Clinical, genomic, and pathological features and therapeutic outcomes of non-small cell lung cancer with MTAP-loss.

8530 Background: MTAP-cooperative PRMT5 inhibitors are under development for MTAP-loss solid tumors. However, clinical, genomic, and pathological features of non-small cell lung cancer (NSCLC) with MTAP-loss are still unclear. Methods: Using the large-scale clinical-genomic database of LC-SCRUM-Asia, the clinical, genomic, and pathological features and therapeutic outcomes of patients with NSCLC with MTAP-loss were investigated. MTAP-loss, CDKN2A-loss, targetable genomic alterations, and tumor mutation burden (TMB) were analyzed using FoundationOne CDx. PD-L1 TPS was evaluated using PD-L1 IHC 22C3. Results: MTAP status was successfully analyzed in 253 samples from NSCLC patients between February 2017 and May 2018. MTAP loss was detected in 54 of 253 (21%) NSCLCs distributed in 33 of 170 (19%) adenocarcinoma, 15 of 60 (25%) squamous cell carcinoma and 6 of 23 (26%) others. The patients with MTAP-loss NSCLC showed no significant difference in age, sex, smoking history, and ECOG PS compared to the patients with MTAP-intact NSCLC. In the patients with MTAP-loss NSCLC, the median age was 68 years old, 63% were male, 78% were ever smokers, and all had ECOG performance status (PS) 0-1. CDKN2A-loss was detected in 100% of MTAP-loss and 12% of MTAP-intact. The frequency of targetable genomic alterations did not differ significantly between MTAP-loss and MTAP-intact NSCLC (44% vs 38%). The frequencies of EGFR and KRAS mutations in MTAP-loss NSCLC were 20% and 15%, respectively, and those in MTAP-intact NSCLC were 21% and 10%, respectively. TMB was significantly lower in MTAP-loss NSCLC than in MTAP-intact NSCLC (Median 6.3 vs. 7.6 Mut/Mb, P = 0.03). MTAP-loss NSCLC tended to have lower PD-L1 TPS than MTAP-intact NSCLC (TPS ≥1%; 50 % vs 63%, P = 0.08). In the adenocarcinoma without targetable genomic alterations cohort, eight patients with MTAP-loss and 47 patients with MTAP-intact received platinum-based chemotherapies without immune-checkpoint inhibitors (ICIs) as the first-line treatment and six patients with MTAP-loss and 45 patients with MTAP-intact were treated with ICIs alone as any line treatment. There was no significant difference in the progression-free survival (PFS) of platinum-based chemotherapies as the first line between the patients with MTAP-loss and MTAP-intact (median 4.7 vs 4.6 months, HR [95%CI] 0.74 [0.35-1.55], P = 0.42). On the other hand, the patients with MTAP-loss treated with ICIs alone showed significantly shorter PFS compared to the patients with MTAP-intact treated with ICIs alone (median 1.9 vs 6.2 months, HR [95%CI] 3.62 [1.05-12.5], P = 0.04). Conclusions: The relatively low TMB and PD-L1 TPS might be involved in shortening the PFS in patients with MTAP-loss treated with ICIs alone. Other than that, NSCLC with MTAP-loss showed no distinct feature in patient characteristics, histopathology, and co-occurring targetable genomic alterations.