Background: NRAS-mutant melanoma is an aggressive subtype with worse prognosis. However, no targeted therapy has been approved to date worldwide. Tunlametinib (HL-085), a novel, potent, selective, oral small-molecule MEK1/2 inhibitor, has showed a favorable pharmacokinetics profile, acceptable tolerability and encouraging antitumor activity in the phase 1 study (BMC Med. 2023 Jan 4;21(1):2). Methods: This is an ongoing, multicenter, open-label, single-arm, phase 2 pivotal registration study. Patients (pts) with NRAS-mutant unresectable stage III or IV melanoma were enrolled and received tunlametinib 12 mg orally twice daily. A historical control of objective response rate (ORR) of 10% was predefined for sample size estimation (100 pts, assuming loss to follow-up rate of 5%) and efficacy evaluation. The primary endpoint was the confirmed ORR per RECIST v1.1 assessed by independent radiology review committee. Results: A total of 100 pts were enrolled. All pts were included in the safety analysis set and 95 pts were included in the full analysis set (FAS) for efficacy analysis. At cut-off date (August 17, 2022), median follow-up was 7.9 months (95% CI: 6.6, 9.8). In the FAS, the median age was 58.0 years (range: 24 to 84). Sixty-four pts (67.4%) had received prior immunotherapy (PD-1/PD-L1 inhibitor). Fifty-six pts (58.9%) were acral melanoma, 16 (16.8%) mucosal melanoma and 12 (12.6%) skin melanoma. Fourteen pts (14.7%) were stage III and 81 (85.3%) stage IV. The most frequent NRAS mutation types were Q61R (40.0%), Q61K (29.5%) and G12D (9.5%). Confirmed ORR was 34.7% (95%CI: 25.3%, 45.2%). Median progression-free survival was 4.2 months (95%CI: 3.5, 5.6), overall survival was immature, and 1-year survival rates was 57.2% (95% CI, 44.7%, 67.8%). Subgroup analysis showed that, in pts who had previously received immunotherapy, the confirmed ORR was 39.1% (95% CI: 27.1%, 52.1%). The most frequent treatment related adverse events (TRAEs) were increased blood creatine phosphokinase (CK), diarrhea, peripheral oedema, facial oedema and increased aspartate aminotransferase. Grade ≥3 TRAEs occurred in 68 pts (68.0%), of which 38.0% (38/100) were increased blood CK. Most of pts with CK elevation are asymptomatic and can be managed by dose interruption or reduction without permanent treatment discontinuation. No treatment-related death was reported. Conclusions: Tunlametinib was well tolerated and demonstrated encouraging treatment response rate in pts with advanced NRAS-mutant melanoma. These results indicate that tunlametinib could be a promising treatment option for NRAS-mutant melanoma, even for those immunotherapy failed pts. Clinical trial information: NCT05217303.