Ajay Karandikar: Why do 90% of promising drugs in clinical trials fail? It’s a staggering number. That’s nine out of 10 drugs that don’t make it past clinical trials that don’t get the necessary funding, the licensing deals that they deserve, that our patients deserve.

Today I’m here with industry legend and a key pharmaceutical stalwart Renata Crome where we’ll be looking at the intelligence gaps that stop these drugs from making it to the market that stop all the innovations reaching the deserving patients. First of all, I’d like to start by saying how privileged I feel to have you, Renata, a part of our scientific advisory board. It’s brilliant. It’s enriching to have a key opinion leader of your caliber as our internal KOL to help our patients to help Omnitheia on its journey to get these promising innovations into the hands of patients.

Renata Crome: Thank you very much for the warm welcome, Ajay. It’s such a privilege to be part of your advisory board. We all know that acquisition of knowledge and access to knowledge and information is more important than ever nowadays. And so I’m thrilled to be part of your journey acquiring data and intelligence so that we can use that to develop agents faster. Thank you.

Ajay: It’s great. And there’s a few questions that we have from the industry which we’ll be discussing today. But I want to begin with my personal question. Something I’ve always wondered with your diverse experience right across the pharmaceutical development where leading global teams at Roche and then Cancer Research UK but also your interesting journey advising the UK government.

I was thinking about your journey thinking about our industry and the first thought I had was it’s incredibly diverse. The next question that automatically came to mind: what have been your learnings working across our industry sectors; health systems, government, large pharma, and then a charity like CRUK? Are there any similarities you feel? Are there any differences as well as any learnings that you would like to share with the industry?

Renata: Gosh, yes. I’ve had the privilege of working across a wide variety of organizations in the life sciences industry. I think the one of the similarities I’ve seen is that people working in these organizations, they join these organizations because they have a passion. They’re all passionate about developing drugs and new treatments to meet unmet needs. So that’s the same.

And I’ve also had the privilege of working with very high-caliber people in all the organizations that I’ve worked in. They’re superb; superb science, superb approaches, superb processes. I think there’s probably a lot of differences between them, but I’ll just pick on one and I don’t think it’s a real difference. I think it might be a perception and it’s a perception that I actually had when I left Roche and I joined the charity Cancer Research UK.

I thought charities are slower in developing new treatments. And coincidentally, one of the first projects that I worked on, Fresh Eyes, I was asked to interview ex-colleagues and collaborators in my network across the pharmaceutical industry and ask questions about how cancer research could partner better with pharma. And I found that all of the pharmaceutical companies shared that same perception that charities are slower.

But working from the inside of the charity, I could see that the processes were cutting edge, the people were very very experienced. It wasn’t so much that the programs were slower. The difference was that cancer research and other charities tend to take on those very difficult very complicated projects which have inherently more issues to solve. And so the perception is that it’s slower because there’s a lot of issues which emerge which have to be sorted out. Whereas the pharma organizations tend to work on the low-hanging fruit, the quick wins, the one could say slightly easier projects that can just steam through. So in my view, the charities aren’t slower. It’s just that they are taking on different type of projects which are more complex which take longer.

Ajay: Yeah, that’s really interesting you say that. That’s the perception I had. And I reflect almost; working helping these innovative companies, one thing was clear. We were always struggling to prove the value. And I think it’s the mindset as well where I personally also believed that these innovative companies, whilst having great science, did not have commercial grounding. It’s a mindset and you use the word there: preconceived notions.

If you and I had that, could we blame large pharma for having them as well? But then an immediate question comes to mind: do you feel these preconceived notions actually play a big part in getting these licensing deals in actually large pharma actually recognizing the potential for these biotech companies? And if so, how can we change it?

Renata: I think these preconceptions do play a large part in successful deals, collaborations and so on. I think the charities should continue to do what they’re doing. They’re doing a fantastic job. But they should highlight the complexity, the issues, and showcase their unquestionable ability to deliver to solve problems and to overcome the issues and demonstrate the successes in this very very difficult area that they’re working in. And I think being able to show that will overcome these preconceptions.

Ajay: Great. And it’s interesting you say that which brings me to the heart of this discussion. The question that many of my consulting clients to begin with and biotech companies have also had and some pharma execs. If you look at the numbers, $2.3 billion it takes today to get a drug from the lab into the hands of the patients and the costs are increasingly on the rise but at the same time the number of innovations making it are falling.

