On Tuesday, November 4th, 2025, Friends of Cancer Research (Friends) hosted its 18th Annual Meeting, bringing together experts from industry, academia, regulatory agencies, and patient advocacy. Over the past few months, Friends members Bernat Navarro-Serer, Hillary Andrews, and Mark Stewart hosted biweekly calls with expert working groups to discuss pressing topics in oncology clinical research that led to white papers released before the event. Each session at the Annual Meeting began with a brief presentation by the moderator, followed by a lively panel discussion and Q&A with participants from across the research community. This year’s meeting centered on three topics and a policy panel discussion:
- Seamless trial designs for rare cancers
- Demonstrating contribution of effect (COE) for combination therapies
- Selecting comparator arms in multi-regional clinical trials (MRCTs)
- Policy panel on the government shutdown, biomedical innovation, and U.S. Food and Drug Administration (FDA) policy
Session 1: Seamless Trial Designs for Rare Cancer Drug Development
The first session, focused on how seamless clinical trials can accelerate progress for rare cancers and was moderated by Allen Melemed (Chimerix, a Jazz Pharmaceuticals Company), with panelists Harindra Abeysinghe (Johnson & Johnson), Lola Fashoyin-Aje (Parexel), Misha Mehta (Friends Advisory Advocate), Chris Turner (Nuvalent, Inc.), and Ying Yuan (MD Anderson Cancer Center).
The White Paper
The white paper defined what makes a trial “seamless” and identified key considerations for implementing these designs. Traditional clinical trials move in a stepwise fashion: Phase I for safety and dose-finding, Phase II for initial efficacy, and Phase III for comparison to the standard of care (SOC). This phased structure provides rigor but often slows trial progress, especially when new protocols, site activations, and administrative steps create long gaps between phases.
For patients with rare cancers, those delays can be devastating. Many rare cancers progress quickly and eligible patient populations are extremely small, meaning even short pauses can cause patients to lose eligibility, or worse, lose their chance to access a potentially life-saving therapy. Seamless trials can help avoid delays by merging these phases into a single, adaptive framework, allowing data from early participants to inform and accelerate later stages without starting a new trial. The white paper highlights strategies for designing these trials effectively, along with examples of ongoing studies already benefiting from this approach.
What We Heard from the Panel
Broadening the Definition of “Rare”
One of the first topics the panel explored was the challenge of defining what counts as a rare cancer. As panelists noted, rarity is not always about how many people are diagnosed; it can also depend on who is affected or how the disease manifests. Some cancers are rare because they affect very small populations, such as certain pediatric tumors. Others are rare subtypes of common cancers, defined by specific biomarkers or mutations. Panelists discussed how these nuances complicate trial design and patient recruitment but also make the case for more flexible and efficient approaches like seamless designs.
The Patient Perspective
A central theme throughout the session was the importance of patient input and the need to design trials that reflect real-world urgency. “Patients can provide a unique perspective that goes beyond the data,” said Harindra, emphasizing that their lived experiences and input should shape how trials are designed.
Misha shared a personal story from a pediatric lens:
“For the condition my son had, the prognosis was 9 to 12 months after diagnosis. There isn’t enough time to wait for a Phase I study to finish and then start a Phase II trial. We lived through that waiting—and we don’t have the luxury of that time. Seamless trials make so much sense because they combine safety and dosing in one continuous process. If there’s a positive signal, you can expand quickly and give families hope sooner.”
These insights underscored how seamless trials can directly benefit patients by reducing delays, maintaining continuity, and responding faster to emerging results, reinforcing that trial innovation must be paired with accessibility strategies so geography, travel burden, and system complexity do not limit who can participate.
Planning and Practicality
Panelists agreed that while seamless trials offer flexibility, they also require careful planning. Everything from choosing endpoints and ensuring drug accessibility to determining how many trial sites to include can influence both the feasibility and quality of the data collected.
Harindra cautioned that streamlining sites to reduce costs and variability can unintentionally limit diversity. “If you’re limiting the number of sites, that also means you may not get a real-world sampling of the population for a given cancer subtype.”
