Before we begin, I'd like to remind you that today's speakers will make statements during their presentations that include forward-looking statements within the meaning of federal securities laws, which are made pursuant to the Safe Harbor provisions of the Private Securities Litigation Reform Act of 1995. Any statements contained in this meeting that relate to expectations or predictions of future events, results, or performance are forward-looking statements. Actual results may differ materially from those expressed or implied in the forward-looking statements due to a variety of factors. For a list and description of the risks and uncertainties associated with Certara's business, please refer to the Risk Factors section of our Form 10-K filed with the Securities and Exchange Commission on March 15, 2021.
We urge you to consider these factors, and you should be aware that these statements should be considered estimates only and are not a guarantee of future performance. Also, in their remarks or responses to questions, management may mention some non-GAAP financial measures. Reconciliations of Adjusted EBITDA, adjusted net income, and adjusted earnings per share to the most directly comparable GAAP measures are available in the appendix of the slide deck accompanying this meeting and in our most recent earnings press release, which is available on our website. Here is our agenda for today. First, we will start off with opening remarks by our CEO, William Feehery. Then we'll move on to our software solutions, which includes an inside look at the Simcyp Simulator. After a brief break, we will delve into our portfolio of technology-driven services. Then we will provide a financial update and guidance for 2022.
There is dedicated time for Q&A after each section. Please keep it to one question and one follow-up. Let's begin.
We need vaccines and medicines that are safe and effective as soon as possible. People respond differently to vaccines and medicines. Different people may need different doses. One size does not fit all. Should the elderly receive a higher dose of a vaccine since they have weaker immune systems? What is the right dose for children? It may be difficult to answer these questions with conventional clinical trials for ethical and logistical reasons. Enter virtual patients. For two decades, Certara and our partners have been using virtual patients to understand how new vaccines and medicines work in different people before we use them on real humans. First, we create individual virtual patients, each with different characteristics such as age, sex, weight, genetics, diet, and medications. These virtual patients mirror real people like you and me. Using virtual patients, we conduct virtual trials.
We test vaccines, medicines, and combinations of medicines at different doses on virtual patients. These virtual trials help us choose the best dose to treat or prevent diseases like COVID-19 while minimizing side effects. We can conduct a virtual trial of the elderly or of children from newborns to teenagers. We can even create millions of virtual patients and virtual cities to determine the fastest path to herd immunity. From cancer to Parkinson's to now COVID-19, the use of virtual patients is helping to accelerate getting safe and effective vaccines and medicines to all patients.
Hello, I'm William Feehery, I'm CEO of Certara, and on behalf of the company, I have to say we're really thrilled that we could hold our investor day in person in New York City. I wanna thank everybody in the room for coming. To those of you online, thank you for joining us. We appreciate your continued interest and support. I'd also like to start out by recognizing and thanking the Certara team and also Nasdaq, who made this event possible with months of preparation and planning. This picture is from a celebration we held a couple of weeks ago here at Nasdaq for the one-year anniversary of our IPO. It's hard to believe that a year has already gone by.
We held our IPO in the thick of COVID, so we were very limited in the number of people who could attend last year, and so it was wonderful to see a larger group come together this time around. I'm incredibly proud of how resilient the global Certara team has been through this ongoing pandemic, how we've been able to deliver on our customer commitments, perform well against our plan and goals, and do our part in accelerating medicines to patients. The video you just saw centers around how we've been using our biosimulation tools to develop vaccines and therapeutics for COVID-19. We're immensely proud that we've been able to contribute our tools and our expertise and collaborate with 35 partners to advance many therapeutics and vaccines for COVID-19. COVID actually reflects just a small portion of what we do at Certara.
Certara works across almost every therapeutic area modality, reflecting what the pharma industry works on globally. Our mission is to accelerate medicines from oncology to rare diseases by using biosimulation and technology. Certara's history stretches back over 20 years of technology development. We've recently surpassed over 1,100 employees worldwide. Those employees are highly educated, with many leading experts who hold doctorate degrees and have decades of drug development experience. We also have approximately 150 software developers and technologists. Our software and technology-driven services allow us to serve drug developers at all stages of R&D, from discovery and development to regulatory and post-approval success. We have a broad and deep base of more than 1,600 customers worldwide, some of which have worked with us for a decade or more. As a result, Certara has an attractive financial profile with a multi-year record of top-line growth.
We're profitable with EBITDA margins in the mid-30s%, and we convert most of that EBITDA into free cash flow, which we use to invest in our many organic and inorganic growth opportunities. At Certara, we transform traditional drug development using leading-edge software to advance drug programs. Our technologies also underpin the services that we deliver to our biopharmaceutical clients worldwide. Using our proprietary biosimulation tools, we mine insights from data and models to inform critical decisions throughout the R&D process, such as what will be the human response to a drug based on preclinical data, or how will other drugs interfere with our client's new drug, or what's a safe and effective dose for children, pregnant women, or patients with renal failure? We also help our clients with regulatory strategy, writing, and submissions powered by technology to increase quality and speed.
Using value communication tools, incorporating large, complex data sets and economic models, we provide support to expand market access of our clients' crucial therapies. Ultimately, we not only accelerate drug development, but we also help to increase safety and efficacy, improving millions of lives each year. Our suite of solutions brings together the best in technology and expertise to advance drug programs. Biosimulation is the genesis and the core of Certara. Results from biosimulation, which is also known as model-informed drug development, have to be incorporated into the regulatory strategy and submission. Beginning in 2014, we made strategic acquisitions to expand into regulatory science, which has grown to be a formidable business today. Three years ago, we made acquisitions to enter into the commercial arena to offer our clients value assessment, market access, and real-world evidence solutions powered by quantitative technologies.
Due to our position as a global leader in biosimulation, we have significant opportunity for growth worldwide. Within current customers, we continue to deepen our partnerships with new capabilities and tools, as well as cross-sell across our integrated portfolio. While Certara has an impressive customer list of over 1,600 companies, there are actually more than 5,000 companies worldwide with active R&D pipelines. There are new and exciting entrants all the time. This means that there are more than 3,000 companies out there that we don't yet work with. Today, we're excited to share with you some recent developments in our business, customer stories, and how we'll continue to achieve strong growth and performance. Biosimulation is an important tool to improve drug development. For more than 20 years, we've been investing in our biosimulation software. The software is based on complex science and technology.
It's been validated against clinical data, and it's been accepted carefully by the industry and its regulators, all factors which make it extremely challenging to replicate. Our biosimulation technology is a computer simulation of what happens when a dose of a drug is introduced into a human body. In particular, our Simcyp Simulator is a complex model that captures the transport of the drug in the body and its absorption, distribution, metabolism, and excretion. While this model would certainly be interesting if we just modeled one idealized human body, drug developers wanna be able to predict what happens when a drug is introduced into different populations of humans, which consist of individual human bodies that vary in many ways. We've developed inputs to our model that allow us to run it again and again, simulating the differences across many different human subpopulations.
Age, sex, and weight are obvious factors in drug metabolism, but so are factors such as ethnicity, genetic markers, or comorbidities like diabetes, renal failure, or liver disease. Using these subpopulations, we can conduct a computer-based trial with our software by modeling what happens in many different human bodies with different characteristics, and then we predict what happens in an actual clinical trial. Certara's biosimulation software has many different use cases during drug development with significant benefits for our clients and ultimately for patients. For example, in discovery, we can help to identify safety risks earlier, or we could streamline an animal study in the non-clinical stage using virtual animal models. In the clinical phases, we can determine the dosing regimen and the optimal clinical study design.
In some cases, our virtual computer-based studies may be used in lieu of a clinical study altogether, resulting in direct time and cost savings. Hannah Jones is going to delve later into the specifics on this when she goes through the Simcyp Simulator demo. Certara benefits from a significant competitive moat around our technology, which is a result of our decades of development, our wide customer base, the amount of time we spent on validation, and the adoption by regulators, academia, and industry. In software, we continue to experience a very high renewal rate. We have expansions in new business, and particularly in our Simcyp and Phoenix biosimulation platforms. As Patrick Smith and Justin Edge will speak later, our technology-driven services have a very high repeat rate, with biotechs fueling increased demand for our differentiated support and larger biopharmaceutical companies steadily increasing their partnership with us.
Our strong competitive position has helped to expand our global reach, which is a key element of our growth strategy. Last year, 75% of our revenue came from work with North American companies, which reflects the significant drug development R&D spend in the region. As you can see, we've made strides accelerating growth in Europe and Asia Pacific, both with acquisitions and to increase our scientific and commercial footprint in those regions. Expanding our global footprint remains a key priority in our growth strategy. Certara's markets are large and growing. We estimate our relevant markets at $13 billion in 2022. Biosimulation is nearing a $3 billion market, with almost half of that attributed to software. In the next five years, our TAM is expected to continue growing from the low- and mid-teens to $21 billion.
Now we use external research to inform our TAM estimates, and one change in note since we reported on this last year is that the CAGR has actually increased from 15%-16%. There are three major secular tailwinds driving the growth in the TAM for our biosimulation tools and also for our technology-driven services. The first is continued global regulatory support. The U.S. FDA has been paving the way for a number of years by incorporating biosimulation into their regulatory guidances. Rob Aspbury is going to speak more about this shortly. Agencies in other regions, including the EMA, Japan's PMDA, and China's NMPA, have been following suit with their guidances. In the U.S., the FDA has used Certara's biosimulation software since 2006 across 12 different divisions and offices.
All told today, FDA scientists hold more than 400 licenses of Simcyp and Phoenix software. Second is the ongoing biotech surge. During the past few years, the fundraising investment in biotechnology have reached record levels. There was $24 billion invested in 2020 in VC deals, and in the first half of 2021 globally, there was $21 billion invested. Now, biotechs have approximately 80% of the R&D pipeline, and many are taking their drugs to market, which require significant end-to-end support. Lastly, with the pressure to deliver returns on R&D investment and increase efficiency, large pharmaceutical companies are expanding their reach by working with external partners. While we gain many new biotech customers, our long-standing partnerships with the large pharma companies continue to grow in the double digits.
We believe our end-to-end platform and growth strategies position us to further penetrate the rapidly growing technology-enabled pharma R&D market of the future. We've all heard that drug R&D is fraught with long timelines, high cost, and failure rate. The fact is that the drugs we're working on today are significantly different than those that were developed in past decades. In 1995, approximately 85% of the R&D pipeline comprised of small molecule drugs. Today, small molecules are less than 60% of the pipeline, with biologics comprising about 43%. The biologics category is actually quite vast with many different types, and we're seeing significant growth in complex biologics beyond monoclonal antibodies, such as CAR T-cell therapies and gene therapies. The cost of developing a biologic can be double or more than the cost of developing a small molecule.
If you consider the cost of failed attempts in the overall cost of developing a successful drug, the total cost has been estimated as high as $2.5 billion or more. Obviously, plenty of room to improve. There's a lot we can do at Certara to support R&D programs for both small molecules and biologics. For example, in cell and gene therapies, it can be critical to determine an individualized dose for each patient because there may be only one chance to treat that patient. For rare and orphan diseases, where this very small patient population makes it impossible to conduct many clinical trials, it's important to make the maximum use of the data that comes from each trial patient.
For trials in difficult populations, say young children or pregnant women, it's important to have a very good understanding of the dose and the effects before clinical trials are even permitted. You can see how being able to predict how a cell or gene therapy would work in a patient with biosimulation could be extremely valuable to a sponsor. We benefit from a powerful regulatory industry flywheel effect, a virtuous circle that fuels ongoing growth in our business. First, I'll start with innovation and biosimulation as we release new products and capabilities which help to expand use cases and increase penetration across therapeutic areas, as well as different modalities and also to a wider range of patient populations. This year, we've launched 10 new products and releases.
