Good afternoon, everyone, and welcome to the Jefferies Healthcare Global Conference. My name is Benjamin Graves with the investment banking team, and it's my pleasure to introduce to you to Archie Sze and Alex Ye with Adlai Nortye.
Thank you, Ben. Good afternoon, everyone. Thank you for joining us. Today, I'm pleased to introduce our two RAS program from our company, Internal Discovery. I'd like to make this presentation more interactive, so I will spend a few minutes talk about the two RAS program, highlight the program, and then I will bring Archie together. We're going to talk about few topic has been frequently asked in recent investor interaction. I want to remind, today's presentation and our discussion may containing forward-looking statement, and also including the investigation of product has not been approved by a regulatory agency. Please refer the further disclaimers for the more details. First, let me give you a briefly highlight for our pipeline. Adlai Nortye has been discovery differentiate oncology portfolio. This including our lead program, AN9025, which is a potential best-in-class, pan-RAS inhibitor.
Right now, the program is in phase I dose escalation study. We expect to complete a dose escalation in first half 2027. By the pan-RAS platform, we also working on a novel ADC, AN4035. This is a CEACAM5-targeting ADC, and will selective deliver the payload to the tumor tissue, further expand the therapeutic potential for the ADC. We also working on the next generation IO, A025. This is a PD-L1, LAG-3 variant, and also CD86 on the C-terminal. The program is currently in phase I dose escalation. We expect to complete the phase I-A by year-end 2026. When we talk about the oral AN9025, the oral pan-RAS inhibitor, we have been showing the differentiate with the market leader for in biochemistry level. We have about over 34 more potent compare with the market leader. This actually improve more durable RAS signal inhibition.
The biochemistry inhibition actually translate pretty well into a cellular potency. We roughly can achieve about 104 more potent in cellular, and additional about 254- 274 in vivo. Our molecule has a longer half-life, roughly about 40 hours in human projection. All those property actually give us a potential hypothesis to starting intermediate dosing in the human study. This molecule also provide a more deeper and more durable inhibition in the RAS signaling. All those property attribute potentially could further open the therapeutics window for the molecule. We have a very strong global IP position and then still retain the ex China right, give us further flexibility moving forward. Our second program, AN4035, is a novel ADC approach. Even the current pan-RAS inhibition has shown very exciting data, but still, people seeing a potential skin toxicity and GI toxicity, even there is a limitation for further combination.
Our solution is use ADC approach. We're able to deliver the pan-RAS inhibitor more selectively to the tumor, which can enhance, open the therapeutics window. Additionally, because this is novel payload mechanism and potentially could also address the lack of diversity of current ADC payload, and potentially can address the ADC resistance. Next year will be our very busy year. We were running to very rich catalyst event for our 2 RAS program. Right now, the AN9025, we starting the first patient enrollment about 4 months ago in both country, U.S. and China. We were planning to starting the intermediate dosing, which will be weekly dosing, by the mid-year 2026, and expect to complete a full dose escalation in first half 2027.
For the ADC program, we have achieved very compelling preclinical efficacy and safety. We expect to filing the IND mid-year 2026. We will expect to enter the human study second half this year in three country, including China, U.S., and Australia. We will also expect to report ADCs data in second half 2027. We believe we have a very strong cash position. We complete two PIPE financing in early this year. The current cash is about $290 million, would good for us until end of 2028. Give us about 2.5 years runway to fully, efficiently execute our RAS development plan. By now I'd like to bring Archie together. We can discuss about several topics being frequently asked by the investor in the last few months. In this, we can provide more color for two of our last program.
Thank you, Alex. My name is Archie Sze, President and Head of R&D of Adlai Nortye. I'd like to thank Jefferies for the opportunity to share our stories. Go ahead, Alex.
Yeah. I think it's lot of investors has been, when we have discussion with them, they always were asking some therapeutic windows and whether we have the opportunity to open the therapeutic window and whether we can improve the therapeutic window compared with the other RAS agents in the market.
