BeOne’s new lung cancer drug: Response rate declines, while durability and brain penetration data support continued development
After the BGB-58067 sample was expanded, the objective response rate fell from 31.7% to 25.3%. However, newly disclosed data on duration of response, progression-free survival and intracranial efficacy led JPMorgan to continue supporting a pivotal lung cancer study in 2027.
Original source：J.P. MorganFirst published on Unhidea：Original source date：
Key takeaway
JPMorgan’s confidence in BGB-58067 extends beyond response rates to durability and brain penetration potential, but differences in efficacy across EGFR subgroups and the conditions for cross-trial comparisons constrain assessments of its competitiveness.
Why does JPMorgan remain positive despite the lower response rate?
BeOne’s PRMT5 inhibitor BGB-58067 achieved an objective response rate of 25.3% among 75 patients with MTAP-deleted non-small cell lung cancer, below the 31.7% reported for 41 evaluable patients in an earlier abstract. Median follow-up was 7.1 months; patients had received a median of 3 prior lines of therapy in the advanced or metastatic setting, and 66% had received immunotherapy.
The key new evidence concerns how long responses last: the 6-month duration-of-response rate was 74.5%, the estimated proportion of responders who remained in response at 6 months; median progression-free survival was 7.7 months. JPMorgan therefore believes that, despite the lower response rate, the durability data still support continued development.
Which patients showed stronger efficacy, and what are the limits of comparisons with similar drugs?
Efficacy differed markedly across groups: the objective response rate was 38.5% among patients with EGFR wild-type disease (39 patients), versus 11.4% among those with EGFR-mutant disease; median progression-free survival in the wild-type subgroup was 8.1 months. The overall response rate of 25.3% masks the difference between the two groups.
For reference, BMS’s similar drug navlimetostat reported a response rate of 34% (12/35) at ASCO in 2026, with median follow-up of 15.9 months and a median of 2 prior lines of therapy; the rate was 29% at the earlier follow-up of 11.7 months. The trials differed in follow-up duration and patients’ treatment histories. On that basis, JPMorgan considers BGB-58067 competitive at an earlier data cutoff, while emphasizing that cross-trial comparisons require caution.
There is also competitive pressure in the EGFR-mutant population: navlimetostat previously reported a response rate of 57% in this subgroup, but this involved only 7 patients, of whom 4 responded. The small sample limits the reliability of the comparison.
What new evidence has emerged on brain penetration?
Among patients with brain metastases evaluable for intracranial efficacy, 4 of 13 had an intracranial response under modified RECIST criteria, for a response rate of 31%. This links the drug’s brain penetration characteristics with efficacy against brain lesions and is a differentiating signal valued by JPMorgan; the direct evidence currently comes from a subgroup of 13 patients.
Does the safety profile support moving to the next stage?
Within the recommended expansion and dose-optimization range of 200–400 mg once daily, common treatment-related adverse events were nausea at 33%, anemia at 24%, fatigue at 21% and decreased appetite at 20%. The more common grade three or higher adverse events listed in the report were anemia at 7%, thrombocytopenia at 5% and neutropenia at 3%.
Of the 87 patients treated within this dose range, 6 discontinued because of adverse events during treatment, or 7%; in 5 patients, the events were treatment-related. JPMorgan considers the safety profile acceptable and, together with the durability and single-arm progression-free survival data, supports advancing a pivotal study in second-line and later-line MTAP-deleted lung cancer in 2027. The company estimates that MTAP deletion accounts for 12%–15% of non-small cell lung cancer.