Two weeks ago, we announced the latest release of Simcyp version 21 with enhanced capabilities that are aligned to two recent regulatory guidances, one on pharmacokinetics and the other on oral contraception. Earlier this week, we announced the latest release of Pinnacle 21 Enterprise with a new analytics module and support for the new and updated CDISC standard. Going back to the flywheel next, as we developed new capabilities and work on client projects, we partner with the regulators, holding training sessions and participating in their workshops. There's also a process of developing and refining regulatory guidances. First, regulatory agencies come out with draft guidances, and they solicit feedback before they finalize them. We are actively informing these guidances in with respect to biosimulation and contributing to them. We have a long list of scientific and regulatory milestones in our history, which Hannah will share later.
The U.S. FDA continues to spearhead the development of guidances that incorporate biosimulation across a broad spectrum of use cases and end markets, which provide direction for the industry. In January this year, the FDA came out with a publication on their focus area for regulatory sciences, specifically including the impact of model-informed drug development, which supports and validates our work. They talk about how biosimulation decreases uncertainty and lowers failure rates and is used to develop information that cannot or would not be generated experimentally. The FDA also lists the wide range of applications for model-informed drug development. As we innovate with new capabilities and regulatory agencies come out with new and updated guidances, which incorporate biosimulation, the industry in turn adopts more biosimulation. On the left, you can see the steady growth in scientific publications on biosimulation every year on ScienceDirect.
From PK modeling, which started more than two decades ago, to physiologically based pharmacokinetic modeling or PBPK, which is more advanced biosimulation that Rob Aspbury will speak to shortly. Experts in academia, industry, and regulatory agencies, as well as our experts at Certara, contribute to these scientific publications, sharing detailed use cases. On the right are just a few of the publications that came out in 2021 that included both our Certara experts and biopharma industry experts. Finally, as our clients receive novel drug approvals, these serve as proof points for our solutions, which then further supports greater adoption across the industry and by regulatory agencies. We advance a wide range of drug programs with our solutions, but we are particularly proud of the impact that we've had on novel drug approvals. Since 2014, our customers have received 90% of novel drugs approved by the FDA.
By using our tools and services to inform critical decisions, we help accelerate R&D, we improve safety and efficacy, and we get these crucial therapies across the finish line with regulatory submission support and market access. Not surprisingly, oncology is a major area of our work, but we support just about every therapeutic area, as you can see here. In 2021, we continue to support novel drug programs, and the drugs on the right are just a sample of the ones we've contributed to this year with our software and services. We have a long history at Certara of successful acquisitions, ranging from small bolt-ons to reasonably significant companies, and they've been well-chosen and important to expanding the technology and our product portfolio.
As a public company, we continue to invest both in organic and inorganic growth as we seek to expand our technology platform and relevance to our customers. Take, for example, our Pinnacle 21 acquisition, which closed at the beginning of this quarter. In services, we look for opportunities to scale, especially as biotech demand continues to be robust. We have an incredible, diverse team of leading experts in their field, and we continue to be a magnet for top talent in the industry. The picture you see here is of some of our talented members of our Simcyp team in Sheffield in the U.K. when we opened up the office a couple of months ago. The Sheffield office is actually our largest office in the world with more than 120 employees. Since the beginning of this year, we've achieved 24% growth in our overall Certara team.
This highly educated and qualified team has more than 350 people who hold doctorate degrees and have decades of drug development experience, and they're sought after to present at regulatory workshops and industry conferences. Many of our scientists not only collaborate closely with customers, but they provide their in-depth expertise into product development. We also continue to grow our team of software developers, which increased by 50% this year. Just recently, seven of our scientists were recognized as being in the top 2% based on standardized citation metrics by Elsevier and Stanford University. Actually, two of them are here with us today, Hannah Jones and Patrick Smith. We continue to invest in our people and our culture to ensure that we remain and grow as an employer of choice within this industry.
We thrive on working on the latest technological advances. We have a really wide variety of customer projects, and we see some of the toughest challenges facing drug developers today. We have a proven growth strategy, which is summarized in the five points on this page. As evidenced by our recent product launches and releases, we invest heavily in technology development to expand the uses of biosimulation and deliver the new features that our customers value. 2022, this will continue to be a top priority, and there are exciting advances in biosimulation and clinical development tools that Rob Aspbury and Leif Pedersen will speak to in a moment. We have a strong track record of cross-selling with our existing customers and landing many of the new and exciting entrants to the biotech world as new customers.
We have a history of successful and well-chosen acquisitions, of which 10 included technology or software. Then, as I mentioned earlier, we have a lot of room to expand globally with the increase in biotech investment and regulatory adoption for biosimulation. Finally, but most importantly, we've created a customer-centric, innovative, and inclusive culture to attract some of the best talent in the industry. Speaking of which, we have several of our management team members and experts with us here today, and I'm excited to introduce them. Starting with our CFO, many of you already have met Andrew Schemick, who has been our Chief Financial Officer since joining Certara in 2014. Andrew brings two decades of financial management experience in both public and private equity-backed enterprises across software and services industries.
We also have Rob Asbury, who started with Certara three years ago and is president of the Simcyp division based in Sheffield in the U.K. He has more than 20 years of experience in the pharma industry and held leadership positions at Covance, including as the general manager of Clinical Pharmacology Services. Unfortunately, he wasn't able to travel to the U.S. to, for today's event, so he's joining us virtually from the U.K. Leif Pedersen, standing in the back, joined more than a year ago as president of our software division. He has more than 25 years of experience in software and technology, and he was previously the chief executive officer of BIOVIA at Dassault Systèmes. He's also held leadership positions at Oracle and Siemens. Justin Edge is our president of...
Sitting in the back there is our president of our regulatory and access division, and he joined Certara three years ago. In his 30-year career, he's held leadership roles with a deep commercial focus in healthcare. Previously, he was global head of the healthcare practice at GFK, where he led a large global services team. Patrick Smith, sitting here in the front, is president of our integrated drug development division. He joined Certara in 2016 with the acquisition of d3 medicine, which he co-founded. With 25 years of experience in pharma, Patrick has worked across all phases of drug development and was formerly the U.S. clinical pharmacology head at Roche. He's also a research professor at the University at Buffalo. Hannah Jones, sitting right in front of me, is head of Simcyp Consulting.
While she originally hails from the U.K., Hannah currently resides in Boston, so we're lucky to have her with us today to conduct a demonstration of our leading Simcyp Simulator. Hannah has 20 years of deep modeling and simulation experience and joined Certara more than 2 years ago from Pfizer, where she was executive director with global oversight of all preclinical, large molecule translational science activities. Previously, she was a modeling and simulation scientist at Hoffmann-La Roche. With that, next, I'm going to invite Leif Pedersen up to kick off the software portion of our presentation. Thank you very much.
Thank you, Will. I will kick off our little software section here. Today, Hannah and Rob and I will review the software section of our products. First and foremost, we're pleased with the results for the first nine months of our software business. They are in line with our expectations for both revenue, for bookings, as you can see, and the overall renewal rate for the business. They are also in line with our expectations for year-over-year growth. As Bill already mentioned before, we're very, very proud of the continued product launches, new product shipments that we've done also at the beginning of this year, including our Synchrogenix Writer product. Naturally, a highlight is the R&D 100 award for the COVID-19 vaccine model and the Pinnacle 21 acquisition.
Let me quickly run through a highlight of the software portfolio we have before we actually dig a little bit deeper into Simcyp. We continue to grow our user base of industry-leading software, and we've passed 60,000 users. If you look to the right here of the chart in the yellow, you can see that our regulatory and market access software and our biosimulation software next to it covers a deeper and wider set of capabilities within discovery, clinical, regulatory, and commercial life science domains. Within the market access area, which is over there to the yellow, our BaseCase products continue to ship quarterly releases, and they provide a foundation for biopharmaceutical companies to make critical decisions and optimize their go-to-market efforts.
GlobalSubmit is a product portfolio within the eCTD submission area, and we continue to gain new customers worldwide and compete very well, and we're globalizing our go-to-market efforts. Synchrogenix Writer, as I already mentioned, is a brand-new product that originally was built for use by our internal regulatory writers, and we are now excited to bring this product to market. With the acquisition of Pinnacle 21 and the recent renewals from FDA and the shipments of market-leading capabilities, we're planning to continue the very, very strong growth we're seeing of the Pinnacle 21 products. Within the biosimulation area, the Simcyp product portfolio continues to be the standard for mechanistic biosimulation, and likewise, our Phoenix and CODEX software products has been very widely adopted across the industry with some of the numbers that you can see.
Within the drug discovery space, D360 is a leading scientific informatics software, and we see very, very strong engagement from our customers to implement and adopt the latest scientific findings into the product, for example, to handle their biologics development. We do have an example outside, here in the coffee room, where we illustrate one of our D360 examples. Down at the bottom, you see our Integral Repository, which we shipped for the first time almost two years ago now. It is a SaaS-based product, and we see very, very strong adoption of this, life science compliant data management piece of software. We are continuing to enhance this software that's actually, 21 CFR compliant to serve as the foundation for all of our software product and being able to cover discovery, clinical, regulatory, and commercial software products as I've covered.
As we talked about and as I promised, the first step is to dig a little bit deeper into Simcyp, and Rob will actually from the U.K. guide us through that. I'll let Rob jump into that, and after that, we'll have Hannah come up here and do a demonstration. You want this back?
Thank you, Leif, and good afternoon, everyone. I'm Rob Aspbury. I've been working in the biopharma industry for over 20 years in both operational and financial roles. I joined Certara almost three years ago to run the Simcyp division, and I'm delighted to be presenting to you today. It's just disappointing that I'm unable to be in New York to meet you all in person. I'm going to provide an overview of the Simcyp software. The Simcyp platform is applied throughout the drug development process. In the early stages of development to help screen drug candidates and assess early safety risks, to determine dosing in the transition from non-clinical to clinical, to assess biomarkers of efficacy and to support label claims as part of clinical programs and even post-approval to support label extensions in special populations such as pediatrics.
The models mature as more data about the drug becomes available in a learn and confirm paradigm, which in turn facilitates wider applications of the software, some of which I'm going to talk about in this section. The Simcyp software can be viewed as three subgroups, each of which is used to predict different outcomes related to drug activity. We refer to these software platforms as being mechanistic because once they've been built and validated, they don't rely on the generation of clinical data to predict an outcome. The models are able to use in vitro lab data to predict pharmacokinetic and pharmacodynamic effects based on first principles. They do this by leveraging an enormous amount of underlying scientific data, which is all built into mathematical equations and assumptions within the software.
Firstly, we use physiologically based pharmacokinetics or PBPK to predict the exposure to drugs within the body at both the systemic and an individual organ level. We do this using the Simcyp Simulator, which is a whole body model able to simulate the absorption, distribution, metabolism, and excretion of drugs for different dosing regimens of the molecule. The graphic here depicts the oral absorption of a capsule, but other routes of administration, such as IV, intramuscular, inhalation, and topical applications, can also be modeled. Secondly, we use quantitative systems pharmacology or QSP to predict the changes in cellular systems which are caused by a drug's interaction with its intended target site. Being able to predict these pharmacodynamic effects, i.e., what the drug does to the body, allows us to begin answering key questions related to drug efficacy using virtual patients.
