Ivonescimab's HARMONi-6 Progression-Free Survival Benefit Narrows With Longer Follow-Up on New China Label, Days After Third NMPA Approval

HARMONi-6 randomized 532 patients against tislelizumab

HARMONi-6 randomized 532 patients with previously untreated stage IIIB to IV squamous NSCLC to ivonescimab or tislelizumab, each with four cycles of paclitaxel and carboplatin followed by antibody maintenance. Akeso's approval release puts 39.0% of the enrolled tumors at a PD-L1 tumor proportion score below 1%, about 63% centrally located, and 33.8% with multi-site, liver or brain metastases. At the prespecified interim PFS analysis the bispecific arm reached a median of 11.1 months against 6.9, with a hazard ratio of 0.60. The Lancet paper on the interim overall survival analysis, with a data cutoff of Feb. 27, 2026, recorded 84 deaths on ivonescimab and 120 on tislelizumab, a median OS of 27.9 versus 23.7 months and a hazard ratio of 0.66, crossing the prespecified boundary of p less than 0.0049. Grade 3 or higher treatment-related adverse events ran 69% versus 59%, and grade 3 or higher hemorrhage 3% versus 1%.

Professor Shun Lu of Shanghai Chest Hospital, the principal investigator, said in the Akeso release: "We are highly encouraged by the dual-positive PFS and OS results from HARMONi-6 and proud to see the approval of ivonescimab plus chemotherapy for first-line sq-NSCLC."

Leerink pushed its modeled U.S. launch to 2029

BioSpace's Annalee Armstrong reported Aug. 17 that "as part of the approval, the NMPA posted some new data on the drug's label showing degradation of progression free survival (PFS) in the Phase 3 HARMONi-6 trial," and that the update drew "increased pessimism" from Leerink Partners, which wrote on Aug. 14 that "FDA's bar for [accelerated approval] based on PFS outcomes in 1L NSCLC is high, in our view." Endpoints News carried the same reading under the headline that the drug "shows smaller benefit with longer follow-up in China." Both outlets' full articles sit behind registration walls, and APO|APE News could not retrieve the Chinese prescribing information, so the size of the updated hazard ratio or median is not reported here.

Leerink's concern, per BioSpace, runs through the global HARMONi-3 trial, which compares ivonescimab plus chemotherapy with pembrolizumab plus chemotherapy in first-line metastatic NSCLC and is meant to support a U.S. accelerated approval. The firm now doubts the final PFS analysis, due in the second half of 2026, will show a "substantial benefit," and it pushed its modeled U.S. launch for the squamous chemotherapy combination to the fourth quarter of 2029 from the fourth quarter of 2027. Summit closed Aug. 14 at $13.35, down more than 4%, and BioSpace noted the company held $690 million in cash at June 30.

Summit's U.S. filing still carries a Nov. 14 PDUFA date

Summit's own Aug. 12 release lists three live programs in the United States. The HARMONi biologics license application, ivonescimab plus chemotherapy in EGFR-mutated non-squamous NSCLC after TKI progression, carries a Nov. 14, 2026, PDUFA date. HARMONi-3 runs squamous and non-squamous cohorts against pembrolizumab plus chemotherapy, and HARMONi-7 tests ivonescimab monotherapy against pembrolizumab in PD-L1-high first-line disease. Akeso's release also restates the HARMONi-2 result in PD-L1-positive NSCLC, median PFS 11.14 months versus 5.8 for pembrolizumab, as part of its case that the bispecific is "a cornerstone of the immuno-oncology 2.0 era."

Why This Matters to the APO|APE Reader

Leerink's own caveat, quoted by BioSpace, allows that VEGF antagonism may front-load benefit and that the longer HARMONi-3 readout could therefore land short of statistical success, a scenario the firm calls unlikely, but one that would make the Chinese label's updated PFS curve a preview of what a global, pembrolizumab-controlled trial produces once the early separation fades. The OS hazard ratio of 0.66 in HARMONi-6 does not move with that reading, and it came against tislelizumab, a comparator that is not the U.S. first-line standard that HARMONi-3 must beat. Whatever the NMPA label shows, the Nov. 14 PDUFA decision in EGFR-mutated disease is the first time a U.S. reviewer puts numbers from this class on a label, and it arrives before HARMONi-3 reads out.