Exscientia Plc

Hello everyone, my name is Chris and I'll be your conference operator today. At this time, I'd like to welcome everyone to Accenture's business update call for the full year ended 2022. All lines have been placed on mute to prevent any background noise. After the speaker's remarks, there will be a question and answer session. If you'd like to ask a question during this time, simply press star, then the number one on your telephone keypad. To withdraw your question, please press star one again. At this time, I'd like to introduce Sarah Sherman, Vice President of Investor Relations. Sarah, you may begin.

Thank you, Operator. A press release and our 20F were issued this morning with our full year 2022 financial results and business update. These documents can be found on our website at www.investors.accentia.ai, along with a presentation for today's webcast. Before we begin, I'd like to remind you that we may make forward-looking statements on our call. These may include statements about our projected growth, revenue, business models, preclinical and clinical results, and business performance. Actual results may differ materially from those indicated by these statements. Unless required by law, Accenture does not undertake any obligation to update these statements regarding the future or to confirm these statements in relation to actual results. On today's call, I'm joined by Andrew Hopkins, Chief Executive Officer, and Ben Taylor, CFO and Chief Strategy Officer. Dave Hallett, Chief Scientific Officer, Gary Paradue, Chief Technology Officer, and Mike Krems, Chief Quantitative Medicine Officer, will also be available for the Q&A session. And with that, I will now turn the call over to Andrew.

Thank you, Sarah. 2022 was another transformational year for Accenture. We continue to validate our AI-driven precision medicine platform and strengthen our business. Accenture's goal is to fundamentally transform the way our industry designs and develops drugs. We believe our unique pairing of excellent science with advanced computational experimental capabilities at every step of the R&D process differentiates our patient-first precision medicine approach. Our approach of model-driven adaptive learning as an overarching technological principle enables us to innovate from discovery and into development. To that end, our remarkable progress to date is a testament to the strength of our company. We are well capitalized with $611 million in cash at the end of the year. This provides us with several years' runway to advance our near-term programs, deepen in our pipeline, whilst also investing the long-term growth. To that end, we marked several significant milestones throughout the year across our internal and partner programs, which we will provide more insight on today. We will also highlight data that illustrates how we're working on designing better drugs for the right patient by combining precision engineering with personalized medicine. At a high level, last year was significant for the trajectory of our business. We started here by signing a collaboration with Sanofi for up to 15 targets. and end of the year, showcasing the value of our multimodal gene signature data for immuno-oncology patient selection at ESMO-IO. Importantly, we've advanced our pipeline. We highlight five programs across oncology and immunology and inflammation, either in clinical stage or IND-enabled studies. We have presented important data from our design and translational platforms across these programs, and we are pleased we can now share these targets and how they combine precision design with personalized medicine. For our A2A candidate, EXS21546 or 546, last year we reported top-line healthy volunteer data in June. The data confirmed our target product profile design, including portency, high receptor selectivity, and expected low brain exposure, with no CNS adverse events reported. This data provided support to move into our patient trial. Late last year, we also received CTA approval to initiate our Phase 1-2 trial. The IGNITE trial will examine the safety, pharmacokinetics, pharmacodynamics, and efficacy of 546 when used in combination with anti-PD-1 therapy in relapsed refractory renal cell carcinoma and non-small cell lung cancer. During the trial, we will be observationally validating our patient selection biomarker aimed at enriching patients more likely to respond to 546 discovered preclinically. The trial will enroll up to 110 patients and we continue to expect the first patient to be dosed in the first half of this year. A vital component of our A2A program is a robust biomarker strategy, as we believe the key to a successful A2A inhibitor is enriching for the right patients with high adenosine in the tumor microenvironment. We presented relevant data at ESMO Immune Oncology Annual Congress in December, identifying a novel patient selection multi-gene transcript signature the adenosine burden score, or ABS. The biomarker was discovered preclinically using our patient tissue platform and multi-omics dataset integration from complex disease-relevant model systems. The IGNITE trial will evaluate this gene signature to identify patients most likely to respond to 506. We look forward to presenting additional data on biological validation of the ABS and our 506 program at the upcoming AACR meeting next month in Orlando. With our CDK7 inhibitor, GTA-EXS617, developed in partnership with GTA Perion, we remain on track to involve a first patient in our planned Phase 1-2 study in the first half this year. This program also showcases what truly makes Nixenti a drug unique. Precision design aiming to transform patient benefit and patient selection strategies. We highlighted some of this patient selection data at the ENA Congress in October to maximize understanding of the effect of 