Yes. You have heard from Alex that our molecule is more potent. I just want to clarify that we never claimed that a more potent drug would automatically become a better drug. The counterargument is, a more potent drug is going to be more toxic does not make sense either. It really depends on the dose and your data. In fact, we have benchmarked pretty extensively against the industry leader, divarasib, pre-clinically. Also we did extensive modeling to estimate the therapeutic index. Here we're showing in the table the therapeutic index estimations head-to-head against divarasib across a number of species. Just without going into the detail, the TI is basically an estimate based on the ratio of the MTD, maximum tolerated dose and also to the MED, minimum efficacious dose.
The MED is estimated based on a xenograft model using the tumor static concentrations concept. Just to show some validity of our model, the MED estimated for divarasib in our model was 105 milligrams. That's actually compared quite consistently with what RevMed had observed clinically seeing some efficacies. The MTD estimated for divarasib was around 550 milligrams. I believe RevMed never actually had reached an MTD, but the highest dose that they look at in the clinic was around 400 milligrams. Bear in mind that our MTD was estimated based on only a 14-day dose range finding study. Now if you extend it to, for example, a 28-day GLP tox, you expect that MTD probably is going to come down and actually to bring it even closer to what being observed in a clinic, 300 milligrams as the RP2D for divarasib.
That shows some validity and give us some confidence about our model. Using this methodology, the TI estimations for our molecule, for example, in mice was 6.6 versus 5.8 for divarasib, so it's comparable. In rats, ours was 3.4 versus 1.6, so slightly better, 6.9 versus 5.3 in projected human TI. Despite the fact that our molecule is about 250 or 274 more potent than divarasib, the therapeutic index is comparable or slightly better. This forms really the base case for our daily dosing. We preserve the therapeutic window. As Alex mentioned, there's potential upside for an even wider therapeutic window using the intermittent dosing.
Yeah. Just further clarify, for the current clinical phase I running in U.S. and China, we are running under the same global protocol. This is under the QD dosing, so oral daily dosing for the first few dose, and then we will continue for the oral daily dosing, but we will also open a new track for QD dosing, which is intermittent dosing, which is a weekly dosing. Archie, I want to see if maybe you can explain a little bit more to the audience why we thinking about a weekly dosing and what's the rationale behind we've been thinking.
Sure. Pre-clinically, we explore several dosing schedule. You can see here in the table on the right that in a xenograft model, we look at 0.1 QD. We also look at a twice-a-week dosing, and the last one was once a week, 0.7 Q week. When you add up the total weekly dose, let's say 0.1 QD times seven, total weekly dose being 0.7. If the total weekly dose is the same, you can see on the left that efficacy was the same. We've reproduced this in three different syngeneic model on top was a cholangiocarcinoma model, below is a melanoma model, and we don't have data to show here. We also had a lung cancer model. In the melanoma model below, you also see we also include a DUSP6 as a pharmacodynamic biomarker.
The same conclusion when the total weekly dose add up to be the same, efficacy was the same, and also in this case, DUSP6 inhibition was also the same. The question is what happened to safety, right? We look at safety in a dose range-finding rat study using the various dosing schedule. You can see when you look at the MTD for the QD dosing, the MTD was 0.2 MPK QD. x 7, that would be total weekly dose will be 1.4 if you expect the same therapeutic or same tolerance, right? If we do the Q-week dosing, the MTD was actually double, 2.8 instead of 1.4. Using the same methodology to estimate the therapeutic index, you can see on the far right that there is corresponding increase in therapeutic index.
