Akeso, Summit's PD-1 bispecific crushes Merck's Keytruda in study, signaling potential new standard in lung cancer

The Keytruda-beating data from Akeso and Summit’s ivonescimab that oncology industry watchers had been awaiting are here. The China-only data are impressive, signaling the PD-1/VEGF bispecific antibody’s potential as a new standard of care in non-small cell lung cancer (NSCLC).

Ivonescimab slashed the risk of disease progression or death by a whopping 49% compared with Merck’s Keytruda in the phase 3 HARMONi-2 study that Akeso is conducting in Chinese patients with previously untreated, PD-L1-positive NSCLC. The number, presented at the 2024 World Conference on Lung Cancer, was highly statistically significant.

Compared with Keytruda, ivonescimab extended the median time that patients lived without tumor progression by 5.32 months, reaching 11.14 months.

The magnitude of ivonescimab’s progression-free survival (PFS) improvement over Keytruda was consistent across key subgroups in HARMONi-2, including those whose tumors were squamous (50%), nonsquamous (45%), PD-L1-high as defined by a TPS score of at least 50% (52%), PD-L1-low (46%), plus patients with brain metastases (45%), nonsmokers (61%), as well as current (49%) and former (43%) smokers.

“These results highlight ivonescimab’s potential as a new standard of care” in first-line PD-L1-positive NSCLC, Caicun Zhou, Ph.D., from Shanghai Pulmonary Hospital, who presented the HARMONi-2 data at the WCLC conference, said in a statement. Zhou is the president-elect of the International Association for the Study of Lung Cancer, the organizer of WCLC.

The idea of dual targeting PD-1/L1 and tumor vasculature is not new. Roche’s PD-L1 inhibitor Tecentriq boasts an approval alongside the company’s VEGF inhibitor Avastin and chemotherapy for the first-line treatment of nonsquamous NSCLC. But Avastin is contraindicated in squamous patients because of an increased risk of bleeding in the lungs known as hemoptysis. 

Ivonescimab, a bispecific, didn’t seem to have the same problem in the study. Hemorrhages were recorded in 14.7% of patients in HARMONi-2’s ivonescimab arm, including in 1% at grade 3 with no more serious cases. The rates were 11.1% and 0.5% for the Keytruda arm. The grade 3 hemorrhage cases reported for ivonescimab were actually in nonsquamous patients.

Akeso’s newly minted CFO, Bing Wang, Ph.D., credited ivonescimab’s ability to leverage an anti-VEGF mechanism without Avastin’s bleeding problem to two factors. First, the PD-1 component of the bispecific helps anchor the whole antibody to the surface of the tumor, whereas a separate VEGF drug would float around in the blood. Second, Akeso’s proprietary Tetrabody technology, which produces antibodies with four antigen-binding sites, can create a very stable structure that enables a “collaborative binding” to further improve the binding specificity of both the PD-1 and VEGF, Wang explained in an interview with Fierce Pharma.

In general, the ivonescimab group in HARMONi-2 experienced higher rates of possible VEGF-related adverse events, most notably proteinuria at 31.5% and hypertension at 15.7%, versus 10.5% and 2.5% for Keytruda, respectively. But no VEGF-related events were above grade 3, and no treatment discontinuations happened because of them. In squamous patients, the rate of all serious treatment-related adverse events was 18.9% for ivonescimab, versus 18.7% for Keytruda.

But before ivonescimab can be crowned the new cancer immunotherapy king, there are few important considerations to note with the current trial. 

“If these benefits are confirmed, it is possible that the therapeutic harmony we’ve had in this first-line space may finally come to an end, and I think that’s a disharmony that we all should be rooting for,” John Heymach, M.D., Ph.D., chair of thoracic/head and neck medical oncology at MD Anderson Cancer Center, said during a discussant presentation of the HARMONi-2 results at WCLC.
 

The new standard of care?


First of all, HARMONi-2 was conducted solely in China, and therefore it likely won’t be able to support a filing with the FDA, which requires global studies. But Wang said Akeso didn’t spot any factors in HARMONi-2 that could have given ivonescimab’s Chinese results an edge that may be hard to replicate in a global phase 3 study.

