OP0088 Immune-related adverse events of cancer immunotherapy – when inflammatory side effects are associated with survival: a single-centre prospective cohort study

Background Immune checkpoint inhibitors (ICI) represent a new standard of care for the treatment of selected advanced cancers and are still being investigated in many other tumour types. By enhancing the T-cell activation, a unique spectrum of inflammatory side effects has emerged, also known as immune-related adverse events (irAEs), including various well-described rheumatic manifestations. Data regarding the association between irAEs and patient outcomes are conflicting. Objectives To evaluate the incidence and characteristics of irAEs in patients receiving ICI, as well as the correlation with tumour response and patient survival. Methods This was a single-centre prospective observational study including all cancer patients receiving ICIs. The occurrence of irAEs, tumour response and patient outcomes were assessed on a regular basis. Overall survival has been considered from the start of ICI. Results From May 2015 to September 2017, 636 patients (70% male, mean age 64 years) have been included in this cohort while receiving anti PD-1 (n=435), anti PD-L1 (n=66) or anti CTLA-4 (n=3) as single agent or as sequential (n=100) or combined (n=32) therapies. Cancer types were mainly melanoma (n=293), non-small cell lung cancer (n=150) and renal carcinoma (n=83). Overall, 274/633 patients (43%) experienced irAEs, either 1 irAE (n=162), 2 irAEs (n=78) or ≥3 irAEs (n=34), with a median exposure time of 52 days30–91 for the first irAE. Dermatological irAEs were by far the most frequent (n=160), followed by digestive (n=80), endocrine irAEs (n=67), rheumatic (n=49) and pulmonary irAEs (n=17). So far, evaluation of tumour response was available for 551 patients, including 190 responders (complete response n=36 and partial response n=154), 192 patients with stable disease and 169 with progressive disease. 122/189 responders (65%) and 107/192 with stable disease (56%) experienced at least one irAE while reported only in 40/169 non responders (24%). Patients experiencing at least one irAE had an increased overall survival (median of 1169 days versus 224 days, p Conclusions Although irAE occurrence is not required for treatment benefit, it strongly associates with overall survival. Optimal multidisciplinary management of irAEs, including rheumatologists when needed, is worthwhile to maintain beneficial responses. Disclosure of Interest None declared