IDEAYA Biosciences, Inc. (IDYA) Earnings Call Transcript & Summary

Good morning, and welcome to the IDEAYA Biosciences Investor R&D Day. At this time, all participants are in a listen-only mode. An analyst Q&A session will follow the formal presentation. As a reminder, this event is being recorded, and a replay will be made available on the IDEAYA website following the conclusion of the call. I would now like to turn it over to your host, Yujiro S. Hata, President and Chief Executive Officer. Please go ahead.

Good morning. This is Yujiro Hata, IDEAYA's Founder and Chief Executive Officer. I wanted to welcome everyone to our 2023 IDEAYA Investor R&D Day event. Note, we'll be making some forward-looking statements, and please refer to our SEC filings as appropriate. I wanted to first thank all of our listeners for dialing in today and provide a special thank you to our guest speakers for their participation, including Dr. Timothy Yap from MD Anderson Cancer Center; and Dr. Ramon Kemp from GSK. For today's agenda, I'll kick us off to walk through IDEAYA's vision and strategy to build a leading precision medicine oncology company. Then Dr. Michael White, our Chief Scientific Officer, will present our efforts in computational drug discovery. Next, Dr. Timothy Yap from MD Anderson will provide a clinical program update on our first-in-class PARG inhibitor, IDE161. Next to Dr. Michael White and Dr. Darrin Beaupre, our Chief Medical Officer, will walk through our multipronged strategy and MTAP deletion. Then Dr. Ramon Kemp from GSK will provide an update on Pol Theta Helicase and Werner Helicase, and then I'll provide our closing remarks. And at the end, we will open up the line for the analyst Q&A portion on IDEAYA management. As an organization, we believe there are several major key themes that will drive the precision medicine oncology field in the next decade. And during the R&D day today, we will highlight several of these key strategic pillars that we believe will continue to guide IDEAYA's growth as a leading precision medicine oncology company. First and foremost is our deep commitment to advancing the scientific understanding of first-in-class targets. As evidence of this, IDEAYA has made important scientific contributions and delivered potential first-in-class development candidates against several important precision medicine oncology targets, including MAT2A, PARG, Pol Theta Helicase and Werner Helicase, all of which we will discuss later today. Next is our resolute focus on predictive biomarkers as a central strategy to enrich for the responder population, as demonstrated by our work on key patient selection biomarkers, such as MTAP deletion, HRD, MSI-High among many others. As further demonstrated by the collaboration we announced with Gilead this morning, as IDEAYA's continued strategic focus to enable what we believe are high conviction clinical combinations. IDEAYA has been an industry trailblazer in enabling potential first-in-class combinations, including PKC and cMET with Pfizer, MAT2A-PRMT5 with Amgen, MAT2A and Trop-2 ADC with Gilead, Pol Theta Helicase and PARG with GSK, a Werner Helicase in PD-1 also with GSK. An additional topic we will focus on today is data informatics. We believe the technological revolution occurring in the computational and drug discovery driven by AI and machine learning, will disrupt the small molecule or drug discovery paradigm. IDEAYA has built significant capabilities in this area for several years, which we are excited to share with you today. The vision and strategy that we've been executing on for the past 8.5 years has resulted in what we believe is one of the most exciting and diverse clinical pipelines in precision medicine oncology today with 5 potential first-in-class programs, which includes PKC inhibitor darovasertib, and a Phase III registrational study targeting first-line MUM. MAT2 inhibitor IDE397 in Phase II, targeting MTAP deletion solid tumors; PARG inhibitor IDE161 in Phase I targeting HRD solid tumors; Pol Theta Helicase inhibitor GSK101 and Phase I also targeting HRD-solid tumors and Werner Helicase development candidate and IND-enabling studies targeting MSI-high solid tumors. Beyond that, we have an emerging next-generation portfolio where we are targeting multiple development candidates in 2024, including an MTAP deletion. With that, we'll start the first portion of the presentation with Dr. Michael White, walking us through computational drug discovery at IDEAYA.

