Summit, Merck must navigate new regulatory terrain in China-first cancer landscape

News Drug Development Summit, Merck must navigate new regulatory terrain in China-first cancer landscape July 27, 2026 | 10 min read | Heather McKenzie Twitter LinkedIn Facebook Email Print Summit Therapeutics revealed positive Western data last week from a Phase 3 trial of lung cancer candidate ivonescimab. In a flipped geographical clinical trials landscape, what more will be required for Summit and China-based partner Akeso—and other transnational tandems—to secure FDA approval? Last week, Summit Therapeutics and China-based partner Akeso revealed new data indicating that the efficacy of their PD-1/VEGF-targeting antibody ivonescimab can translate across geographies. This is a promising start, as China-partnered cancer assets typically must show efficacy in U.S. or global trials to attain the pinnacle: FDA approval. Why they must do so—and how sponsors should go about this—is more complicated.“Probably the biggest thing to keep in mind is this is just a flip of a process that has existed for, I don’t know, the last 20 years in drug development,” Daina Graybosch, senior managing director, immuno-oncology at Leerink Partners, told BioSpace. “It used to be that global studies were bridged to Asian countries.”Graybosch pointed to Merck’s immuno-oncology blockbuster Keytruda as an example. Merck ran a global study, KEYNOTE-407, of Keytruda in first-line non-small cell lung cancer (NSCLC). The company then conducted a bridging study—or a smaller substudy—for Chinese regulators, “with the theory that there might be something different about the disease” in this population. Today, China hosts more of the world’s clinical trials than the U.S. or any other country, and the gap is widening, with more and more valuable assets being tested there first.“What I feel has completely transformed the conversation is that many of the assets and the trials that are conducted exclusively in China are quite novel and innovative therapies that are not necessarily follow-on drugs,” Harpreet Singh, chief medical officer at Precision for Medicine and former division director of Oncology at FDA, told BioSpace. A prime example is Summit and Akeso’s ivonescimab, a PD-1/VEGF-targeting antibody that beat Merck’s blockbuster PD-1 blocker Keytruda in a Phase 3 NSCLC trial hosted in China. In that study, ivonescimab was 49% more effective at reducing the risk of disease progression or death. Then, in May, Akeso presented overall survival data from the Phase 3 HARMONi-6 study, touting a 34% reduction in the risk of death in Chinese patients. Still, questions swirled around whether these sparkling data would translate to the American population.“The risk that China data would not translate has been a lingering bear argument,” Stifel analysts wrote in a July 22 note. The newest data, from a two-year follow-up analysis of North American and European patients in the global Phase 3 HARMONi study, are “a strong counter to that pushback,” they added. Lung cancer Summit and Akeso sharpen case for ivonescimab with new Western data Summit Therapeutics and Akeso are answering questions raised about ivonescimab’s ability to translate to Western populations, though the partners have yet to share if the new findings are statistically significant. July 22, 2026 · 2 min read · Gabrielle Masson Read more Summit and Akeso don’t stand alone in this new terrain, however. Earlier this month, Merck and China-based Kelun-Biotech reported positive Phase 3 results for their antibody-drug conjugate (ADC) sacituzumab tirumotecan, or sac-TMT, which, when combined with Keytruda, showed the potential to replace platinum-based chemotherapy in first line NSCLC. Again, however, these data come from China, so Merck will need to replicate the results in a global population.It’s a dilemma that’s becoming more common as biopharma turns to China for innovation and clinical research. U.S. approval is not only critical for a Big Pharma company—it’s key for their Chinese partners as well, according to Boyang Wang, founder of purpose-driven longevity fund Immortal Dragons.Overall, the biopharma and healthcare sectors are “booming” in China, Wang told BioSpace. “But when it comes to overall economic value for Chinese companies, [that] is still lacking because it is very hard for them to actually go to the primary market, which is the U.S. North American market.” As a result, these companies have to “half-way sell their pipeline to another international Big Pharma and collaborate,” he said.So what do transnational pairs like Summit and Akeso or Merck and Kelun need to show to attain Western approvals, and how can regulators ensure that U.S. and European patients don’t miss out on transformational drugs? ‘Variability creates noise’ The most important factor to keep in mind under this new landscape, Graybosch said, is that in China, “you’re starting with a homogeneous population,” and then moving into a more diverse global setting.