The US Food and Drug Administration has declined to approve ivonescimab, a first-of-its-kind antibody that blocks both the PD-1 immune checkpoint and the blood-vessel growth factor VEGF, for patients with EGFR-mutant lung cancer whose disease has progressed after targeted therapy. In its pivotal trial, the drug added to chemotherapy cut the risk of disease progression by 48%. But it did not produce a statistically significant improvement in overall survival, and the FDA said that was not enough. The decision, disclosed by Summit Therapeutics on 15 November, is the first US regulatory verdict on a closely watched class of cancer drugs.
In cancer medicine, two measures dominate the evaluation of new treatments: how long patients live without their disease getting worse (progression-free survival, PFS) and how long they live (overall survival, OS). Drugs are often approved on PFS when the evidence is strong and the treatment is not unduly toxic. The FDA’s decision on ivonescimab shows where that latitude ends.
What happened
Summit Therapeutics, which holds rights to ivonescimab in the United States, Europe and several other markets, said on Sunday that it had received a complete response letter from the FDA. This is the agency’s formal notice that an application cannot be approved in its current form. The decision date had been 14 November.
According to Summit’s account of the letter, the FDA concluded that “substantial evidence of effectiveness is not established” without a statistically significant improvement in overall survival, and that PFS alone did not support a positive balance of benefit and risk in this setting. Summit’s release cites no safety, manufacturing or data-integrity problem. The FDA does not publish complete response letters, so its reasoning is known only as the company reports it.
The application covered ivonescimab with platinum-based chemotherapy for adults with EGFR-mutated, non-squamous non-small cell lung cancer (NSCLC) whose disease had progressed after an EGFR tyrosine kinase inhibitor.
The trial behind the decision
The application rested on HARMONi, a Phase III trial in 438 patients. Ivonescimab plus chemotherapy was compared with chemotherapy alone, and the results reported by Summit and Reuters are as follows:
- Progression-free survival: the primary endpoint was met. Median PFS was 6.8 months with ivonescimab against 4.4 months with chemotherapy alone, a hazard ratio of 0.52, meaning a 48% lower risk of progression.
- Overall survival (primary analysis): 16.8 months against 14.0 months, a hazard ratio of 0.79. The p-value of 0.057 narrowly missed the conventional threshold for statistical significance.
- Overall survival (longer follow-up): Summit says it showed a hazard ratio of 0.76, consistent in Asian and Western patients. This was a post hoc analysis, not the pre-specified test.
- Safety: treatment-related adverse events led to discontinuation in 7.3% of patients on ivonescimab and 5.0% on chemotherapy alone. Treatment-related deaths were 1.8% and 2.3% respectively.
Why survival mattered here
The FDA’s position reflects a recurring concern. Adding a drug to chemotherapy adds cost and side-effects, so regulators want to know that patients gain something they value. A delay in tumour growth on scans does not always translate into longer life. In later-line lung cancer, after targeted therapy has failed, the agency evidently judged that a survival benefit had to be demonstrated, not inferred.
Post hoc analyses, however encouraging, cannot replace a pre-specified result, because looking again at accumulating data raises the chance of finding a favourable result by chance. A survival trend that falls just short of significance is therefore suggestive, not conclusive.
The FDA does have a route for approving drugs on early or surrogate evidence. Its accelerated-approval pathway allows conditional approval on endpoints “reasonably likely” to predict clinical benefit, provided confirmatory trials follow. That pathway, however, is intended mainly for conditions with unmet need, and it has come under growing scrutiny after several cancer approvals were later withdrawn when confirmatory trials failed to show benefit. For a regimen that adds a new drug to established chemotherapy, the agency appears to have set a higher bar. (This description of FDA pathways is general regulatory context, not drawn from the letter itself.)
What makes ivonescimab different
Most immunotherapies for lung cancer block a single target, PD-1 or its partner PD-L1, releasing a brake on the immune system. Ivonescimab is a bispecific antibody: a single molecule that also blocks VEGF, a signal tumours use to grow new blood vessels. Combining the two in one molecule is intended to concentrate both effects in the tumour.
The drug was developed by Akeso in China, where regulators approved it for this indication in May 2024. Summit licensed it in 2022 in a deal reported by Reuters to be worth up to US$5 billion. Reuters also reports that AstraZeneca said in September it would invest US$2 billion in Summit, an indication of how much commercial weight rests on the molecule. Many companies are now developing PD-1/VEGF bispecifics, so the FDA’s first decision on the class is being closely studied.
What it does and does not mean
The rejection applies to one indication, based on one trial. It is not a judgement that the drug class does not work. Summit says ivonescimab is being tested in other settings, including first-line lung cancer in the HARMONi-3 trial, colorectal and urothelial cancers, and in combinations with partners’ drugs. Summit has not said whether it will seek a meeting with the FDA or resubmit with more mature survival data.
The India angle
EGFR mutations are considerably more common in lung cancers among Asian patients, including Indians, than among patients of European descent. Patients whose disease progresses after targeted therapy have limited options. The FDA’s reasoning is relevant to India’s oncologists and to the Central Drugs Standard Control Organisation, which often considers drugs already approved, or rejected, elsewhere. NSH found no Indian filing for ivonescimab. The figure on EGFR prevalence is a general characterisation and should be sourced to Indian data before publication.
What to watch
- Whether Summit requests a Type A meeting or plans a resubmission.
- First-line results from HARMONi-3.
- Regulatory decisions on competing PD-1/VEGF bispecifics.
– Sai Chaitanya Puligadda