Alright, thank you Yujiro. Excited to tell you guys a little bit about our drug discovery efforts. Now, as everybody knows on this call, drug discovery is very hard, fraught with unknowns that can hamper a progress, yet obviously, the need for effective cancer therapies is broad and urgent. And to my mind, a real strength that IDEAYA brings to the table to help address unmet need for oncology is the breadth, quality and pace of the drug discovery enterprise here. Some examples are on the left. All challenging targets, multiple target classes and all potentially first-in-class. Now, part of the IDEAYA enterprise leading molecules like these has been the early and productive adoption of a computer-assisted discovery. We're obviously not a software company or a GPU company. We're a company that seeks to compress time to IND for new chemical entities. And to do that, we've really invested in battle-tested technologies that, in the right hands, can help deliver high-quality molecules with fewer design cycles. In the next few minutes, I'll give you a high-level view of our internal cheminformatics and computational chemistry teams, what they're doing to help prioritize and optimize small molecule designs to get us to the right compound quickly and efficiently. So on the next slide, Yujiro. We'll start with prioritization of compound synthesis. Now, for that, drug-like property prediction has obviously been a widespread use for quite some time as an important approach to triage design ideas that may produce a compound that's in soluble and stable and active, some combination of all 3, partially squares, you can see that as a literature example on the top left is really a conventional statistical method that uses linear regression to assess where new design might fall in chemical property space based on what has been observed with similar looking compounds. Now, limitation of this method for first-in-class drug design, is the need for the user to handpick parameters believed to be correlated with the response based on prior experience, and our team found this can work well if you're starting with a heavily benchmarked model, but it performs poorly once your MedChem team begins to design highly original molecules in order to process and prosecute novel target classes, and we do that all the time. So, to solve for this, we turn to advances in the field of artificial neural networks, in particular, multilayer neural networks that build models, connecting observations to molecular features with no preselection bias. And in our case, this is a feature space with a commentarial complexity across about 2,000 dimensions. And this is in a deep neural network we call HARMONY AML. You can see the results of implementation of this approach on the right. We have substantially reduced the frequency with which we make compounds with unattractive properties across our programs. And I've highlighted a particularly challenging program with respect to avoiding poor solubility labeled Project A on the bar graph, where this method was highly impactful for acceleration of lead optimization compare the red bar before versus after implementation of HARMONY AML. We're making really a lot of nice compounds against this program now. On the next slide, Yujiro, given our focus on structure-based drug design, we've also invested in approaches like free energy perturbation that allow us to model how a small molecule of protein will interact with each other in space over time and, therefore, design molecules that will optimally relax into the protein target to block its function. This is a computationally expensive approach that solves a discrete quantum equation across incredibly high [ commentorial ] complexity of molecular interactions to predict binding affinity. And it's only relatively become practical at scale with the advent of cloud computing. And one of my favorite examples using this approach here at IDEAYA is illustrated here. We arrived at a really potent compound against a first-in-class target labeled IDC-A in the table on the right, but it was greasy, unstable, poorly soluble, hard to move forward. And to get past this road block, hundreds of design ideas from our MedChem team were evaluated by FEP, clustered within 2,000 dimensional chemical features space. And as indicated in the t-SNE plot on the left, this revealed a predicted binder, where we had no previous experience that was potentially within an attractive chemical property space. So we made it and it works. That one labeled, IDC-B giving up only a little potency, and that was corrected with a single modification to give IDC-C. So only 2 molecules made to solve a tough problem. Next slide Yujiro, this last example comes from our new MTAP program that I'll introduce later today. Here is our FEP prediction performance at the beginning, and it really kind of fell off a cliff, which can happen when dealing with something very new. And the solution was parameter optimization via machine learning. This delivered a new model that returned accurate predictions for the subsequent designs for new molecules that have advanced this program at a high pace. And I'll stop here, Yujiro so we can transition to the PARG program.

Great. Thank you, Mike, for that walk through. For the next section, we'll have Dr. Timothy Yap from MD Anderson, walk through IDE161 clinical program update. Dr. Timothy Yap has been on a lead investigator on the study. And I have to say for this year, 161, we've made absolutely tremendous progress. We went from IND filing, IND clearance, initiating Phase 1 as well as a dose expansion all in 2023. So, Dr. Yap with that, we'll head it off to you for the next section.