Ying added that although adaptive trial elements are appealing, they come with trade-offs. “Everyone likes flexibility, but every adaptation comes with a price. Those risks add up quickly, so it’s important to strike a balance between flexibility and statistical and regulatory credibility.” Simulations, thoughtful pre-planning, and strong communication with regulators were all noted as ways to manage that balance.
Chris shared that successful seamless trials must also be dynamic. “If you learn new information that you hadn’t anticipated, you can thoughtfully work that in and potentially amend the trial later. It’s challenging, but it can be done.” Lola emphasized that this adaptability is part of what makes seamless trials so powerful. “These studies are iterative. As you generate data and learn how patients respond, you adjust and evolve. Discussing specific examples and seeing what worked, what didn’t, and why—that’s what will move this approach forward.”
Looking Ahead: Collaboration and Transparency
The panel concluded that strengthening seamless trial design for rare cancers will require collaboration among researchers, regulators, and patients worldwide. Chris noted that engaging with global regulatory bodies will be key to aligning expectations and ensuring that evidence generated through seamless designs is robust and credible.
Panelists agreed that sharing real-world experiences—both successes and setbacks—will be crucial for advancing understanding. As Lola put it, “more collaboration and transparency about what worked and what didn’t work, what was acceptable and what wasn’t, and why, will really help move this approach forward.”
While seamless trials will not be the right fit for every situation, they offer a promising path toward faster, more efficient, and more patient-centered cancer research, particularly where time and participants are limited.
Session 2: Trial Designs for Combination Drug Development
The second session focused on one of the most complex questions in modern oncology drug development: how to efficiently determine each drug’s contribution of effect (COE) within combination therapies. The panel was moderated by Gideon Blumenthal, (Merck & Co., Inc.), with panelists Julie Bullock (GSK), Gary Doherty (AstraZeneca), Elad Sharon (Dana-Farber Cancer Institute), and Carol Vallett (Friends Advisory Advocate).
The White Paper
Over the past decade, the pace of oncology drug development has accelerated, and combination therapies have become a cornerstone of that progress. By targeting multiple pathways at once, combination regimens can improve outcomes and expand treatment options. However, combining drugs also increases the risk of toxicity, making it essential to ensure that each therapy in a combination adds value.
To study this, researchers examine COE, or the impact of each drug in a combination therapy. The traditional approach to evaluating COE is a factorial trial design, where each drug is tested on its own, in combination, and against a control. For example, a four-arm study might include Drug A, Drug B, A+B together, and a control arm.
While this structure can yield strong evidence, it also presents real challenges, particularly as many new therapies are now co-developed from the start. Large factorial trials can require huge sample sizes, long timelines, and substantial resources, which may delay patient access to promising treatments. The white paper explored these challenges and discussed when alternative trial designs might be appropriate, provided they generate robust evidence and are aligned with regulatory expectations.
What We Heard from the Panel
Practical Challenges and Alternatives
Panelists acknowledged that while factorial designs are the gold standard, they are not always practical. Gary noted that “full factorial designs can lead to very large studies with very long timelines, which ultimately can delay patient access and approvals. There are real feasibility issues—not just recruiting enough patients, but ensuring that each arm makes sense clinically for the patient.”
Carol emphasized how study complexity affects patient participation:
“It can be a tough job to recruit the number of patients needed for a trial like that. It’s absolutely necessary that patients who are asked to participate understand the risks and possible rewards—both to themselves and to others in the near future. The more complex a trial is, the more this needs to be communicated.”
Because of these challenges, sponsors are increasingly considering modified or adaptive designs that can still demonstrate COE in the registrational trial without requiring every study arm.
Exclusion of Monotherapy Arms
A recurring discussion was when it is appropriate to exclude a single-drug (monotherapy) arm from a combination trial. Gary explained that the decision must depend on scientific rationale. “It’s very situation dependent. It depends fundamentally on the mechanism of action of each individual drug and what we expect from the combination. There are instances where you simply wouldn’t expect monotherapy activity, or where it’s known to be poorly effective.”