“Whenever you do any sort of clinical experiment in humans, variability creates noise, and you’re probably going to see differences in degradation of results,” she continued. This isn’t isolated to China, Graybosch emphasized. “I think if you went and you did a study only in people in Norway and then you did the same global, you’d find the same thing.”The FDA signaled its position on new drug applications based on single-country data in 2022 with the rejection of Eli Lilly’s PD-1 inhibitor sintilimab in NSCLC, developed in partnership with China’s Innovent Biologics. Importantly, sintilimab’s Phase 3 study was conducted only in China. Although the primary reasons behind the drug’s rejection had to do with the fact that it did not offer a significant advantage over existing PD-1 inhibitors for NSCLC, such as Merck’s Keytruda and Bristol Myers Squibb’s Opdivo, the FDA did recommend an additional multiregional trial if Lilly and Innovent wanted to resubmit. According to an SEC filing, Lilly terminated ex-U.S. rights to sintilimab in 2022. Drug Development In Expected Decision, FDA Rejects Lilly’s Sintilimab in NSCLC The U.S. Food and Drug Administration rejected Eli Lilly’s Biologics License Application, issuing a Complete Response Letter for the company’s PD-1 checkpoint inhibitor sintilimab. March 24, 2022 · 3 min read · Mark Terry Read more Within heterogenous geographical populations, there are several factors that can lead to different trial results. These include drivers of the disease or the genetics of the cancer, Graybosch noted, pointing to squamous NSCLC as one example.In HARMONi-6, “there was a big segment of nonsmokers”, Graybosch noted. “You don’t see nonsmoker squamous [NSCLC] in the U.S.” For the Chinese nonsmokers, was their cancer caused by something else, she asked, or might the discordance stem from trial enrollment differences?Analysts were asking similar questions following Akeso’s presentation of the data at the American Society for Clinical Oncology (ASCO) annual meeting in May.“Does the totality of the invonescimab data from HARMONi-6 point to broad efficacy across patient types? Simply put, we do not know,” BMO Capital Markets wrote to investors on May 31.One critique of HARMONi-6, Singh said, is that not enough older adults were enrolled. However, “Most, if not all . . . of our [multiregional] clinical trials in lung cancer would suffer the same critiques.”Some of these questions may be answered later this year as Summit expects a data readout from the multiregional HARMONi-3 trial in the second half. And a lot could ride on those results; the FDA is expected to render a decision on ivonescimab by Nov. 14.Another potential difference between China and U.S.-based trials could be the socioeconomic makeup of the trial population, Graybosch suggested. “In some ways [a] China patient is maybe more like a U.S. [or] global community patient” because more Chinese trial participants tend to come from the general community, whereas U.S. patients who are offered the chance to participate in trials “are somehow privileged,” which equates to a better prognosis, she said.Disparate overall survival results can also stem from post-trial care, according to Graybosch. Chinese patients often go back to their villages after the trial, she noted. “If you look at post-study treatments, you’ll see a certain percentage get Chinese traditional medicine. That’s not happening in a clinical trial, and that has some impact.”Finally, Graybosch said, differences in biology may be “much more important” with one mechanism, such as a VEGF bispecific, than with a PD-(L)1 drug alone. “We’re just going to need to see more of these studies replicated in the global settings.” Global complexity: FDA and EMA There is precedent for FDA granting approval to drugs based on data from a non-U.S. study.In 2024, the FDA approved four drugs based on trials without any U.S. participants, PharmaVoice reported. These included therapies for cancer and rare disease. BeOne Medicines’ Tevimbra, got the FDA’s nod for patients with unresectable or metastatic esophageal squamous cell carcinoma based on data from 0.3% U.S. participants, according to the Center for Drug Evaluation and Research’s 2024 Drug Trials Snapshots Summary Report.The FDA has a few options to regulate these drugs, Singh said. If a treatment shows a “truly transformative overall survival benefit” in another country, the FDA could grant approval based on those data but tied to “very substantial post-marketing requirements” to prove that efficacy does extend to U.S. patients.For a medicine that could be “potentially transformative,” a second option would be to conduct a bridging study, like Merck did with Keytruda, Singh added. In this scenario, instead of the FDA requesting a complete parallel multi-regional trial—which Singh said would cost “tens of millions of dollars” and four to six years to complete—a sponsor could instead be asked to generate a small amount of multi-regional or U.S.