Good morning, everyone. Welcome to today's webcast, where we will review the preclinical data from our CDC7 program that was presented on Saturday at the virtual AACR annual meeting. We'll also provide an overview of two other internal programs, MALT1 and WEE1, as well as highlight the role of our computational platform in accelerating the discovery of our novel molecules. Please note that portions of today's discussion, including our responses to questions, may include forward-looking statements about our future expectations and plans, including the speed and capacity of our platform, the clinical potential of our internal programs, the favorable properties of the inhibitors we've identified, our clinical development plans, and our operational plans and strategies.
Our actual results may differ materially from what we describe today due to a number of important factors, including the considerations described in the Risk Factors section of our Form 10-K and in other filings that Schrödinger makes with the SEC from time to time. These forward-looking statements represent our views only as of today, and we caution you that we may not update them in the future, whether as a result of new information, future events, or otherwise. Before we begin our discussion, I'd like to point out the Q&A box on the bottom of the website platform. You have the opportunity to type in questions in this box at any time throughout our discussion, and we'll do our best to answer everyone's questions. If we don't have time to get to your question, we'll try to follow up with you after the program has ended.
Additionally, there will be a full recording of today's webcast archived in the Investors and Media section of the Schrödinger's website for the next week. Now I'll turn the program over to our CEO, Ramy Farid.
Thank you, Jaren. Thanks everyone for joining us on today's webcast. Let me start by giving you a high-level overview of Schrödinger. We've developed a computational platform that is transforming the way therapeutics and materials are discovered. The platform is enabling our customers and our internal drug discovery team to discover high-quality molecules for drug development and materials applications faster and at lower cost with a higher success probability compared to traditional methods. We have a software business where we license our platform to pharmaceutical companies, biotech companies, materials companies, universities, and government labs worldwide. We're also leveraging our platform in a number of drug discovery programs in collaboration with pharma companies and biotech companies, some of which we've actually co-founded. We also have an internal drug discovery pipeline, as Jaren mentioned, that we'll be telling you much more about shortly.
The computational platform that we've developed is having a significant impact on drug discovery programs, as you can see in this slide. In traditionally run drug discovery, it's typical for about 5,000 molecules to be synthesized, and when successful, it can take four to six years to get to a development candidate, and it's still often the case that those development candidates have issues. By leveraging our platform at scale, and that's a very important point, and you'll see that shortly, we and our partners are able to explore billions of molecules computationally, synthesizing far few, only about 1,000 molecules, and it takes about half the time to get to a development candidate, and this is key with high-quality molecules.
We're continuing to invest heavily in advancing the science underlying our computational platform with the aim of approaching the ultimate goal of computing with high accuracy, and again, that's very important, you'll see that shortly, all relevant molecular properties on as much of chemical space as possible. The point here is that the more molecules you can explore and the more properties we can compute with high accuracy, the more likely it will be that we can rapidly discover high-quality drug candidates. To give you a sense of what the next decade might look like, it's useful to see how we got to where we are now. As you all know, of course, computer performance has been increasing exponentially for a long time. That we're all very well aware of.
What may be less well known is that, for example, the number of proteins in the human genome for which we know the three-dimensional structure has also been increasing exponentially. The number of compounds we can explore computationally has also been increasing exponentially, we're at around 100 billion or so, and again, Robert's going to cover that in more detail shortly. As has the number of properties that we can compute with near experimental accuracy. Again, we'll cover this in more detail shortly. What does all this mean? What it means is that we can reasonably predict that within the next decade or so, computers will be potentially more than 100 times faster than they are today, which is kind of incredible. High-resolution protein structures, and that point is very important.
High-resolution protein structures of approximately 50% of the human genome will be available. There'll be approximately 10 or so molecular properties that we can compute with experimental accuracy. This may all lead, in the next decade, to us being able to reliably discover extremely high-quality development candidates, potentially within a year from program launch. It's clearly a really exciting time for computationally driven drug discovery, and in general, molecular design. We'll continue to advance our computational platform to realize the vision for the future that I just described on the previous slide. We'll continue to license our computational platform to pharma and biotech companies. We'll continue to advance our collaborative drug discovery programs and initiate new ones. We'll continue to license our platform to material science companies. We'll continue to advance our internal drug discovery programs and initiate new ones. Going to hear about that shortly.
