Improving Targeted Delivery to Enhance Personalized Medicine (with Podcast)

Once CRISPR was popularized, gene therapy companies focused on developing therapies because they knew it would be so important, but the technology required to target specific cells had not caught up, so there really wasn’t any way to get them to market.

Time in podcast: 4:45

Ligandal is working to create delivery that could be uniform for various types of cells, helping improve on current methods that can only target specific types of cells.

Time in podcast: 8:00

Ligandal has developed a method that is improved over time with computational tools. They are recognizing patterns that could reduce the time of some gene editing processes that typically take months to days - or even as little as hours.

Time in podcast: 11:50

Soon, treatments like CAR T-cell therapy will revolutionize cancer treatment, and the treatment of other complex diseases like sickle cell anemia or beta thalassemia.

Time in podcast: 15:00

Andre: Sure. I was studying Biomedical Engineering at RPI when I decided that gene delivery was this big missing piece within medicine. So we had all these tools for sequencing the genome, for understanding what causes various diseases, and yet even with the advance of tools like CRISPR, there weren't effective ways of actually reprogramming that molecular information to rid the body of negative pathology.

I became fascinated with the idea of how you could actually deliver genetic cargo and different genetic engineering tools to the appropriate places in the body, and create a platform for targeting specific cells. I spent two years during my undergrad building what ended up being the first proof of concept for delivering a guided nuclease, which is CRISPR, into targeted cells, without using a virus.

Andre: So the first work really centered around delivering TALENs, which were kind of a precursor to CRISPR, and were still quite a good gene editing tool. We did a series of studies all the way to in vivo mouse studies, where we injected adult mice with a bone fragility disease, with the gene editing constructs carried by these nanoparticles, into their bone marrow.

Around 2013, I took it to Silicon Valley, and it was really a choice between do I go get my Ph.D., spend another six years in academia, or do I take the data and intellectual property and use the knowledge that I have in this burgeoning industry and create a missing piece for delivery. So, we've been working on creating iterative delivery systems for the last five years at Ligandal.

Minu: So that's how you ended up founding Ligandal after your undergrad, just deciding not to continue in academia anymore. That's a big step going from undergrad right into a company. But I guess the interesting part here is really the technique which you developed or proof of concept which you developed during your undergrad studies.

Andre: Well, that's exactly what we've really developed at Ligandal. The idea is that anytime you have a cell, it's going to have a certain profile of receptors, so, we have two pieces to really develop in our platform. One is knowing that you can actually package up a given payload. The second is knowing that you can actually get that payload to the target itself. So we develop both of those things. We create an iterative process to take a range of polymers that are actually based on peptides, and condense them around the given payload they want to deliver, such as CRISPR, so that you form a stable package.

Then we decorate that package in targeting ligands, and we have a discovery platform for really rapidly creating ligands. So we can take an unspecified cell, and based on the knowledge of which receptors it has, we can create targeting molecules, targeting ligands, faster than anyone else. We've created an advancement not just in the delivery system and the packaging, but also in the targeting and being able to create ligands in hours or days, instead of months.

Andre: The computational component is very important, but it's not all encompassing. One of the great things about the idea of a computational platform is it can simulate a lot of things. But at the end of the day, you're going to have to interact with biological systems which are highly dynamic, complex, and really are more amenable to brute forcing methods of screening. So, you need to find the right balance between simulating things and actually performing the experiments.

That's the balance that we've tried to find at Ligandal, where there is a simulation component, so we do look at creating ligands that haven't been used before. But we're starting from native proteins, so we're not starting completely de novo, trying to make sequences that didn't exist before. That's the first piece.

The second piece is actually making the nanoparticles, and there are many components that go into making a nanoparticle. In our case, we use electrostatic interactions, which are charged interactions between negatively charged things, especially with gene that you're delivering. And positively charged things, such as the peptide sequences that actually package up that payload.

What we actually developed is a software interface for rapidly designing those experiments and having robots perform those experiments. So we tied the computational elements, which are really simulating and designing the targeting molecules, to the physical studies of making nanoparticles. And that is something that also, much like biology, you can't fully predict. So we need to screen hundreds of things, but we end up iterating and finding patterns that work.

And once we lock in those patterns and find the things that are consistent in their formulations for the size of those particles, how well they packaged the materials, etc., we then do a biological study and look at how well those nanoparticles performed in a cellular context. So every time we perform an experiment, we can take the top 10 or 20 percent of performing groups, wipe those out, and then the rest are iterated around. So it continues to get better and better.

That's really the novelty of our approach. We're not just trying to take an off the shelf delivery system, put something inside of it, and hope that it targets the right cell. We're actually tuning and fine tuning all of the aspects of how you'd want to interact with a cell down to the level of what enzyme is going to break apart my nanoparticle. Is it going to go to the nucleus? Is it going to release quickly or slowly? Is it going to be a big particle or a small particle? Is it targeting one receptor or another? All of those things are what we’re looking to nail down.

Andre: Well, I think there are different degrees of personalized medicine. I think what we're going to see sooner rather than later, is diagnostically responsive immunotherapy. So taking a T-cell, for example, and reprogramming it to attack a specific kind of cancer cell, because the FDA's already approved the first gene therapy or immuno-oncology, as is Novartis' Kymriah in 2017. We're now looking at an exponential advancement in the field around autologous cell therapies.

So, in this case, you have to take an immune cell, a T-cell, that normally surveys your whole body, and you have this huge variety of these so-called T-cell receptors that each respond to different antigens or different markers that shouldn't belong in your body. So that could be a virus, that could be some sort of bacteria, or it could be cancer that you just developed with its own set of mutations, and results, and new sequences of proteins on the surface.