617 using primary patient material. For the first time, we showed how we integrate machine learning, data from primary human tumor samples, and multi-omic sequencing capabilities to predict tumor efficacy of 617. Using our deep learning AI and high content imaging platform, we've previously confirmed 617 activity in primary human samples Data presented at AACR led us to generally define two groups of patient samples, effectively high and low responder groups, when focusing on ovarian cancer. We believe that leveraging this information will enable us to identify responders and non-responders to 617 across tumor types. This is a key component of how we designed our Phase 1-2 study. We look forward to sharing more detail on the tumor types we'll investigate in shortly. With both our CDK7 and 8-way programs, we can now begin to see the hallmark of what an Accenture drug looks like. AI and machine learning is applied not just in the process of how we design a drug, but also how we identify the right patients for that drug. Accenture invented the first AI-designed drug to ever enter the clinic. Since then, we have made significant advancements in our technology and AI capabilities. We have developed a comprehensive physics-based platform encompassing molecular dynamics and quantum mechanics, which is combined with our AI, generative, and active learning capabilities. We have also significantly progressed our engineering and data platforms, enabling scalability and robustness. As you may have seen, over the past several weeks, we have highlighted three new targets that are progressing. EXS4318, our clinical stage PKC theta compound that was in-licensed by BMS. EXS-74539, our LSD-1 inhibitor, and EXS-73565, our MALT-1 protease inhibitor. These embody what Accenture can do in terms of AI-generated molecular design, where we truly lead the field. These molecules are great examples of how our drug hunters using our AI can solve complex problems such as kinase selectivity in the case of PKC-beta, brain penetration coupled with reversibility in the case of LSD1, and allosteric inhibition in the case of MALT1. I'll now spend a few minutes on each of these important programs, showing you how we designed these compounds and what we have seen to date. In February of this year, Bristol Myers Squibb initiated a first-in-human study of AXS4318, a potential first-in-class selective PKC fetal inhibitor. BMS will oversee clinical and commercial development, and Xcentia is eligible for milestone payments and, if approved, tiered royalties on the net product sales. PKC-FETA is an attractive immune-modulating target, which plays a critical role in the control of T-cell function and is a key driver of several highly common autoimmune diseases, but has proved challenging to dose. The target product profile was particularly challenging due to the need to balance the stain, high levels of target inhibition to drive efficacy, with low daily dosing in humans, PKC-FETA is structurally similar to several related kinases, making it difficult to achieve the high levels of selectivity required to avoid off-target effects. Our team of experts, leveraging our AI design platform, delivered a balanced candidate with potent on-target activity while maintaining high selectivity and a favorable fabric index, as demonstrated in the IND enabling studies. As you can see here, Previous molecules have failed at the challenge to design a candidate with required potency as well as selectivity against other closely related kinases. We believe a well-balanced molecule meets required properties to potentially provide benefit in patients. This molecule is our first immunology and inflammation candidate to enter into a clinic. This is a significant milestone, illustrating Accenture's strength, efficiency, and flexibility to precision design high-quality therapeutic candidates. Last week, we shared an update on our next generation LSD1 and MALT1 inhibitors. I'll highlight a few details, but encourage everyone to visit our website and watch the video which detail these candidates further. 539 is a differentiated lysine demethylase 1 or LSD1 inhibitor, precision designed to improve patient benefit and solve challenging design objectives. It promises strong potential in both hematology and oncology. To date, Other LSD1 inhibitors in development elsewhere have failed to achieve the combination of appropriate pharmacokinetics, good brain penetrance, and reversible mechanism of action. Our candidate has been designed to achieve suitable CNS penetration to target brain metastases common in certain cancer subtypes. In vivo studies have shown favorable activity also in small cell lung cancer xenograft models with dose-dependent inhibition of tumor growth. in vivo studies have also shown a favorable absorption, distribution, metabolism, and excretion profile with shorter predicted human half-life than some LSD1 inhibitors currently in clinical trials. We believe this may benefit on-target tox management, allowing for platelets to recover following dosing given the reversible nature of 539. We believe the exquisite control of LSD1 inhibition and the superior management of platelets will be a critical differentiator for 539 in the clinic, particularly in combination with a standard of care that often has negative effects on platelets. The flexibility to genuinely explore intermittent dosing regimens in the clinic and thus maximize the therapeutic window is another reason we believe that 539 is differentiated from other compounds in development. Here you can see the properties of 539 against two other LSD1 candidates, specifically looking at factors such as CNS penetration, mechanism of action, and predicted, as in the case of 539, or published clinical dosing regimens. As you can see, only 539 achieves a unique combination of a