age was 76 (range 70-90) y. 22 pts (34%) had 31 (48%) grade 3/4 side effects, 25 pts (39%) 29 (45%) dose reductions and 25 pts (39%) 32 (50%) treatment delay. 9 pts (14%) needed unplanned hospital admission. Most frequent side effects were neutropenia: 15 pts (23%), 20 ev and gastrointestinal 9 pts (14%) 12 ev. Hurria score was significantly correlated with incidence of grade 3/4 side effects, both for pts HI/i 14/52 versus Hh 8/12 pts (p=0.01) as for events Hl/i 18 versus Hh 13 ev (p= 0.03). G8 score was not significantly correlated with grade 3/4 side effects for patients: G 8 <14 12/37 versus G 8 ≥ 14 10/27 pts (p=0.70) nor for events: G 8 < 14:16 versus G 8 ≥ 14 15 ev (p=0.45). There was no significant correlation between neither Hurria nor G8 score and treatment delay, reduction or unplanned hospital admissions neither for number of patients nor events: delay Hl/I 19/52 versus Hh 6/12 pts (p=0.39), Hl/i 24 versus Hh 8 ev (p= 0.41) , G 8 <14 14/37 versus G 8 ≥ 14 10/27 pts (p=0.95), G 8 < 14 18 versus G 8 ≥ 14 12 ev (p=0.80), reduction: Hl/I 20/52 versus Hh 5/12 pts (p=0.84), Hl/i 22 versus Hh 7 ev (p= 0.52) , G 8 <14 16/37 versus G 8 ≥ 14 9/27 pts (p=0.40), G 8 < 14:20 versus G 8 ≥ 14 9 ev (p=0.16). Hospital admission: HI/I 7 versus Hh 2 (p=0.77), G8<14:4 versus G8 ≥ 14:5 (p=0.38). Conclusions During chemotherapeutic treatment of older patients with gastrointestinal cancer the incidence of side effects, dose adjustments and unexpected hospital admissions is high. The Hurria predictive score can define a group of patients with higher risk of grade 3/4 side effects. Performing this score can guide clinicians in treatment decisions in older patients and can be a basis for further research on optimizing treatment strategies in the elderly.