Questions like what's the right dose level or what's our level of confidence that a particular compound or target has the ability to modulate a particular disease. The third piece of the jigsaw, our newest area of research, is quantitative systems toxicology and safety or QSTS, which looks at the off-target effects of drugs. Simplistically, as a drug travels through the body, it will interact with target sites that we want it to and potentially many that we don't, and when the latter occurs, it can result in unintended adverse reactions or side effects. Being able to flag the safety risks associated with a drug early in the development process allows more informed decisions to be made by our clients before moving into the clinic. Together, these three disciplines make up the Simcyp platform, which is being used by biopharma companies and regulators across the world.
Eighteen of the top 20 biopharma companies use the Simcyp platform, as well as 11 regulatory agencies. We've also received multiple grants from the FDA and European bodies to advance the capability of our software, aimed at addressing specific challenges being faced by the industry. We take these grants, and we invest them directly into research, which helps to build our capabilities. A data point that we're very proud of is that Simcyp software is being referenced in over 250 label claims, where biosimulation was used in lieu of conducting a clinical trial. This has directly supported the approval of 85 novel compounds by the FDA. We believe this demonstrates the quality of the scientific knowledge that underpins the platforms we're developing, and it is a key differentiator of what we do.
If you think about it, being able to replace a clinical trial essentially with math is a significant achievement. Biosimulation is complex, but its use is becoming more widespread, and one of the key drivers behind its increased adoption by the biopharma industry is the implementation of regulatory guidance supporting its use, as William mentioned earlier. The number of guidance documents released by global regulatory agencies has accelerated over the last 5 years. What's particularly important relative to Simcyp is that many of the documents issued in those last five years specifically include PBPK recommendations. The FDA is arguably leading the way, but the other regions are following a similar path. For example, the first regulatory guidance addressing the use of PBPK to determine DDIs was released in China as recently as 2020.
In contrast, there is currently very little regulatory guidance covering the application of QSP and QSTS, but we believe that's only a matter of time. These tools are being used today to support submissions, albeit to a lesser extent, and the adoption of them is likely to follow a similar path to PBPK. It's worth noting that we've worked collaboratively with the FDA for many years, and based on number of licenses held, the FDA is actually our biggest user. We continue to invest a lot of time educating the agency on the technical components of Simcyp, advancements made to the software, sharing new use cases, and providing feedback to help inform the guidance where applicable. As the knowledge base continues to grow across the industry, it drives the number of applications where the use of biosimulation is accepted as a validated tool.
This makes the market and opens up new opportunities for biosimulation. In terms of generating new use cases and applications of PBPK, the development of the Simcyp Simulator has benefited enormously from using a very successful consortium model, the so-called Simcyp Consortium. It has 35 members, including 17 of the top 20 pharma, who have been directly involved in helping to guide the development of the platform for many years. The company logos shown here are just a representative sample of the full membership, but this is unique to Simcyp and means the investments we make into the software are tightly aligned to resolving the challenges being faced by our current and our future clients.
We have direct insight into how many of our consortium members are applying PBPK in new and innovative ways to support decision-making, regulatory submissions, and the generation of case studies validating new applications of the Simcyp Simulator. We take that insight, and we use it to make the simulator better for our clients. It's a symbiotic relationship which has been in operation for over 20 years. Interest in PBPK is also growing in the world of academia. The graph on the right shows the number of peer-reviewed scientific publications referencing the use of Simcyp over the last 15 years. The figures clearly show that while the number of publications by Simcyp scientists has increased steadily each year, the number of papers written by independent contributors is increasing exponentially.
This helps the credibility of our software, but also many of these external publications include experimental data which we can incorporate into the simulator and refine the modeling assumptions. It's fair to say that we're a very data-hungry business. This next slide provides additional detail behind the 250+ label claims supported by Simcyp in lieu of running a clinical trial, which I referenced previously. It's data we've shared before, and we continue to update this list as we become aware of more examples. It's a high bar to obtain this level of acceptance from the regulators, but our track record of success continues to grow. A significant proportion of the examples listed here relate to the prediction of drug-drug interactions across a broad range of therapeutic areas, but not all, and the landscape's changing as the science progresses.
This is just to show an example of where Simcyp was used to determine drug-drug interactions for a recently approved drug that's now used in the treatment of lupus. The FDA multidisciplinary review document includes full details of how the simulator was used, the model outputs, and we're just showing a couple of extracts from it here. The impact of all the work involved is really summarized in the prescribing information at the bottom, which references the adjusted dose recommendations for when this drug is taken at the same time as other medications, which then alter its metabolism and change the exposure levels from those observed when the drug is taken in isolation. Essentially, sometimes a higher dose is required for the drug to work. Sometimes a lower dose is needed to avoid safety issues.
Using PBPK in this way not only saves our clients time and money, it avoids the general risks associated with running any clinical trial and the challenges of recruiting large patient groups onto clinical studies. For drug-drug interactions in particular, multiple studies are often required to cover the anticipated risks associated with a particular compound. The benefit of using PBPK can be considerable. Simcyp can also be used to extrapolate predictions into what we call special populations. We generate a PBPK model based on healthy volunteers, validate it with available clinical data, and then extrapolate into other populations of virtual patients, such as pediatrics or geriatrics, where it can often be challenging to collect clinical trial data. This requires advanced population models, and Simcyp has market-leading capability in this area.
There are drugs approved for use in pediatrics today based on data generated from Simcyp. In many scenarios, replacing a clinical trial in its entirety is not currently a realistic option. Simcyp also has the potential to reduce the number of real patients needed in a particular study. For example, where certain patients suffering with impaired renal or hepatic function are required, predictive models are developed with virtual patients and then validated using clinical trial data collected from a small number of real patients. Large amounts of virtual patient data can be generated and then combined with the real clinical data to generate the statistical power required by the study design, which ultimately saves clients time and money because they need to recruit fewer patients. The final area I wanted to highlight in relation to PBPK is complex generics.
For a generic drug to be approved, it must be shown to be bioequivalent to the reference or innovator compound. This can be difficult when systemic exposure to the drug can't easily be measured simply by taking blood samples or when the trials need to be run in challenging patient groups. The added complexity creates additional risks for generics companies and disincentivizes the development of generic alternatives to many of the off-patent drugs currently on the market today. Regulators are keen to find new ways of supporting their development, and PBPK can definitely be part of the solution. As an example, the FDA has frequently cited the approval of generic diclofenac using virtual bioequivalence as a success story.
In this case, a generic to the marketed drug Voltaren, which is applied to the skin and intended only to act locally to relieve pain, was approved using Simcyp PBPK modeling in lieu of a clinical trial. We have a poster in the room today highlighting this same example. The use of PBPK to demonstrate bioequivalence virtually is an area of increasing interest and one which we're actively involved in. Switching gears from PBPK, the other two emerging technologies being developed in the Simcyp platform are QSP and QSTS. As I mentioned previously, QSP examines what happens after a drug candidate interacts with a target site and predicts its ability to modulate a disease. It relies on incorporating a deep understanding of biological systems into computational software.
Our goal is to generate products which can be applied across specific therapeutic areas or a particular modality. To achieve this, we're currently focusing on the key areas highlighted on this slide. For example, our immunogenicity platform predicts the body's immune response to large molecule drugs. Whenever a large molecule is injected into a human, there will be an immunological response to some degree, whereby the body produces anti-drug antibodies which can reduce the effectiveness of the drug. Understanding that reaction in a virtual environment using different populations and asking different questions can help inform decision-making in the design of drug trials. For example, implementing a different dose level or dosing regimen can help minimize the negative effect of the body's immune response on the drug. Obviously, testing what if scenarios in a virtual space is potentially a lot quicker than doing it for real.
Interestingly, we recently adapted our immunogenicity model to successfully predict the immunogenic response to several COVID vaccines. In this case, assessing how to maximize the immune response rather than trying to reduce it. We continue to work with several clients on their COVID vaccine programs using this model. In many ways, QSP is viewed as the logical next step for mechanistic biosimulation. Most large pharma companies are investing in this area, but it isn't an easy thing to do. Our immunogenicity and immuno-oncology platforms have been in development for almost five years now and are our most advanced models. There's also demand developing in many other therapeutic areas, such as gene therapy and rare diseases. Finally, a comment on our first QSTS tool. Most drugs don't just interact with their intended targets.
They also bind to other sites in the body, and those unintended reactions can cause side effects. To assess the level of inherent risk, most pharma companies screen their compounds of interest against a broad range of off-target receptors using a specialist vendor. Interpreting the results that this screening process produces is often subjective and can be inconsistent between individuals. To address this, we recently developed a piece of software called Secondary Intelligence. It's designed to streamline the analysis of the safety data and to facilitate consistent decision-making. It can be used both prospectively to predict outcomes in non-clinical and clinical studies, or retrospectively to help explain findings in studies which have already been conducted. The need to understand safety pharmacology is not unique to pharma.
It's important in some adjacent industries as well, the chemical, agrochemical, and cosmetic industries, for instance, all of which use receptor binding assessments to determine the risk profiles of their products. These industries are also faced with restrictions on their ability to use animals in testing, so the use of in silico solutions may well be attractive to them in the future. Hopefully, that helped provide a high-level overview of the Simcyp platform. I now hand you over to my colleague, Hannah Jones, our VP of Consultancy for the Simcyp business, who's going to walk you through a more detailed demonstration of the Simcyp simulator.
Thank you, Rob. I'm Hannah Jones. I lead the consultancy group within the Simcyp division of Certara. I've been in the pharma industry nearly 20 years in various translational modeling roles, mainly in large pharma, but more recently at Certara. Before getting into the Simcyp Simulator demo, let me first walk you through some of its capabilities and how we use it to explore a range of applications. Typically, before running a clinical trial, we want to understand the pharmacokinetics of the drug. What we mean here is what the body does to the drug, and in particular, how it is absorbed, distributed, metabolized, and excreted over time. This is important as it allows us to predict the exposure of drug in the body relative to its efficacy and safety profile, and ultimately, to understand dose and dosing regimen.
More specifically, we can predict blood and tissue concentration time profiles in virtual patients using the mechanistic, physiologically based pharmacokinetic model within Simcyp. We do this by integrating population data, compound data, and information on trial design. Simcyp is applied across different stages of the drug discovery and development process. In discovery and non-clinical stages, we can use Simcyp models iteratively to understand pharmacokinetic risks, to guide clinical candidate selection, to predict pharmacokinetics in animals to inform dose selection for toxicology studies, and ultimately, to predict pharmacokinetics in humans to help select the starting dose for the first in-human trial. Once clinical data are available, the model can be refined as needed and used to guide the next study. The model can then be used to predict drug interactions, food effects, the effects of different formulations, and potentially differences between healthy volunteers and patients.
At this stage, Simcyp is used to guide the clinical pharmacology plan and in particular, which studies need to be performed. In certain cases, the model can be used in lieu of clinical studies. At later stages, the model can be used to extrapolate and predict scenarios that are either logistically or ethically challenging to perform, such as studies in children and organ impairment patients. For many of the drugs on the slide that Rob shared, Simcyp has been used beyond the initial approval to support prescribing to additional groups. It is a learn and confirm cycle. As you move across to the right, you generate more data, which increases the confidence in the models, the range of applications, and the ability to answer additional questions. Shortly in the demo, I will show a case study with ibrutinib.
In this case study, Simcyp modeling was used to reduce the number of clinical pharmacology studies required and to inform the drug label. Pharmacyclics and J&J sought to bring ibrutinib, known by the trade name Imbruvica, a small molecule for cancer targeting rare B-cell malignancies, to the market. Typically, cancer patients will take multiple medications already, so it is important that we understand the potential interactions of these drugs with any new medication, such as ibrutinib. Ibrutinib is metabolized by an enzyme called CYP3A4. Ibrutinib is therefore susceptible to interactions with drugs that inhibit or induce this enzyme. This can lead to higher or lower exposure of drug in the body, which may impact its safety and efficacy profile. To understand this further, a Simcyp model for ibrutinib was developed in healthy adults using in vitro lab and clinical data.