Thank you very much, Yujiro, and thank you for the invitation to present here in your R&D Day again. Good morning, everyone, and thank you all for being here. I presented the rationale preclinical data and trial design for IDE161 at the last IDEAYA R&D Day. And so, it's a real pleasure for me to now share the early exciting clinical data from the IDE161 Phase I trial and to provide you with an update on the program. So, the slide that you can see here shows how PARG is a mechanistically differentiated target along the DNA repair pathway, which is, of course, a clinically validated pathway, given the success that we have seen with PARG inhibitors. And in the panel on the left, you'll see how PARG activity is required to resolve DNA repair. PARG is essentially an enzyme involved in DNA damage and repair, which helps remove PARG chains once DNA repair is completed as shown in the bottom figures. PARG inhibitors, therefore, prevent the removal of PARG chains from areas of DNA damage and IDE161 is a first-in-class PARG inhibitor. The panel on the right shows how PARG inhibition is mechanistically distinct from PARP inhibitors, which is a critical point as we think about the clinical application and line of sight for IDE161. Firstly, it's key to note that the mechanism of inducing cell death for a PARG inhibitor is a distinct from that of a PARP inhibitor. PARP inhibitors trigger unrestrained fork progression, resulting in a single strand and double-strand DNA gaps and breaks that trigger cell death. In contrast, PARG inhibitors induces fork reversal, which results in nuclearization leading to mitotic catastrophe and subsequent cell death. And therefore, one may expect overlapping and/or distinct cell killing between the 2 different classes of agents, depending on the tumor landscape. Next slide, please. So, here's some background on IDE161, which is the potential first-in-class PARG inhibitor. And as you can see on the left, IDE161 is a potent and selective inhibitor of PARG with an IC50 around 2 nanomolars. And in the bottom row, you can see that treatment of tumor cells with IDE161 results in PARP accumulation, very consistent with its mechanism of action. IDE161 also induces a high independent increase in markers of DNA damage, such as [ H2AX ] ATM, H2AX KAP1 and H2AX RPA. The panel on the right on the slide shows the waterfall plot from a large panel cell line screen, demonstrating that tumors with HRD and other mechanistically related biomarkers was sensitive to IDE161. Next slide, please. This slide shows how IDE161 is active and well tolerated in HRD ER-positive HER2-negative breast cancer models. In the panel on the left, you'll see how in preclinical studies comparing sensitivity between IDE161 and the PARP inhibition niraparib and various tumor cell lines, IDE161 was active in cell lines that were both sensitive and resistant to niraparib. But a few lines were sensitive to PARP inhibitors only consistent with the idea that PARP and PARG inhibitors have distinct mechanisms of action. And in the panel on the right, these in vivo PDX models demonstrate that IDE161 is highly active in tumors that are relatively insensitive to PARP inhibitors. You can see tumor regressions with IDE161 versus niraparib, which leads to only modest growth invasion in BRCA-altered breast cancers. These data is first strongly supported us proceeding to the Phase I clinical trial of IDE161 in the human setting. Next slide, please. And here is the IDE161 Phase I first-in-human trial design. The study started with dose escalation using a BOIN, also known as Bayesian Optimal Interval Design, expansion into ER-positive HER2 of breast cancer, ovarian cancer, both with HRD was planned based on preclinical models. A basket cohort for patients with tumors harboring HRD was also included to identify activity in other cancer types. Also plan was a cohort for dose optimization. Dose escalation proceeded and based on evolving clinical data and expansion quarters opened to further dose optimize and focus on tumors of highest priorities that included not only HRD-positive breast and ovarian cancer, but also prostate, gastric, CRC and endometrial cancer. The breast and ovarian expansion cohorts were also opened to obtain more efficacy data and these indications at a therapeutic dose level. Next slide, please. Here are the PK data of IDE161 at the expansion cohort dose. Based on preclinical data, the exposure seen in the dose expansion are at or higher than the therapeutic exposure expected to be needed in humans. What is interesting, too, is that the patient with the highest exposure had less [ holding ] way of adverse events, while the patient with the lowest exposure had a tumor response with a minus 31% tumor reduction on their second scan. Next slide, please. And here are some preliminary pharmacodynamic data assessing power accumulation in PBMCs in patients treated with IDE161 at the expansion dose. You can see that all subjects in the expansion dose showed evidence of PD [ effects ] and surrogate tissue, providing merging evidence or proof of mechanism. Next slide, please. To provide further evidence of PD activity and patient tissue, the following images are of nuclear high levels assessed by IHC on skin biopsy samples collected pre and on treatment from a patient treated at the IDE161 expansion dose. You can see that nuclear staining with [ IHC ] for PARP positive nuclei showed power accumulation on treatment, as shown by the brown staining consistent with that observed in the PD assessment in PBMCs that I showed on the previous slide. Next slide, please. The preliminary clinical experience of IDE161 shows a manageable safety profile at the expansion dose. The majority of treatment-emergent adverse events were low grade and no drug-related discontinuations or treatment-related SAEs have been reported. Those evaluation is currently ongoing to determine the Phase II expansion dose or doses to move forward. As mentioned, dose expansion was initiated in prior to HRD solid tumor types, including HR-positive HER2-negative breast, ovarian, prostate, colorectal, gastric and endometrial cancers. Next slide, please. This is the preliminary clinical efficacy at the expansion dose of IDE161. We have observed 2 confirmed resist partial responses, when in a patient with metastatic colorectal cancer with a BRCA1 mutation and another in a patient with endometrial cancer harboring both BRCA1 and BRCA2 mutations who also had a GCIG CA-125 response. The observation of these clinical responses is encouraging with more patients coming on to trial with indications of interest based on [ preclude ] data and clinical data from the first few cohorts. Fast Track designation was recently granted for IDE161 and BRCA1 and 2 mutated HR-positive HER2 of breast cancer and ovarian cancer post-PARP inhibitor therapy. Next slide, please. Here is one of the patients who achieved the confirmed RECIST partial response. She is the patient with advanced endometrial cancer who received multiple prior therapies, including carbotaxol, bevacizumab followed by paclitaxel maintenance and then the PD-L1 inhibitor Atezolizumab. The patient had a reduction in a cervical lymph node target lesion as indicated it under scans and a complete response in a nontarget media sinus lesions and overall, a RECIST partial response that was confirmed. Next slide, please. This is the other patient who achieved a confirmed RECIST partial response. This is a patient with advanced colorectal cancer who had previously received FOLFOXIRI chemotherapy. The patient had a minus 31% tumor shrinkage in their pelvic mesenteric lesions as shown on the slide and overall a RECIST response. So, in summary, IDE161, as a first-in-class PARG inhibitor that first far has shown great PK properties as evidence of PD proof of mechanism in surrogate tissue as what appears to be a manageable safety profile and are showing early exciting signals of antitumor activity. Thank you very much for your kind attention, and back to you, Yujiro.

Great. Thank you so much, Tim, for that wonderful walk through. And again, we really appreciate all your contributions to the program. Thank you very much. For the next section, we'll be moving into the topic of emerging therapeutic opportunities for MTAP deleted cancers. Over the last several years, IDEAYA has had several significant efforts in the area of MTAP deletion. In addition to our efforts on IDE397, which is currently in a Phase II expansion as well as clinical combination studies with Amgen, we've had multiple initiatives in the MTAP deletion space spanning multiple programs at preclinical phase, including programs where we're targeting a development candidate in 2024. In addition, we've been evaluating multiple what we believe our first-in-class clinical combination opportunities. And as I stated earlier, we announced on earlier this morning our partnership with Gilead to evaluate the combination with Trodelvy specifically, and MTAP deletion bladder cancer. It gives me great pleasure to introduce our next set of speakers, Dr. Michael White, our Chief Scientific Officer; and Dr. Darren Beaupre, our Chief Medical Officer, that will now walk through this next section.