And if we at the start of the discussion I mentioned nine out of 10 drugs don’t make it. And on the other hand if we’ve seen there’s more funding available, there’s some really promising molecules and some really innovative companies that want to make a real difference. We have all of these. We have the perfect ecosystem, one might even say. Why are costs increasing and innovations falling? What do you think the key hurdles are here? Why is this happening?

Renata: I like to call it the three Ps. The patients, the payers, and the potential. So, first of all, the patients. I don’t think that we are involving the patients and their journey and their needs early enough in the development process. So that we might be developing drugs which the patients find difficult to access, find difficult to take and so on. So that’s the first P.

The second P is the payers. I think it’s very important to develop drugs which are clearly differentiated. A lot of people think that that differentiation is just solely focused on the science, the unmet need, getting a drug which is more efficacious, more safe, more tolerable and easier to make. But it’s not just that. It’s also the; will it be something that payers will reimburse? And that needs to be factored in right at the beginning of the journey.

And the third thing, the potential. And by this I mean the realizing the full potential of the drug. And for me this means understanding all that information that’s possible to harness about that agent, about that indication, about the therapeutic area from different sources; the competitive intelligence, the research. At the moment, I think that a lot of this information is held in different silos and it’s quite difficult to access and a lot of it is untapped. There may be some hidden nuggets there which I think it’s very very important to try and access and collate together.

Ajay: It’s bringing it into our world here, right? You’re absolutely right. So in terms of, you mentioned payers, you mentioned showing the evidence, showing the data. Currently, in my 15 years of experience, this was literally my frustrations as well having worked on both sides of the spectrum; working with pharma biotech companies to solve their intelligence and insights questions, but also working on the other side, licensing analysis or also investment analysis.

There was always this scenario where I almost think of it like a restaurant. Going to a restaurant and you wish to order some food. But imagine instead of a menu, one menu, you’ve been presented with six. Here’s one option. Here’s your starter. Here’s your main course. Here’s your dessert. This is the world I think we live in right now. The fragmented nature of this industry where a biotech comes in wishing to prove, wishing to have that evidence back. But there’s five or six business intelligence platforms out there. Some of them offering clinical trial intelligence. Some of them offering pipeline or you mentioned competitive intelligence.

It’s almost impossible to purchase everything. But also it’s impossible to have that manual hours put in. So that’s your 10 menus you’re talking about, right? You have to integrate that together. And when I think about the, you mentioned payers, KOLs, the actual real world evidence. I think when it comes to primary research, the fragmented nature of the industry gets even worse. So you have a company bringing you the experts, then someone else interviewing them; moderation as it’s known. And I’d say about majority of the industry, the moderators that actually do the moderation are not a part of the industry. Therein lies the other issue.

And then finally you get all of these multiple transcripts and recordings together. And guess coming back to the restaurant analogy: if you ordered a pie you’ve picked your potatoes, you’ve picked your sauce, you’ve picked your all sorts of ingredients that go in the primary research world. Then you were presented with make your own pie option. Somehow you have to put all of these transcripts together in another report. And there’s just six steps for something that should be one single step. And if that’s the nature of the industry, that’s the options we are presenting the industry with, it’s no secret, it’s not a surprise that the industry is actually presenting fragmented sources.

And this is one of the key things we set out to change at Omnitheia where we wanted to integrate everything together. There’s a lot of evidence out there. There’s a lot of data out there. We just need to integrate that. We just need to stop working in silos. But then I further go into if that’s the case with large indications, your popular indications if you may, the data and the fragmented nature and the lack of ready availability of data for smaller niche orphan indications, it gets even worse.

And these are indications as we were discussing earlier. This is where these are life-threatening patients. Patients die. There’s nothing available at the moment. So, if that’s the nature of the industry here and large pharma always seem to have those questions around orphan indications, “we don’t see enough evidence,” “I don’t see enough value.” Now, we want to get these indications, we want to get these drugs out to those deserving patients.

But let’s say the question now taking a question from a couple of our biotech companies: I have a nonpopular indication I’m working with, an orphan indication even, but I come to you as large pharma and you don’t see commercial value in it. If I flip that question, Renata: yes, we know there’s value in saving patients, but if there’s no commercial value, why should you as a large pharma bother to work with orphan drugs then? Are there any benefits? And if so, what are they?

Renata: Well, really, really important question. So, firstly, there are massive benefits to working in an orphan indication. Firstly, if we speak about the commercial benefits, whilst these agents may not generate billions, they might not be blockbusters, but there’s still a large number of patients belonging to the rare disease community. And so in the majority of cases, these drugs to treat orphan indications do generate commercial value and return on investments. It may only be in the millions, but they do return value and a return on investment.

