Naji Gehchan: Welcome to SpreadLove in Organizations, the healthcare leadership podcast where we explore leadership with purpose.
I’m Naji, your host, and I’m delighted to be joined today by Pedro Coelho, Founder and CEO of BIORCE, an AI-native company transforming how clinical trials are designed, optimized, and executed.
After years advising pharmaceutical and biotech companies as a strategy consultant in the life sciences sector, Pedro saw firsthand how inefficiencies in clinical research were slowing access to life-saving treatments. The consequences weren’t just operational, they were human. Drugs were too expensive, timelines were too long, and too many patients were left behind.
In 2024, he founded BIORCE to change that. BIORCE is building what it describes as the world’s first true AI assistant for clinical trials.
Pedro is passionate about the intersection of technology, data, and human impact, and believes AI has a central role to play in reshaping how treatments reach the people who need them most.
Pedro, it’s such a pleasure to see you again and have you with me today.
Pedro Coelho:
Thank you, Naji. It’s a pleasure to be here. I’m looking forward to this discussion.
Naji Gehchan:
Pedro, I always start where I always do. Take us through your personal story and what led you to found one of the hottest AI-native companies in clinical research today.
Pedro Coelho:
I think it dates back to a random opportunity I had earlier in my career. A biotech company was running a Phase 3 study in Europe that wasn’t going very well, to say the least. I came in as a consultant to understand what was going wrong.
After a deep dive, we identified a number of issues, but they all boiled down to what I call a lack of love in the trial. The CRO running the study simply wasn’t giving it enough attention. It had become a slow and painful relationship between the sponsor and the CRO.
That experience eventually led me to start a CRO focused on rescuing struggling studies. We successfully turned that study around and brought it to a good place, and then we went on to rescue many others.
Eventually, what I describe as both the worst event and the best thing that ever happened in my life occurred.
My father was diagnosed with melanoma.
At first, I wasn’t overly concerned. Melanoma is often one of the more treatable cancers when caught early. I remember telling him not to worry because, compared to many cancers, melanoma is often manageable.
Unfortunately, I ended up eating my words.
Three months later, it had progressed to Stage 4 and had metastasized extensively. I had never seen anything like it, and neither had his oncologist.
At that point, I became obsessed with finding options that might save his life.
We found two studies involving BRAF and MEK inhibitors. One was from Pfizer and one from Novartis. We were able to secure compassionate access to Pfizer’s treatment and quickly get him started.
The results were remarkable.
His scans went from looking like a Christmas tree full of metastases to showing almost complete remission. Even his doctors said they had never seen a recovery like it.
But there was one concern. These therapies did not effectively cross the blood-brain barrier. If the disease had already spread to the brain, we would only be extending life rather than solving the root cause.
Unfortunately, that concern became reality.
My father eventually passed away from a brain tumor that originated from the melanoma.
That experience put me through what I call an early-life crisis.
I realized I had done everything I possibly could. Probably more than most sons would ever be able to do. Yet I couldn’t stop thinking that if treatment had started even a month earlier, the outcome might have been very different.
It was a bittersweet feeling. We extended his life by ten months, but it wasn’t enough.
At the same time, I began questioning whether the CRO model truly had aligned incentives. I started falling out of love with my own company because I felt I was part of the problem rather than the solution.
Eventually, we sold the business and decided to build something that could fundamentally improve how drugs are brought to market.
At the same time, AI technology was reaching an inflection point. My personal journey and the evolution of technology aligned in a way that gave me the opportunity to take that leap and build something for the greater good.
That became BIOS—the first operating system designed to manage clinical trials from end to end.
That’s really the story behind the company.
Naji Gehchan:
Pedro, thank you for sharing that story with us.
As you described it, it sounds like your technology met your purpose.
There’s obviously a lot of discussion today around AI. And when I think about clinical research—and certainly what you’ve experienced personally—millions of people are facing similar situations.
