AI-Driven Chemistry: The Design Process Behind Zasocitinib

Takeda vice president and global program leader Graham Heap, MBBS, PhD, recently spoke with Dermatology Times on the design process and current data available on zasocitinib, a selective oral TYK2 inhibitor. Heap, a trial investigator, sharedbackground on the discovery of the drug and its potential as a ‘next-generation’ TYK2 inhibitor.

Dermatology Times also spoke with Christopher Bunick, MD, PhD, foradditional context on the significance of these results.

Transcript

Graham Heap: My name is Graham Heap. I'm the vice president and global program lead at Takeda.

Dermatology Times: Can you talk about the unique design process behind zasocitinib?

Heap: Zasocitinib was designed by a really great team of chemists over at Nimbus Therapeutics, and as part of their development process, they really wanted to understand how well the molecule they were designing fit into the area of the enzyme they needed to block. To do that, they collaborated with a company called Schrodinger, who uses AI assisted design to figure out whether the compounds that are being designed fit into the specific pocket of the enzyme that they're trying to do. By leveraging AI, they can massively increase the throughput of the number of compounds they start to design and screen. It's a really efficient tool to increase the throughput and to try to really get a molecule that fits as well as it possibly can into the specific site of the enzyme they're trying to block.

DT: What data have you seen with zasocitinib so far?

Heap: This is an area where millions of people are affected. The advantage of the kind of compounds we're talking about is they impact a range of immune mediated inflammatory diseases, of which psoriasis isjust 1 of those. There's really a still ongoing need for potential compounds that are balanced in their benefit risk profile and also balance convenience and tolerability. That's really what the goal of the development process for this compound was; To try and drive toward a compound that looks like that. What we really wanted to do was to do was to understand what the compound could do at a really fundamental level.One of the things we've always thought about in this development process is to have a really good molecule. It's our view that selectivity is really important to just inhibit the thing you want to inhibit and avoid getting off target effects by inhibiting other compounds.The study we did here was an in vitro study. It was performed in a lab. We assessed zasocitinib, which is the compound that we're developing, and we compared it to a range of other compounds, including a TYK2 inhibitor that's already approved to deucravacitinib, and then some JAK inhibitors that are approved (baricitinib, upadacitinib, and tofacitinib). The goal was to really understand how muchTYK2 inhibition these compounds give, and how much inhibition of JAK1, JAK2 or JAK3 these compounds give. The first thing we did was to look at how much does that the compound bind to the thing you want it to bind to. How tightly does your drug, or in this case, zasocitinib, bind to the TYK2 enzyme.What we found was that zasocitinib binds really strongly to that specific site, which is kind of what we expected based on the AI design features and what the compound had been designed to achieve. The next step was,asking how well does it bind to things we don't want it to bind to? If you want a specific TYK2 inhibitor, you don't want your drug to bind to JAK1, for example. What we found that was the zasocitinibbound to the TYK2 site 1.7 million times more than it bound to the JAK1 site. That is a huge range, and it means this is a very highly selective inhibitor. We then compared that to deucravacitinib, which is the TYK2 inhibitor that's currently available, and we found that that margin for deucravacitinib was only 87 times. What that means is that zasocitinib in vitro data shows that it's19,000 fold more selective than deucravacitinib for binding for TYK2 than JAK1. That's a great in vitro experiment. That's what the science tells us, but doesn't really necessarily tells us what happens in the body. We wanted to take it 1 step further and think about what does this mean? We took some some blood from some healthy volunteers and started to run some experiments where we incubate that blood with the compounds that we're testing, and then try and work out how much does this inhibit the cytokines, which actually cause the body to react to things. By doing that and combining it with some pharmacokinetic models, we can start with assess using predictions: How much inhibition of TYK2 or JAK1 will you get in a person when you give these drugs a clinically relevant doses, because the doses we tend to use in a lab tend to be doses way above what you give to patients. What we found in that pathway is, if you look at how well zasocitinib inhibits the TYK2 pathways, it generates about 91% inhibition of TYK2, and that 91% inhibition stays above what we call the IC50 so that's about the 50% inhibition level for 24 hours. It gives you this continual level of inhibition over the course of a day. If you compare that to the results we saw with deucravacitinib, the dose that's used in the clinic, which is 6 mg, only produced a 24% daily inhibition and it only stays above that 50% inhibition level for 3 hours. That's the TYK2 inhibition piece. The other piece we really wanted to think about was, what does it mean for other JAK inhibitors, the JAK target, so JAK1, 2, or 3. If you look at ducravacitinib and zasocitinib, neither of those drugs reached the IC50, the 50% limit for JAK1 or JAK2. Which is kind of what we expected to see and the percent daily inhibition for both drugs was it was zero for zasocitinib and 3% for ducravacitinib. Almost no JAK1 inhibition, JAK2 inhibition, or JAK3 inhibition for deucravacitinib. What I think is interesting is you can then compare that to some of the JAK inhibitors that are available, things like upatocinib, baricitinib, topacitinib, and if we assess those in a similar manner, you see that they actually havevery limited TYK2 inhibition. As you would expect, when you assess JAK1 or JAK3 inhibition, and you can start to see that these drugs have pretty different inhibition profiles for cytokines. You get this distinctive profile of what the TYK2 inhibitors can inhibit versus what the JAK inhibitors can inhibit. Those were kind of the key takeaways, and I think it was a really nice validation of the design process that went into the drug.