We expect in the future to initiate material science collaborations, as we did about 15 years ago with drug discovery collaborations. I'd now like to hand it over to Karen Akinsanya, our Chief Biomedical Scientist, and Robert Abel, our Chief Computational Scientist, who will tell you more about our platform and our collaborative and internal drug discovery programs.
Thank you, Ramy. As described, our computational platform supports and underpins all the activities Ramy was describing. We have here depicted how we use our computational platform to accelerate drug discovery. On the far left of this slide, you can see how our technologies support hit identification using combined methods that leverage both physics-based methods and machine learning capabilities. Once we find those hits, the first order of business is typically to determine binding modes for those hits. That can be done either by way of working with partners to experimentally determine structures or to use the physics-based methods themselves to determine the binding modes of the molecules.
Once we have the binding modes of the hits in the context of the protein target, we can then use the full power of the platform to perform large-scale ideation, where we can enumerate all the various synthetic derivative molecules that a project team might want to consider to further optimize those hits into molecules that might be advanced into in vivo studies or clinic with confidence. Once we've ideated those molecules, we can use combined physics-based and machine learning methods to do multi-parameter optimization, where we are finding those few molecules that will have the right potency, selectivity, and other property balance, such that they can be advanced forward into synthesis and assay with confidence. This cycle is iterated as the project team obtains more experimental data regarding the particular constellation of properties that are necessary to have those molecules that can be advanced in the clinic.
We iterate around the cycle doing new large-scale ideations and new multi-parameter optimizations to triangulate those molecules that will be most exciting to advance the discovery project. A key capability that allows us to progress these projects is to combine the accuracy of physics-based methods with the speed of machine learning. We have here depicted an example of how we combine these methods. For example, if we want to evaluate a space, a chemical space of 1 billion molecules, advancing 1 billion molecules into full physics-based computational methods, full free energy calculations, is not computationally feasible. We can easily evaluate a representative sampling of 1,000 such molecules in a day.
Once we have this sampling of the full space of interest, we can build machine learning models, approximate machine learning models, which then can advance an improved subset of that full chemical space into full free energy calculations, in this instance, 5,000 molecules. Then based on retrospective profiling of the accuracy of these methods, of those 5,000 that are advanced into full free energy calculations, we would easily find 10 that look very promising. Then based on the accuracy of the physics-based methods, typically about eight out of those 10 would be optimized along the property dimensions most interesting to the drug discovery project and would materially advance that discovery project toward its endpoints. Now I'd like to hand this discussion over to Karen Akinsanya, who will describe how we use these methods to advance our active discovery projects.
Thank you, Robert. As Ramy alluded to in his comments, we have been deploying this platform for a number of years. In fact, over the last 15 years, we've engaged in collaborative drug discovery with a number of companies. On this slide, you can see the large number of programs that we've been working on in those collaborations. I think just to point out a few things, as Ramy described, these projects have an average life cycle from initiation to development candidate declaration of about two to three years. You can also see that we've worked with small biotechs as well as large pharma, who in many cases, have access to the software themselves, but using it at the scale we do, is how these collaborations have been conducted.
I think the other thing to point out before I move on is that the targets that are the subject of these drug discovery projects cover a wide range of different target classes, protein classes. Some of these are GPCRs, kinases, integrins. It's a very large array of different targets. I'm going to move on from this slide just to describe, as Ramy also indicated, that we have our own internal drug discovery programs. We do obviously still collaborate, but we have initiated an internal pipeline, and today we'll be describing three of our advanced programs, CDC7, WEE1, and MALT1. In addition, since we kicked off our internal program efforts just under three years ago, we had also worked on a number of other projects, several of which actually were subject to a collaboration that we announced last year with Bristol Myers Squibb.