The model was then verified using clinical data with known inhibitors and inducers of CYP3A4. The model was then applied to untested clinical drug interaction scenarios in cancer patients to establish risk profiles. Now I will go into the Simcyp simulator demo to show you how we can do this. Before jumping into the case study, I will first show you a few key features of the simulator. As I mentioned earlier, we have three key components to the Simcyp model: the population data, the drug data, and trial design information. Starting with the population data, I have it set on healthy volunteers. As you may already know, drugs are typically first tested in healthy volunteers, so we use this population to predict the first in-human dose for a new drug.
Clearly, healthy volunteers are not representative of wider society, so we've developed populations that enable us to extrapolate to other patient groups, such as cancer patients and children. Each population is defined by a range of physiological factors that affect the exposure of the drug in the body and consequently dose. For example, on the Demographics tab, we define the age range, sex, weight, and height for this population. We can also consider genetic differences. These play an important role in setting the right dose for each patient. Reminding you of the importance of CYP enzymes, we can evaluate how patients that are metabolized by these differently will react to the new drug. For example, in this healthy volunteer population, 8% are poor metabolizers of the CYP2D6 enzyme.
For a drug metabolized by this enzyme, this results in higher drug levels in the body and the possible requirement for a lower dose. Now I'm gonna take you through some of the sub-models in Simcyp. These allow us to confidently predict how the drug will perform in key organs. Here in the liver model, we define the abundance of the different drug-metabolizing enzymes. These are used together with in vitro data to determine the role the different enzymes play in the clearance of the drug. This helps define drug interaction risk factors. The gastrointestinal tract model is important as it allows us to predict the rate and extent of absorption of the drug. We can do this under fasted and fed conditions.
As you may know, this can be important as when you get a drug from your doctor, you are sometimes asked to take it with or without food. Important parameters to define these potential differences include the gastric residence time, pH, and bile salt concentration. Moving to the Tissue Composition tab, we don't just simulate the exposure of drug in the blood, but we also simulate drug concentration in different organs in the body. This allows us to assess issues related to both efficacy and toxicity. Taking biopsies is an invasive procedure, so the concentration of drugs in human organs can really only be explored via simulation. On this page, we show the water, lipid, and protein content of different organs. Bringing in lab data, we can predict how drugs will partition into these different tissues.
We also have highly complex mechanistic models for the skin, the brain, lung, and even for tumors. We can use the brain model to predict whether a drug will cross the blood-brain barrier, and we can use the lung model to predict exposure of drug in particular regions of the lung, and we have done this to support COVID therapies. Moving to pediatrics, we recognize that children need to take medications, but including them in clinical trials is challenging. Children are physiologically not just small adults, so you can't just scale the dose. Simcyp has developed a unique pediatric model that incorporates parameters related to organ growth and drug-metabolizing enzyme expression. These enzymes are low in babies and take time to reach adult levels. These factors and others affect the amount of drug the baby is exposed to and consequently the dose requirements.
Given that it is very difficult to perform clinical trials with not only young children, but with pregnant or lactating women, the impact of such modeling approaches can be really informative. To address how drugs will perform in each of these populations, we have also created a pregnancy and breast milk model. We can use these together with our pediatric model to explore the exposure of drug in the pregnant mother, the fetus, as well as drug concentrations that the baby may be exposed to, either via breast milk or via regular drug administration. The Simcyp Simulator has allowed us to move from the one-size-fits-all drug dose from years ago to stratifying populations by age and other demographic factors to smaller patient groups such as the renally impaired, and ultimately to personalized dosing.
We can simulate populations of virtual patients by accounting for the distribution of these different demographical, physiological, and genetic parameters for each population. We also consider the covariations between these different parameters to simulate realistic populations of virtual patients. The Simcyp engine, which is behind this interface, represents more than 20 years of data curation. The model has evolved over this time period with leading pharma companies, academic institutions, and global regulators. I've highlighted some of the key population inputs, but I've really only touched the surface. The Simcyp engine behind this interface consists of thousands of differential equations and millions of lines of model code. This is why the Simcyp Simulator is so unique. The complexity is hidden behind the user interface. Now I'll move on to the drug-specific data input and describe these in the context of ibrutinib.
Remember, the overall objective here is to integrate drug, population, and trial data to predict outcomes in virtual populations. The input parameters for ibrutinib are used to characterize its absorption, distribution, metabolism, and excretion properties. We start by adding some basic in vitro data relating to its physicochemical and binding properties. These, together with information on solubility, permeability, and metabolism, allow us to predict how it is absorbed, distributed, and metabolized. Looking more specifically at the Absorption tab, ibrutinib is administered as a tablet, but we can also simulate this part of the simulator to refine the formulation of a drug, for example, moving from a tablet to a capsule. On the Elimination tab, we input in vitro data relating to the metabolism of ibrutinib. These data help inform drug interaction risks.
Ibrutinib is mainly metabolized by CYP3A4, but other clearance mechanisms are involved, and these are shown here as inputs. Our next goal is to simulate the clinical trials for ibrutinib. Within the Simcyp Simulator, we can simulate as many trials and include as many virtual patients as we want, including looking at different age ranges or the percentage of males and females. We then select the route of administration, the dose, and whether we want to simulate a single or multiple doses. We then run the simulation by clicking the green arrow. In this case, Simcyp will predict the exposure of ibrutinib in healthy volunteers after a single oral dose. The results are output into Excel. First, I will show you some examples of the kind of things that we can see.
On the first page of the output, we have a summary of the trial design, the compound parameters, and the results. You can also view the demographic data and CYP enzyme levels for each virtual patient that's been simulated. Now to the key results. Here you can see the predicted concentration of ibrutinib in the body with time on the X-axis and concentration on the Y-axis. The simulated concentrations compare well to the observed data. On another sheet, you can visualize the predicted pharmacokinetic parameters and how they vary across the different simulated trials. Finally, you can also pick out particular virtual patients to help understand extremes and risks. For example, virtual patient nine has a higher exposure than some of the others. Looking back at the enzyme tab, this virtual patient has a low CYP3A4 level.
We repeat this step to confirm that we can capture the observed ibrutinib exposure in other studies with other populations and dosing regimens. We predict scenarios that have not been studied. Recall that ibrutinib is metabolized by CYP3A4. We observed a large drug interaction with ketoconazole, a strong CYP3A4 inhibitor. What happens with other CYP3A4 inhibitors? To explore this further, we can update the population to a cancer population and include an inhibitor model. In this case, we are using diltiazem as a moderate inhibitor drug. Diltiazem is one of more than 100 drug models we've created within the simulator. On the Interaction tab, we describe the inhibitory potency of diltiazem against the CYP3A4 enzyme, along with its metabolite that also inhibits CYP3A4. Next, we set up the trial design.
Here we add the dose and dosing regimen for ibrutinib, but also for the co-medication diltiazem, and then we run the simulation. We can view the simulated exposure of ibrutinib in the body in the absence and presence of diltiazem. As you can see, the predicted exposure is about 5-6 times higher when co-administered with diltiazem. This represents a strong drug interaction. As our goal is to determine appropriate dosing recommendations for different patient groups, we can view the degree of interaction predicted across the different trials. This allows us to suggest alternative doses in the presence of diltiazem. The cadence is to select a population, run the simulations to determine the drug interaction risk, and then we can do this with as many other drugs and populations as required. Moving beyond drug interactions, we can take this model and explore other scenarios.
These may include looking at exposure in different populations across different age groups, exploring sex differences, looking at different doses and dosing frequencies, and large versus small trials. The choices are endless. One Simcyp Simulator and many different submodels. Just moving back to the slide deck. Here we summarize all of the results of the ibrutinib model for the known and other untested CYP3A4 inhibitors and inducers, which were used by the sponsor and accepted by the FDA for the final drug label. On the Y-axis, we have the different interacting drugs. On the X-axis, we have the magnitude of interaction. On the left, we have the inducers. These drugs reduce the ibrutinib exposure and may impact efficacy. On the right, we have inhibitors. These drugs increase ibrutinib exposure and may impact safety. To avoid this, we may need to adjust the dose of ibrutinib.
The blue boxes in the corner show the clinical results for a strong inhibitor and strong inducer. In the middle we have the simulations for the untested scenarios. Essentially, we've created a yardstick that allows the clinician to provide dosing instructions based on patient type and other drugs, the patient may be taking. Based on these results, the Simcyp model, together with relevant efficacy and safety data, was able to inform the ibrutinib drug label as shown here, and provided guidance to clinicians around dose adjustments in lieu of clinical drug interaction studies. Ibrutinib was recently approved by the FDA for its eleventh indication. The use of the Simcyp simulator to predict drug interactions, inform drug labels, and eliminate the need for clinical trials was once quite novel, but today is an expected and encouraged approach. Regulators have used the ibrutinib example in several best practice presentations.
Consequently, models using Simcyp are now routinely used and developed to simulate drug interactions. The use of the Simcyp Simulator has informed 85 drug labels and facilitated more than 250 label claims in lieu of clinical studies. This is just the beginning. There are now a number of evolving applications, including special population predictions, bridging formulations, and many other what if scenarios. Finally, the application of the Simcyp Simulator and other mechanistic modeling tools, such as quantitative systems pharmacology and quantitative systems toxicology and safety, allows us to streamline the clinical development process of drug programs, which accelerates drug approvals and gets life-changing medicines to patients faster. Adoption of these approaches by industry and global regulators is accelerating, paving the way for a bright future. Now I'll hand you back over to Leif. Thank you.
Thank you so much. Good. Thank you for doing that demonstration of the Simcyp Simulator. Naturally, the beauty of these mechanistic biosimulation models and the simulator and QSP is their ability to actually provide predictive value significantly before you even get into clinical studies. There is a very large and untapped opportunity within the clinical space itself, where more than half of the money is being spent in human clinical trials. I'll take a second here to look into where Certara is specifically heading in this area and what we cover. Let's look a little bit at this workflow and how this scientific workflow works and what Certara can actually do and how we can help resolve pain points.
Today, when clinical studies are started for a new promising candidate, biopharmaceutical companies are very, very quickly receiving a lot of data from many different sources, as you can see here on this diagram. This might be clinical trials using the clinical trial software of the biopharmaceutical that they have in place in-house, or it might be receiving data from multiple different CROs that has their own software systems, or it might even be hospital or manual systems, for example, in third-world countries. These ever-growing number of data sources needs to be brought together, and they need to be validated. This effort today is largely manual, and it lacks a very systemic governance, and they are the foundation for the simulation that we just saw.
It is important for biopharmaceutical companies that this ga-data gets to the highest level of standardization and validation across all of these data sources as quickly as possible, so that different scientific roles can actually take that data and use it for a vast amount of decision-making related to drug dosing, as we just saw, the potential optimization of clinical trials, and also identify and figure out how to handle potential side effects. As part of this entire process, you naturally want to see if it's possible to use virtual patients. You know, what do you know from earlier clinical trials? What's publicly available? Today, biopharmaceutical companies have a lot of opportunities to optimize within this space. Also illustrated with the demonstration that you just saw a minute ago.
Last but not least, all of this data that you see here in the first little bubble on the diagram here needs to be managed. The analysis, the conclusions, the patient narratives, and everything eventually needs to be submitted into a submission of the highest quality, free of errors, and consistent to achieve the highest possible approval, and secondarily, make it easy to make very vast amount of information publicly available. Today, also, this work is very manual, very error-prone, and very often very inconsistent. Let's take a look at how Certara can help through this clinical workflow process. Here are some of those data points that I just walked through, actually outlined here, and it's the material that you have available to you as well.