DT: What are the next steps in research for zasocitinib?

Heap: I thinkwe're really pleased with the development process we've been through so far with zasocitiniband we've been very excited by the phase 2 results we've seenin both psoriasis and psoriatic arthritis. We're particularly excited in the psoriasis program by our our12 week results, where we had about a third of patients achieving clear skin at that time point. Obviously, the next step we need to do is to take this into phase 3 trials and see what the drug can produce in a much larger patient population with much longer endpoints. Those trials are ongoing at the moment, and we've been very pleased with the recruitment in those trials, and they're moving along at a really nice pace. Within the psoriasis space, those large phase 3trials are ongoing at the moment, and we're looking forward for our phase 2 results to hopefully be validated in the phase 3 trial.We'll see what those results look like. That's really the next step in psoriasis. We're also evaluating the drug in psoriatic arthritis. There's phase 2 results that we've previously presented andthedata from that is going to be presented at the European dermatology meeting in the coming weeks. Those phase 3 trials are going to start in the coming year, and we're excited to see those get up and running. Then the other thing we're doing is we believe there's a lot of potential of TYK2across a range of immunomediated diseases. We're also assessing the drug in inflammatory bowel disease, and we have phase 2 studies in both Crohn's disease and ulcerative colitis that are already ongoing, and they've started recruiting as well. It'sa it's a big program, and we're excited to see it move forward.

DT: What are you most excited about as Takeda enters the Dermatology space?

Heap: Takeda is a relatively newcomer to the dermatology space, and we're really excited to introduce Takeda to the dermatology audience. This is a company that was founded in 1781, that's 243 years. This is a company with a deep understanding of pharmaceutical research, and we've been working in immune related diseases for over 3 decades now. We're really excited to bring that experience to dermatology, but also to learn about the dermatology community and understand how we can help.One of the things we're really excited about is advancing this compound.We think selective inhibition is really the key to achieving patients goals here, and the selective inhibition is a really, I think, a promising approach to trying to maximize, what we think is the potential benefits, while hopefully potentially minimizing some of the off target effects that you could start to see with less selective inhibition. We were really encouraged by the phase 2 results. Obviously, this is an investigational compound. More studies are needed. We're looking forward to phase 3 results coming out. We don't have any head-to-head studies yet, but we're looking forward to getting the results of the 1 we've started and some others that are potentially coming in the future.

[This transcript has been edited for clarity.]