Protocol Design as Business Strategy: Q&A with Mark Freitas, Alvarez & Marsal

As drug development grows more competitive and capital more constrained, the consequences of early protocol decisions have never been higher—particularly for the small and midsize companies making bet-the-company choices without adequate board-level visibility.

To explore this further, Applied Clinical Trials spoke with Mark Freitas, managing director and life sciences practice lead at Alvarez & Marsal, about what's actually driving late-stage failures, why reimbursement strategy needs to be integrated earlier into trial design, and whether artificial intelligence (AI) is delivering on its promise of shorter development timelines.

ACT: From your vantage point, how has the role of clinical trial design evolved over the past few years?

Freitas: That's a good question. On the surface, the evolution has been well documented—adaptive designs, decentralized trials, real world evidence integration. But if you dig beneath the surface, the more consequential shift isn't in the trial design methodology itself. It's really more in who's making the design decisions and what's driving that.

We see a lot happening particularly in the small and midsize market cap companies. There are much more significant consequences around those design decisions, and with access to capital being somewhat constrained, companies have to make choices and strategic trade-offs. That's where we see a lot of opportunity and evolution happening.

57% of protocols get at least one substantial amendment, and nearly half of that is avoidable. But those amendments come with costs and sometimes delays, and those can be critical to a company that's smaller, cash constrained, and trying to manage its runway. We did a report where we looked at complete response letters released by the FDA—about 200 CRLs—and what we found was that the delays were enormous, but what was actually causing them was more operational than scientific. Just operational decisions and questions.

That leads to a critical dynamic around trial design evolution: does leadership understand the consequences of the big decisions being made early in a company's life cycle? Particularly in the small and midsize markets, you have bet-the-company decisions being made often without meaningful board or C-suite involvement. That's a part where we're seeing much more active engagement now.

ACT: Obesity and oncology are often cited as examples of especially crowded pipelines. What makes trial design in those therapeutic areas uniquely challenging right now?

Freitas: The competitive dynamics are really hard. If you look at obesity at a broad level and double-click on GLP-1s specifically, there are 39 GLP-1 drugs in development across 34 companies. So what makes you different? Obesity is a little bit easier in terms of the patient population you're targeting, but you are competing with all those companies to find those patients. When you get to oncology, that problem just magnifies. And when you look at obesity more broadly, you're talking about roughly 300 drugs being developed across the market—about 10x the GLP-1 segment alone.

So it's not just, can you show weight loss? It's can you do it in a way that's competitively differentiated—from a regulatory evidence standpoint, from a commercial differentiation standpoint, and fundamentally from a business strategy standpoint. Given the rate of M&A in the industry, you have to think about protocol design and clinical trial strategy as your business plan, to the extent that you could become an acquisition target or are trying to create some type of exit for shareholders.

Whether it's obesity or oncology, if you pick the wrong tumor type, don't build out the right evidence generation plan, or your trial design is too restrictive, it can be really difficult. And the small and midsize companies have a challenge that big pharma doesn't—they have to make choices. They can't necessarily do a comparator trial, a label expansion trial, and an outcomes trial. They have to take strategic bets and advance one path at a time, all while the market is moving ahead of them.

It used to be that you could rely on the science of your drug. But what you're seeing now is that science is outpacing the IP associated with it. Even if you're first to market with a novel mechanism, you don't have enough patent life before the next company comes up with a more novel approach—roughly two and a half to three years before the next mechanism is on the market. That speed and pace makes it much more complex than some other disease states.

ACT: You've spoken about access and downstream adoption shaping protocol decisions much earlier. What does that look like in practice from a clinical development standpoint?

Freitas: Fundamentally, the thing we talk to our clients about is PTRs—probability of technical and regulatory success. And we advocate for adding another R into that for reimbursement. There's an initiative out there around PPAS, probability of pricing and access success, but that's usually dealt with separately. What we're advocating for is integrating it into the PTRs process.

