Table of Contents
Dr. Ahmet Can Berkyurek, CEO of CamMed, sat down with Onyx for a feature-length interview on non-viral self-amplifying RNA, an AI-driven hunt for ancient RNA-replicating proteins, and his case for the UK as a place to build an RNA therapeutics company.
Can you introduce yourself, give a bit of background, and explain what you're working on at CamMed?
Thank you, Adil. I'm Dr. Ahmet Can Berkyurek, a scientist by training with a PhD in stem cell biology from Japan, and I also worked at the University of Cambridge as a fellow. During the COVID-19 pandemic, I worked with mRNA vaccines, targeting SARS-CoV-2 infections.
At that time I saw some pharmaceutical companies stopping or closing their RNA therapeutics programmes because of certain issues they faced with canonical mRNA technologies. That was when I came up with an idea that I thought could be a good solution to those problems, and this became the idea behind the company.
My background is RNA biology, and I started the company in those times during the pandemic. I raised money with that idea. Now we have a patent in place, and we are talking to those pharma companies about how to solve the issues in their therapeutic portfolio programs.
For readers who aren't well acquainted with RNA biology, can you explain what you're making — and what does non-viral self-amplifying RNA let a clinician do that normal mRNA can't?
First, let's focus on what an mRNA is, why it is a good technology, and how what we're doing is going to change the shape of RNA therapeutics. RNA is a molecule that can instruct your body to produce a certain medicine, for example. The drug development field is moving more and more towards protein therapy. And, mRNA gives instructions to your cells and tissues to produce a specific medicine, at a specific time, in a specific tissue. The advantage is, we can control it, we can delete it, we can stop it, or we can make it work even faster, compared to other technologies that integrate into the genome or that cause other side effects. That's the main advantage of mRNA, and that's how we use it.
But what we are doing is generation number four of mRNA technology. We used mRNA in the COVID-19 pandemic as a vaccine. But when it came to advanced therapies, or really hard-to-target biology, we needed improvements. We needed it to stay longer in the human body — not a few days, but maybe months. We also needed it to work impeccably in delicate tissues, like the brain, cardiovascular, or kidney.
So we took that technology and changed the core architecture. It's still the same thing that instructs your body to produce a specific medicine, in a specific tissue, at a specific timing, but we prolonged the time — not just a couple of days, but six to ten weeks staying in your body in a specific tissue. And we did it in a way that somehow bypasses the immune system. It tricks your body, so it doesn't trigger the immune system — or, compared to other RNA technologies, it's very, very low. It doesn't really cause any side effects.
What underpins your work is a set of conserved RNA-related motifs you found inside enzymes and systems. Could you explain that in more detail?
We have an internal joke in the company. We say that if someone were to do what we did experimentally, it could have taken them between 8,000 and 40,000 years. Literally. Because we screened thousands of proteins, we looked into their structures, and then we compared them to other motifs that are known to replicate RNA. Experimentally, it takes three, four, or five years for a single protein, and given thousands of them, it could have been thousands of years. Not possible in a lifetime.
But thanks to artificial intelligence programs, we were able to reduce this to a couple of months with 90 to 95% accuracy, which was good enough to boil down candidates and start testing them in the lab. We identified a couple of them, started testing them, and one was successful, which is now in our product pipeline.
I can also explain this in a different way for a non-expert. I always explain it in an artistic way — I love paintings, I love surrealist paintings. One surrealist painter is Salvador Dalí, the other is Picasso. Both have the same method, right? But when you look at the paintings, you'll know that they are different, because they use different motifs, and you will identify who the painter is. Similarly, proteins have motifs in themselves which define function. What we did is look into each protein as if it's a painting, and tried to identify those specific motifs that replicate RNA or amplify RNA signals. We identified a couple out of thousands, and now it's in our product pipeline, and it's working.
The industry standard in the clinic is generally replicases derived from alphaviruses, which limit how large a payload you can carry. What does your replication machinery look like, and how are you achieving amplification without viral replicons?
Viral replicons are great tools, actually. We already have a couple of products on the market for COVID vaccines and flu vaccines, and we're now seeing a couple of therapeutics programmes coming up as well. But what viruses do is use a multi-component machinery. There isn't only one, but three or four different proteins required to do this function, and then the length of the vector becomes really, really long — approximately 10 to 11 kilobases. It's achievable, but when it comes to manufacturing stability and drug delivery, we face issues. It becomes more challenging.
In our case, our replicase does all the functions with a single protein, which is five times smaller than alphavirus proteins. So that's already a big advantage when it comes to delivery, formulation, stability, and everything. That's one advantage.
The second thing is that viruses, when they act as a replicon in the human body, create an antisense genome. So first of all, you have a really huge double-stranded RNA that creates some inflammation or other side effects. In our case, the enzyme is already bypassing that — we already skipped that phase. It's producing the product without creating a double-stranded RNA product. That's the second advantage.
The third thing is, it's not a virus, but it's something close to human biology from an evolutionary perspective. It's close enough to human biology that it doesn't create side effects, but distant enough to do its unique function in the cytoplasm of cells. That's the third advantage.
Are you co-delivering or recruiting RNA binding proteins?
This is an RNA binding protein itself, but we don't have anything additional — just an enzyme. We have an RNA and a specific sequence that makes our enzyme recognise it.
You've described RNA as being tissue-aware — constructs whose folded shape is stable in one cell and degrades in another. What's your design handle for that?