Thank you, Yujiro. I'd like to begin this section by just reminding everyone why I am tackling cancers, are one of the most exciting precision medicine opportunities out there to address unmet clinical need. And that's because loss of MTAP activity results in 2 distinct targetable vulnerabilities, reduced capacity for the nucleotide synthesis required to support tumor cell DNA replication and repair and reduced capacity for the mRNA splicing activity required to support tumor cell growth and survival. As outlined in the schematics shown here, leveraging the synthetic interactions within this system to kill MTAP tumor is a major focus for IDEAYA. On the left, inhibition of MAT2A with IDE397 creates a double hit on the production of metabolic intermediates required for de novo synthesis, de novo nucleotide synthesis, and on the right, inhibition of MAT2A with IDE397 creates a double hit on the PRMT5 dependent methyl transferase reactions required for spliceosome function. So, in the next few slides, we'll take you through how we're exploiting these mechanistic relationships to bring monotherapy and combination therapies to the clinic that fully leverage the precision medicine opportunities associated with MTAP cancer. On the next slide, let's start with our lead molecule in this space, IDE397, an allosteric inhibitor of MAT2A enzymatic activity. As shown on the upper left, consistent with the synthetic legal hypothesis, the cell biological consequences of MAT2A inhibition are highly dependent on MTAP status. There's at least a 100-fold selectivity window for perturbation survivability in MTAP versus wild-type cells and at least a 1,000-fold selectivity window for perturbation of PRMT5 activity. We know from functional genomics efforts that MAT2A and PRMT5 have highly correlated dependency profiles across 1,000 cancer cell lines. And consistent with IDE397 biochemical activity, the same is true for IDE397 and PRM5T, you can see that in the panels on the lower left. Now, as many of you know, the mechanistic basis of this synthetic breakbulk interaction is accumulation of the MTAP substrate MTA, which is a direct partial antagonist of PRMT5. Now this means that the strength of the synthetic lethal relationship depends on accumulation of cellular MTA, which is a function of both MTA production and MTA clearance via active transport. You can see an example of that for the HCT-116 isogenic pair in the middle panel top row. Now lymph have node tumors, variability in NTA accumulation is a major source of variability of PRMT5 pathway suppression because MTA competes with SAM for binding to PRMT5. Thus, the higher the MTA SAM ratio, the deeper the suppression of the PRMT5 pathway activity. As shown on the upper right, IDE397 through suppression of SAM production can dramatically increase that ratio selectively in MTAP null cells, which corresponds to dose-dependent inhibition of mRNA splicing and MTAP null tumors in vivo. The volcano plots in the upper right show drug-induced intron retention events, the block production of functional proteins. Now, as we've noted in the past, this mechanism delivers a monotherapy opportunity in tumors with optimal NTA accumulation as shown in the squamous lung cancer embodies PDX models on the lower right. And as we noted at the AACR meeting earlier this year, this mechanism delivers a very important opportunity more broadly across MTAP null tumors with variable MTA accumulation when combined with an MTA cooperative PRMT5 inhibitor. More on that on the next slide. As we heard earlier this year, both Amgen and Mirati have established clinical proof-of-concept with MTA cooperative PRMT5 inhibitors and have noted some key mechanistic features of these molecules on the left. All disclosed next-generation PRMT5 inhibitors reinforce the MTA bound inactive confirmation of the PRMT5 active site by filling an open space in the MTA bound pocket as indicated in the top left. They also avoid SAM occupied PRMT5 via charge, charge clash with SAM itself, as shown in feature 2. Finally, some compounds, perhaps most notably the Mirati compound additionally incorporates super cooperativity with MTA through charge mediated attraction to potentially maximize selective inhibition of MTA bound PRMT5. Now, the take-home message here is that combination benefits with IDE397 is observed irrespective of the balance of those features that I've shown on the left. As shown on the right, with the Mirati compound and the Amgen compound, we see durable complete responses with IDE397 combination therapy using exposures that are a fraction of the clinically relevant monotherapy doses in humans. On the far right, I'd like to point out that we're not seeing any preclinical tolerability issues with these combinations, including in a 14-day rodent hematology study using the Mirati preclinical compound MRT9768. On the next slide, the next question is why does this combination work so well? Now, we think there are 2 important mechanisms at play. First, shown in the top row, is the ability to extinguish PRMT5 activity selectively in MTAP null tumors to safely deliver a deep response. As shown in the upper left, using a cellular target engagement assay based on displacement of a fluorescent analog of a first-generation PRMT5 inhibitor, we find that IDE397 enhances the binding of MTA cooperative PRMT5 inhibitors to the target. You can see that as a function of the IDE397-dependent increase in the Mirati compound target binding potency as indicated by a reduction in the area under the dose response curve. This is exactly what you would expect from an IDE397 dependent increase in the MTA [ SAM ] ratio in MTAP null cancer cells summarized in that schematic in the middle. And this translates to amplified PRMT5 pathway suppression in MTAP tumors as indicated by single agent versus combination drug effects on both gene expression and mRNA splicing shown on the far right. Second mechanism in the bottom row is the ability to deliver a durable response. And we find that the low-dose combination produces biological responses in the tumor that are not achieved with the maximum effective monotherapy doses. Notably, IDE397 uniquely delivers pathway modulation that we think contributes to a durable antitumor response by inhibiting the function of adaptive resistance pathways. On the next slide, based on everything I just summarized, we're very excited about our collaboration with Amgen that's evaluating the IDE397-AMG193 combination as a potentially transformative therapy for MTAP null lung cancer. Also, as Yujiro noted, we're continuing to build a multipronged approach to address unmet need across MTAP cancers. And to that end, our abilities conferred by MTAP loss have delivered a new target hypothesis. Chronological evaluation of that opportunity is shown here and indicates the potential for meaningful therapeutic benefit that is mechanistically distinct from PRMT5. We're seeing durable complete responses in a challenging [ entity ] in the setting of MTAP loss and we anticipate a tolerability profile that will combine well with other assets in the MTAP space. Now the next slide, Darrin and I will wrap up this section by turning to our freshly announced collaboration with Gilead to evaluate a Trodelvy-IDE397 combination that we believe will address unmet need for bladder cancer patients harboring MTAP null disease. Now, the mechanism underpinning this opportunity really turns on the combinatorial roles that MTAP and MAT2A play in 1 carbon metabolism and how that can be targeted by IDE397 to disrupt the de novo nucleotide synthesis. As illustrated here, there's a tight coupling of the methionine and folate cycles to replenish nucleotide pools. This is an essential activity in cancer cells to meet the demands of accelerated DNA replication, transcription and DNA damage repair. You can see the relevant endpoints highlighted in those red boxes. Now upon loss of MTAP, MTA can no longer be recycled into adenine and methionine, thereby cutting off resources to that arm appearing synthesis. Furthermore, MAT2A inhibition in this context interferes with SAM dependent generation of homocystine, thereby cutting off the key resource for production of tetrahydrofolate, a metabolic and immediate endopholate cycle for production of both purine and pyrimidines. Now, the coincidence inhibition of PRMT5 makes halves worse from the standpoint of genomic instability due to the replication stress that occurs when mRNA splicing is inhibited. This all creates a perfect storm, if you will, that sets up MTAP null cancer cells to be killed by an IDE, therapeutic index can be maximized by employing an ADC like Trodelvy to deliver the topoisomerase inhibitor in the tumor-selected fashion. And you can see the relevant IDE397-MOA in action on the right, so here, global metabolic profiling of IDE397 drug effects is on the concentration of a number of key cellular metabolites as highlighted in this plot. Now, in particular, we saw a significant depletion of pyrimidines and purines in MTAP null cells that was not observed in the wild-type setting. This effects underpins the strong correlation of cellular sensitivity to IDE397 with sensitivity to antifolate, topoisomerase inhibitors and inhibitors, nucleotide synthesis that we reported at the AACR meeting earlier this year. And it also translates, obviously, into induction of DNA damage in MTAP null tumors by IDE397 in vivo, as shown on the bottom right. Next slide. As Darrin and I will describe in a moment, bladder cancer, as Darrin actually will describe in a moment, bladder cancer is a priority indication for IDE397 in the clinic. And we believe this mechanism that I just reviewed is particularly relevant in that setting. A key example illustrated on the left, it is work from MD Anderson that noted a compelling enrichment of pemetrexed antitumor activity and MTAP null bladder cancer patients. Recall pemetrexed directly interferes with de novo nucleotide synthesis. Preclinically, we see tumor regressions with IDE397 in bladder patient drive xenograft and we see combinatorial benefit with a topoisomerase inhibitor together with IDE397 in a bladder cancer model that's difficult to control with either single agent as shown on the far right. Last slide, for me, I'll turn the floor over to Darrin. And in the interest of time, I'll just close by iterating with the biology outlined in this schematic, underpins our mechanistic conviction that IDE397-Trodelvy combination has a potential to create a synthetic lethal pair that fully capitalizes on the mechanistic vulnerabilities associated with MTAP loss. So Darrin, please?