I'm going to move on now to describe our CDC7 program. This was the subject of an AACR presentation this weekend. CDC7 inhibition, we view as an interesting mechanism that allows one to exploit replication stress. CDC7 itself is a kinase that is important in the initiation of DNA replication and we view as an attractive target for cancer therapy. Overexpression has been noted in a number of solid tumor types, including ovarian, lung, and triple-negative breast cancers. There are a number of studies that have been conducted with CDC7, both in CDX, PDX, and in phase I-B studies. CDC7 is interesting because if you inhibit this kinase, it prolongs the S-phase progression, which actually allows you to sort of influence the fate of cancer cells.
Cancer cells are particularly vulnerable because they have a lot of DNA damage repair, and when you inhibit CDC7, this can drive them towards mitotic abnormalities and ultimately apoptosis. It's important to note that CDC7 has been of interest for quite some time. If you look back to the 2000s, there are descriptions in the patent literature of CDC7 inhibitors. Interestingly, these were not as potent and indeed had a challenge in terms of PK and selectivity. The last generation of CDC7 inhibitors discovered were definitely more potent, in addition, still had some challenges with regard to PK and/or poor selectivity. The goal of our program has been to identify picomolar inhibitors with improved drug-like properties. I'm going to hand back to Robert now, who will describe how we arrived at our lead molecules.
To meet the challenge outlined by Karen to identify these very tight binding CDC7 inhibitors, we used our computational platform to ideate and triage over 74 billion derivative molecules of the initial hits. We had used our platform to discover versus this target. Those 74 billion subsets, almost 6,000, were progressed to very sophisticated physics-based modeling techniques, which were utilized not just to optimize potency, but to simultaneously optimize the balance of the potency, the selectivity, and the permeability of these molecules. Leveraging the accuracy of those physics-based methods, only 226 compounds had to be advanced to synthesis and assay in order to identify DC quality matter. In fact, those compounds include what is, to our knowledge, the most potent and ligand-efficient inhibitor ever discovered for CDC7.
Now I'll hand things back to Karen to describe in detail some of the experimental profiling of these compounds.
Thanks, Robert. Over the next few slides, I'll describe the characteristics of the compounds that we're advancing. First of all, you can see in this slide that we have very potent inhibitors that are picomolar in nature. This is a survey of several compounds that we have characterized, where you can see that both by biochemical assay, in terms of the ADP-Glo, we've seen a very nice potency, but in addition, we've characterized the biophysics that shows that if you look at binding of CDC7-Dbf4, again, you can see by SPR, very nice 10 picomolar KD. It's also worth noting the t halves where we see a variety of different residence times for these inhibitors. Further, we have characterized the effects of these inhibitors in terms of MCM2 phosphorylation.
You can see here that we are in COLO 205 cells, which is a colon cancer cell line, where you can see we have a variety of our different compounds here, a nanomolar IC50 in terms of the MCM2. In the Western blot, you can see dose-responsive effect on MCM2 as well. We have compared our compounds with a reference standard, TAK-931. Again, here you can see a fold potency difference between our compound and TAK-931. Further, we've looked at the impact of CDC7 inhibitors in terms of anti-tumor cell growth activity. Here you can see again a range of compounds, and their effects on tumor cell growth. Compound 1 through Compound 3 showing a difference in terms of potency and compared again here to TAK-931 in terms of potency. We're very pleased to see very active compounds in our lead molecules.
We've also characterized the activity in a large range of cell lines. Here you can see an example with compound three, where we've looked at several hundred cell lines, actually, and you can see there's a variety of response across those cell lines. Of particular note is the AML cell lines, which do appear to be more sensitive to CDC7 inhibition, here on the lower right, relative to solid tumor cell lines. We've also been very interested in characterizing the somewhat selective nature of CDC7 inhibition on cells. Here you're looking at the induction of apoptosis, again in COLO 205, the colon cancer cell lines, versus normal fibroblast WI-38. What you can see here is that 24 hours after treatment with our CDC7 inhibitor, you're seeing an impact on COLO 205, but nothing on the normal fibroblasts.
