Association Of The Lung Immune Prognostic Index With Outcome In Patients With Metastatic Urothelial Cancer Treated With Immune Checkpoint Inhibitor

545 Background: Prognostic factors for survival in metastatic urothelial carcinoma (mUC) have been reported in patients (pts) treated with chemotherapy (Bellmunt and al, JCO 2010). However, no biomarkers have been clearly identified in the setting of immune-checkpoint inhibitors (ICI). The Lung immune prognostic index (LIPI) was associated with clinical outcomes for ICI in lung cancer (Mezquita et al, Jama Oncol 2018) and other tumor types (Varga et al, TAT 2019). In this multicenter retrospective study, we correlated LIPI with outcomes in pts with mUC treated with ICI. Methods: Pts with mUC enrolled from May 2013 to July 2018 in 7 high volume centers were analysed. LIPI score includes: LDH > upper limit of normal and neutrophil/[leukocytes minus neutrophils] ratio (dNLR) > 3. The following LIPI subgroups were defined: good prognosis (0), intermediate prognosis (1) and poor prognosis (2 factors). Median (m) progression-free survival (PFS) and median overall survival (OS) were calculated using Kaplan-Meier method, log rank test was used for statistical comparison. Cox model was used for multivariate analysis. Results: To date, 152 mUC pts have been enrolled and preliminary analysis have been performed in 135 pts. Median age was 67 years, 111 (82%) pts were male, 106 pts (79%) had ECOG PS 0-1, 31 pts (23%) had liver metastasis. Median follow-up was 21.1 months (mo) (95% CI; 16.3-24.5), mPFS was 3.6 mo (95% CI; 2.6-6.0) and mOS was 13.8 mo (95% CI; 11.5-23.2). LIPI classified the population in good (56%), intermediate (35%) and poor (9%) prognosis group. In multivariate analysis, estimated mPFS in good, intermediate and poor prognosis were 5.8 mo (95% CI; 2.6-6.0), 3.6 mo (95% CI; 3.3-16.2) and 1.2 mo (95% CI;0.1-NR), respectively (p = 0.001). Estimated mOS in good, intermediate and poor prognosis were 17.3 mo (95% CI; 13.0-36.8), 16.9 mo (95% CI; 6.9-NR) and 5.4 mo (95% CI;2.5-NR), respectively (p = 0.19). OS data will be validated on larger cohort. Conclusions: The LIPI score is associated with clinical outcome to ICI and may be a useful tool for identifying patients who may not benefit from ICI. Validation in independent prospective cohort is ongoing.