You’re trying to solve a very difficult problem in one of the most complex industries imaginable.
Where do you believe AI can create the most immediate impact in clinical research?
Pedro Coelho:
Our thinking has evolved significantly over time.
When I started BIOS, the original idea was simple.
I had seen countless protocols that required multiple amendments because they were fundamentally flawed from the start. At the CRO, we specialized in complex Phase 2 and Phase 3 studies, particularly in oncology and cell and gene therapy.
We saw enormous amounts of time and money wasted on avoidable protocol amendments.
The original vision was to build an AI model capable of identifying weaknesses in protocols before studies even began.
The idea was straightforward: upload a protocol, and the system would tell you what might go wrong and how to fix it.
The long-term dream was even bigger—to build a system capable of generating a complete clinical trial protocol.
At the time, most investors thought that was unrealistic.
Fortunately, we proved otherwise.
Within a little over a year, we launched the first protocol-generation platform, and the first AI-generated protocol ever submitted to regulators was generated using BIOS technology.
But today our vision is much broader.
We see BIOS as an operating system for clinical development.
It starts with evaluating assets and mechanisms of action. It helps determine which indications make the most sense from a scientific and economic perspective. It allows teams to model different development strategies, simulate outcomes, evaluate risk, and optimize investment decisions.
From there, it supports protocol generation, feasibility assessments, site selection, enrollment modeling, contract negotiations, budget negotiations, and site activation.
Today, BIOS supports organizations through site activation.
The next step is study execution.
The goal is a single operating system that manages the entire clinical development lifecycle.
Naji Gehchan:
That’s brilliant.
As you’re describing it, I’m realizing that this isn’t just an assistant for one function. It’s an assistant for virtually every function involved in clinical development.
Let’s take protocol amendments as an example.
How can AI predict the need for amendments before a trial even begins?
Pedro Coelho:
Interestingly, this started with research conducted by Kenneth Getz and others who studied protocol amendments.
One of the findings that really stood out to me was that approximately 85% of protocol amendments are avoidable in hindsight.
About 45% are driven by human error.
Another large portion comes from poor study design and feasibility planning.
That insight led us to ask a simple question:
Could we train a model on historical protocols and amendments so it could recognize patterns before they happen?
So we gathered large datasets of original protocols, amendment rationales, and updated protocol language.
The model learned the original language, the reason for the amendment, and how the protocol changed afterward.
Over time, it began identifying recurring patterns.
One of the advantages of AI is that it can learn across therapeutic areas.
A lesson learned in respiratory disease can be applied to oncology if the underlying operational issue is similar.
Humans rarely have that breadth of experience.
Today, our models have been trained on more than one million studies, over 300,000 protocols, and millions of amendments.
Naji Gehchan:
And you’re using publicly available data?
Pedro Coelho:
Yes, publicly available data forms the foundation.
But equally important is our scientific team.
Today we have approximately 40 scientists who help annotate and contextualize the data.
For example, if a protocol amendment occurred because an infusion time was not properly specified, our scientists explain not only what changed, but why it mattered clinically and operationally.
That context is critical.
Without it, the model might simply learn that infusion times are important in every protocol, which obviously isn’t true.
The scientific reasoning behind the amendment is what allows the model to truly understand the situation.
Naji Gehchan:
So humans are effectively teaching the AI the context behind the data.
Pedro Coelho:
Exactly.
The AI learns patterns, but the scientific experts help it understand why those patterns matter.
Naji Gehchan:
Looking five to ten years ahead, if BIOS and companies like yours are successful, what does the future of clinical trials look like?
Pedro Coelho:
I believe clinical development becomes much more focused on value creation rather than administration.
AI is already transforming drug discovery. We’re seeing systems generate candidate molecules at unprecedented speed.
The challenge is that we’re still testing those molecules using processes that haven’t fundamentally changed for decades.
That creates a bottleneck.
The future is an integrated operating system that takes a molecule from discovery through development and eventually to market.