Thanks a lot, Mike, for providing the background that supports the IDE397 and Trodelvy combination. It's really intriguing to consider the opportunity for these 2 agents to serve as a synthetic lethal pair and tumors associated with MTAP loss. Also and importantly, the fact that Trodelvy allows tumor-specific delivery of a topoisomerase inhibitor provides the opportunity to maximize the therapeutic window with this novel combination. What I'll be doing over the next few slides is providing some of the clinical correlates that suggests that this combination is well positioned to potentially make a significant impact in MTAP-efficient urothelial cancer supported by some of the early efficacy data that we've seen in our IDE397 early phase clinical study, which I will also share. On this particular slide, what I'd like to do is summarize for you some of the key findings from the UNITE study that was presented at ASCO and ASCO GU this past year. What the authors did in this particular retrospective analysis was to review the data from patients with urothelial cancer who were treated with standard of care therapy and determine whether molecular correlates from the subject's tumors could predict patient outcomes. On the left is a data set from approximately 90 patients treated with Trodelvy where the authors correlated a number of biomarkers to treatment response. Interestingly and consistent with some of the biological rationale that Mike shared for Trodelvy in the MTAP deficient setting, patients who were MTAP deficient had a much higher response rate to Trodelvy than those patients who are wild type for that marker. The response rate was 50% in MTAP deficient subjects versus 19% in those subjects who are MTAP wild type. On the right, in contrary to the find being seen with Trodelvy, subjects treated with enfortumab who are MTAP deficient actually had a worse progression-free survival and overall survival relative to those who had wild-type MTAP status. Enfortumab, as you know, is an ADC that targets Nectin-4, which is currently approved for urothelial cancer subjects who have progressed after chemotherapy and checkpoint inhibitor therapy, but also as a relevant treatment option for subjects that are platinum ineligible. The clinical data presented on this slide, however, implies that subjects with MTAP deficiency may benefit more from Trodelvy or Trodelvy based regimen relative to additional information you may not be aware of in that similar findings to that shown for enfortumab have also been reported in urothelial cancer subjects treated with checkpoint inhibitors. This is a study that was published in Nature Communications in 2021, looking at urothelial cancer patients and disease-specific survival after therapy. Interestingly, those patients who are MTAP deficient did significantly worse with Checkpoint inhibitor therapy compared to those subjects who were MTAP proficient. Combined with the information shared with enfortumab, these data once again suggests that urothelial cancer subjects who are MTAP deficient may require a novel treatment approach. Next slide, please. Also highly relevant to the results just shared on the findings shown on this slide from a study that was recently presented at ESMO, evaluating the enfortumab and pembrolizumab combination versus standard of care first-line chemotherapy in advanced urothelial cancer. Here, it was found that EV+P had a better progression-free survival and overall survival versus a platinum-based regimen. And importantly, based on data from this study, many treating physicians feel this is a practice-changing result, implying that EV+P will become the new standard of care for first-line treatment of advanced urothelial cancer. Again, highly relevant to our discussion today is the fact that patients who are MTAP deficient tend to do poorly with either of these agents as compared to those who are MTAP wild type, suggesting that urothelial cancer patients with MTAP deficiency will likely need additional options. Next slide. So you may wonder what proportion of patients with urothelial cancer have MTAP deficiency. And this slide shows that MTAP deficiency occurs in approximately 25% to 30% of subjects with urothelial cancer, and this disease represents one of the most common tumors for which MTAP deficiency is found. Therefore, this is a relevantly sized subject population where additional therapies will likely be needed. Next slide. So, therefore, you may ask, what have we seen in our own early phase trial of IDE397, the IDEAYA-MAT2A inhibitor in subjects with MTAP-deficient urothelial cancer. This slide represents very early data from a small number of urothelial cancer subjects treated at therapeutic doses who were evaluable for disease response. On the left 3 subjects showing evidence of relevant tumor shrinkage. In fact, one subject had a complete response to therapy as a third post baseline scan. And the other 2 subjects who remain on study have shown improvement in their tumor measurements as treatment has continued. And it's our hope that these subjects will achieve a partial response after further continued therapy. Also of importance to note is that all 3 subjects have received prior standard first-line chemotherapy and checkpoint inhibitors and have progressed. On the right is data from 3 subjects who had cell-free DNA available for evaluation. And importantly, once again, all 3 subjects showed evidence of a molecular response post treatment. These clinical findings are encouraging and appear consistent with the concept that IDE397 is itself an active agent in advanced MTAP deficient urothelial cancer. Next slide, please. So, this now sits the stage for the IDE397 and Trodelvy combination, which we plan to evaluate in the context of our current ongoing clinical trial. This is a collaboration with Gilead, where the objective will be to demonstrate a significantly improved response rate and duration of response than that seen with either single agent alone. The target patient population will be subjects with advanced relapsed urothelial cancer with MTAP deficiency. The benchmark for Trodelvy is the patient population that's based on their registration trial, which received accelerated approval, which showed an overall response rate of approximately 27%. And the median duration of response of approximately 7 months. And in this line of therapy and the median progression-free survival of approximately 5 months is expected. As shown in the scheme on the bottom right, the plan is sufficient urothelial cancer, a patient population who generally does poorly with current standard of care agents such as enfortumab and checkpoint inhibitors. And our hope is this could represent a novel opportunity for urothelial cancer subjects with an unmet medical need. And finally, not only is the biology supporting this novel combination intriguing, but the fact that based on the current clinical data, the agents appear to have nonoverlapping toxicity is further support for testing this novel combination. And with that, I'll pass it back to you, Yujiro, for the next speaker.