Within this data standardization and compliance workflow, we offer Pinnacle 21, and we offer the Integral Repository. These products help ease the management of information as earlier discussed, and it also helps manage the actual integration and management of these data in a 21 CFR Part 11 compliant fashion. The next step when you've got all this is to actually standardize and validate this data through a set of state-of-the-art application to eliminate these manual errors, the quality of the data, and the lack of the consistency of data. The final step here in this first process is to actually prepare the clinical data for biosimulation and further analysis, which is helped with different data mapping algorithms and preparation applications. This is often referred to as CDISC, SDTM, and ADaM data set conversion.
Let me mention a couple of points specifically in this situation related to Pinnacle 21. The U.S. FDA and the Japanese PMDA leverages Pinnacle 21 Enterprise to validate all clinical data submissions in the world. Also, 22 of the 25 top biopharmaceutical companies by R&D spend and six of the top 10 CROs leverages Pinnacle 21 Enterprise daily to optimize processes that we have just reviewed. As already mentioned, the combination of Certara's and Pinnacle's 21 software asset makes it possible to leverage this Integral Repository. Last but not least, we are very excited to actually roll out the services and different technology-driven services to help the adoption of this clinical data management platform. If we go back to our little workflow. No, let me actually show a little graph here. The
What you see here on this is that, the impact. Oops, that went a little bit fast. Okay. The graph that we just saw before actually illustrates the impact that it has to actually use these type of tools to help remove these errors and apply that consistency, to the data and how it actually lifts those, and you saw that on the different graphs. After the completion of this data standardization and compliance, we actually move into the actual data analysis and the biosimulation, and this is where the data scientists will start to perform and make these important decisions within the preclinical and the clinical phase I to IV using these biosimulation and data analysis. We've already done a comprehensive review of this that was presented by Rob and demonstrated by Hannah.
What I will quickly do is jump in and talk a little bit specifically about the Phoenix PK/PD software. Our Phoenix platform is the go-to market for non-compartmental analysis PK/PD and toxicokinetic modeling used by more than 13,000 users globally. This Phoenix platform consists of multiple integrated modules that you can see here to the left, that uniquely manages the entire empirical biosimulation workflow and include all needed aspects of data processing, graphical tabulation, and report generation. Within this biosimulation workflow is not only important to perform the actual analysis, you need to validate all incoming data and convert and perform the different points of analysis and make it possible to manage iterations of the work. If you're in a situation in the future where you need to return to this work, it's actually important that you can also manage this.
Phoenix is being used by key regulatory agencies, as already mentioned, and by far the majority of all biopharmaceutical companies today. For the life sciences industry, it's important to work with a validated software platform that can be trusted and where there's alignment with the regulatory agencies in place for these software pieces. The very last step that we want to look at here is the actual regulatory preparation and submission. Within this step, the biopharmaceutical company needs to manage all the data that we've earlier talked about, the analysis, the conclusions, the patient narratives, and so forth, into a regulatory submission that must be of the highest quality, must be free of errors, and be consistent to achieve the fastest possible approval.
As mentioned before, this work is very labor-intensive, and today it is very manual, and there's often very little technology or advanced technology uses for this. Therefore, this phase is very error-prone and often very inconsistent as well. Certara offers our Synchrogenix portfolio of software to help automate, simplify, reduce errors, and leveraging state-of-the-art machine learning and artificial intelligence that have been inserted into a set of applications that are very easy to use, like the Synchrogenix Writer that you see here, and where you can easily, for example, create consistent patient narratives, that helps reduce errors by leveraging template models.
If you look at this in summary, right, the clinical data workflow is very comprehensive and it's important that you understand the cyclical nature of this and how we can actually help reduce these very complicated steps that are many of and reduce their complexity and risk. The last little thing I wanted to do here is to kind of like make it a little bit more complex. This is an example from a top biopharmaceutical company, and it kind of like makes the example I gave a little bit for way more complex, but it illustrates exactly the same point, the ability to actually prepare the data, the ability to analyze the data, and the ability to consume and actually make the decisions you need. With that, I wanted to thank you.
This was the overview for the Simcyp and for the software pieces of our software from Hannah and Rob and I. With that, I think we go over to the Q&A.
Can you go to the next slide?
If I can go? Of course, I can go to the next slide.
We're gonna bring Rob back in here.
Bring the guy from the U.K. back online here.
We're gonna bring Rob in a second. Should we take questions?
There we go. Oh, here we are.
Here's Rob.
Hey, Rob.
Great. Can you hear me?
He can hear us when we're miked.
Okay.
Any questions from the audience?
If you haven't met Rob before, by the way, he's taller than I am. You have no feel for it.
You want me to start?
Yes, please start.
Hey, it's John Kreger from William Blair. Question about the demo. Can you just help us better understand the data feedback loop? You have these predictions that you're making, and it seems like you're comparing it to observed data. How do you do that? Are you getting that 100% of the time? Do you sort of rely upon your clients to sort of push that data back to you?
Why don't we let you-
Maybe Rob, I can take this one.
Yeah.
Typically we start off by developing the model with one set of clinical data, and we use in vitro data to feed into the model. Assuming that the model is able to capture that first set of clinical data, we then look at other pieces of clinical data and verify the model. We perhaps start off with the first clinical study that is being performed, maybe the single ascending dose, check that we can predict pharmacokinetics for the first few doses, and then use the model that we've developed to then predict other PK studies. For example, the multiple ascending dose, maybe an initial interaction study that's being performed or some organ impairment studies.
Should we think about the model accuracy as dramatically different today versus 5 or 10 years ago? Or once you sort of, kind of lock it in, is it fairly stable over time?
Obviously we've generated more understanding over the years in terms of the data that goes into the model. Things like hepatic blood flow, for example, in hepatically impaired subjects or renal function in renally impaired subjects. Yeah, the data that goes into the models has improved dramatically over the last five years, 10 years, which has enabled us to kind of develop more populations and predict other kind of scenarios that we weren't able to do in years gone by. I don't know whether, Rob, you have anything to add.
No, I think that's a fair summary, Hannah.
Yeah. Hi.
Did that answer your question or?
Yeah.
Okay, great.
Hi, thanks. Nick from Baird. You know, it looks like you all released an update since earlier this month, and in general it seems like refreshing the software is a pretty key part of the strategy here. I'd love if you could just maybe give one or two examples of some of the changes that you've made to Simcyp over time, and in general, just how your consortium strategy really informs the evolution of that product. It'd be great to understand some of that a little more. Thanks.
Rob, can you take that question?
Rob, did you hear the question? Do you want me to repeat it?
Yeah, I think the question was around the evolution of Simcyp over the years and the different models. You know, if we go back 20 years, you know, the Simcyp Simulator wasn't even mechanistic. It was a static model to predict sort of point in time outcomes. So since then it's moved on a lot. Hannah talked around the different populations. Obviously, that increases the application of the models. Also we've added in organs as well. Instead of just systemic exposure, we look at exposure in an individual organ level. Then we're you know over the years we're going deeper into the mechanisms of metabolism as well.
You know, a lot of that feedback and interaction does come from the consortium clients and the challenges that they're trying to address in their drug development. But also from academic users and just the scientific community in general. You know, as we advance the models, different users try different applications and it improves over time. You know, there are other drivers as well. You know, as in vitro data has become available that we can use or just general biological data. Computing power increases, so the models can get more complicated and run more quickly. There are lots of things, lots of drivers that are improving the simulators over time. Did that answer your question?
I think there was a question also on some of the latest changes, right?
Yeah. It'd be great if you can touch on some of the changes that were made in the most recent update or some of the more recent versions of it.
Some of the updates are quite technical, but you know, from the feedback from the consortium members and also from regulators, you know, we have some line of sight to where the usage will be going over the next few years, you know, from the draft guidances. Hannah mentioned renal and hepatically impaired patients, for instance. You know, that's something that's on the agenda for the FDA, and so to become, I think more focus on that over the next couple of years. Pediatrics is significant. Obviously, it's difficult to run trials in children. Those have been strong developments for us. Pregnancy as well, you know, females that are pregnant not being able to take drugs.
We're developing or we have developed models in those areas as well.
Some of the other areas I guess we've expanded on is predicting PK for large molecules as well.
Michael Ryskin, Bank of America. I actually wanna ask on exactly that, sort of the evolution of the software as it comes to different drug modalities. If we think about, you know, what's required for PK/PD and from Simcyp and Phoenix for small molecule versus antibodies versus mRNA versus cell and gene therapy, I imagine there's different parameters and there's different underlying science that goes into it. Could you talk about your capabilities for some of those new drug modalities? You know, are you able to answer all those questions? Is there applicability right now? Are the guidances there for those, and sort of when can we expect updates and progression into that?
Maybe I can just start off, Rob, and then hand over to you. I guess most of the... Well, obviously, the demo I showed today was for small molecules, but we have an antibody model as well, and we use that quite routinely for pre-predicting first-in-human PK and pediatric dosing is also a common kind of application. We can also combine our antibody and small molecule model to make an ADC model, and we use that model quite frequently, again, to predict human PK, pediatric dosing, and obviously drug interactions as a result of the payload. We are looking into developing kind of other models as part of the consortium. I don't know whether, Rob, you want to comment on other modalities.
It's really around. You know, you mentioned a lot of different modalities there and the areas are quite diverse, so it depends really what you're trying to look at. I mean, the Simcyp Simulator, you know, we have a large molecule module, which can look at various things. Like, as Anna mentioned, antibodies and other proteins. You know, at some point oligonucleotides come into it. Antibody drug conjugates. But that's really looking at the metabolism of it. Once you get into gene therapy, you're really looking at dosing and more on the QSP side of the platforms.
How is the gene therapy applied and what dose and the efficacy of that? Obviously it's quite important 'cause it's usually a single time dose. It's not really just the Simcyp Simulator looking at PBPK. Some of the therapeutic areas you referred to fall across different models that we build in the different platforms.
Any other questions?
Could you just maybe talk about the typical users or your customers, sort of what their educational background experiences look like? Where they're getting that experience from that sort of user base.
Rob, can you speak to the user base? Who are the ones using the software?
You know, there are modelers on the client side. I mean, they can sit in different departments. Could be in the DMPK department, clinical pharmacology. It's really a mirror of the talent that we have within Simcyp, but just on the client side. You know, they're highly educated, skilled users of the software tools.
I think-
I think across the software tools that, you know, there is more variety to it, right? Naturally, the modelers. What you also see, for example, for the Pinnacle 21 tools is it's a lot of biostats and stats and those types of backgrounds that comes into the picture. If you look at the early discovery tools like, D360, it's MedChem, little bit of computational science. Then naturally on the commercial side, it's more the, you know, market access, health economics, background that uses those types of tools.
I guess just.
Is that becoming an area of specialty in?
PBPK?
Yeah.
Yeah, for sure. I think one of the advantages of the interface is that you don't need to kind of see the equations behind the software. We have quite a lot of kind of non-modelers using the software as well. As long as you have kind of a good understanding of general pharmacokinetics and drug metabolizing enzyme kind of efficiencies and capabilities, you can actually kind of use the modeling software as a non-modeler.
We do, as part of that, have a center of excellence model with the educational institutions. You know, we have that pretty much all over the world, whether it's here in the United States, Europe, Japan, all over Asia, to help the academic community use the technology, yes.
There are many universities as well are using Simcyp as part of their courses, so many graduates are kind of learning how to use Simcyp as part of their degrees and certainly postgraduate degrees.
That answer your question?
Yes.
Thank you.
Okay. Another question. Right there.
Thanks. Towards the end of the presentation, you showed us sort of six software platforms.
Yes.
If I recall.
Yes.