This is supporting the hypothesis that the intermittent dosing can further widen the therapeutic window, and the rationale behind that is that normal tissue, because they are less addicted to the RAS signaling pathway, if you give normal tissue in drug holiday, they can recover faster from the on target of tumor toxicity. Whereas cancer cells are more addicted to the RAS signaling pathway. If you pause and blast them with a high dose and shut down the RAS pathway, they tend to die faster. And at least this is supported by our preclinical pharmacology and toxicology experiments. As Alex mentioned, we would start off with just a QD dosing, and at some point, we're going to branch off to initiate a separate and parallel track with the intermittent dosing.
Two potential outcome with the intermittent dosing if our hypothesis translate into the clinic, and that is we achieve comparable efficacy but with better safety, and that allow us to do combinations, a more tolerable combination. The second is if we can achieve comparable safety, but we can go up with the dose further to get better efficacy. I think that would be meaningful, especially in the setting of non-small cell lung cancer, where sotorasib and adagrasib somehow landed on a lower RP2D, 200 milligrams as opposed to 300 for PDAC. If we can optimize the dose using an intermittent schedule and achieve better efficacy, I think that would be very meaningful.
Thank you, Archie. This is very informative. Just a follow-up question I think it's been also very interest for the investor regarding the data release. We would expect to release the data potentially could be first half 2027, and I would assume that would be including the full QD dosings phase I-A data. Are we also expect to be a fully weekly dosings data or could be just partial?
We will have some of the Q-week data.
Okay. It would be very looking forward to see those data. Another question that I think is also people quite frequently ask about the GTP hydrolysis. Can you also comment about this area, whether it's very critical for a RAS inhibitor's activity and what's your view about this hypothesis?
Yeah, I think this is an interesting question. I have to say that the verdict's probably not out there yet. We believe that there are more than one way to skin a cat, so there are more than one way to inhibit the RAS signaling pathway. You can do it catalytically if you have a drug that can induce GTP hydrolysis. You can also shut down the RAS pathway quite effectively if you have a very potent molecule like AN9025, and you can do it stoichiometrically inhibiting the pathway. Our molecule has a slightly lower GTP hydrolysis compared to divarasib and sotorasib . We actually have not shown that data.
We do have data indicating that the GTP hydrolysis may not be so critical because if that's the case, then you expect that a less GTP hydrolyzing molecule would have lower efficacy in those mutants of RAS that are capable to hydrolyze GTP, for example, in the G12X variant. When you look at here the cytotoxicity of our molecule compared to divarasib and sotorasib across a panel of RAS mutant cell lines, you can see below on the left that the hierarchy of sensitivity is actually preserved between us and RevMed. G12X is still most sensitive, followed by G13 and then Q61, wild type being least sensitive. If GTP hydrolysis is so important, then you expect a drop in efficacy in the G12X, but we didn't see.
That's, to me, indicating that the GTP hydrolysis may not be as critical if you have a very potent inhibitor like ours that can stoichiometrically inhibit the RAS pathway.
Thank you, Archie. In the second half of this year, we're also going to progress the AN4035 into a human study. I wonder, since we have put both RAS programs in the clinical, what's your view about those programs? Do you see both programs as competing for a similar indication, or how do we differentiate the development resource for both programs?
Yeah, I think that's a great question. We really view these two programs as complementary than cannibalizing each other. The reason is the small molecule is more advanced, right? The mechanism has been proven. It's also a convenient way to administer the drug. The ADC program can, on the other hand, address some of the potential limitations of the small molecule, namely by selectively delivering the inhibitors to tumors through the ADC approach. It can avoid the systemic toxicities from pan-RAS(ON) inhibitor, especially in the skin and the GI tract. An ADC approach, we believe that can widen the therapeutic window. Also more importantly, it can also enable some rational combination that is otherwise challenging for a small molecule to do. For example, in combination with EGFR for colorectal cancer.
We really think that these two programs are complementary to each other, and we are equally excited about both programs.
That's great. For ADC, we see this potentially could be also a platform play because, Archie, you oversee the whole IND capability and also the whole IND activity inside the company. What's your view about the additional ADC program, and what area potentially could be, you feel, could be a good fit for the ADC approach?