Great. Thank you, Mike and Darrin, for that traffic walk-through. And again, thank you also to Amgen and obviously, our new collaboration with Gilead to enable what we believe are both first-in-class potential best-in-class combinations in the area of MTAP deletion. Now, going on to our next section. It gives me really tremendous pleasure to introduce Dr. Ramon Kemp. GSK has been a collaborator with IDEAYA for several years. And we just wanted to make several comments before Ramon starts here first. I believe there's been really just terrific strategic alignment between the 2 organizations in terms of pursuing target areas where we believe we have the opportunity to be both first-in-class and deliver tremendous patient benefit. In addition, the last several quarters in our partnership with GSK has been tremendously productive. We recently delivered the development candidate for Werner Helicase and Pol Theta Helicase recently just entered Phase I. As far as we're aware, we believe IDEAYA, GSK, we are, from our understanding, the first group that has delivered back-to-back Helicases in terms of development candidate. And hopefully, we'll have her second here in the clinic shortly. So, with that, Ramon, we're going to pass it to you.

Great. Thank you, Yujiro. It's my pleasure to be here [indiscernible] on behalf of GSK, discuss 2 of the key programs that we have in collaboration. At the outset, I'd say that we have a very strong collaboration, as you've already highlighted, and our partnership with GSK and IDEAYA, there's quite a bit of excitement around these 2 programs that I'll talk about today. Both the polymerase Theta program as well as the Werner helicase programs are both very much aligned with our GSK priorities and strategies that we have, and ambitions we have for precision medicine, developing medicines that are really tailored to specific patient characteristics to ideally enhance the efficacy of our medicines while minimizing toxicity and tolerability concerns. Secondly, they really leverage 2 of our more established medicines that we have within our GSK pipeline for use in combination, notably our GSK-PARP inhibitor, niraparib for a potential combination with polymerase theta program and also our GSK PD-1 inhibitor dostarlimab, potential combinations with Werner Helicase. Next slide, please. So, the hypothesis behind the polymerase data program really focuses on reducing the resistance to PARP inhibitors and therefore, extending the depth and duration of PARP inhibitor response. In this vein, our niraparib PARP inhibitor is well positioned as a combination partner. Key mechanism of resistance to PARP inhibitors is reversion of BRCA mutations, and this BRCA reversion is mediated by a process known as microhomology end joining or MMEJ, the polymerase data mediates this MMEJ repair. Therefore, the inhibition of polymers data prevents the PARP resistance by this mechanism, which is estimated to account for about 3% of PARP inhibitor progresses. The data slides below highlighting 3 different preclinical BRCA mutant models, 2 of which are CDX models and one of which is the patient-derived model, that we not only see inhibition of tumor growth, but also evidence of tumor regression when GSK-101, our polymerase data is combined with niraparib. Next slide, please. And here, I'm simply highlighting our initial first time in human clinical trial, which you've highlighted. We've just started our Phase I dosing. This is a Phase I design in monotherapy as well as in combination and our trials are open right now. Next slide, please. The Werner helicase program is exciting for several reasons. And notably, we believe that we may potentially have the best-in-class mono plus PD-1 combination. Our program centers on a target patient population with MSI-high, dMMR cancers, which, as you know, our patient population is very much aligned with our overall strategy in GSK for dostarlimab. And so, we know this patient population very well. We're very familiar with the patient needs. Our Werner inhibitor is very potent and selective to MSI-high, dMMR sales, as outlined in the Cronos score in the bottom left. Werner is a multi-domain protein and the helicase domain is the best domain to target for activity, as shown in the center panel. And on the right, these drug properties translate to robust antitumor regressions in preclinical animal model experiments as illustrated by the red line as well. Next slide, please. And lastly, our development strategy combined Werner helicase with our PD-1 inhibitor, with tremendous belief that dostarlimab is a top-tier industry-leading PD-1 inhibitor in this regard. As you probably have noted in the past 12 months, dostarlimab has distinguished itself, it's a very effective PD-1 inhibitor in MSI-high tumors. We've reported on our RUBY study in first-line endometrial cancer in combination with chemotherapy as well as our data in rectal cancer neovad adjuvant setting, demonstrating 100% clinical complete responses obviating the need for difficult surgeries. And also in non-small cell lung cancer, the PERL study, which demonstrated robust and comparable results to pembrolizumab in non-small cell lung cancer. So, we believe Werner plus dostarlimab has the potential to be a highly effective combination and potentially best-in-class combination as well. We're currently executing the preclinical IND-enabling studies and plan for IND submission and first-in-human dosing next year. And with this, I'll turn this back over to Yujiro.