It will allow companies to evaluate scientific, clinical, regulatory, and economic considerations in a single environment.
Most importantly, it will significantly reduce the amount of administrative work required today.
The workforce won’t disappear, but it will shift toward higher-value decision-making.
Experienced development leaders will continue to play a critical role.
AI may provide recommendations, but human judgment, experience, and governance will remain essential.
The question won’t be whether AI can make a recommendation.
The question will be whether leaders agree with it.
Naji Gehchan:
That’s an inspiring vision.
And given your personal story, I’d love to talk about patients.
Where is the patient in all of this?
Pedro Coelho:
The patient is at the center.
One area we’ve focused heavily on is patient matching.
Historically, patient matching has been expensive and difficult to scale.
We wanted to build a system that could evaluate patient populations continuously and at massive scale.
If you’re designing inclusion and exclusion criteria, you should immediately understand how those decisions affect patient access.
You should be able to test hundreds of scenarios and identify the criteria that provide the best balance between scientific rigor and patient availability.
That’s already happening.
The next evolution is creating intelligent patient interactions throughout the study.
Today, many patient-reported outcome systems are static.
Patients answer the same questions repeatedly.
But that’s not how physicians think.
A physician adapts based on what they’re seeing.
Future AI-driven systems will do the same.
They’ll ask different questions based on previous responses and gather richer, more meaningful insights about patient experiences and quality of life.
Ultimately, this should improve both clinical research and patient outcomes.
Naji Gehchan:
I love that vision.
Especially in areas like cell therapy, where we’re seeing transformative outcomes that don’t always fit traditional frameworks.
The ability to better understand patient experiences could be incredibly valuable.
Pedro, let’s shift gears.
I’m going to give you a word, and I’d like you to tell me the first thing that comes to mind.
Leadership.
Pedro Coelho:
People.
And hard.
Leadership is difficult because every person is different.
There is no one-size-fits-all approach.
Great leaders adapt to the people they’re leading.
They also lead from the front.
I never ask someone to do something I wouldn’t do myself.
Sometimes leadership means rolling up your sleeves and getting involved in difficult situations.
Too many leaders become disconnected from reality.
That’s why leadership is hard.
Naji Gehchan:
AI.
Pedro Coelho:
The future.
And maybe magic.
There are moments when what these systems can do feels almost impossible to explain.
Naji Gehchan:
CROs.
Pedro Coelho:
Inefficient.
Slow.
And often operating with misaligned incentives.
Naji Gehchan:
Spread Love in Organizations.
Pedro Coelho:
The future.
I genuinely believe that in the age of AI, culture becomes even more important.
People have more choices than ever before.
If organizations fail to create environments where people feel valued, motivated, and connected to a mission, they’ll leave.
Life is too short to spend it doing work that doesn’t matter.
Companies that truly care about their people will attract and retain the best talent.
Naji Gehchan:
That’s powerful.
Any final words of wisdom for leaders listening today?
Pedro Coelho:
Leadership is changing rapidly.
More than ever, leaders need to be agile and comfortable with constant change.
I tell my team all the time that I can change my mind in five minutes.
Don’t treat everything I say as permanent truth.
You’re here to think, challenge ideas, and help move us forward.
If you’re leading as though every decision is carved in stone, you’re putting yourself and your organization at risk.
The only sustainable approach today is to embrace change and continuously adapt.
Naji Gehchan:
Pedro, thank you so much for being with me today. It was a great conversation.
Pedro Coelho:
The pleasure was mine. Thank you, Naji.
Naji Gehchan: Thanks for listening to the show! For more episodes, make sure to subscribe to Spreadloveio.com or wherever you listen to your podcasts. Let’s inspire change together and make a positive impact in healthcare, one story at a time.
Follow us on LinkedIn and connect with us on spreadloveio.com. We’re eager to hear your thoughts and feedback. Most importantly, spread love in your organizations and spread the word around you to inspire others and amplify this movement, our world so desperately needs