That's in contrast to staurosporine, which is a non-selective kinase inhibitor. What you can see here is that you're getting effects on both the cancer cells and the normal fibroblasts. And we view this as important in terms of therapeutic index. Further, we've looked at the cell cycle dynamics, and we've studied, again in COLO 205 cells by flow cytometry, the impact of these inhibitors on cell cycle dynamics. What you can see is that at lower concentrations and earlier time points, we see an increase in cells in the S phase, and later this transitions into an increase in G2/M cell population and robust induction of apoptosis. We feel this is represented by the sub-G1 phase cell population, which you can see here in yellow.
We did not observe this in normal human bone marrow mononuclear cells, which we think is again, an important distinction when it comes to the selective nature of CDC7 inhibition on cancer cells. Moving on, just to describe some combination work that we've been doing. We think CDC7 inhibitors could be potentially utilized in combination with both approved and investigational drug candidates. Here we can see that in combination with venetoclax in terms of BCL-2 inhibition, olaparib, a PARP inhibitor, an ATR inhibitor as well as WEE1 inhibition, you see very nice synergistic effect on inhibition of cancer cell viability. Finally, we've taken our inhibitors in vivo, and we've been able to now show strong anti-tumor activity in colon cancer xenograft models at very low doses.
In fact, these are very potent compounds as we described, therefore, you can see that at 2 - 10 mg/ kg BID, you're seeing very nice anti-tumor activity. It's pretty clear from the PK of these studies that we are well above the IC50 required to drive these effects, even at these very low doses. We also see very nice target engagement in terms of phosphorylated MCM2 in tumor tissue, six and 12 hours after dosing on day 14. We've also investigated this in AML xenograft models. Here you can see in the MV-4-11 tumors a similar story in terms of anti-tumor activity.
Here you can see at 1, 2.5, 5, with different dosing schedules, very nice anti-tumor effects, with very robust tumor inhibition, where you can see with a dosing holiday, actually, we are able to maintain efficacy as well as you can see here, very well tolerated in terms of the animal body weight, and the effects on MCM2 are as we would have expected. In summary, I'd like to share that we are very pleased with our CDC7 inhibitor leads. We have potent selective compounds that show robust anti-tumor activity. We believe these are the most potent CDC7 inhibitors reported to date. Importantly, as I pointed out in the earlier slide, they have excellent PK and drug-like properties.
We have shown nice target engagement through MCM2, both in vitro but more importantly in vivo, and the potential for combination of CDC7 inhibitors with a number of other agents. Our in vivo data, we think, gives us great support for moving forward with this mechanism. I'm next going to move on to MALT1, which was presented at ASH in December. MALT1 is another of our advanced programs where this is in the BTK and NF-kappa B pathway. MALT1 is one of the key regulators in the BCL10- MALT1 CBM complex signalosome. In fact, 30%-40% of DLBCL patients currently experience progression or relapse following standard of care R-CHOP. We believe that MALT1 is going to be a very interesting mechanism with regard to suppression of NF-kappa B signaling in such patients.
Mutations in MALT1 have been shown to trigger constitutive activity in combination with these fusions where you see really very robust drive through NF-kappa B. Our data suggests that MALT1 may be an interesting mechanism for patients who suffer with B-cell lymphoma, in particular ABC- DLBCL. I'm going to hand over to Robert to describe the discovery of our MALT1 inhibitors.
Thank you, Karen. To support the progress of this drug discovery project, we've been able to use our computational platform to identify and advance multiple novel and promising series. This involved computational exploration of billions of compounds, which allowed the team the license to be highly efficient with regard to their execution of the project chemistry, really only go after synthesis of those molecules most likely to further advance the discovery project. Key piece of this was utilizing the technologies to support multi-parameter optimization of the matter that we could achieve very high potency while also having good drug-like properties, good ADME properties. By leveraging our platform in this way, we're now on track to initiate IND-enabling studies in the first half of 2021.
