Clinical predictors of long-term responses to immune checkpoint inhibitors (ICI) in a multi-tumor cohort

BackgroundLong-term responses to ICI are observed in various solid cancers. While extensive research has been conducted to identify molecular biomarkers, additional efforts are needed to optimize patient selection in clinical practice.MethodsReal-world study of patients (pts) treated with ICI in a tertiary hospital (2015-2022). Pts with progression-free survival (PFS) >24 months (m) were selected. Baseline clinical features, exposure to steroids (ST), and immune-related adverse events (irAEs) were analyzed.ResultsWe evaluated 475 pts (median age: 67.5) with advanced solid tumors (NSCLC: 34.5%; urothelial: 17.3%; renal: 13.7%; melanoma: 11.2%; HNSCC: 10.7%; others: 12.6%). 84.8% received anti-PD1/PDL1, 13.3% anti-PD1+anti-CTLA4, and 1.9% anti-CTLA4 (42.1% in 1st, 43.2% in 2nd, 14.8% in further lines). 71 pts (14.9%) had a PFS >24 m, most of them (69%) having discontinued ICI (26.5% due to irAEs) after a median follow-up of 42.9 m. Median duration of ICI was 21.3 m. Only 15 pts (21.4%) developed PD after 24 m. Among pts with PFS >24 m, there was a higher proportion of baseline ECOG PS = 0 (53.5% vs 27%, p<0.01), autoimmune disorders (9.9% vs 5.9%, p=0.22), and exclusive lymph node metastases (mets) (22.5% vs 12.9%, p=0.03), contrary to liver (18.3% vs 26.5%) and bone (21.1% vs 29.1%) mets. Brain and lung mets were similar in both groups. 54.9% of pts with PFS >24 m received ICI in the 1st line (vs 39.9%, p=0.017). Best responses for pts with PFS >24 m (vs <24 m) were: CR (33.8% vs 1.2%), PR (53.5% vs 22.5%), and SD (11.3% vs 14.4%) (p<0.01). IrAEs were more frequent in pts with PFS >24 m (54.9% vs 29.5%, p<0.0001), and this group was more exposed to ST for irAEs (38% vs 17.6%). Late exposure to ST (>6 m of ICI) prevailed in pts with PFS >24 m (54.9% vs 35%, p<0.01), contrary to ST within 30 days (11.3% vs 26.5%, p<0.01) and 3 m after ICI initiation (16.9% vs 35.4%, p<0.01).ConclusionsOur results from a multi-tumor cohort suggest that a baseline PS = 0, using ICI in the 1st line setting, exclusive lymph node metastases, CR/PR as best response, and occurrence of irAEs are significantly correlated with a PFS >24 months. The early use of ST is negatively associated with PFS >24 m, contrary to the late exposure to ST and the use of ST for the management of immune-related toxicity.Legal entity responsible for the studyThe authors.FundingHas not received any funding.DisclosureA. Cortes Salgado: Financial Interests, Institutional, Advisory Board: AstraZeneca, PharmaMar, Daiichi Sankyo, MSD, Eisai, Accord Healthcare. P. Garrido Lopez: Financial Interests, Advisory Board: Janssen, MSD, Novartis, Medscape, Takeda, AbbVie, Amgen, AstraZeneca, Bayer, BMS, Boehringer Ingelheim, Daiichi Sankyo, Lilly, Pfizer, Roche, Sanofi. P. Gajate: Financial Interests, Advisory Board: BMS, Roche, Pfizer, Ipsen, MSD, Merck, Jannsen, Astellas, Eisai, Novartis. All other authors have declared no conflicts of interest.