Our immune system can recognize all of these things. With immunotherapy, the goal in the case of T-cell immunotherapy is to take those T-cells and to give them targeting characteristics, where you genetically engineer the cell so that it can stick to a specific kind of cancer cell or a specific kind of virus.

There are companies now trying to focus on the personalized aspect of that. One of them is called Pact Pharma, for example, where they're actually sequencing your tumor, they're identifying which neoantigens or new peptide sequences that tumor has. And then they're actually building this microfluidic platform where they'll pull out your cells and circulating T cells will go through this chip, and they will stick to those peptide sequences that you identified in that patient.

And the T-cells that get pulled out are the ones that already have recognition for that neoantigen. Then you have the sequence of the T-cells, identify which T-cell receptors they have, and then you have to genetically engineer a larger population of that person's T-cells to have the same sequence.

That's a complex process, but that's kind of the first foray that we're seeing into personalized gene editing immuno-oncology. We're also going to see a lot of stuff in the genetic disease space. So, you have a disease like Sickle cell disease where you have a single base pair that results in the disease. That's a pretty easy off-the-shelf fix, if you can actually deliver to the right cells or even perform in a target cell transplantation.

But then you have diseases like Beta thalassemia or Von Willebrand disease where you have hundreds of different mutations underlying even a single genetic disease. Where, yes, it is the one gene that’s affected, but it can have many, many different mutations that result in the pathology. So how do you treat those diseases? You're going to have to actually stratify patient population within that single disease into different gene editing modalities, if you're going the gene editing route.

And then we have more complex, multi-factorial diseases also. So we're going to be treating the disease of aging itself, where you're going to say that an older person isn't just sick because they have Alzheimer's disease, you're going to say they're sick because they got older, and maybe there's a comprehensive set of things that you need to do to reverse the biological age of the human. So that could take decades to really mature, but as it matures we're looking at actually saying here's a diagnosis, here is the set of genetic changes that I'd like to affect in the patient, whether they are temporary or permanent. And here are the tissues in which I'd like to create those changes.

I think delivery is a bottleneck to being able to do that. Because you can't always do a cell therapy, you can't always just target the liver. You're going to need to actually predictably say I'm going to fix X, Y, Z cells to choose an organ with A, B, C, D, E, F, G, different genetic traits.

Andre: Well, certainly in vivo is a huge bottleneck. But we're seeing that the bottleneck is even in the ex vivo market. So when I mentioned earlier about it taking 12 months to even get past the wait list for manufacturing a virus? That applies to ex vivo, as well. So we're finding that, from a demand perspective in the industry, ex vivo delivery systems are still very valuable. Of course the eventual goal is to do in vivo, and we do have select applications that we're developing data fits for within the in vivo context, as well.

So, it's a fine balance, but from a regulatory perspective, being able to start with cell therapy approach where the FDA is already comfortable with the end product, so that you're putting the cells back in - that is a good point to show that your delivery system is safe and effective. And it’s also something that there are a number of partners where you could have up to 100 or more different genetic modifications you're performing in the cells, and the goal is just to provide the supply chain and the manufacturing for all those therapies, as far as the delivery systems go.

What we're finding is that in theory, it seems easy to just modify a gene in a cell and put in back, but in reality, there are many steps to that cell therapy process. You have to take the cells out. You have to use nucleofection or electroporation where you expose the cells to a big electromagnetic field in order to permeabilize the cell membrane. And that's where you can introduce CRISPR or whichever tool you're doing the gene editing with.

But CRISPR performs cuts, it doesn't perform insertions. You're basically using molecular scissors to create a double stranded break at some part of the genome. We then need to actually insert a new DNA sequence into that spot. The industry is typically using electroporation for the deletion part, and then they're manufacturing viruses to insert the genes back. Within that process, you also have to have very specialized conditions for growing those cells, because you actually have to grow them for typically three to six weeks.

So, you need the time between taking those cells out of the patient and putting them back in. And during that period, there are a number of steps for activating T-cells, that are also very complex and cumbersome. And the whole process ends up costing about $270,000 just for the manufacturing component for a player like Novartis.

And the drug's going to price at $470,000. So that is a really big bottleneck. Where if you can reduce the amount of steps required to engineer the cells, and make it more that you're taking the cells out, putting some nanoparticles on them, and then putting them back in; where there are only one or two steps, you're looking at major cost and time saving across the whole industry. Then you can also use those techniques for in vivo therapies, so the delivery system itself is designed to be used in vivo as well.

Andre: That's an interesting question. I'd probably kick Paul out of his spot at Synthego.

Minu: Okay. Perfect. I think he would take it as a compliment, hopefully. Great.

Andre: Love you, Paul.

Avery Posey | CAR T Therapy: Treating Cancer with Engineered Cells (with Podcast)

This topic was mentioned a number of times in this article, so if you want to learn more about T-cell therapy and how it will replace chemotherapy as the main treatment for cancer, check out our conversation with Avery Posey.

Genetically Engineered Cows: How CRISPR Could Improve Cattle Farming (with Podcast)

In this CRISPR Cuts podcast episode, learn about how a project to genetically engineer cows without horns is helping to save the beef and cattle industry, that could be in great danger without this revolutionary tool.

TED Talks Science & Medicine | How CRISPR Lets Us Edit Our DNA

In this general intro to CRISPR given by Jennifer Doudna in a TED Talk, learn the basics of how CRISPR was discovered, how it works, and what it means for the future.

Andrew Porterfield | CRISPR in the Grocery Store

In another episode of CRISPR Cuts, learn more about how soon CRISPR foods could be hitting the grocery store shelves near you, and what this means for the future of the food industry.