1419P - Basal lymphopenia as a potential predictor of benefit from immunotherapy in metastatic non-small cell lung cancer

Abstract

Background: Immunotherapy (IO) is an established treatment (tx) option for metastatic non-small cell lung cancer (mNSCLC). Nonetheless, as only a minority of pts shows tumor response, research is focusing on identification of predictive factors to improve pts’ selection. The potential role of blood cell count alterations has been proposed, but evidence is contradictory. We aimed at studying the role of lymphopenia in a cohort of mNSCLC pts treated with IO. Methods: We retrospectively collected data about all mNSCLC pts treated with IO at Istituto Nazionale dei Tumori, Milan, Italy, from 04/2013 to 01/2018. Basal lymphopenia (Lp) was defined as a lymphocyte count (LC)≤900/mL at the first administration of IO and was considered as a categorical variable. Survival was estimated with Kaplan-Meier method; log-rank test was used to compare curves. Multivariate analyses were performed with Cox proportional model. Results: We identified 150 pts, with a median age of 66.5 years. IO was an antiPD1 in 64.0% of cases, an antiPDL1 in 31.3% of cases, and a combination antiPDL1/CTLA4 in 4.7% of cases. IO was administered as a first line tx in 23 pts, as a second line tx in 66 cases, as a more advanced line of tx in 61 cases. Median progression free survival (PFS) and overall survival (OS) of the global population were 3.2 and 11.2 months (mos), respectively. Though non-statistically significant, there was a tendency towards a lower response rate (RR) for cases with basal Lp (10.0% vs 25.0%, p.0881); disease control rate (DCR) for the same group was significantly worse (30.0% vs 58.4%, p.0074) than for cases without Lp. Pts with Lp also showed shorter PFS and OS than cases with normal LC (PFS 1.9 vs 3.0 mos, p.0010; OS 4.5 vs 13.5 mos, p<.0001). The impact of LC on OS retained significance after correction for the effects of performance status, which was the only other variable influencing this endpoint. Conclusions: The presence of Lp at the beginning of IO was related to inferior disease control and shorter survival in the analyzed cohort. Given the limitations of a retrospective study, these results need confirmation in larger case series. Nonetheless, the suggestion that Lp may predict poor benefit from IO in mNSCLC warrants further investigation. Legal entity responsible for the study: Istituto Nazionale dei Tumori. Funding: Has not received any funding. Disclosure: G. Lo Russo: Honoraria: Eli Lilly, BMS, AstraZeneca. C. Proto: Honoraria: BMS, MSD, Eli Lilly. D. Signorelli: Honoraria: AstraZeneca, Boehringer Ingelheim, BMS. M.C. Garassino: Honoraria: Boehringer Ingelheim, AstraZeneca, Roche. M. Imbimbo: Honoraria: BMS. All other authors have declared no conflicts of interest.