Keytruda’s PFS performance in HARMONi-2, at a median of 5.82 months, was roughly in line with past experience with that drug. In the China extension component of the phase 3 KEYNOTE-042 trial in first-line, PD-L1-positive NSCLC, Keytruda monotherapy demonstrated a median PFS of 6.3 months, which matched the global trial population’s 6.4 months. Compared with HARMONi-2, that Merck study and its Chinese portion enrolled a slightly higher percentage of PD-L1-high patients, who generally respond better to PD-1 inhibitors. 

Xia
Akeso CEO Michelle Xia, Ph.D. (Akeso)

Wang acknowledged that Asian NSCLC patients tend to have better prognoses than non-Asians. So, the median PFS number may not be the best way to interpret ivonescimab’s efficacy. However, PD-1/L1 inhibitors’ treatment effects—at least against chemo—tend to be similar for Asian and non-Asian patients, he noted, pointing to an FDA analysis published in 2019. That’s why the 49% reduction in the risk of progression or death versus Keytruda gives Akeso confidence: it’s a relative comparison between ivonescimab and Keytruda in the same patient population. 

On the safety side, Leerink Partners analysts in a Sunday note flagged that Asia has a higher adoption of anti-VEGF drugs than Western countries, especially in lung cancer. The team therefore wondered whether doctors’ familiarity with the mechanism positively impacted treatment discontinuation.

Based on HARMONi-2’s findings, Akeso’s partner Summit plans to start a global phase 3 trial, HARMONi-7, pitting ivonescimab against Keytruda in first-line PD-L1-high NSCLC, Summit’s CEO Maky Zanganeh told Fierce Pharma in a separate interview.

“The key was willingness to go head to head with [Keytruda] and and to be superior, too; that takes courage, but what’s the alternative?” Bob Duggan, Summit’s executive chairman, said during an interview. “We’ll kind of sneak up on them over a four- or five-year period of time.”

Still, with better efficacy, Keytruda’s various combinations with chemotherapy have become the true standard of care in patients who can tolerate them, especially among those with PD-L1-low or -negative disease, as opposed to Keytruda monotherapy.

Summit has an ongoing phase 3 trial, HARMONi-3, that is combining ivonescimab with chemotherapy in first-line squamous NSCLC, with Keytruda and chemo serving as the comparator. The reason why Summit started in squamous “had nothing to do with lack of confidence in nonsquamous as much as exceedingly high confidence in the squamous population,” Dave Gancarz, Summit’s chief business & strategy officer, said during an interview, pointing to how Roche wasn’t able to reach squamous patients with Avastin.

Ivonescimab’s current data bode well for a potential showdown with Keytruda and chemo, even though ivonescimab doesn’t have to face Keytruda-chemo as a monotherapy. In fact, ivonescimab single agent’s 11.14-month median progression-free survival (PFS) length in an inclusive patient group is in the same ballpark of the 11.1 months that Keytruda and chemotherapy showed in PD-L1-high nonsquamous NSCLC patients in the landmark KEYNOTE-189 trial in an updated analysis. That’s a high bar because Keytruda and chemo have historically shown notably shorter median PFS time in PD-L1-low or in squamous patients.

However, such cross-trial comparisons should be viewed with a grain of salt because of different patient characteristics. For example, in a small subgroup of Japanese patients enrolled in the global KEYNOTE-189 trial, the Keytruda-chemo regimen’s median PFS reached 16.5 months.
 

Overall survival data


Besides being a single-country study, HARMONi-2 has yet to report data on whether ivonescimab can extend patients’ lives. That datapoint remained immature, as the current PFS analysis came after a median follow-up of just 8.67 months. In KEYNOTE-042, Keytruda showed a median overall survival of 16.7 months, or 20.2 months in the China extension portion.

Both Akeso and Summit were tight-lipped about how the overall survival (OS) number looks right now or its maturity level before a final analysis. But Wang did note that a PFS improvement above 40% will give some physicians confidence in ivonescimab’s strength. As Keytruda secured its various NSCLC indications based on OS improvements, ivonescimab likely needs OS results to win over the FDA and doctors.