I would want to highlight that the full breadth of capabilities of the platform were utilized to support this work, including supporting the hit finding activities of the projects, doing iterative optimization of those hits once they've been identified, and doing multi-parameter optimization using our most sophisticated computational analysis techniques to ensure that these molecules were evolving along property dimensions that would support them being advanced into in vivo studies. The deployment of the technology in this way allowed us to ideate and triage a idea space of 8.2 billion compounds. About 12,000 of those were advanced into atomistic physics-based modeling. Then from that computational data, through synthesis of only 78 compounds, we were able to arrive at DC quality matter within 10 months.
Within the first two months of the project, we were able to identify matter that could be advanced into in vivo studies to establish good PK of the compounds we were synthesizing. With that, I'll hand things back to Karen to discuss more detailed experimental characterization of these compounds that we've been able to discover for this target.
Thank you, Robert. I'm going to give you a sample of the data. The ASH presentation went into a lot more detail, I will just share a few slides. One of the key characteristics as we described was the suppression of NF-kappa B transcriptional activity. That's the goal for the MALT1 inhibitor program, you can see here in the western blot that we have been able to characterize the cleavage of a number of substrates, RelB, N4BP1. In terms of our inhibitors, 7055 is shown on this slide, where we've been able to see a dose-responsive effect. Further in NF-kappa B reporter assays in Jurkat cells, we've been able to demonstrate a very nice dose-dependent inhibition of NF-kappa B activity.
In terms of IL-2 secretion, we have a number of markers that we're tracking in this program, but you can see here, again, very nice inhibition of IL-2 secretion in Jurkat cells. In vivo, we've been able to characterize several of our leads, and again, referencing the ASH presentation where we showed anti-tumor activity in the OCI-LY10 CDX models, where we've been able to characterize the characteristics, the anti-tumor activity of 7055 administered BID. You can see a nice dose-responsive reduction in tumor volume. We use uncleaved BCL10 as a biomarker of MALT1 inhibition. You can see again, a very nice dose-response increase in uncleaved BCL10. In keeping with my comments about interleukin-2, here we're tracking IL-10 as another biomarker, and you can see very nice dose-responsive inhibition of IL-10 in vivo.
We've also looked at combination opportunities for MALT1. Here you can see two graphs on the lower panels describing combination of our compounds with ibrutinib in one case, where you see additive effect for MALT1 and ibrutinib. On the last panel there, a combination of MALT1 inhibitor in our hands with venetoclax also showing a very nice additivity. We're going to move on now to the WEE1 program. I'll start by describing WEE1. This is sort of in keeping with our interest in DNA damage repair and replication stress. WEE1 is a very important kinase involved in G2/M and S-phase checkpoint activity. It's actually a gatekeeper of the G2/M cell cycle checkpoint. In normal cells, DDR or DNA Damage Repair is mediated by checkpoints, which either activate DNA repair or induce cellular apoptosis or senescence.
In the case of our WEE1 program, we're looking to discover very selective inhibitors of WEE1. As we described for CDC7, we believe that if you're able to inhibit WEE1, this will make cancer cells a lot more vulnerable because of the buildup of replication stress. It's important that one does that in the context of very selective inhibitors so that you don't get a lot of off-targets or toxicity. I'm going to just briefly describe some of the data that's in the public domain for WEE1. This is obviously now a clinical asset in other companies' hands. This is a phase II study from AZD1775 in uterine serous carcinoma. This was presented, in fact, last year at ASCO, where there was very nice data demonstrating monotherapy efficacy with an ORR of 30%.
There have been a number of studies, obviously, with AZD1775 in a number of different tumor types, including lung, ovarian, and some recent data in pancreatic. We think this is a very interesting mechanism. We do think that the optimal WEE1 inhibitor profile will be one that supports dosing flexibility and combination opportunities. And as such, as I alluded to, we think minimizing kinase off-targets is going to be key, and also avoiding any ADME or PK challenges, including the potential for inhibition of CYP3A4 which has an impact on elimination of the drug or potential for accumulation, and also may make it challenging to combine with other agents like PARP inhibitors. Those are the goals of this program, really coming up with a very selective inhibitor with excellent drug-like properties.