Secondly, they have regulatory benefits, accelerated approval and so on. And also may provide opportunities to gain exclusivity in other larger indications to ensure continued returns and exclusivity in some of the big indications.

Ajay: So interest, you know, that’s a really good point, Renata. And this is close to heart because you mentioned the opportunities with working on orphan drugs. And I say close to heart because I actually know someone where they suffer from an orphan blood disorder. The opportunities there are more than just on the patient level.

We know how clinical trials always get delayed. It’s difficult to find patients. But in orphan indications, we do have patients that really really wish to participate in the trial. They’re willing to. They want to. Why? Because they could get a therapy for a disease or an indication that they struggle with. They face daily issues. There’s no therapy there. So by participating in the trial, they might get that cure. But also they want to help the other patients. So there’s the availability of patients there as well.

So there’s a lot that we have at our disposal. And while this happens, while there’s this opportunity, there’s always almost a concern. The second concern that biotechs have where what they seem to think is large pharma only typically focus on these big blockbusters, the big threes in oncology if you may. Is that true? And again goes back to the first question as well: what can we do to change their mindset?

Renata: So in part it is true that pharma companies do concentrate on for instance in oncology; colorectal, breast and lung cancer. But very often they also will look at the orphan indications like glioblastoma because they do see the potential benefits of working in these areas. They do recognize that it’s important to not only address the big three but also the smaller more niche indications. They want to help those patients as well.

So in effect they do tend to work in these as well to some extent but I think it’s important for biotechs to bring that information to their attention on the potential commercial benefits of this. I’ve had a recent experience where I was working with a smaller charity which recognized that one of the big pharma companies had a cancer drug in use which could potentially benefit this orphan indication. And they took the numbers. They did the market assessment, took the numbers to the big pharma and showed them that the numbers proved that there would be commercial benefit to working in this indication and persuaded them to work on it. So there are potential opportunities when you have the data and the information to demonstrate the benefits.

Ajay: That’s promising to hear and I’m really happy to hear that as well seeing this happen in real life. Now we’re at this great point in time I believe, strongly believe, where you mentioned it’s important to have that right data, right insights, the right evidence. Data and evidence has always existed but again it’s scattered in poster presentations, in conference journals, in clinical trials.gov, public sources, regulatories.

But the limitation always has been how much data and where do you access it from. And I say we’re at a great time because we have the emergence of AI where if used correctly we have the ability to go into all the publications to all the journals even for orphan indications. Yes, there hasn’t been a lot of research on those, but we have the molecular pathways. We have the receptor level data and AI enables us to go into the hundreds of thousands of millions of documents and access that data in good time.

It enables us to analyze our data against it. And I believe in 2026 and onwards, the emergence of AI will help us to back our promising innovations further to access and harness that power because if harnessed correctly, as you say, as long as we can prove the value with real evidence, these orphan indications will get approved. Patients will see those drugs as well, and we will get those big licensing deals. So I’m very optimistic and very positive.

I mean we’ve seen real evidence working with Omnitheia with our promising clients, with our innovative clients, with even stepping out of orphan indications, even the indications say in oncology and CNS that are not these top indications. We’ve seen the promise. But I want to almost ask this question to someone with R&D experience such as yours. It’s still new. What’s your experience been with AI? Have you seen any tangible benefits yourself?

Renata: Yes, I have actually. Just very very recently, I’ve seen benefits again in the orphan disease space. I’ve been involved with an organization that’s been looking at treatments for a particular orphan disease. And for the last six or seven years, the team have been mining all the information more manually; googling and using various data sources to find new treatments, new potential leads.

And then more recently we embarked on an AI search and very very quickly found a large number, dozens of potential leads that could be useful to treat this particular rare disease. I liken it to the London Underground map and traveling on the Piccadilly line and when you get to King’s Cross it intersects with the Northern line, with the Victoria line, with the metropolitan and circle lines. So the power of AI is such that it sees these biological intersections and can potentially make links or show potential pathways where they intersect, where one particular pathway might have an influence on the pathway that you’re exploring. So yes, I see boundless potential with AI. I’ve got a question for you now, Ajay. I see a lot of potential here. As a scientist, I always think about the quality, the consistency, the accuracy of data. So, can you tell me how you ensure confidence in the quality, the consistency of your data?

Ajay: Oh, I’m glad you asked that question, Renata. It’s a great question for two reasons. Firstly, I’m a scientist myself and we walked into the space with the need to use AI seeing the power. I had even worked with companies having amazing advanced analytics at their disposal. Did I understand all of it? Absolutely not. Did I understand most of it? Absolutely not.