-only data to prove that the efficacy in one region will translate to the U.S.This option would also address the challenge of equipoise by ensuring that U.S.-based patients don’t miss out on a potentially life-extending drug. “If you already know how the control arm and experimental arm are going to perform, is it appropriate to repeat this experiment?” Singh asked. The answer, she said, is very nuanced.Here, Europe comes into play. Singh pointed to a hypothetical scenario in which the European Medicines Agency (EMA) approves a new front-line regimen for NSCLC and the FDA does not. If a company were to then include this new regimen as a partial or full control arm in a global trial, you’d have “an immediate problem in running a trial with one single control arm,” she said.Indeed, EMA guidance is ambiguous on the subject.“EMA does not have a dedicated policy for the evaluation of medicines that have been tested only in China, nor does it have one for cancer treatments,” an agency spokesperson told BioSpace in an email. “The guideline explains that if either the regulatory authority or the study sponsor have concerns that ethnic factors may influence a medicine’s safety, efficacy or dose response, additional data may be required.” Path to FDA approval So, what’s a China-partnered drug sponsor to do? “It’s all about the timing,” Singh said. Companies embarking on a development program outside of the U.S. that also plan to seek FDA approval should “start expanding your regions early in the course of drug development.”For sponsors with completed single-region trials who have seen “potentially transformative results” in a single population, Singh recommends engaging with the FDA and considering a robust post-marketing requirement or a bridging study.“I would be approaching FDA to talk about options A or B and really doubling down on the notion that the bridging strategy is a well-traveled regulatory pathway to make therapies available across different populations,” she added. Current FDA guidance also recommends a proactive approach.“The FDA encourages applicants to meet with agency officials in a ‘presubmission’ meeting when approval based solely on foreign data will be sought,” an agency spokesperson told BioSpace in an email.Both Summit and Merck appear to be taking a comprehensive approach. Summit and Akeso are awaiting data later this year from the multiregional HARMONI-3 trial, which compares ivonescimab and chemotherapy to Keytruda plus chemo in first-line NSCLC. The study, which comprises approximately 1,600 patients, is expected to readout in December 2028. Meanwhile, Merck and Kelun’s sac-TMT is currently being evaluated in five global multicenter Phase 3 studies, according to Kelun’s July 14 release.There is also considerable precedent for the design of such trials, according to Graybosch. “In some ways, it’s no different from taking a small, randomized Phase 2 proof of concept, which we know is a biased patient population, and then designing a Phase 3 trial—and we don’t always do a good job at that.”Graybosch pointed to Roche’s anti-TIGIT drug tiragolumab as a case in point. In July 2024, Roche halted the Phase 2/3 SKYSCRAPER-06 study after the drug combined with Tecentriq failed to significantly improve progression free and overall survival versus Keytruda and chemo in patients with NSCLC. This was after the combo succeeded in the Phase 2 CITYSCAPE trial.“The small Phase 2 [study] showed a tremendous benefit, and it degraded in a more heterogeneous Phase 3, even in the same regions, and they simply underappreciated it,” Graybosch said. Clinical trial design, she added, “is an art, not a science.” Drug Development Roche Axes Phase II/III NSCLC Study as TIGIT Drug Loses to Keytruda-Chemo Combo Roche’s tiragolumab, when combined with its PD-L1 blocker Tecentriq, did not significantly improve progression-free and overall survival versus Keytruda and chemotherapy in patients with non-small cell lung cancer. July 8, 2024 · 2 min read · Tristan Manalac Read more Twitter LinkedIn Facebook Email Print Cancer Lung cancer FDA Regulatory China Phase 3 Merck & Co. Heather McKenzie Heather McKenzie is senior editor at BioSpace. You can reach her at heather.mckenzie@biospace.com. Also follow her on LinkedIn.