With that, I'm going to pass this over to Robert, who will describe how we used our technology to arrive at some very interesting leads.
Thanks, Karen. As highlighted, selectivity was a major challenge and goal for this project. We were able to adapt our computational platform to identify those residues that maximally differentiated this particular target from the rest of the gene family, then further use our computational platform to optimize the molecules such that they would maximally engage the intended target while minimally binding to the rest of the kinome. We have here examples where we had initially quite promiscuous binding molecules, where through only a handful of rounds of chemistry, we were able to take those molecules from being promiscuous binders to be exquisitely selective for the target of interest. We were able to leverage this capability for WEE1 to identify a matter that really binds with very high selectivity toward the intended target.
In fact, we were able to do this for multiple different lead series, such that we could find those molecules that would both manifest the binding selectivity we were after, as well as the constellation of other properties we were seeking to optimize to have as high quality matter as possible to advance into follow-on experimental studies. With that, I'll hand things back to Karen to sort of summarize some of this progress.
Yeah. As Robert said, with a breakthrough really provided by the technology, we have been able to identify WEE1 inhibitors that have no PLK1 activity, in fact, as well as exquisite selectivity against a number of other kinase hits that have been seen in prior series of WEE1 inhibitors or generations, I should say. That allows us now to move these molecules that are potent on WEE1 with very limited activity against any other kinases into in vivo studies. We have now generated very nice data in vivo, which we'll be sharing at a later time point. We've been able to show a very favorable ADME and PK properties, as well as importantly, very robust antitumor activity in relevant CDX models.
In summary, for the internal pipeline, we've been able to deploy our physics-based methods really to a number of programs that we've shared with you today. Of course, both the collaborative portfolio, but we focused today on the internal programs. In each, the technology had a profound impact on the discovery of unique molecules. We've been able to now build the capabilities to support the clinical execution of our programs. That's an ongoing process. We plan on moving forward with IND-enabling studies for our internal programs. We expect to submit up to three INDs next year, with our first IND submission expected in the first half. We'll also be expanding into additional disease areas and initiating new programs this year. With that, I will hand the platform back to Ramy to close.
Thanks a lot, Karen. In closing, and before we open the call up then to the Q&A session, I'll just leave you with this slide that I showed you earlier, which again provides an overview of our business and how we intend to continue to grow by not only investing in all of these areas, but also by leveraging the extraordinary synergies between all these areas of our business. Thank you everybody, and we'll now open the call up to Q&A.
Thank you. To ask a question, you will need to press star one in your telephone. To withdraw your question, press the pound key. Please standby while we compile the Q&A roster. Our first question comes from Michael Yee with Jefferies. Your line is open.
Hey, guys. Good morning, and thanks for hosting this. I had two questions. On CDC7, I guess, which was of course the highlight of the AACR for you, can you kind of compare and contrast the potencies and the differentiation versus Takeda? I think you showed Takeda in many of those as a control. Maybe just talk about the potencies there and how to think about differentiation. Secondly, on MALT1, I think there's a lot of interest here, of course, because everybody knows the BTK market. Do you expect to see strong response rates as a single agent post-BTK, and do we expect that J&J would be a good read-through given I think they're in the clinic? Thanks so much.
Yeah, thanks for the question, Mike. First of all, on CDC7, you are correct that TAK-931 has been shown to also have a very good potency and we believe that our compounds perhaps are the most potent inhibitors, and we've shown this both by biophysics data as well as binding data. As you move across from biophysics into cell-based assays and then in vivo, I think what we're seeing is that we're able to dose at a very low milligram per kilogram dose to be able to achieve a very robust antitumor activity. We do believe that the potency we see really does translate through to in vivo activity. I'll also say that from a selectivity standpoint, selectivity is obviously important, and we do think TAK and our compounds are sort of on par with regard to selectivity.