Great. Thank you so much, Ramon, for that terrific walk through, and we'll go into now the final session with closing remarks. So thank you to all the presenters, and we're excited about the progress we have made in 2023, and we believe we are positioned for another transformational year in 2024. We have built a world-class organization, a truly diversified first-in-class precision medicine oncology pipeline with multiple high conviction clinical combinations underway and built an industry-leading drug discovery platform that has a demonstrated track record, including 2 first-in-class helicase development candidates. As we look forward into 2024 and beyond, we will focus on the following key investment priorities to further extend IDEAYA's leadership and precision medicine oncology, including computational drug discovery to enhance our IND engine, structural biology to continue to unlock challenging first-in-class targets, neoadjuvant and adjuvant indications that have the opportunity to transform the patient journey and our continued commitment to excellence in translational research and development to both the discovery and advancement of first-in-class targets and transformative combinations. IDEAYA was founded over 8.5 years ago with a singular mission of improving lives through transformative precision medicine, and I truly believe our journey has only just begun. We're now finished with our prepared remarks, and we'll open up the line to the operator for the analyst Q&A portion of our webcast.

Thank you, Yujiro. The first question comes from Anupam Rama at JPMorgan.

Two quick ones for me. Is there anything in the baseline characteristics or maybe any emerging predictive biomarkers for the 161 program from the endometrial and CRC patients that showed responses. My understanding is these tumor types may be potentially less sensitive than, say, ovarian and the subsets of breast. And then second question, what are the ongoing preclinical activities for the Werner's program and the sort of final gating factors to getting that program into the clinic in 2024?

Sure. For the first one, Mike, do you want to start with that one in terms of potential predictive biomarkers in CRC and endometrial for 161?

Yes, Anupam. So, the situation there, we think, is the presence of HRD. You'll know that both of those patients that Darrin described as partial responses, had BRCA mutations. So, that is in frame with what we have been seeing with respect to the role that BRCA plays in replication for stability. And we think we're going to particularly see advantageous activity in diseases where that BRCA status is on top of replication stress and that replication stress is really dependent upon the collateral features that are present in those patient populations, and we're excited about the fact that there might be some enrichment with respect to overlapping HRD status with replication stress in those tumor types. We've shown that in the context preclinically with the ER-positive HER2-negative breast cancers, and I suspect that those mechanisms are in line with what we're seeing in the endometrial and colorectal cancer patients.

Mike, if you don't mind, I would just add one thing to that. When you say insensitive tumors, I presume you mean PARP in sensitive tumors, which they are. As Tim pointed out in the presentation, the mechanism of action of PARG inhibitors is not precisely the same as a PARP inhibitor. And so, there may be an opportunity to be in tumors where PARP inhibitors aren't, which is part of the excitement there. So, that's another thing to keep in mind and that was the purpose of the basket part of the study to look at tumors outside of what you typically see, but fit within the context of what Mike's described.

Great. And then, Anupam, for the second part of your question, in terms of Werner Helicase and sort of I think it was really around any final fitting items from now to the IV filing. Here, I think all we could say is the IND-enabling studies are well underway, and we look forward to give continued updates here as we get into 2024. But as you can see, based obviously on the discussion today, it's full step ahead.

The next question comes from Maury Raycroft at Jefferies.

One of the potential benefits for the 397 plus AMG 193 combos that you can decrease the respective doses of each drug due to synergy between the 2 drugs. What are your expectations for what the 397 plus Trodelvy combo doses could look like based on preclinical modeling? And based on Slide 39, can you talk about how ctDNA response for UC matches with clinical responses?

Great. Yes. Darrin, why don't you take both of those?

Yes. So, listen, I think it's reasonable to think that perhaps the doses may not be sort of the standard doses. But you know what, you have to put the 2 molecules together clinically to understand that better. So, within the context of that study, we'll have the ability to do whatever is necessary for the doses of both Trodelvy and our molecule in order to make sure we have a combination that works. As I mentioned, the fact that we don't have overlapping toxicity, based on what we've seen between the agents we plan to combine, makes it hopefully a lot easier, and we'll be able to allow to potentially give the full dose of both agents. But the expectation is it may be based on, for mechanistic reasons that less will be fine. So that remains to be seen. And then with respect to your other question, yes, some of those molecular responses did indeed correlate with the tumor responses we saw on the left-hand side, the tumor shrinkage.

The next question comes from Gregory Renza at RBC.

Just Yujiro, on the new target opportunity and then tumor, solid tumors. Just curious, what is the new target have to achieve to be either complementary to or differentiated from the current peer approach? Is it faster kinetics, deeper antitumor response, maybe further combo opportunities. I'm just curious how we should think about hearing more on that. And then, just secondly, on the Trodelvy combo study. Just curious just the concept of adding a PRMT5 inhibitor, would that make sense scientifically for a triplet? And then lastly, just I know separately outside item today, Yujiro, I'm just curious if you could provide an update on the [indiscernible] programs.

Sure. So Mike, do you want to take the first 2, and then I'll take the last.