Are those all integrated in on a common platform? Is it important if they're not?
Yes, there is a high degree of integration between those different software platforms, but there's always opportunity to do more. There is opportunity to do more within the cross-section of domains between clinical and regulatory, as Bill also mentioned in the early part of the presentation, and we're always exploring those opportunities with our customers as well. I think specifically the model that you see with Simcyp, where we have these consortium, and we have that with others as well, the very tight working relationship with the customers really help us explore those, what those opportunities are.
I don't know, maybe this is for you, William, but I think early on in the presentation, you kinda highlighted the geographic mix, and you've talked a lot about how adoption by regulatory agencies in the U.S. internationally has spurred growth. You're seeing some of that faster growth in Japan and in China, parts of Europe, but it's still off a very small base. Is there anything impeding, you know, international markets becoming a bigger part of the mix? Is it just timing of getting it through the regulatory agencies, getting adopted, getting the guidances? Could you just talk about the geographic mix, you know, going forward?
Yeah. Thanks for the question. It varies by region. In Europe, it's not that different from here. There's a greater R&D budget than in the U.S., so we have a bigger presence, but we've been expanding in Europe because there's a great talent base and maybe we're a little bit under-penetrated there relatively. Asia's really a very different situation. The amount of money going into China, in particular, has increased very, very rapidly, probably faster than the amount of talent really available. It's a little bit of a challenge for us around growing there. We've, I think as I mentioned, we opened an office in Shanghai last year, and we've been adding to that even in the pandemic when it's hard to travel.
Probably the biggest challenge for us there is bringing in enough qualified experts to keep up with the industry. It's a good problem to have. I think there's gonna be a lot of growth in the future. If you go to Japan, it's a little bit different situation. It's you know, a much more mature market and you know, a number of companies that are serving the Japanese pharmaceutical market more exclusively. We have a pretty good position there, and we've had that for a while. You know, actually some of our biggest clients are in the Japanese market. It really kind of depends on what's happening around the world. I think the interesting one to watch will be China going forward.
You know, a lot of the European and North American ones are pretty related. You know, a lot of the work we do in Europe, for example, might even be, you know, a global company, for example. Hopefully that helps.
Thanks.
Okay. That's it for. Do you have another question? Okay.
Yeah. We've got a question from Dave Windley, who asked a question via email. His question is: regulatory guidances appear to usher in new accepted use cases and broader use of the software. What's the revenue mechanism, more users within a client or use of more modules within the platform by the same users? Can you help expand on the revenue profile of, the adoption of the software?
Is that specifically directed for Simcyp or broadly?
Why don't we use Simcyp as an example?
Okay, good. Rob, we'll let you go first then. The question.
Okay.
I don't think he heard. Your mic is not-
Well, the question was centered around pricing models.
Yeah
what the scaling factors are. Is that fair?
No, I think the question's around as regulatory guidances expand, how does that translate to more revenues in Simcyp? How does that translate to expansion?
Right.
Right.
Yeah.
It certainly generates more use cases. The you know once we have case studies that are adopted and are accepted by the regulators, then other companies want to do the same, obviously. That can drive certainly the tech-enabled consultancy part of our business and the services for clients that don't necessarily hold licenses. I think we talk a little bit about that in the next section. In terms of existing users, it certainly increases their use of the licenses internally. It drives the need for greater access and that can increase revenues we drive from those clients.
Sometimes, depending on what the regulatory change is or what the application of the simulator is, it can extend out into other user groups within the same company. We can expand internally with existing users. Was that your question? Did that answer your question? Sorry.
Yes.
Hmm.
Okay. Now we're gonna take a brief break and then for five minutes, and then come back for the tech-driven services portion. Thank you.
Thank you.
Thanks.
Thank you, Rob.
Thank you.
Just wait a minute, Patrick. Okay, everyone. Welcome back after our brief intermission. My name is Justin Edge. I lead our regulatory and access business, and it's a real pleasure to speak to you today. Let's talk about services, our technology-driven services. Our services are predominantly powered by our technologies. What has this delivered for us? Well, our first nine-month revenues of almost $150 million represents about 22% growth over the same period in 2020. Our bookings grew by 10% in the same period, and in part, due to the timing of contract renewals.
Our net repeat rate has held steady at 115% when compared to 2020. This continued growth in repeat business has been enabled by software innovation and the addition of 18% more experts supporting our services, with significant headcount growth in Europe and APAC regions. Our teams have worked on more than 1,200 projects year to date. They continue to support significant numbers of regulatory submissions. There are macro and micro factors driving this growth. First, continued investment into the biotech sector has deepened the resources of many more biotech companies. This trend is especially pronounced among China-domiciled biotechs pursuing ambitious global drug development and regulatory programs worldwide. Second, we continued to benefit from regulatory tailwinds. Clients are pursuing parallel or closely sequenced global approvals in multiple jurisdictions, including the U.S., the EU, U.K., China, and Japan.
Next year, we'll see the implementation of the EU's Policy 0070 that governs clinical trial transparency and disclosure following several years of Brexit and COVID-related delays. The FDA's Project Optimus provides yet another lift by encouraging dose optimization in the early stages of oncology clinical trials. Finally, we've been able to differentiate our Certara services with our technologies such as RegTech, that Leif spoke about, that drives speed and quality, along with our proprietary biosimulation tools that we have been discussing earlier this afternoon. The services themselves, they span the continuum from early stage discovery and clinical development to regulatory and all the way through to commercial and access solutions. We harness proprietary Certara technologies to differentiate our services throughout the customer journey. For example, we support regulatory submissions using our own eCTD publishing suite of software called GlobalSubmit.
We communicate the value of new medicines and devices to payers using our BaseCase visualization software. Our Simcyp and Phoenix software underpins our biosimulation, clin pharm, pharmacometrics consulting services. Patrick is going to elaborate on some of these services in a few minutes, but I'd first like to provide a more granular view of how we power our regulatory writing with Synchrogenix Writer. We talked a lot about safety and drug-drug interactions earlier, but patient safety narratives are undertaken by sponsors to communicate serious adverse events to regulatory agencies. Synchrogenix Writer was developed at Certara to draft and complete patient safety narratives at scale, at speed, and with zero defects.
Our regulatory teams around the world use it to produce thousands of safety narratives with impeccable consistency because any downstream changes in format are applied to the entire batch of narratives in the production cycle. This allows our clients to customize documents to their own preferences and internal guidelines, again, at scale. Let's get a wee taste in a short video I'm going to queue up.
Patient safety narratives are critical to clinical study reporting across every therapeutic area throughout drug development. These narratives summarize adverse events experienced by participants during a clinical study. Successful narrative projects require cutting-edge tools and experienced professionals to manage moving timelines with attention to detail and accuracy. Enter Synchrogenix Writer. Spearheaded by Certara's regulatory experts, Synchrogenix Writer software uses automation and advanced technology to enable earlier drafting before database lock and streamline the narrative writing process. Now, organizations can successfully manage thousands of patient narratives in a fraction of the time while ensuring quality, speed, and scale. Certara's regulatory team has already used Synchrogenix Writer software to deliver more than 10,000 high-quality patient narratives for leading sponsors. In today's world of increasing clinical trial complexity, Synchrogenix Writer helps to ensure quality, speed, and scalability needed to meet regulatory requirements and tight timelines for patient narratives.
Hi, I'm Patrick Smith, and I lead our integrated drug development consulting services. I'd like to start by sharing another example of how we use our software to offer differentiated technology-driven consulting to our clients. You may not be aware, but publicly available clinical trial data happens to be substantially underutilized in drug development. To address this problem, we've created CODEX, which is a software product which includes large databases of more than 55 therapeutic areas and more than 10,000 clinical trials. The data is continuously updated from the medical literature and from observational studies. Importantly, clients can also add in their own proprietary data to see how their drug stacks up against the competition. The decisions that CODEX informs include the most important decisions in the entire drug development process.
It allows customers to correct for differences in clinical trial designs and patient populations to determine how their drug compares to other drugs in the market in terms of both safety as well as efficacy. As such, it's often used at the end of phase II for a company to make a go, no-go decision in terms of investing in a full and expensive phase III program, as it can tell you if you're likely to be successful against your benchmark competitor. It also allows one to conduct clinical trial simulations to optimize clinical study designs, which will maximize the probability of success of one's program, and to make decisions about the likelihood, for example, of a head-to-head trial succeeding against a competitor. Currently, Codex is licensed by nine of the top 15 pharma companies. I wanted to share a recent example of a Codex project.
This one was funded by the Bill & Melinda Gates Foundation around COVID. In this project, we used Codex to curate almost 500 trials and real-world observational studies to evaluate the impact of various therapeutic options which were being tested as potential treatments for COVID-19. This was also very interesting. As you may know, the pharma industry is making a large push to utilize real-world evidence and observational arms in regulatory decision-making. Using Codex, a meta-analysis was conducted across a variety of potential drug treatments for all of these trials, and for the first time to directly compare the quality and the level of evidence that was coming from these real-world studies versus the gold standard randomized controlled trial. The results are shown here on this slide, both overall and by treatment type.
On the bottom, the odds ratio of death is at the bottom with one meaning there's no impact of the treatment. If it's less than one, it means the treatment reduces death. If it's above one, then mortality rates were actually higher with that treatment. The real-world observational studies here are shown in red and the randomized trials in green. If you'll notice, the width of the diamonds correspond to the 90% confidence intervals as well. What's really fascinating here is that every single time the real-world data actually overestimated the treatment effect, and as you can see by the width of the diamonds created much less confidence in the results.
With Codex, we're actually able to adjust for this level of overestimation with real-world data and link the results of these early observational studies to what we would expect their performance to be in randomized clinical trials to better inform decision-making should we ever find ourselves in this situation again. If you recall, early on in the pandemic, there were lots of small, poorly controlled observational studies, that came out, and the randomized trials didn't come out until later in the process. This helps us put all of that early data into better perspective. We can also gain additional insights by looking at the generation of knowledge over time as the pandemic progressed. This is also very analogous to a drug development program where data is being generated as the program and the trials progress.
Here we're looking at two therapies, anti-IL-6 drugs on the left and glucocorticoids like dexamethasone on the right. The odds ratio for death in this case are on the Y-axis, and the X-axis is time as the pandemic progressed. The black line and the green or blue clouds indicate the odds ratio estimate and 95% confidence intervals, again, in chronological order as evidence became available. You can see the real-world evidence and the randomized trial data. Looking at anti-IL-6 data, you can see it looked good from the very, very beginning, but there wasn't a lot of confidence. Mostly the evidence was coming from real-world studies. You can see that RCTs started to emerge around day 300, and there was good confidence that these agents actually worked and made a clinically meaningful improvement for COVID-19 patients.
You can imagine if you were a regulator or a physician wanting to treat someone with these drugs, you can use this kind of an analysis to understand how much confidence we have in the decision itself and how it changes over time as data become available. On the right, we see the steroids or glucocorticoids like dexamethasone. Initially, it started to look based on limited data that steroids might actually make things worse. But as we learned more and data started coming out, this turned around by approximately day 200, and now we have substantial evidence that these drugs are highly effective. This is a great example of the limitations of some of the observational and sometimes poor quality data, and reminds us that one must proceed with caution until sufficient data is generated to make confidence decisions.
Client journeys can start anywhere along our end-to-end platform with many different on-ramps. Many new clients land at Certara for biosimulation software, where our regulatory science, drug development expertise, and market access solutions attract not only additional new logos, but offer many opportunities for cross-selling of other solutions. As we have grown, we have more commonly leveraged the full suite of service offerings with more clients relying on Certara to both strategically design and to implement a biosimulation-based approach to drug development and to help execute on that strategy. For example, 90% of our top 50 clients use both our biosimulation solutions and tech-driven services in regulatory science and market access. We engage with our clients continuously and tend to grow with our clients as they grow. Our success is embedded and aligned with their success.