So my challenge as a scientist had always been yes we need to use AI but when someone asked me the question as you have: “what’s the accuracy, how confident are you?” I never was because I never understood it. So this journey led me to work with industry stalwarts on the technology and the analytic side. So at Omnitheia I partnered with the best minds and experts in technology that had built analytics frameworks for Dow Jones, for build life science analytics data management companies like InfoDesk.

And what I realized, and I’m going to explain this in essentially lay man terms: as a scientist AI is amazing as long as we control two things. Number one, the input. What I mean by the input? Currently, and you know, we all use OpenAI, you have ChatGPT, you have Gemini, Perplexity and all the rest. Now the way I see this and I visualize this as almost a fishing net. So when we search for something, we put in a query. All of these have certain number of tokens. The fishing net goes out and Gemini has a larger token count than ChatGPT i.e. it has a bigger net.

But what this all means is the fishing net goes out and whatever it catches are the sources. Now in our industry as you know life sciences clinical trials change paths. Some trials pause then they start again. Trial designs change, APIs get broken, there’s more regulatory documents that are put in. So if your fishing net has only caught on the first instances it finds, your AI is then analyzing those sources only. You are only looking at a section of the input to generate your output.

This is where errors come in. This is where data hallucinations come in. So what we did firstly; control the input. We have ingested original documents, original sources. We do not let AI choose the sources. We curate the sources. We put them in. Then secondly, we let AI do the analysis. And this AI has been built specifically for the life sciences industry.

So I mentioned about the need to control these sources. So yes, we’ve got the sources, but can they still change path? Of course, they can. So we then secondly, once we’ve had the input, we’ve controlled the output by putting a control center where scientists such as you and I do not need to understand the intricacies of technology. We have the experts handle it. What we see is live sources, accurate sources, but a scientifically validated confidence score, a confidence certificate that ensures that the end outcome and the end result that we get is backed by science, is backed by evidence.

This is how we’ve controlled the AI hallucination and data. And this is how we intend to help our promising innovations get there. Now, we’ve discussed a lot about with all of these great insights that we’ve discussed and how we’ll be using AI, how we need to almost change our mindset to change the large pharma company’s mindset. That’s a lot of promising points that we’ve discussed here.

But to summarize this then as a key takeaway for our industry: I’m sure our biotech companies and consulting clients that have post questions would have been happy to get some of these insights here. But as a takeaway for them let’s say I’m a biotech company, Renata: I have a phase 2 phase 3 asset in clinical development and I’m coming to you at a Roche or an AstraZeneca looking for a licensing deal. What are the three things I need to ensure I bring to the table to get this successful licensing deal?

Renata: First point: starting off with a high value problem. I’ve already mentioned differentiation and this is not just scientific differentiation but also developing an agent that fills an unmet need for patients but that payers are very willing to pay for. So start off with that and ensure that that’s what you’ve got. Otherwise you won’t be successful.

Then know your biology; understand the mechanism of action, the pathways, the assays that are needed, the biomarkers, the patient population. Really understand your biology and develop your program to demonstrate that biology. And then finally the input from the stakeholders. And by input I mean really get those stakeholders on board very very early on so that you’ve integrated and co-created the development plan so that for instance the key opinion leaders, the regulators, the payers, the patients so that you’ve introduced their passions, their needs, their vision for the agent and got alignment on all of those across all of the stakeholders. Because just seeking their input and possibly taking it on board or possibly not just won’t lead to success.

Ajay: Thank you. That’s really great. In fact, Renata, I’ve been taking notes. Whatever you said just now, it just gives me renewed zest here. And I talk about mindset; the renewed zest to ensure that we all have a positive mindset when we’re bringing our innovative drug.

That’s very important from an Omnitheia perspective. Yes, we’ve been we’ve discussed all the AI, all the data there, but my key takeaways have been: yes, we’ve proven the value, but what we will also be doing and also ensuring is that we have all the biomarker data, the target data. We are even validating all our early stage clinical claims. But then we are lucky enough to have internal key opinion leaders such as yourselves but also our global KOL networks, our payers that work with us to ensure that we onboard them early. We are in alignment with them early enough and then we are also involving the health systems early.

Because what this is telling me and telling our hopefully our biotech industry is that you’ve got an innovative asset, you’ve got all what it takes, all you need is that extra 25% that’s going to get your drug over the line. And we are committing to helping our clients with that. Renata this has been great. Thank you for joining. It has really inspiring insights and like I said, I’ve learned a lot and I’ve got the renewed energy to make it happen. Thank you.

Renata: It’s been a pleasure. Thank you.