Yeah. Thanks to our small molecule program, we have a suite of highly potent small molecules that can be modified to turn into ADC payload. The one we pick here itself is more potent than divarasib. It itself has nanomolar to picomolar IC50. A very potent molecule by itself. One very compelling piece of pharmacology for our ADC platform, by the way, it's called RASiCA, which stands for RAS inhibitor conjugated antibody. It's been trademarked. That is the retention of this class of drugs, presumably because of cyclophilin A binding. In this slide, you can see when we treat with ADC, and then we look at the amount of free payload inside the cell versus outside the cell. The red line on the graph on the right, you can see a high and sustained level of free payload inside the cell.
When you look at the ratio between amount of drug inside the cell versus outside the cell at 24 hours in the table below, you can see that the ADC payload achieved a 44-fold enrichment. In contrast, when you treat the cells with a topoisomerase I base ADC that also targets TROP2, that ratio was only 1.0. Meaning the topoisomerase I inhibitor, after it's got internalized and cleaved, it would rapidly exit the cell. In comparison, the RAS inhibitor ADC payload would be retained and kill the cell, presumably because of binding to cyclophilin A. That translated into a very promising in vivo efficacy. You can see in the xenograft model, CL-40 is a colorectal cancer model below. The comparison here actually is more important in comparison with the small molecule payload.
You can see that when you give the animals the small molecule free payload at 1 MPK daily, this is close to its MTD already. You cannot push the dose much higher. You can see the ADC achieved much better efficacy compared to giving the free payload. The ADC can achieve that in the absence of any weight loss. This really has proven the ADC approach could widen the therapeutic window. In this biodistribution study, it further supports that hypothesis. Here, we're looking at a tumor-bearing mouse, and you can see on the right when we measure free payload after administration of just the payload, tumor in red has some enrichment of the free payload over the blood. Compared to other normal tissues, actually, tumor did not contain the highest concentrations of the free payload.
On the other hand, when you look at free payload after administration of the ADC in the graph on the left, you can see in red tumors indeed contains orders of magnitude higher concentrations of the free payload compared to blood. Not just compared to blood, but also compared to other normal tissues. This is, again, has shown that the ADC approach can push more drugs into the tumor while lowering the systemic exposure. By the way, this payload also had a pretty robust bystander killing effect. All these together really make it very ideal ADC payload and really complements our current arsenals of ADC payload, which is the topo and antimitotic agents.
Thank you, Archie. I think this is few topic we've been discussed, Right now we would be happy to open the floor to the audience, We happy to take any questions. Go ahead. Okay, please, Amy.
Can you explain how dosing is going to be different between ADC and the pan-RAS inhibitor? What's the expectation of the amount of drug that's going to be delivered between the two products?
Yeah, maybe I take that. The small molecule, it's a oral drug, right? It's going to be dosed by two different dosing schedule. As I mentioned, one is the QD, once daily, continuous dosing. That's sort of like the base case, what the RevMed, Erasca, other companies is doing. The alternative dosing schedule is the once a week dosing schedule that we're going to test as well. We are hoping that with the once a week dosing, it potentially can widen the therapeutic window. We'll see. ADC is intravenous so it's going to be given once every two weeks. That's more sort of aligned with the other colorectal cancer regimens. As Alex mentioned, the ADC prime indications is going to be colorectal cancer and to some extent, PDAC, and then followed by lung. We're sort of targeting a more colon-centric approach.
Good.
for your ADC strategy, primary focus is going to be combination like your phase I and II different combinations?
Yes. The ADC probably more amenable to combinations, including EGFR combinations. The first-in-human study, an EGFR combination has been shown to be efficacious in combination with various MAP kinase inhibitor, right? Including KRAS and also BRAF inhibitor in colorectal cancer. That's a very validated biology. The first-in-human study is going to be designed such that there will be a monotherapy dose escalation and in a staggered manner, there will be also a EGFR combination.