Our solutions become embedded within their R&D process as well. Finally, customer migration in the industry helps to spread adoption of our software. As clients move from one company to another, they often bring Certara with them. As we have grown into a more mature organization, we have also developed specialized practice areas. These practice areas allow us to pair up our technologies with deep drug development expertise, which is specifically tailored to address the special challenges that are associated with developing drugs in these areas. We've combined our capabilities in these high-growth areas with robust go-to-market strategies. For example, if a company is developing drugs for pediatrics, we want them to see Certara as the go-to group with expertise, experience, and tools to do this more efficiently and more effectively than anyone else in the industry.
As you can imagine, it's not easy, not an easy thing to take a drug that has been studied in adults and figure out how to dose that drug in children, such as a neonate. The tools that we use in this area, like Simcyp, which Hannah demonstrated, are very well suited for the special challenges in pediatric drug development, and we have more experience than just about anyone in the industry in doing this. Pediatrics is an area of fast growth within the market, as every new drug requires a pediatric development plan to be conducted or for a waiver to be submitted to avoid trials. Furthering the need is the recent RACE Act, which is now requiring pediatric programs in oncology, where previously these programs were largely exempt from doing pediatric trials.
To demonstrate our thought leadership, we recently organized a global pediatrics drug development conference, which was attended with KOLs worldwide, including academics, pharma, and FDA. Here's an example of a recent project from our pediatric practice area. We worked with Mirum, which was a startup biotech company who acquired a drug called LIVMARLI that had been previously shelved by a pharma company. Mirum wanted to develop this for Alagille syndrome, a rare pediatric disease. Again, with a small internal team, they took advantage of many of our integrated service offerings at Certara. One key differentiator of our consulting practice is our ability to pair substantial strategic drug development capabilities with our technologies that helps bring those technologies to life and to deploy them in new and novel ways to solve some of the most challenging drug development problems.
To that end, we attended FDA meetings with Mirum. We helped them set up a biosimulation-based strategy for drug development and brought in multiple service offerings and technologies like Phoenix, like Simcyp, to support the eventual approval of LIVMARLI. More than 35 scientists and regulatory experts from Certara worked on this program. At times, there were even more Certara employees than employees from Mirum working on the program. The result was an approval and the first and only treatment option for which is a terrible disease, and this drug has had an enormous positive impact for children and their families. This next example comes from our global health practice area. You may not be aware, but river blindness is the second leading cause of blindness globally. It's entirely preventable.
It's a parasitic infection which is carried by the black fly, which is endemic to sub-Saharan Africa, and there are more than 16 million people infected. We partnered with a nonprofit company called Medicines Development for Global Health to support them in getting moxidectin approved, a drug which is significantly more potent than the standard drug, ivermectin, which has been used for the last 30 years. Medicines Development for Global Health at the time only had six employees. We partnered with them to create a virtual drug development team to implement biosimulation approaches to satisfy regulatory hurdles and supported them to gain FDA approval.
We continue to support the program going forward. Now, here's a brief video with Mark Sullivan, who's the Managing Director of Medicines Development for Global Health, on what it was like to work with Certara.
The role that Certara has played has really been as an integral part of our drug development team. Forming part of the project team, indeed being part of our governance structure, really bringing that global drug development expertise into our group. We're located in Australia, which is a little bit isolated from some of the core aspects of pharmaceutical development, and so we really needed that expertise that Certara brings. The strength of the group more broadly than just the people who are located in your particular region, we brought that to bear on our program. Certara's role was as an integral part of our program.
They fulfilled the clinical pharmacology function, but also then brought in a number of other really regulatory expertise aspects to the program right across the board from clinical pharmacology, even into our CMC program. In fact, you know, Certara sat alongside of me at FDA meetings and really gave me a great deal of confidence about our approach, the strength of our data and the information that we were sharing and how to share that was really a critical part of that particular story. Moxidectin as a treatment will be used in adults and children and particularly some of the indications such as soil-transmitted helminths, which affects 1.5 billion people, most of whom are children.
We are really focused on delivering a pediatric formulation and Certara have been a core part of that process of developing a pediatric formulation. Also modeling out, you know, the information that we would need to select the dose appropriately to help us in making very good decisions about how we go forward in pediatric use. Also, you know, in global health, drugs are used very broadly in the field, often in mass drug administration programs. It's incredibly helpful to have really well-modeled data in populations such as pregnant women and in breastfeeding, you know, the type of real-world use where it's very difficult to get data. The modeling data helps us to understand a lot more about that use. That's been an important part.
Some of these things all fit together, the pediatric formulation and pediatric development, dose selection, risk associated with populations such as pregnant and breastfeeding women. Certara, honestly, how else can you do this? You need a real engine of clinical pharmacology and the modeling built into your program. This was such an important role for us to fulfill, and Certara has done that beautifully for us.
I just wanna share a few more examples of how we've recently delivered our technology-driven services. I don't have a video, so it's just me. First of all, the first of these case studies really exemplifies the growth of Chinese biotechs that we alluded to earlier. This client was a long-time Phoenix software customer with several promising cancer therapies in development. We partnered with the company from early-stage development all the way through to late-stage regulatory phases, culminating in significant support preparing and publishing the submissions to the FDA. Right. We've subsequently been engaged to conduct a lot of HEOR modeling to support its payer engagement strategies. The second case also illustrates how we evolve from a point solution to providing a broader spectrum of services.
In this example, a publicly traded biotech client relied on us for NDA and BLA submission support that included lead authoring of the regulatory documents and the publishing of the entire submission packages to the FDA, again, using our GlobalSubmit software that Leif talked about earlier. We helped the client navigate some last-minute hurdles, actually using a lot of automated process-processing and QCing to avoid missing a submission date following an additional late-stage FDA stipulation. Okay. This relationship started with our Synchrogenix software, but it's matured into drug development consulting and tech-driven regulatory services as I described. The third example demonstrates the growth of our Access services and software.
In this case, it's a large global pharma client that has partnered with us for more than a decade and has been expanding our work in market access and HEOR across multiple therapeutic categories, some of which we've listed here over the past several years. For example, our health economic models support the client's engagement with payers and also health technology assessment bodies. The impact of these models can be demonstrated to payers using our BaseCase software. That's the primary value communication platform that this client uses. We've also actually co-published extensively with the client in the last decade. Recently they successfully received coverage in both major and secondary EU markets for the work we did.
All three of these examples really underscore the importance of landing and expanding both in biotech and in large pharma accounts, pursuing multiple development and submission programs, actually in many different geographies as well. Back to you, Patrick.
Thanks, Justin. Most importantly, as William mentioned previously, Certara is about its people and the entrepreneurial culture that is focused on making a difference for patients through the success of our clients. Collectively, we have more than 350 scientists with doctorate degrees and have exhibited very strong growth globally, including in Europe and Asia Pacific. We've also been winning awards as an employer of choice. We've been expanding our commercial team, which all puts us into a very strong position to continue scaling Certara into the future. Our talented scientists are the core behind our efforts to move the industry forward and to influence change in terms of the adoption of biosimulation and our other technologies, both by regulators and the pharmaceutical industry.
We do this by providing extensive training and certification for users of our tools, by ensuring that our tools are available at the universities so that new scientists are trained as users. We also offer webinars and conferences and have a very impressive track record of innovation via scientific publications, leadership presence within key organizations, and are frequently engaging with regulatory agencies as invited speakers at seminars and at workshops. These activities are key to driving the adoption of what we do and for expanding the market for our future growth. Thank you for your attention, and we'll now open it up to Q&A.
No questions? Yeah.
Can you just help us understand to what extent you're interacting directly with the innovator versus the CRO? Are they your partner in that? Are you guys both sort of dumping data into the client? Just how does that work? I'm trying to figure out if the CRO is sort of a partner in your efforts or more of a competitor.
We sit on the same side of the table with the client. It's the innovator. We're focused on making the innovator successful. We interact with CROs because, you know, data comes in oftentimes through the CRO data clients on the other line who are running, you know, the actual studies themselves. That's the limit to our interactions with them. It's almost all on the innovator side.
There's a question at the back.
Yeah. Hey, thanks. It's Vikram Kesavabhotla with Baird. You know, you've talked a few times throughout the presentation about the quality of the team and the expertise of the group at Certara. You know, it seems like the pandemic the last couple of years has definitely brought more attention to this industry overall. I'm curious from your perspective, you know, is the hiring environment today any more or less competitive than it's been in the past? As you think about, you know, growing the business in the future, how you think about navigating that going forward?
Happy to-
Why don't you start?
Happy to lead off.
Yeah.
I'll let Patrick correct me. So there certainly is plenty of competition for the skill sets that power our business, which is a reflection, though, on the growth of the space. We've done a lot of things to meet those challenges and stay ahead of them, I would say, since the onset of COVID. For example, we have instituted a lot more programs to develop talent from within, to incubate talent from within. We've got a residency program for regulatory writers. We're now onto our third intake this year, so that's been very successful for us. We have improved on significant apprenticing programs. I believe, in fact, McKinsey talked in an article about this a few months ago, the importance of upskilling from within the regulatory space.
I would also add that we continue to perform very well relative to the turnover benchmarks we keep an eye on at CROs, for example, for the same talent. We are certainly winning a lot more battles than we're losing, I would say, in the space. I don't know if you want to add to that.
Yeah. Same. I mean, it is competitive out there, but we have. I think as a leader in the industry, Certara is a really attractive place that people want to come and learn how to do biosimulation-based drug development and apply these techniques.
Right.
the technologies.
Perhaps I'd add one final point as well. As we've expanded our geographic footprint, that also affords us many more hubs, actual or virtual, to hire into and incubate talent as well, and we've really driven hard on that as well, I would say in Europe and APAC in the last year. Any questions online? No? Thank you.
We have the financial.
Don't forget your phone. Thanks.
Testing. Don't peek ahead. Should I get going in the interest of time to leave more opportunity for questions? No picture of the CFO, but I got a building.
That's our headquarters.
As William mentioned earlier, our financial performance for the first nine months of the year was strong and ahead of plan. Through September, we delivered revenues of $210.8 million, representing 18% revenue growth. The revenue growth for software was 10% and for services was 18%, resulting in a mix of 29% software and 71% services. Net loss of $3.6 million, which includes $20.8 million of equity-based compensation and $12.7 million of one-time costs, which are primarily related to M&A and transaction expenses. Adjusted EBITDA was $75.5 million, up 15%, and includes approximately $4 million of ongoing public company costs. Diluted loss per share of $0.02. Adjusted diluted earnings per share of $0.16 per share.
Our bookings growth drives strong visibility, given the majority of our bookings convert to revenues in 12 months or less. It's similar to an ACV. Our trailing twelve-month bookings total $322 million, up 18% versus the same period last year. That's through October. The bookings were driven by 16% growth in software bookings and 19% growth in services bookings. As we mentioned in the earnings call, after a slow slowdown in the summer months, we saw a recovery in activity starting in September, which has continued through November. Year-over-year growth in November further increased the TTM growth rate from what's reported in October, which provides confidence in the outlook for next year. We closed the acquisition of Pinnacle 21 at the start of the fourth quarter, the largest acquisition in the company's history.