Carol added, “if the data clearly show that an arm is ineffective, it wouldn’t make sense—and it wouldn’t be ethical—to randomize patients into that arm or expose them to unnecessary toxicities.” Safety, panelists agreed, must always take precedence. Elad underscored this point, noting that safety is an obligation both throughout clinical trials and beyond.
Interpreting Non-Powered Analysis
The discussion then turned to non-powered analyses, smaller or exploratory analyses that can still provide meaningful insights when full statistical power isn’t feasible.
Julie shared:
“When it comes to non-powered analyses, it really depends on what you’re trying to conclude. Are you trying to show that a monotherapy arm isn’t effective? That there’s no activity, so you don’t want to randomize patients to that arm? We make a lot of decisions in drug development based on non-powered analyses—but we need agreement with regulators on how those results will be used.”
Carol added a patient-centered perspective, “If it’s a tiny number of people—like five—we’re not going to make decisions based on that. But if it’s a larger group and the results look promising, even without full statistical power, that’s something patients and caregivers can understand and value.”
Patient-Centered Trial Design
Panelists emphasized that maintaining patient trust and safety must guide every design decision. Carol noted that “patients need to be assured they’re not being unnecessarily exposed to toxicities. Gathering information on quality of life, including patient-reported outcomes, is incredibly important.”
Julie encouraged greater use of adaptive designs that protect patients as data accumulate:
“Adaptive combination studies can allow us to modify randomization as we go, making sure patients are placed into the arms with the best risk-benefit balance. Arms that show poor efficacy or high toxicity can be stopped early, and patients can cross over to more effective treatments.”
Elad added that using early endpoints, such as ctDNA, that are sometimes used for treatment decision-making in clinical practice can strengthen the overall evidence for determining whether a monotherapy arm is effective or whether a patient should be switched to a more effective treatment arm.
Ultimately, understanding COE is not just about data; it is about ensuring that every component of a therapy benefits patients without adding unnecessary risk or delay.
Session 3: Control Arm Selection for Multi-Regional Clinical Trials
The third session examined a key challenge in global oncology drug development: how to select appropriate comparator arms in Multi-Regional Clinical Trials (MRCTs). The session was moderated by Harpreet Singh (Precision for Medicine), with panelists Judd Englert (Amgen Inc.), Sumithra Mandrekar (Mayo Clinic), Kristin McJunkins (Friends Advisory Advocate), Raymond Osarogiagbon (Baptist Hospital Group), and Kathleen Winson (Genentech).
The White Paper
MRCTs accelerate global access to new therapies, improve the diversity and generalizability of clinical data, and enable more efficient regulatory review across regions. A central challenge in MRCT design is selecting an appropriate standard of care (SOC) comparator, which anchors interpretation of a new therapy’s efficacy and safety relative to existing treatments. SOC can vary widely across regions due to differences in regulatory approvals, clinical guidelines, real-world practice, access, and reimbursement. It is often not a single treatment, but a range of acceptable, context-dependent options that evolves over time as new evidence emerges. This presents two key challenges for trial design: (1) SOC may shift during an ongoing trial, and (2) multiple SOCs may exist simultaneously, complicating the selection of a single comparator.
Beyond the scientific complexity, comparator selection raises ethical and operational considerations. Patients and investigators must view the control arm as acceptable and relevant to current practice; otherwise, trials risk poor enrollment or high dropout rates. Trial sponsors must therefore take a thoughtful approach that balances scientific rigor, ethical integrity, and global feasibility.
What We Heard from the Panel
Balancing Interpretability and Global Applicability
A key theme from the panel was the need to balance interpretability and applicability across regions. Judd noted that as more SOCs become available, how and where they are used, approved, and sequenced can vary dramatically around the world. Yet, sponsors still aim to conduct a single global Phase III trial that remains relevant to as many patients as possible.