A practical example would be a phase three placebo-controlled trial that clears the FDA—great, you got approval—but then you can't get reimbursement for it. Drugs are getting to market but can't actually generate value. A good example of that is the Roctavian asset with hemophilia. It was a gene therapy that offered some pretty remarkable benefits for certain patients, but it was super expensive, couldn't justify the economics, and ultimately got pulled from the market. Patients are being deprived of really meaningful medicines in some cases because the value story doesn't hold up.

Integrating the reimbursement angle is probably one of the more important things we see. Everyone's tackling it in different ways, but the failure point is that there isn't a standard approach. There's a lot of fluidity around what it actually means to be reimbursement-successful. You're looking at global launches with different payers, or even in the US navigating the way PBMs handle formulary negotiations. What one company looks for to justify formulary placement is very different from what another would look for.

It's hard, but it's critical to the success of any development program. A curative therapy or breakthrough therapy—everyone will find a way to get access to that. It's the marginal improvements where the value equation gets murky, and that's where it becomes much trickier to figure out the access and evidence piece.

ACT: Many organizations are using AI and advanced analytics to accelerate protocol design. Has that actually shortened development timelines?

Freitas: It's funny, because of the way the process works and the timing element of it, the short answer is yes, it has shortened timelines. The longer answer is we'll see—because it takes a while for drugs to go through the process and get approval. You don't have that many success stories yet of a drug that went through development quickly, got approval, and is now on the market helping patients. There are some, but it's not the norm quite yet, because it's only been a few years and it still takes a few years to get through the process.

That said, AI is absolutely accelerating upstream tasks. Just a couple of days ago, BMS announced their partnership with Faro for protocol design—a perfect illustration of where some interesting concepts are coming together. Faro has some good data showing the ability to speed up protocol design. But BMS has also entered into partnerships with Evinova for AI-native study design, and other partnerships that aren't quite as public. So you start to look at the infrastructure play and the investments being made, and yeah, they're looking for ROI. Some of that is still experimentation to figure out what it is and how much can be justified. But they're finding pockets of innovation, and that's true across the industry.

You get this divide again between big pharma and the small and midsize companies. The smaller companies can't invest in large-scale infrastructure bets. But it might actually end up being better for them, because what they can do is leapfrog and pivot faster. A lot of the evidence coming out of those experiments is public, so they can learn from it and benchmark against a leading practice or new standard. As those standards take shape, they'll be in a better position to ingest that into their processes and go faster.

Ultimately, I would say it's the promise of shortening development timelines more than the actual shortening of them, at least right now. We certainly see elements of it. But can you say it always results in a faster approval and patients are benefiting? Still too early to tell. But it certainly seems like it's going to break through in the next year or two.

ACT: When AI and analytics speed up protocol development but internal governance and review cycles haven't kept pace, where does execution actually break down?

Freitas: A faster bad decision is still a bad decision. That's part of the confluence of the small and midsize company issue, where you have decisions being made in a more democratized way. People wear many hats, you're trying to go fast, you have all these insights and data to help you, but you might still not have the right expertise and leadership in place to act on that decision.

The converse is also true. You see both ends of the spectrum, just for different reasons. Big pharma has a lot of process and bureaucracy to work through, so faster insights don't help much if they're stuck in committees and monthly review cycles. The ability to act on decisions and drive consensus across stakeholders—the process hasn't caught up to the technology, and that's where they struggle.

So both sides of the spectrum have a challenge around execution breakdown. When you think about it, AI isn't just accelerating things—it's adding dimensions that didn't exist before, and that requires you to execute in a very different way. How do you rewire your operations to facilitate better and faster decision making? That combination of getting the right experts in the room, bringing the best minds together, but not so many minds that it bogs down the process—that's the holy grail that everyone's searching for.

Editor's note: This Q&A is a lightly edited rendering of the original audio/video content. It may contain errors, informal language, or omissions as spoken in the original recording.