HARMONi PFS Is Strong but November 14 Hinges on Fragile Nominal OS Claim

FDA Nears Decision on Summit/Akeso's Ivonescimab for NSCLC

Summit Therapeutics and Akeso's bispecific antibody, ivonescimab, is awaiting an FDA decision by November 14 for EGFR-mutated non-small cell lung cancer (NSCLC). The application is supported by data from the global Phase 3 HARMONi trial, which showed ivonescimab plus chemotherapy significantly improved progression-free survival (PFS) compared to placebo plus chemotherapy. While initial overall survival (OS) data fell short of statistical significance, a later analysis in September showed a nominally significant OS benefit in the overall global population, addressing a previous FDA requirement for approval. If approved, ivonescimab would be the first PD-1xVEGF bispecific antibody in the U.S. market.

• The FDA is expected to deliver its verdict on Summit Therapeutics and Akeso’s ivonescimab by November 14 for EGFR-mutated NSCLC. An approval would be significant as it would introduce the first PD-1xVEGF bispecific antibody to the U.S. market, potentially establishing a new backbone therapy in oncology, as suggested by analysts. This decision follows ivonescimab's prior approval in China for various NSCLC indications.
• The regulatory application for ivonescimab is supported by data from the global Phase 3 HARMONi trial. Initial findings from May 2025 demonstrated that ivonescimab combined with chemotherapy significantly improved progression-free survival (PFS) compared to placebo plus chemotherapy, meeting one of the study’s primary endpoints. This positive PFS outcome is a crucial component of the drug's efficacy profile.
• A key hurdle for U.S. approval was the FDA's previous requirement for a significant overall survival (OS) benefit. While initial OS data lacked statistical significance, an additional analysis presented in September showed a nominally significant OS benefit in the overall global population. This later data point is critical for addressing the FDA's concerns and improving the drug's approvability prospects in the U.S.

Addressing Unmet Needs in EGFR-Mutated NSCLC

Despite improved survival outcomes, acquired drug resistance to epidermal growth factor receptor tyrosine kinase inhibitors (EGFR-TKIs) is inevitable and the 5-year overall survival rate remains <20%. Central nervous system metastases remain a therapeutically challenging subset to treat owing to the blood-brain barrier (BBB), while genetic heterogeneity further impacts clinical responses and outcomes.

• Acquired T790M resistance limits early-generation TKIs, found in over half of patients with NSCLC progressing on first-generation TKIs, with T790M detected in 47.6% after acquiring resistance to afatinib and in half of lung adenocarcinoma after acquiring resistance to afatinib.

• Acquired resistance to osimertinib still emerges after approximately 10 months, with median duration before disease progression reached 14.2 months in an 86-patient cohort, driven by heterogeneous mechanisms including C797S mutations in EGFR (23.3%), amplification of MET (15.1%), amplification of HER2 (8.1%), mutations in PIK3CA (7.0%), transformation to small cell lung cancer (9.3%), and epithelial-mesenchymal transition (12.8%).

• Multiple concurrent resistance mechanisms complicate subsequent therapy, detected in 24.4% of the patient population, including EGFR-dependent and EGFR-independent mechanisms, bypass pathway activation, RET rearrangements co-existing with activated EGFR mutation, and histological transformation.

• Primary resistance restricts clinical benefits, with potential mechanisms including CDC73, SMAD4, RB1 and PIK3CA, BIM deletion polymorphism and EGFR exon 20 insertions, and concurrent TP53 mutations observed in 72.5% of patients.

• CNS progression remains a major limitation, with brain metastases complicating clinical evolution in approximately 25-40% of cases, leptomeningeal metastasis diagnosed in approximately 3%-5% of patients, and limited potential of first- and second-generation TKIs to cross the blood-brain barrier rendering them less effective in management of CNS metastases.

• Post-resistance treatment options are limited, with research on fourth-generation EGFR-TKI drugs with a C797s mutation at various experimental stages and no drug approved for clinical use.