The expected 2021 pro forma revenue from Pinnacle 21 is in the range of $23 million-$24 million, with $6 million of GAAP revenue for Certara in the fourth quarter. Looking forward, we are expecting Pinnacle 21 revenue for 2022 in the range of $28 million-$31 million at a 43% Adjusted EBITDA margin. Quick follow-up from the earnings call. At the time of our earnings release, we indicated that as a result of the accounting standard for business combinations, that a large percentage of the revenues from Pinnacle 21 would not be recognized due to a deferred valuation adjustment that's required in purchase accounting. Subsequent to the third quarter, the FASB issued final guidance on this topic, which requires companies to apply ASC 606 revenue recognition. Certara has elected to early adopt this guidance.
As a result of this, there'll be no deferred revenue valuation adjustment. Certara will not be providing guidance or reporting an adjusted revenue metric. Said another way, Certara will recognize the revenues from Pinnacle 21 on the same basis as they were recognized before the acquisition was closed. Looking forward to 2022, and as Bill mentioned, our goals of mid-teens revenue growth, we're expecting the following. Revenues to be in the range of $350 million-$370 million, which is 20%-25% growth. The revenue guidance assumes revenue growth excluding Pinnacle 21 in the range of 12%-17%. Adjusted EBITDA to be in the range of $127 million-$135 million, with an adjusted EBITDA margin of 36%-37%.
The adjusted EBITDA guidance includes an additional $2 million-$3 million of incremental investment in research and development to accelerate software products, many of which were mentioned earlier. This reflects approximately 100 basis points of reinvestment of what would otherwise have been margin expansion. Adjusted diluted earnings per share in the range of 48-53 cents per share. Going forward, we are revising the calculation of adjusted diluted earnings per share to exclude amortization related to M&A. Based on the preliminary purchase accounting for Pinnacle 21, we expect acquisition amortization to total approximately $45 million in 2022, up from approximately $37 million in 2021. Fully diluted shares in the range of 156-158 million, and the GAAP tax rate of 40%-45%, cash tax rate of 20%-25%. I'll pause there.
Do we wanna go through William's section and go to questions? Okay. That's the financial update. I'll hand it to William to make concluding remarks.
Okay, I'll just make a couple remarks before we do Q&A. Today, we provided a more in-depth look at Certara and how we are passionate about accelerating medicines with our great technologies and our great people. I think that the problem that Certara solves is an important one. We can predict the effect of a drug on populations even before clinical trials are performed. For our clients, this is potentially worth millions of dollars per drug program and savings of time and resources. For patients, it means better therapies that are delivered more cost-effectively. Biosimulation is here today, and it's a real and growing part of the drug development process. We work with nearly all the top 50 pharma companies, and in most cases, those relationships date back many years.
Some of those companies spend millions of dollars a year with Certara, and all of them could benefit and will benefit from the trend of increasing use of biosimulation in drug development. More recently, biosimulation has begun to penetrate biotech companies, which give us hundreds of new clients a year and counting, and who are in many cases, well-funded, smart, and moving very fast. There is tremendous opportunity for Certara to bring our multiple software and tech service products to their drugs as they move through the drug development stages. Our vision, as you can see from today, is more expansive than just biosimulation. At Certara, we see drug development as a data and analytics-centered process, a series of make or break decision by our clients, the drug companies, on molecule selection and trial design, followed by a data-driven rationale to regulators on why that drug should be approved.
Biosimulation is a key tool, and we seek to use it along with other data analysis and modeling tools and services to serve our clients end-to-end during their drug development process. Let's see if I can go to the next one here. Our focus on biosimulation and innovative technologies to make a difference in drug development has led to an attractive investment profile. We're a recognized leader in biosimulation, and we continue to expand the use cases and achieve new regulatory acceptance. Our regular cadence of releasing new products and capabilities is critical to increasing adoption, and it supports our land and expand strategy with customers. We benefit from an attractive large market that's growing in the mid-teens, far faster than pharma R&D spend because of the need to improve drug development efficiency, especially with more complex and costly therapies.
Certara itself benefits from a significant competitive moat, which is created by the 20 years of investment in our technology, our extensive experience in biosimulation, the growth in adoption by regulators and the overall industry worldwide. Our customers embed our technology in their R&D processes, and that results in very high renewal rates. Our 1,600 customers cover a broad portion of drug developers worldwide, but there are at least double that number of companies in the pharma industry with whom we have not yet worked. I believe that within the industry, we're known for what we do, and we're being encouraged to develop new products and capabilities. The result of all of this is a growing, profitable business that generates cash flow to fund further investment. On behalf of the entire Certara team, thank you very much.
Now we'd like to turn it over to some questions. Thank you very much.
Wanna come up?
Take your mic so they can hear you.
Okay.
Questions?
Andy, you called out that $2 million-$3 million of R&D. Is that more one-time in nature, or what is it specifically for?
I would consider it a two-year incremental investment in R&D. It's primarily related to the data platform that Mike was discussing, as well as virtual bioequivalence to SPD.
Next.
Yeah. Hey, thanks for taking the question. You know, if I go back to the third quarter call, I think you all talked about, you know, some dynamics around your counterparts taking vacation days and that, you know, having some impact to bookings. You know, now we're a couple months, you know, since that time. Have things normalized from that standpoint? And then secondly, just given the recent rise in COVID cases, you know, in speaking to your customers, is there any reason to think that would be either a headwind or tailwind to bookings going forward, just around the timing of some of your deals closing?
I mentioned earlier, November was better than October, and October was a great month. The bookings performance has been ahead of our plan. The biggest issue I see is, you know, for December, it's not a big issue, is timing. We have large high visibility renewals, and whether they close on December or they close on January, has no impact on revenue recognition. From our perspective, the momentum has picked up and remains strong.
Test.
Just one quick follow-up, if that's okay. Just the revenue guidance for 2022. Anything in there in terms of the mix from software and services or the growth rate across those two lines that you're expecting next year?
On, on a ex-exclude-
Pinnacle 21. Is this working? No.
What do you-
I need to find out more about this whole thing.
Okay. The growth rate excluding Pinnacle 21 was 12%-17%. Reminder that we build our forecast based on visibility, so it's based on the bookings exiting the year, the deferred revenue and the renewal rates. We have 80%-85% visibility going into the year. Does not include some of the growth or new product revenues to a material extent to achieve that number. The growth rate for software is low to mid-teens, services mid to high teens. Then Pinnacle 21, we stated that guidance separately, so you could have visibility into that. I widened the range a little bit, not because of a change in expectations, because of making the forecast approach consistent from a visibility perspective.
The impact of Pinnacle 21 and the software growth will be 400-500 basis point shift towards software from services, so 30%-35% software mix for 2022.
Hey, it's gonna be a little bit of a sort of back-end question on margin opportunity and the shift mix going forward. Where I wanna attack it from is, you know, you talked a lot throughout the entire presentation on hiring and aggressive hiring throughout the organization. You said headcount was up by over 20% this year across software services, U.S. international. How should we think about that 20% or 24% growth in headcount versus high teens growth in bookings and revenues? What I'm getting at is shouldn't the workforce scale better in terms of how that translates into revenues, and what are the moving pieces in that?
Yeah. I think there is a ramp up time for consultants, so that can impact margins if, you know, on a quarter-to-quarter basis for back-end hiring. We're recently seeing a lot of hiring. There can be some impact, but that's a good thing for us because it creates more opportunity for next year. The apples to apples comparison on the headcount growth is the 20%-25% growth 'cause the headcount included Pinnacle 21.
Yeah.
They're aligned.
Should those be more in line going forward, sort of staying at that same rate?
Yeah. We target them to be in line as a, at a minimum to ensure that we have margin expansion opportunity.
Yeah.
for reinvestment and maintain the mid-30s% margin.
Okay.
Yeah. Hey, thanks. Two more questions from me. I guess one, going back to the very beginning of your presentation, you talked about the biosimulation TAM, and you mentioned the uptick from 15% to 16% in terms of the CAGR. Just curious, one, in terms of what's your sense for what drove that increase and the reason for why those it sounds like an external service is kind of causing that acceleration. Would be great to get some color on that. And then, associated with that, you know, you referenced the size of biosimulation TAM now about $3 billion. Like is your sense based on the current Certara product portfolio that that full $3 billion is addressable at this point with what you currently have?
Do you think you're gonna have to engage in more, you know, M&A in order to kind of fill out the entirety of that market opportunity? It'd be great to get some color from a higher level perspective.
Yes. On the first one, Gian can assist, but I'll give you my perspective on it. The increase in the TAM is correlated to increased use cases for biosimulation. If you look back when those CAGRs were developed 2, 3 years ago, things like the virtual bioequivalence and the QSP and the QSTS would likely not have been incorporated in that. Increased use cases from adoption and innovation. The second question was, can we go after the $3 billion biosimulation TAM organically? My answer would be yes, it is an organic opportunity. We could accelerate that through M&A, but it's our core opportunity.
Can we assume that the bookings growth that you're kind of thinking about as you give us the guidance for next year is effectively the same as revenue growth?
Yes.
Okay. A follow-up to that, or should we think about those growth rates as being similar across biosimulation, regulatory, and market access? It seems like from today.
Yeah.
We heard a kind of a less mature and therefore bigger opportunity in regulatory and market access and maybe a little bit more established presence in biosimulation.
The biosimulation is growing in the mid- to high-teens% and the more innovative biosimulation approaches, when we talk about the QSP, for example, is growing faster. The reg services and the EVA services, I would look to those in the low-teens%.
One more. What about global? Should we assume that the same kind of dynamic plays out with Asia a little faster than Europe, faster than North America?
Yes.
Hey, thanks. Just a follow-up on Pinnacle 21. You referenced today, you know, they work with 22 of the top 25 biopharma companies. I think you said 6 of the top 10 CROs. What do those remaining biopharma companies and CROs use right now for that type of capability? And is it reasonable to think that you could further penetrate that kind of top bucket of customers in the coming years?
I think maybe Leif, do you might wanna take that to start and then we can, you know, for the best answer.
Most of the remaining biopharmaceutical and CRO companies use homegrown solutions. It'll be an upgrade from a homegrown solution to a commercial software product. Is that a good? Thank you.
Any other questions? Questions? No questions? We're good? Great. Bill, you, there's a little bit more.
Well, I just have what I'd just like to say is thank you for choosing to spend your valuable time with us today to learn more about Certara. As I said when I started, it's really great to for the people in the room, it was really great to meet you, especially after one year of being public and seeing many of you on only Zoom. I appreciate you coming and it's been great to see you in face to face. We've talked a lot today about our talented team, so I'd like to end with this video of some of our team members sharing why they like to work at Certara. Thanks very much. Start. One of the best parts of being at Certara is the opportunity to grow and learn from each other and our partners.
I came here because there was a great sense of purpose and mission here. I wanted to be in a growth environment. I wanted to be in a place where we were doing really important things in life sciences.
During the recruitment process, I had the opportunity to interview with a lot of different people at Certara. Every single person I talked to, the passion they had for what they did for their clients and the people in the company and the mission, it really shined through. It was that passion that sealed the deal for me.
I have had a long career at Certara. One constant that I see throughout my career are the opportunities for people to see their potential and realize it.
Colleague experience is at the forefront of everything we do at Certara. Our people are our most important asset. The people I get to work with at Certara are the best and the brightest in their fields. They are great teammates. They're fun to collaborate with. I thoroughly enjoy working with them every day.
I often tell my clients that when they engage with Certara, they are not just engaging with me, but are essentially engaging with a thousand great minds to support the drug development program.
At Certara, we have a smart and intellectually curious team with over 300 PhDs. Collaboration is key, and when our teams come together to help our partners solve their unique challenges, patients win.
What differentiates us is our people and our purpose. We bring our value to our clients through the collective subject matter expertise, market-tested methods, and our overall program management skills. We never, ever let our clients down.
We are doing interesting and important work. I am inspired every day about what I see around me. I am excited about the road ahead.