An analysis of past studies done in July by Citi analyst Yigal Nochomovitz, Ph.D., found that a PFS improvement above 30% for an immuno-oncology trial in NSCLC is “very likely” to translate into a statistically significant OS benefit. A separate calculation by Citi analyst Wangbin Zhou, who highlighted the challenge of mounting an OS showing when subsequent therapies may blur the front-line treatment effect, argued that a PFS advantage of around 40% would likely bring ivonescimab a “clinically meaningful” OS benefit that’s stronger than 20%.

In a Monday note to clients, Nochomovitz said ivonescimab exceeded “just about all clinical expectations” against Keytruda. He highlighted that HARMONi-2 doesn’t allow crossover, “which should protect an expected OS benefit given the highly clinically meaningful PFS.”

“[W]ith necessary validating data (OS in HARMONi-2, then HARMONi-3, HARMONi-7 readouts) ivonescimab is poised to potentially deliver a practice-changing novel therapy for lung cancer,” Nochomovitz wrote.

Merck has previously tried to improve upon Keytruda with Eisai-partnered VEGFR inhibitor Lenvima in first-line PD-L1-positive NSCLC. In the phase 3 LEAP-007, the doublet led to a statistically significant 22% improvement in PFS but was linked to a 10% potential detriment to OS compared with Keytruda alone.

Nasty side effects played a role in the negative OS finding for the Keytruda-Lenvima pairing, Akeso CEO Michelle Xia, Ph.D., pointed out during an interview. A drug may be efficacious at shrinking tumors, but toxicity—and hence how long patients can stay on that drug—is very important in helping translate the antitumor effect into good OS, she said.

In HARMONi-2, treatment-related adverse events led to 1.5% rate of discontinuation and a 0.5% rate of death in the ivonescimab arm, versus 3% and 1%, respectively, for Keytruda. In LEAP-007, the treatment-related discontinuation rate was 29.1% for Keytruda-Lenvima and 11.2% for Keytruda.
 

Accelerating a 'cornerstone drug'


Akeso’s goal is to turn ivonescimab into a “next-generation I-O cornerstone drug,” Xia said. The Chinese biotech has performed several combination studies of ivonescimab with other in-house molecules, including its PD-1/CTLA-4 bispecific cadonilimab and its CD47 antibody ligufaliumab, and has seen acceptable safety profiles, Xia said. 

Thanks to positive data from the HARMONi-A trial in China, ivonescimab in May got its initial approval in its home country as part of a combination with chemotherapy for previously treated EGFR-mutant NSCLC.

Meanwhile, Summit is conducting the international phase 3 HARMONi trial evaluating ivonescimab plus chemo in NSCLC patients who progressed after treatment with a third-generation EGFR inhibitor such as AstraZeneca’s Tagrisso. The trial will include those patients in HARMONi-A who fit the description. 

Duggan
Summit Executive Chairman Bob Duggan, Ph.D. (Summit Therapeutics)

Summit obtained rights to ivonescimab in the U.S., Canada, Europe and Japan in late 2022. At that time, the hefty $500 million upfront payment and the potential deal value of up to $5 billion was a shock to many.

Now, as ivonescimab is the first to beat Keytruda in a head-to-head phase 3 trial, it could come into question whether Summit, which counts ivonescimab its lone asset, is the best company to manage the drug against a standard-of-care that generated $25 billion sales last year. Looking at Merck, the Keytruda maker in-licensed a TROP2-directed antibody-drug conjugate from China’s Kelun Biotech in late 2022 and has already launched 10 global phase 3 trials—including five in NSCLC—for the drug, which has not been approved anywhere.

Xia and Wang said Akeso partnered with Summit because of its leadership team’s experience. Duggan was the CEO of Pharmacyclics until the Imbruvica developer was bought out by AbbVie in 2015. Zanganeh was then Pharmacyclics’ chief operating officer.

In a “100-meter-race, we’re at the 10-meter mark now,” Duggan said in an interview. “We will accelerate, not decelerate. We have access to the funds [and] access to people that are as good as or better than any Big Pharma, and we’re all-in on this.”

As to whether Summit would seek out help from a larger company, Duggan said, “we’re going to optimize patient benefit, and we don’t need money. We don’t need partners per se. We need to focus on our business.”

Editor's Note: The story has been updated with a comment from John Heymach, M.D., Ph.D., during a WCLC presentation, plus additional input from investor notes by Leerink Partners and Citi analysts Sunday and Monday.