Exscientia Plc

These documents can be found on our website at www.investors.accentia.ai, along with the 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 Professor Andrew Hopkins, Chief Executive Officer, Dr. Mike Krems, Chief Quantitative Medicine Officer, Dr. Nicholas Krall, EVP Precision Medicine, and Ben Taylor, CFO and Chief Strategy Officer. Dr. Dave Hallett, Chief Scientific 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. The first half of 2023 have seen a major step forward in our development pipeline with four compounds progressing in clinical trials. We recently dosed the first patients in both elucidate our Phase I-II trial of our CDK7 inhibitor, GTA-EXS617, partnered with GTA Perion, and IGNITE, a Phase I-II trial of our A2A receptor antagonist, EXS21506. Earlier this year, two partnered compounds, a PKC fetal inhibitor, in-licensed by Bristol-Myers Squibb for immunology, and a bispecific psychiatric compound designed for sumatoma pharma, also started phase one clinical trials. We have also made significant progress in other parts of our pipeline. Notably, we initiated a prospective observational study called EXCITE-1, evaluating the predictive power of our precision medicine platform in ovarian cancer. The study has the potential to validate further our platform for wider use across a variety of solid tumors. Building a robust pipeline and advancing our wholly owned and partner development candidates is a testament to the strength of Accenture's business model, our capabilities and strategic collaborations. As we continue to bring differentiated compounds to the clinic and to further strengthen our AI-led end-to-end discovery process, we are solidifying our leadership in AI-enabled drug design and development. This summer, We also opened our automation lab outside Oxford here in the UK, which will enable us to integrate AI and automation to drive faster, high-quality experimentation. We are building our own hardware and software solutions to automate a wide range of experimental laboratory processes, including chemical synthesis and biochemical and biophysical screening. We expect all our new capabilities to be online later this year, and we look forward to sharing our progress. The integration of AI with automation to drive ultimately autonomous experimentation is, we believe, the next frontier in improving productivity in drug discovery. I'll speak in a moment about our core capability to adapt and rapidly integrate technological advances into our broad platform. Technology is at the center of our strategy to change the way drugs are invented and developed. To maintain nimble product development in our technology platform, we have promoted three industry leaders to our executive committee. Professor Charlotte Dean is our new chief AI officer. Dr. John Overton is our new chief data officer. And Eileen Jennings-Brown, our chief information officer, all of whom are proven leaders and innovators in their fields. We also appointed Professor Francisca Michael to our board of directors in May. The extensive experience in cancer research using computational and mathematical methods will be invaluable to us as we continue to advance our pipeline and platforms. We remain well capitalized with $509 million in cash at the end of the quarter. This provides us with several years of runway to advance our near-term programs. We look forward to achieving our upcoming milestones and sharing more details of our clinical development plans and progress in the second half of the year. Our computational platform is designed to learn and solve problems that have been too complex for traditional methods. In order to make that process faster and more efficient, we must be able to generate relevant, high-quality, proprietary biological data to inform our models. The wide variety of digital data we generate reflects a complexity of biology. This integrated data is shared across our models to drive system learning and better results. Importantly, our modeling is data agnostic and our generative design technology is model agnostic. This is important as target product profiles we design to are not defined by one data type, such as a protein structure or a high content screen, but by a wide variety of data types. Some of our experimental systems are the first of a kind, such as our precision medicine technology using AI to assess live patient samples to predict patient response. We conducted a prospective clinical trial showing that it was able to improve outcomes in selecting the right drug for a specific patient. Other capabilities are focused on capturing the most data possible from experiments and or integrating it with automation. This gives an advantage in both understanding the model system as well as speed and cost efficiency. The power of our computational platform is in its integration and breadth. There is no single algorithm that defines our operations, but a wide variety of proprietary algorithms and data sets. Our platform is constantly evolving and our capabilities are expanding as we invent and adopt new technology. The overarching technological approach that unifies what we do from discovery to development is model-driven adaptive learning. We pioneered the use of generative AI and active learning to design drugs, and we continue to build on our leadership position in those capabilities. We have now also built thousands of predictive models that can empower our workflows to evaluate novel chemistry in a virtual environment. In addition, we've now built a world class physics based platform that we integrate into our generative and modeling systems for structural analysis. It is a natural next step for us to integrate AI and physics based design methods together. Over the last year, you have now seen how we apply the same methodology of model driven adaptive learning to improve clinical trial design and execution. We utilize clinical trial simulation to better understand the most important variables in clinical trial design. We then create a statistical plan that evaluates results in real time as the trial progresses in order to make better decisions sooner on the clinical trial progress. We've also laid the groundwork to integrate our precision medicine biomarker capabilities into our clinical trial execution in the near future. We believe This can lead to far higher probabilities of success as we're able to better understand which patients will respond to our drugs. Most importantly, we have demonstrated that our platform works. Our AI-based platform has delivered eight development candidates. In other words, novel drugs that are expected to enter clinical trials. The physical properties of these