Really it's this low dose that we think is going to offer us the opportunity for a decent therapeutic index. Also I would say that the dosing schedule that we've been working on, I think also offers up some very interesting opportunities for the mechanism. Then with regard to your question on MALT1, you are correct that Janssen is in the clinic studying this not just as monotherapy. It seems from recent trial submissions in clinicaltrials.gov they are also looking at this in combination settings, perhaps with ibrutinib, I believe was the compound named. We do believe actually that based on our data, at least in preclinical CDX and PDX models, that you will see some monotherapy activity from MALT1. Really I think the question is what does the combination do in terms of the ability to really push cancer cells into regression?
That's work that we're doing right now. We are looking forward to seeing, obviously, more disclosures around MALT1 at other meetings, AACR, ASH, to get a sense of how this is all translating into the clinic. We do think that you will see monotherapy efficacy with MALT1.
Great. Thank you. Appreciate it.
Thank you. Our next question comes from David Lebowitz with Morgan Stanley. Your line is open.
Thank you very much for taking my question. When you look at other CDC7s, what characteristics do you think those molecules have, I guess, given them weaknesses that you think your technology can allow you to exploit?
Yes. A little hard to hear you. I think I captured it. The weaknesses in other molecules. What I'll just say is that when you look across the patent publication literature for CDC7, as I was alluding to in one of the early slides, being able to get really potent inhibitors that also have great PK as well as limited ADME liabilities has been a challenge. Some of the very first CDC7 inhibitors that went into the clinic essentially were not able to progress through phase I because of limited PK. The real goal here was to be able to get the potency, the selectivity, and the ADME properties that will include things like solubility also limited or no implications in terms of CYP3A4 or other DDIs.
We believe that our molecules have that sort of balanced set of properties as well as this potency that allows us at 1 and 2 mg/ kg to see really robust antitumor activity. When we've compared with reference molecules, we do believe that this potency has led us to be able to go in with human dose predictions that will be quite low in terms of a milligram per kilogram basis.
David, were you also asking about the role the technology might have had in getting to that profile? It was hard to hear you. Was that also part of your question?
I was really more focused on the actual target themselves and specifically on the molecule.
Great. Okay.
Less on the technology.
Got it. Okay. Sorry.
If we flip over to WEE1.
Yes
There's been some new clinical data this weekend with a new molecule. Beyond AstraZeneca's out there putting out some data, and I'm wondering how you view the new molecule vis-a-vis AstraZeneca. As far as how your molecule might fit versus this new one.
David, what I'll say there is that we've been really focused on this question of selectivity from the outset of the program. We believed that therapeutic index, because clearly there were signals of efficacy early on actually for the earliest WEE1 inhibitors in the clinic. We believed that having a very selective compound was going to be able to open up the therapeutic index. That was really the goal of our program. Obviously as our compounds have moved forward and we've been able to solve that challenge, one of the other things we've been really focused on is the PK/PD relationship and really what that leads to in sort of in vivo.
We have benchmarked our molecule against AZD1775 at this point and believe that we have a PK/PD relationship established that gives us a lot of confidence in moving forward with the set of novel WEE1 inhibitors into the clinic. With regard to more recent entrants, we don't know the structure of those, and so it's more challenging to benchmark them. We are very pleased with the validation that we're seeing for this mechanism through others' work in the clinic. We believe that this is a really a strong mechanism and one that we're excited to take forward.
Thank you for taking my questions.
Our next question comes from Do Kim with BMO Capital. Your line is open.
Hi, this is Jameson on for Do. Thanks for taking the questions and congrats on the data and progress. First one on the CDC7 program. To clarify on the dosing, does the preclinical data support the potential for once-daily dosing versus the twice-daily used in the xenograft model? Takeda's, excuse me, CDC7 inhibitor had a once-daily dosing profile in its phase I. Any additional color on how your compound's profile will compare when you do eventually take it into the clinic? I have a follow-up. Thanks.