HARMONi Trial: Ivonescimab's Efficacy in EGFR-Mutated NSCLC

In patients with central nervous system involvement after prior EGFR tyrosine kinase inhibitor therapy, the pooled analysis of KCSG-LU20-15 and KCSG-LU21-01 evaluated lazertinib in 72 patients, including 45 patients (62.5%) with leptomeningeal metastases (LM). Among 46 patients evaluable for intracranial response, the intracranial ORR was 50.0% and the intracranial DCR was 97.8%, with median intracranial PFS of 15.2 months. Among 66 patients evaluable for systemic response, overall ORR and DCR were 22.7% and 97.0%, respectively, with median PFS and overall survival of 10.6 and 17.8 months, respectively. A prospective single-arm two-cohort study (NCT03257124) evaluated osimertinib 160 mg in T790M-positive brain metastases (BM) or LM patients who progressed on prior EGFR TKI. In the BM cohort, intracranial ORR and disease control rate were 55.0% and 77.5%, respectively, median progression-free survival (PFS) was 7.6 months [95% confidence interval (CI) 5.0-16.6] and median OS was 16.9 months [95% CI 7.9-not reached (NR)]. In the LM cohort, intracranial disease control rate was 92.5% and complete response rate was 12.5%, median OS was 13.3 months (95% CI 9.1-NR) and median PFS was 8.0 months (95% CI 7.2-NR). The most common adverse events were decreased appetite, diarrhea, and skin rash; however, most were grade 1-2. In first-line EGFR-mutated advanced NSCLC, the randomized phase III AENEAS study (NCT03849768) compared aumolertinib versus gefitinib in patients with baseline CNS metastases. Treatment with aumolertinib significantly prolonged median CNS PFS compared with gefitinib in both cFAS (29.0 vs. 8.3 months; hazard ratio [HR] = 0.31; 95% CI, 0.17-0.56; P < 0.001) and cEFR (29.0 vs. 8.3 months; HR = 0.26; 95% CI, 0.11-0.57; P < 0.001), with confirmed CNS overall response rate of 85.7% and 75.0% in patients treated with aumolertinib and gefitinib, respectively, and no new safety findings observed.

For first-line combination strategies, MARIPOSA (NCT04487080) assessed amivantamab-lazertinib versus osimertinib in previously untreated EGFR-mutated advanced NSCLC, where amivantamab-lazertinib significantly improved overall survival versus osimertinib (HR, 0.75; P=0.005). MET amplifications (3.4% vs 13.1%; nominal P=0.002) and secondary EGFR mutations (1.4% vs 7.6%; nominal P=0.01) were significantly reduced with amivantamab-lazertinib versus osimertinib, and median second-line PFS was 8.4 vs 5.3 months (HR, 0.72; nominal P=0.02) among participants who started a first subsequent therapy. The open-label single-arm RELAY+ study evaluated ramucirumab (RAM) (10 mg/kg every 2 wk) plus gefitinib (GEF) (250 mg once daily) in East Asian patients with untreated NSCLC having EGFR ex19del or L858R mutations, with median (95% CI) OS of 47.4 (35.9‒57.6) months and 3-year OS rate of 61.8% at median follow-up of 42.5 mo. Grade 3 or higher treatment-related adverse events (AEs) were reported by 51 (62.2%) patients, with the most common grade 3 or higher treatment-emergent AE of special interest of hypertension (25.6%; grade 3 event only). A network meta-analysis of 5 trials involving 1791 patients found that the combination of osimertinib with chemotherapy and with ramucirumab, as well as the combination of lazertinib with amivantamab, significantly improved PFS compared to osimertinib monotherapy, with osimertinib plus pemetrexed-based chemotherapy demonstrating PFS benefits in the whole cohort and almost all subgroup analyses, while the combination of osimertinib with chemotherapy and the combination of lazertinib with amivantamab resulted in significantly higher incidence of treatment-related adverse events compared to osimertinib monotherapy.

After progression on EGFR tyrosine kinase inhibitors, the ORCHARD (NCT03944772) osimertinib plus savolitinib module enrolled 32 patients with MET gene amplification (≥4 copies of MET over tumor ploidy) after first-line osimertinib, with confirmed ORR of 47% (80% CI: 35-60), median duration of response of 14.5 months (95% CI: 5.6-18.7), median PFS of 7.6 months (95% CI: 3.2-15.9) and median OS of 20.7 months (95% CI: 9.9-34.8). Fourteen patients (44%) had grade ≥3 treatment-emergent adverse events; most commonly pneumonia (n = 3; 9%). The phase III CheckMate 722 trial (NCT02864251) evaluated nivolumab (360 mg once every 3 weeks) plus platinum-doublet chemotherapy versus chemotherapy alone, with median PFS of 5.6 v 5.4 months (HR, 0.75 [95% CI, 0.56 to 1.00]; P = .0528), median OS of 19.4 months versus 15.9 months, ORR of 31.3% versus 26.7%, and median duration of response of 6.7 versus 5.6 months, respectively. Grade 3/4 treatment-related adverse events occurred in 44.7% and 29.4% of patients treated with nivolumab plus chemotherapy and chemotherapy alone, respectively. The open-label phase 2 ILLUMINATE trial (NCT03994393) evaluated durvalumab, tremelimumab, and platinum-pemetrexed after EGFR TKI progression, with ORR of 31% (95% CI: 20-45) in T790M negative and 21% (95% CI: 12-34) in T790M positive cohorts, median durations of response of 9.5 months and 6.3 months, and median PFS of 6.5 months and 4.9 months, respectively.