e20604 Background: Oncogene-addicted non-small cell lung cancer (NSCLC) is an heterogeneous group including different molecular alterations. Most single agent (IO) and combination immunotherapy (CT-IO) trials excluded specific oncogene-addicted NSCLCs such as ALK and EGFR; however, it is likely that other oncogene driven tumours were not tested for but included in these trials and could be actually treated with IO in clinical practice. Methods: Among 428 patients treated with IO or CT-IO as first-line treatment, 30 patients were identified to harbour rare molecular alterations (37% non-smokers, 63% former or current smokers): METex14 skipping mutation (n = 10), MET amplifications (n = 4), BRAF other than V600E mutations (n = 7), HER2 exon 20 insertions (n = 5), RET fusions (n = 2), EGFR exon 20 insertions (n = 2). We analyzed progression-free survival (PFS) and overall survival (OS) in patients harbouring these mutations who underwent first-line IO (if PD-L1≥50%; n = 23) or CT-IO (if PD-L1 < 1-49% n = 2 or < 1% n = 5). Results: In whole population median PFS was 5,1 months (95%CI: 2,39-11,29) and median OS 25,5 months (95%CI: 8,80-60,98); 13 patients experienced progressive disease as best response, the disease control rate (DCR) was 56,7%. Patients treated with IO (n = 22) had a mPFS of 4,7 m (95%CI: 2,13-23,48) and a mOS of 29,8 months (95%CI: 10,08-60,98); those treated with CT-IO (n = 8) had a mPFS of 5,1 m (95%CI: 2,29-6,01) and a mOS of 8,5 m (95%CI: 7,68-12,51). Among the most represented molecular alterations, MET alterations (n = 14, 3 treated with CT-IO and 11 with IO) had a mPFS of 5,1 m (95%CI: 1,67-23,90) and a mOS of 25,5 m (95%CI: 8,80-60,72). In particular METex14 skipping mutations had a mPFS of 5,1 m and a mOS of 25,5 m, while MET amplifications had a mPFS of 2,4 m and mOS of 10,3 m. BRAF nonV600E mutations (n = 7, 2 treated with CT-IO and 5 with IO) had a mPFS 5,3 m and mOS not reached. HER2 exon 20 insertions (n = 5, 2 treated with CT-IO and 3 with IO) had a similar mPFS (6,0 m) but a poorer OS outcome (8,5 m). ( Table 1) Conclusions: Our results show a limited effectiveness of both IO alone and CT-IO as first-line treatment in oncogene addicted mNSCLC, even taking into account survival outcomes of upfront CT compared to targeted agents in pivotal trials. [Table: see text]