We're seeing this especially with lipid nanoparticles. Most of them go to the liver — if they're delivered through intravenous injection, they mostly go into the liver and not other areas. But now we see specific LNPs, lipid nanoparticles, that can target cardiovascular systems, kidneys, and spleen, and we're now also seeing preclinical data on central-nervous-system-specific delivery systems. That kind of determines where the RNA is going to end up. That's the first part.
The second thing is that we can implement intelligent sequences, tissue-aware sequences, on our vectors that make them function better in certain tissues and then stop functioning in others. That's an improvement of the technology — the second way we can control tissue-specific function.
With self-amplifying systems and prolonged protein expression, one safety concern is that it's harder to facilitate natural clearance and finite, gated expression. What's controlling the off switch?
The first thing is that RNA is a very fragile molecule. So even if it's there for a few days or a few weeks, we initially see a huge spike of expression, and then it gradually degrades. It's never in the tissue constantly. There's always a degradation step, and eventually it goes to zero. So what we do is redose patients so that they regain the levels of expression. That's the first control mechanism — the nature of RNA molecules. Unlike DNA, RNA degrades, but we can control how long it stays, by increasing or reducing the dose. RNA modifications, or those intelligent sequences, are another way of controlling the duration of effect. Compared to DNA vectors or DNA-based systems, that's one advantage of RNA, and we can control it.
The second thing is the nature of the therapeutics, or the proteins. Some proteins are more stable, some less stable, but eventually they also degrade. They don't stay forever. However, if this were gene editing technology, for example, it would be constant — once changed, it just stays there. But in our case, we can control how long it stays, and it just clears out.
The saRNA field is getting quite crowded. As an early-stage UK company, are you looking more towards a platform you license out, or your own lead programmes?
Both, actually. We already see the value of licensing the technology to other companies — which we saw a couple of years ago, when they faced certain issues with other RNA technologies. I think we can help them overcome certain issues. But we also see the value and opportunity of developing our own pipeline in-house, because most of the time we see a big unmet need in rare disease areas. A, for commercial reasons. B, because they're hard-to-target, difficult biology.
I think we can help those big unmet needs with in-house programmes, especially in rare diseases. It's a small market, a small number of patients, which is an advantage for us in getting to the clinic — we can accelerate the route through regulatory agencies. And at the same time, it's an advantage for those really difficult patients, especially kids, who have a lifespan of 15, maximum 20 years, where we can help increase their lifespan by 5 or 10 years. We see the value in both approaches. At the moment we have two potential deals in discussion, but we are also continuing our animal models for Sanfilippo syndrome, which is a paediatric rare disease in the central nervous system.
What does your fundraising status look like, and what are the plans over the next year?
It's a never-ending process — it's always fundraising, of course. But we closed our pre-seed round in February with Zinc VC, SFC Capital, the Francis Crick Institute, and some angel investors. Now we're going towards our seed round.
What we're working on — we've been communicating with investors and understanding their expectations. They want to see animal, mouse, proof of concept, and benchmarking to other RNA technologies. That's what we're working on. The plan is to get initial animal study data in late summer, then start communicating with investors again in autumn, and open our seed round after that. So R&D data is going to be the determining factor for our fundraising.
You built CamMed out of Stevenage and the Cambridge ecosystem, but your own background includes research stints in Japan and at Cambridge. How does the UK environment for RNA biotechs compare to Japan, or the US?
Yes, I have some experience. We've been attending industry-specific events in the US, and I have some connections in Japan. But first of all — not just for RNA, but for biotech overall — the UK is one of the best places, especially Cambridge, London, and the Oxford Triangle. One of the best places in the world to establish a biotech company. Investment, talent, support, and advisors — from all these angles, we're in a really good space, and I feel really fortunate to be here.
I was able to raise venture capital money just with an idea on paper, which is not often the case. Generally, people need a spin-out, or really advanced-stage data. But I was able to convince investors here in the UK, because there are some investors who are open to risk, with a high risk appetite. Compared to the US — Boston, San Francisco, New York — London, Cambridge, and Oxford have a very similar level of RNA therapeutics companies and programmes. In the US, we see bigger investment rounds and stronger government support, for example match funding and other tax schemes, compared to the UK. But from a talent and idea perspective, I think they're very equal. We've been attending events in Boston, San Francisco, and New York, connecting with other founders, and they have really interesting ideas — we see the potential for collaboration with those companies.
Taking this evolution-informed philosophy and looking ahead, what do you think will be possible in RNA medicine in 10 years' time that's impossible to design today?
Great question. Maybe not just 10 years — but let's say I have a dream. I hope I see it in my lifetime. In physics, and in other industries, we see a term called singularity. The space industry, the automobile industry, physics — we always talk about technological singularities, but not in biology. So what I'm saying is kind of against the nature of biology, but let's also keep in mind that, as human beings, we are more intelligent than biological evolution, and now we are in the age of artificial intelligence. I believe we are going to be facing technological singularities in the biotech and medicines industry very soon.
It means there will be one thing that works for all disease indications. RNA is a prime candidate, because most diseases happen because of loss of function or gain of function. Either you need to inhibit something, or you need to activate something — and you can do both with RNA as a technology. So if there is a genetic mutation in a rare disease, you can activate it as a protein replacement therapy. If it's a gain of function, like in cancer, you can inhibit it with small RNAs, temporarily or for a long time. You can do most things.
At the moment, we have limitations. It's a partial technological singularity, let's say, for specific diseases using RNA. But I believe we are going towards a future where this is one thing for everything. That's the way I would put it — not 10 years, but even longer. And I hope I see it in my lifetime.
Comments