Yeah, thanks for the question. I think it's worth noting that in vivo in animal models, a number of compounds, not just CDC7 inhibitors, show more rapid clearance. I think we fully expect, based on the human dose predictions that we're working on, that this compound will not necessarily need to be dosed BID. We think actually we're going to have a very nice half-life in humans. It is sort of common to see BID dosing in mouse models to maintain exposures. As you saw in the studies we showed, we're expecting to be well above IC50 with a number of different dosing schedule options. If we're right about the predictions of our human PK, that should open up some very interesting opportunities in the clinical setting.
Great. Thanks for the clarification. One more on CDC7. In the sensitivity analysis evaluating the different cancer cell lines, is there any underlying reason why B-ALL would be more sensitive to CDC7 inhibition compared to solid tumors? Does that broadly apply to other hematological malignancies? Thanks.
That's something that is currently under investigation by our research team. We believe that tumors that have a high degree of replication stress are potentially more sensitive to CDC7 inhibition. That's something that we're working through right now to be able to compare and contrast the responses that you see, not just between solid tumors and liquid tumors, but also between different solid tumor types. That's the subject of ongoing research by our team.
Got it. Thanks again for taking the questions and congrats on the data.
Thank you.
Once again, ladies and gentlemen, if you wish to ask a question, please press star then one. Our next question comes from Michael Ryskin with Bank of America. Your line is open
Hi, this is Wolf Chanoff on for Mike Ryskin. Thanks for taking my question. Starting off with CDC7 places in the deck, we noticed that you provided data for a number of different compounds, some places CPD I, some places CPD II. Just wondering if you're seeing kind of similar effects across the compounds whenever you aren't listing those assets? Then I have a few follow-ups.
Yeah, it's a great question, Michael. One of the things about the approach that we take to drug discovery is that we are able to rapidly identify a number of different series of molecules. We were very fortunate, actually, in a number of these programs to have really great characteristics of these molecules early on in the programs. We have been fortunate, and I think Robert touched on this in one of his slides, that early on in the programs, within months of initiation, we were able to take a number of these different compounds into in vivo settings or into these cell line screens. As you can imagine, we are looking across our different molecules and deciding which of those has the best balance of properties to move it into IND enabling studies. We do have a number of options across our series, to consider.
Just to sort of round out, for a number of these lead series, we have seen very nice in vivo data. There's no sort of different sort of compound one, three. We're seeing nice data across them all.
Okay, that leads directly to my next question. That sounds like you guys have not selected a lead compound yet. That's still down the line?
It is in progress.
In progress. All right, I appreciate that. Just for one more general follow-up, could you provide a little color on the status of the IND enabling studies? Are there any steps that you guys need to take for those three programs before the studies kick off? If so, when can we expect that?
As we've discussed previously, obviously one of the key things was booking slots, that's work that we've completed, booking slots for our programs. Everything else is moving along quite nicely. Of course, API and CMC is something that's critical at this stage, and that's something that's also making very nice progress. I think as we've shared previously, we are kicking off IND enabling studies in the first half of this year. The number of programs that we're fortunate to be looking at here, we expect to be doing that over the course of the next months into 2021. As we've shared, again, we expect to be through IND enabling studies with our first IND opening in the first half of next year.
Okay. Thank you. I really appreciate all the content.
Thanks.
Thank you. I'm showing no further questions at this time.
Thank you. I also have no further questions on the webcast, so I'll turn it back over to Ramy to wrap up.
Thanks. Thanks everyone very much for joining us today. I just wanted to just say a few words just to sort of close things out. While we focused today on just three of our internal programs, I think you saw that we have quite a diverse portfolio, both collaborative and internal programs advancing toward the clinic. It's really been extremely gratifying to realize the power of our computational platform as we advance our own pipeline. We're really looking forward to updating you on our R&D activities throughout the year. Thanks again very much, everyone, for your time.