Prognostic score for second or further line immunotherapy in advanced non-small cell lung cancer (aNSCLC): An external validation

e14077 Background: Beyond PD-L1, nowadays oncologists can only use clinical characteristics to candidate patients for immunotherapy (IO). Previously, a clinical prognostic score composed by ECOG performance status (PS), sex, histology, stage, uses of platin-based therapy at first-line (1L) and response to 1L categorized 3 different prognostic groups for patients treated with second-line (2L) chemotherapy (CHT) (Di Maio, EJC. 2010 Mar;46(4):735-43.). The aim of this study is to assess if the same score is able to discriminate the outcome of aNSCLC pts treated in 2L or further-line IO, potentially helping decision making. Methods: We recorded data of patients collected from two institutional databases: Istituto Nazionale Tumori of Milan and IRCCS Oncologico Giovanni Paolo II of Bari, Italy. Overall survival (OS) was the primary endpoint and also progression-free survival (PFS) was assessed. Prognostic score was generated, and pts were divided into 3 prognostic groups: best (B: < 5), intermediate (I:5-9), worst (W: > 9). Results: Overall, 347 pts were included in the analysis (193 from Milan and 154 from Bari). Median age was 66 years (y) (30 – 88y), most were < 70 y (67.5%), male (70.7%), smokers (79.5%) and adenocarcinoma (74.6%). ECOG PS was: 0 (23%), 1 (54.5%) and 2 (22.5%). Pts distribution was: 28%, 51% and 21% in the B, I and W groups, respectively. Median OS was 18.0 months for B group, 8.5 months for I group (HR vs B 1.83, 95%CI 1.35 – 2.47, p < 0.001) and 2.6 months for W group (HR vs B 5.77, 95%CI 3.99 – 8.33, p < 0.001). Median PFS was 3.4 months for B group, 3.7 months for I group (HR vs B 1.35, 95% CI 1.03 – 1.77, p = 0.032) and 1.9 months for W group (HR vs B 2.51. 95% CI 1.80- 3.50, p < 0.001). Similar results were obtained stratifying the model by Institution. Conclusions: This clinical prognostic score, that was generated in patients treated with second-line chemotherapy, is able to highly predict outcomes of patients treated with IO. These results demonstrated that in pre-treated aNSCLC pts, the worst category has a dismal absolute life expectancy, and probably would not benefit from any active systemic therapy (independently if CHT or IO). Perhaps for these pts best supportive care could be the best choice.