How SOC Is Selected
Kathleen explained, “A lot of times, the treatment landscape will inform the SOC.” Sumitra added, “You could do region-specific trials—and I know nobody likes that because it complicates everything. It makes trials larger, and when you have to combine data across them to come up with one overall answer, that’s very difficult. If you can plan upfront that you may have to pivot, what are your options?” The challenge, she noted, is that flexibility can introduce heterogeneity into the trial.
Kathleen emphasized that control arms should reflect how patients are treated in the real world, “We’re trying to design these trials to reflect real-world treatment settings. We want to be able to enroll these trials. We need to have control arms that mirror how patients are getting treated now.”
Patient-Centricity and Ethical Considerations
Raymond reminded the audience, “there is no such thing as a perfect trial. In a world of widening options, we have to make decisions in designing and choosing when we do our clinical trials. You can’t ask and answer every question in every clinical trial.” Sumitra added, “Not anticipating change is the worst thing you can do when you embark on a clinical trial today.”
Kristin highlighted the importance of transparency with patients, “You really need to be transparent about the evolving options, helping them understand that changes could happen, and making sure they have the information they need to decide whether to enroll.” Raymond added that no protocol should move forward without the engagement and feedback of patient advocates and community providers—input that has made a significant difference in choosing control arms.
Given the challenges of selecting a SOC that applies across multiple regions, one might question whether it would be easier to conduct separate trials in each area. However, as Judd explained, the purpose of MRCTs is to bring medicines to patients more quickly and to generate evidence more efficiently. “If we go back to doing a U.S.-only trial, or back to only doing a European, or, you know, in Asia. The challenge then becomes every one of those is going to be slower. Each study will take longer, and we’ll delay patient access,” he said.
Ultimately, the discussion reinforced that while selecting a single SOC to use across regions is complex, thoughtful planning, transparency, and engagement with patients, investigators, and global regulators can ensure that MRCTs remain both scientifically credible and ethically grounded.
Policy Discussion: Government Shutdown, Biomedical Innovation, and FDA Policy
At a lunchtime policy discussion moderated by Jeff Allen (Friends President and CEO), panelists Anna Abram (Akin), Remy Brim (BGR Group), Roy Herbst (Yale Cancer Center), and Amy McKee (AstraZeneca) explored how government funding uncertainty and evolving FDA policy influence biomedical innovation.
The Policy Context
Panelists agreed that while the FDA and medical research remain bipartisan priorities, government shutdowns and short-term budgeting disrupt public health planning. They emphasized that sustained innovation depends on stable funding and continuity, not crisis-driven cycles.
“Every time we face a potential government shutdown, it puts critical public health work on hold—and that has real consequences,” said Remy.
User Fee Reauthorization
Discussion turned to the next Prescription Drug User Fee Act (PDUFA) reauthorization, now in negotiation. Panelists viewed it as an opportunity to modernize regulatory processes and build on pandemic-era lessons.
“User fee reauthorization isn’t just a funding exercise; it’s about setting a vision for how we regulate innovation,” said Roy. Anna added, “PDUFA VIII is a moment to think about what kind of FDA we want for the next decade—one that’s science-led, flexible, and transparent.”
Collaboration and Lessons from COVID-19
The panel emphasized that public–private collaboration remains central to progress. The COVID-19 pandemic proved that flexible regulatory science and real-world evidence can accelerate lifesaving treatments.
“We’ve seen incredible progress when FDA, industry, and patient groups come together,” said Remy. “The challenge is sustaining that momentum when politics gets in the way.” Anna noted, “The agency’s ability to adapt during COVID shouldn’t be an exception; it should be a model.”
Protecting Science from Politics
Panelists urged insulating scientific work from political disruptions.
“We need to protect FDA’s core scientific work from becoming collateral damage in political fights,” said Amy. “Patients can’t wait for Congress to get its act together. Drug development timelines don’t pause.”
Despite challenges, participants expressed optimism about continued cross-sector collaboration and praised Friends for fostering nonpartisan dialogue.