Yes, that was a lot of good questions there. I think the first question, Greg, is, you were asking what do we feel like we really need to do with this new target and how that would play with respect to the IDE397-MTA cooperative PRMT5 inhibitor combination. So, on the one hand, I think what we have here is a brand-new opportunity, and we're not stopping until we really understand how to do the best we can for these patients. So, we are pursuing this as a key opportunity to be able to evaluate how much we can have an impact in the space where we are otherwise perhaps in a situation where we have to rely on the optimal MTA accumulation in these tumors. So this is an exciting opportunity that may be distinct from that particular need. We just don't know yet. But certainly, it's an important thing for us to evaluate. It's a very exciting opportunity for the field, the patients in general that we have this new target opportunity to go after. Now having said that, there also is obviously the possibility for the triple combinations. You mentioned some thoughts about triple combination with respect to Trodelvy. I certainly think that there's going to be a triple combination opportunity with this new target based on the mechanism of action. It is distinct and that's always a good thing. If you're putting it together with other things that are known to have activity in that particular setting. Also based on the mechanism of action, we are not expecting to see overlapping toxicity. And obviously, it's early days, and we'll learn a lot more about that very, very soon. But certainly, that's not off the table. But suffice it to say, Greg, that we just don't know yet how well the IDE397-PRMT5 combination is going to serve the full spectrum of [indiscernible] patients. This is an additional opportunity in that space, and we're pursuing it very aggressively. With respect to your next question, asking about perhaps the mechanistic relevance of combining Trodelvy and PRMT5, with, I believe you're asking also with, MAT2A, Trodelvy and PRMT5 triple combination. That's something from a mechanistic standpoint, Greg, but I guess, it would probably make sense. I mean you're talking about the fact that inhibition of PRMT5 causes inhibition of splicing, which causes RNA polymerase stalling, which in turn causes R-loop formation. R-loops are a big problem with respect to replication stress because they create these transcription replication conflicts during S-phase and this needs to be resolved by Topoisomerase in order to be able to release that before the replication fork comes through and creates havoc. So, from a mechanistic standpoint, that does make sense. And certainly, that's something that I'm sure Darrin and his team, together with Gilead will be looking towards as the evaluation of the combination continues in the clinic.

Yes. The only thing I'd add to that, Mike, is, again, we have to think about the overlapping toxicity. And the MAT2A inhibitor brings the dual-pronged approach, as Mike pointed out, the effect on nucleotide pools and alternative splicing, of course, that could be amplified in the alternative splicing side, perhaps with the PRMT5 inhibitor. But then you have to think about what baggage it brings along in terms of toxicity. So, where we're starting, I think, is a good place to start being open-minded, of course, for the long run is fine. But I think the hope is that as Mike pointed out, the MTAP space is going to be very, very large. And it's possible that a PRMT5 inhibitor and a MAT2A inhibitor isn't going to solve all of our problems in an MTAP deficient setting. So, this is another opportunity to get perhaps some tumors and perhaps it's bladder cancer, where we think this is really the lowest hanging fruit for us for this novel combination, it could be several combinations that are necessary to get many of the tumors or cover most of the tumors that are MTAP deficient. But we're taking our first step in that direction in any case.

Great. And I think the last question was related to, Greg, just to rigosertib, any updates there. I'll be brief. But I think there, it's just full execution mode. As you know, we've announced dosing our first set of patients for the registrational study, teams working very hard on the international site activation. We anticipate 2024 will be a big year for the program, both in terms of enrollment, hitting our enrollment targets for the registrational trial. And then I would say, second, as we noted earlier, around neoadjuvant application, both in terms of a potential clinical efficacy update in 2024 and potential also regulatory update as it relates to what could a potential accelerated approval path look like in that setting.

The next question will come from Charles Zhu at Guggenheim.

I just have a couple across the MAT2A and Trodelvy as well as PARG, regarding MAT2A and Trodelvy. So you mentioned Trodelvy's benchmark of a 28% response rate in about 7 months of DOR. Given that was in all comers and Trodelvy looks like it may have better activity in MTAP deletion, would you have to basically re-baseline the [ ADC ] single-agent activity with respect to determining contribution of components? Or is that the all-comers are valid? And regarding PARG, could you also, I may have missed it, but could you perhaps remind us which Phase I dose you have expanded so far as well as the potential or possibility of expanding future higher doses in context of the Phase I.

Sure. Maybe I'll start with the second one first. So, Charles, we haven't disclosed what the expansion dose is. So, first, I would just highlight that for IDE161, and that's mainly for just future IP reasons. Darrin, you want to just finish off that one with 161 and then we can get into the Trodelvy and kind of what our baseline view as it relates to response rate and contribution components.

Yes. I mean the only thing I'd add on 161 is that we're still in the phase of optimizing the dose. We're expanding out a dose that, as you saw, based on what Tim presented, we think is in the therapeutic range, as pharmacodynamic activity likely to be active. But it's still early days. We don't have a huge safety set to evaluate. So we'll continue to enroll patients at this dose level. And if all works out, maybe we'll stay here. But we're being open-minded to what's necessary to deliver that molecule in the best way possible. And as you saw, there was one patient who had exposure that was much lower than the group and that patient responded. So, we're keeping an eye on where we're at and what the toxicity is, and we'll modify if necessary as we go forward. With respect to the Trodelvy combination, you're right. I mean it's possible that -- to be quite frank, our expectations are fairly high with the combination based on what we saw in the MTAP-deficient setting with Trodelvy. Now admittedly, as you saw, it was a small sample size, retrospective analysis, et cetera. But that all being said, based on mechanistic reasons, we think that the opportunity here is great. And as we establish the dose and schedule for the 2 molecules, if we see the kind of spectacular activity that we hope, based on what we see with both molecules as a monotherapy, we'll be able to make some statements about what the contribution of components is similar to what we've done with darovasertib, right? We had 2 molecules there. We had to come up with a contribution of component argument there. As we talk to the FDA, I think a similar thing will happen here. But really, our expectations are quite high with this combination. And then hopefully, with a relatively small number of patients, they'll be realized. So, it's possible a year from now, we'll be able to tell you a lot about what that combination is doing.

Okay. The next question will come from Yigal Nochomovitz from Citibank.