drugs can be clearly measured, and that shows they have achieved complex design goals where traditional methods did not. Importantly, our generative molecule design technology has been core to delivering the eight development candidates so far. Accenture pioneered the use of generative AI and active learning for better drug discovery since we first published our revolutionary approach in the leading scientific journal Nature. The combination of our deep bench of experts in both tech and drug discovery allows us to lead the field with our proprietary algorithms and deep learning models. We have also demonstrated we can rapidly adopt new technologies with customization for our platform as we learn and grow as a company. For example, we believe that physics-based modeling should be integrated into AI-driven systems to take advantage of regenerative power and multi-parameterization optimization that AI allows. In a period of just about a year, we were able to deliver a physics-based system that benchmarks in line with industry leaders. However, because it's purposeful for our systems, we're able to customize it for better results within our own architecture and are even now working on how it can be seamlessly integrated with our generative algorithms. Similarly, we have industry-leading protein modeling capabilities that are optimized for antibodies and other related biologics. This allows us to use generative methods for de-novel biologics design at scale, and we'll tell you more about this program later this year. Before turning over to Mike Crams and Nicholas Kroll, I'll now take a moment to highlight where we are with our clinical and near-clinical programs and the important progress we've made this year. As you know, by next year, Accenture has committed to advancing at least four molecules of meaningful economics into clinical development. We're well on our way to achieving this goal with five programs, either in clinical stage or entering into IND enabling studies as of now. All of which have been designed using our AI platform in much shorter time frames than industry average. And we believe maximum quality using our AI-led discovery platform. both our CDK7 and A2A programs, and now in Phase 1-2 clinical trials with patients enrolling and similar timelines. IGNITE is a Phase 1-2 clinical trial of 546, our A2A receptor antagonist, in combination with anti-PD-1 therapy for renal cell carcinoma and non-small cell lung cancer. ELUCIDATE is evaluating our novel CDK7 inhibitor, 617, for the treatment of advanced solid tumors both as monotherapy and in combination with standard of care. The elucidate phase 1-2 trial will evaluate the safety, efficacy, and pharmacokinetics of 617 across multiple ascending doses in six indications, including head and neck cancers, pancreatic cancer, non-small cell lung cancer, and HR positive HER2 negative breast carcinoma and ovarian cancer. In both the elucidate and IGNITE trials, we use simulation-guided trial design to determine the operating characteristics and adaptive design to evaluate statistical results in real time when the trial is running. Both of these programs demonstrate the true hallmark of an Accenture drug candidate. Precision design compounds using AI and ML combined with novel patient selection strategies with the goal of identifying the right patient for the right drug. Our precision-designed PKC fetal inhibitor, 4318, partnered with Bristol-Myers Squibb, continues to advance through Phase I clinical trials in the United States. This is another example of using our AI-driven technology to design against complex, multi-parameter challenges where others have failed. Our two wholly-owned precision-designed LSD1 and MULT1 inhibitors, 539 and 565, are continuing to progress through IND-enabled studies and we'll share more detail on the clinical plans later this year. I'll now turn over to Dr. Mike Rams, our Chief Quantitative Medicine Officer, to talk a bit about our CDK7 program. Mike. Thank you, Andrew.

Today, we'll highlight more details of our CDK7 program, how we're optimizing our clinical development strategy, and how we choose the patient population that may benefit from this molecule. Additionally, Alongside the evaluation of 617 in the elucidate trial, we will partner with GT-APERON to generate data and correlate data and response to previously collected ex vivo results to substantiate further the value of our precision medicine platform. Here, we highlight the design of the elucidate trial for 617. As Andrew mentioned, we're looking at six different tumor types and we'll be studying 617 both as monotherapy as well as in combination with standard of care. One thing to note is that CDK7 is a broader biological mechanism than A2A. This is why we're including more tumor types and also why we're using our platform differently in this trial. Nikolaos will talk about this more. As with IGNITE, elucidate is also following the principles of model-informed drug development. We started with simulation work to understand the key variables, then structured the phase one trial to see what dosing may be most impactful. We are moving away from traditional 3 plus 3 designs to focus on what we believe will be more informative in learning about the investigational compound. Our precision medicine platform will help us assess the best potential combinations for dose escalation. Importantly, we don't expect to be going into specific subtypes in dose escalation, as a good majority of patients are expected to be responsive to this mechanism of action. Instead, we will retrospectively assess if there are biomarkers that impact the level of response in patients to help inform the Phase II or later clinical development strategy. As I mentioned on the prior slide, CTK7 is a broad mechanism that we believe can apply to a large number of cancer types. On this slide, you can see the six indications we have selected for our Phase I-II trial based on experimental work we have done to date as well as peer-reviewed literature. Highlighted here is the U.S. incidence for these indications, which are all relapsed refractory patient populations. In the U.S. alone, there are 75,000 patients each year that fit the criteria for which we are enrolling in our elucidate trial. One place of potential benefit for this mechanism is within CDK4-6 refractory patients. But through our platform work, we believe it could be much broader and look forward to learning more in the clinic to see which patients may benefit most from CDK7 inhibition. I will now hand the call over to Nicolas Krall, our EVP Precision Medicine, to walk us through how we're using our precision medicine platform with this program.