reversible mechanism, suitable CNS penetration to target brain metastases, and a predicted human half-life aligned with once-daily dosing. Our molecule, also met a long list of other criteria such as selectivity against related enzymes, high bioavailability in preclinical species, and in vivo efficacy in relevant models of SCLC, a potential indication where 539 may have benefit. Importantly, we were able to use AI to find this highly differentiated molecule with a specific target product profile ultimately exploring new chemical space. Using our 3D mapping algorithms, we identified targetable features of each region of a protein. We then used our 3D generative AI design algorithms to produce prioritized populations of molecules, meeting specific optimization criteria. Machine learning models efficiently scored the compounds of CNS penetrants, alongside optimizing multiple parameters, including portency and admin properties. By then applying active learning methods, we were able to select the most information-rich molecules to make and test at each design cycle, usually around 10 to 20 compounds. This allowed us to find novel molecules outside the established domain of applicability that were counterintuitive, which enabled us to find a new starting point for design, and covering this chemotype ultimately led to 539. The result was that we were able to create molecules which achieved a target product profile that no competitor molecule had exhibited. We believe 539 has the potential to become the first potent, selective, reversible, and brain-penetrant LSD-1 inhibitor to meet significant unmet need in a range of oncological and hematological indications. We look forward to highlighting the precision design of this compound, as well as the latest in vivo data at the AACR next month. I'll now highlight another candidate we've recently unveiled, our MALT-1 inhibitor EXS-73565 or 565. MALT-1 or mucosa-associated lymphoid tissue lymphoma translocation protein 1, is an important protease target with potential applications in hematology. It aims to inhibit the uncontrolled proliferation of malignant T and B cells in hematological cancers. Accentia's AI-driven precision design approach was able to optimize the safety profile of agents targeting MOT1, whilst also generating potency and selectivity. When considering an optimal target product profile, the team took into account the likely use of a MALT1 inhibitor in combination therapies such as BTK inhibitors. Therefore, in addition to portency, selectivity, and a balanced set of overall properties, we were mindful of potential drug-drug interactions. In these two charts here, you can see that in vivo studies of 565 have shown anti-tumor activity in mouse models and favorable pharmacokinetics both as monotherapy and in combination with abutranib. Toxicology studies have also shown an acceptable therapeutic index with the ability to maintain high levels of potency, selectivity, and safety benchmarks, whilst avoiding meaningful inhibition of UGT1A1, which can lead to excessive levels of bilirubin and is a known cause of drug-drug interactions. This chart here compares 565 directly with published and patented MALT1 scaffolds from various groups. 565 compares very favorably across all parameters, examining potency, cellular activity, and drug-like properties. 565 has very little activity at UGT1A1 and is highly differentiated in this respect. We would predict that many of the other compounds would likely inhibit UGT1A1 to a meaningful degree and thus present challenges in clinical development. It's important to note, in designing this compound, it was the first time that we merged molecular dynamics with AI at Accenture. Molecular dynamics, a physics-based method, is just one of the tools that we have in our tech stack today. In fact, molecular dynamics simulations provided additional insights into critical binding interactions within the allosteric site. Using hotspot analyses allowed us to understand the allosteric binding pocket, highlighting key interactions needed for design. Molecular dynamics then enable us to understand the dynamic motion of the binding pocket and develop a design strategy to improve the potency and broader properties of our compounds. Using data and knowledge of other allosteric multiple inhibitors, our generative design algorithm, Gambit, was used to evolve novel molecules. This resulted in a suite of promising compounds. We believe that 565 can be developed to meet the significant medical need that exists today. through potent and selective MALT1 protease inhibition with the potential for a meaningful safety differentiation. In summary, these compounds demonstrate the potential of the Accenture platform to efficiently deliver precision design compounds that may provide substantial benefit to patients. Better design, we believe, improves the probability of success of reaching patients. And while the average industry timeline to discover a development candidate takes around four and a half years, and synthesizing between 2,500 to 5,000 compounds, it is remarkable that these two candidates were designed in 15 to 20 months respectively, and from synthesizing 344 and 414 precision design compounds. These stats underscore how our AI-driven approach is not only differentiated in terms of precision design, but also faster and more efficient than conventional methods. IND enabling studies are underway for both these inhibitors, and we expect to provide an update on clinical development plans, leveraging Accenture's personalized medicine platform in the second half of 2023. These compounds have potential broad application in oncology and hematology. Overall, we are thrilled with our recent advancements and look forward to sharing more details on our progress. I'll hand over now to Ben to walk through our financials.