Item 1. Business
Item 1. Business.
Overview
We believe that the surest path to a better future is through technological advancement and that the new frontier of technology lies at the interface of computation, engineering and biology. Our mission is to improve health with technologies that transform the way that antibody-based therapies are discovered. We aim to become the centralized operating system for next generation antibody discovery.
Our full-stack, artificial intelligence-, or AI, powered antibody discovery platform searches and analyzes the database of natural immune systems to find antibodies that can be developed as drugs. We believe our technology increases the speed and the probability of success of therapeutic antibody discovery, including enabling discovery against targets that may otherwise be intractable. We forge partnerships with drug developers of all sizes, from large cap pharmaceutical to small biotechnology companies. We empower them to move quickly, reduce cost and tackle the toughest problems in drug development. As of December 31, 2021, we had 156 discovery programs that are either completed, in progress or under contract with 36 partners. As a recent example, in a collaboration with Eli Lilly and Company, or Lilly, we applied our technology stack to co-develop two antibody therapies to treat and prevent COVID-19. At the start of the pandemic, when bringing a treatment to patients quickly was the priority, we discovered bamlanivimab from a single blood sample obtained from a convalescent patient, and with our partners, advanced into clinical testing 90 days after initiation of the program. It was the first monoclonal antibody therapy to reach clinical testing and the first to receive Emergency Use Authorization, or EUA, from the U.S. Food and Drug Administration, or FDA, for the treatment of mild-to-moderate COVID-19 in high-risk patients. Since receiving authorization in November 2020, bamlanivimab alone and together with other antibodies has been used to treat approximately a million patients, potentially preventing tens of thousands of hospitalizations and deaths. Knowing that additional antibodies would be needed to combat emerging variants, we deployed our platform to identify bebtelovimab (also known as LY-CoV1404), a highly potent antibody that neutralizes Omicron and all currently known variants of concern (VOC). Lilly advanced bebtelovimab into clinical testing in May 2021 and received an EUA from the FDA in February 2022. Lilly progressed into these clinical trials at a greatly accelerated pace as a result of the Coronavirus Treatment Acceleration Program, which is a special emergency program for possible coronavirus therapies created by the FDA in 2020 to expedite the development of potentially safe and effective life-saving treatments to combat the COVID-19 pandemic. With respect to other or future product candidates, there is no assurance that any of our partners or collaborators will be able to advance a product candidate into clinical development on this timeframe again in the future. We initiated our partnering program in 2015 and have only had this one program result in clinical milestone payments to us to date, and we have not yet had a program receive marketing approval.
Antibodies, which are proteins generated by natural immune systems to fight infection and disease, are amongst the fastest growing class of drugs and are used across multiple therapeutic areas, including oncology, inflammation, neurodegeneration and many others. In 2019, antibody-based therapeutics accounted for over $140.0 billion in sales worldwide and represented five of the top 10 selling therapeutics. The rise of genomics, high-throughput biology and genetic engineering has greatly expanded the opportunity and the ecosystem of innovators working to advance the development of antibody-based therapeutics. There has been a proliferation of biopharmaceutical companies pursuing innovative drug candidate formats and new targets. As new entrants continue to emerge, we believe the total addressable market will continue to expand.
As the field of antibody therapeutics evolves, finding novel antibodies with desired therapeutic properties has become increasingly competitive and demanding. We believe that there are two fundamental problems hindering the discovery and development of next generation antibody-based therapeutics. The first is the state of technology: because of the limitations of legacy discovery approaches, there are many well-validated targets for which suitable antibodies cannot be found. The second is access: most companies are forced to cobble together fragmented solutions and lack the facilities and expertise needed to prosecute their antibody programs. Both of these problems contribute to the rising cost of drug development and delay bringing needed therapies to patients.
Many emerging and established life sciences companies have been built around technologies that focus on one or a limited number of steps in the discovery process, including immune repertoire sequencing, or RepSeq, single-cell analysis, AI, and transgenic rodent platforms. We believe we uniquely integrate proprietary technologies that address each of these steps, creating a complete solution for our partners. Over the last 10 years, we have developed and assembled technologies that unlock the database of natural antibodies. We are democratizing the industry by providing our partners of all sizes with access to our centralized operating system.
1
As depicted in Figure 1 below, our technology stack is a chain of interlocking technologies that is designed to enable the identification of antibodies with desired therapeutic properties.
Figure 1: Our Technology Stack
Some notable technologies within our stack that compound the productivity and efficiency of each step of the discovery process include:
•
Source. We combine proprietary immunization with genetically engineered mouse technologies, including the proprietary suite of humanized mice we acquired in November 2020 in connection with our acquisition of Trianni, to provide a diverse source of human antibodies.
•
Search. Our patented microfluidic single-cell screening technology combines speed, throughput, efficiency, resolution and versatility, enabling rapid and deep searches of natural antibody responses.
•
Find. Following the acquisition of Lineage Biosciences Inc., or Lineage, in March 2017, we integrated high-throughput RepSeq technology with our single-cell screening technology to provide leading capabilities for the comprehensive profiling and functional characterization of antibody diversity.
•
Analyze. Our internally developed platform, Celium, a powerful computational engine for mining, interacting and visualizing the terabytes of data generated during an antibody discovery campaign, combines software, AI and visualization tools to organize, compute and interactively explore large multidimensional data sets.
•
Engineer. We acquired rights to the OrthoMab bispecific technology in June 2020, which is a versatile and clinically-validated protein engineering solution to design and produce bispecific antibodies.
The marriage of advanced data collection and computation creates a flywheel effect that augments our technology. As we run our partnership business, we are amassing unique, multi-dimensional data sets that link measurements at the level of single immune cells with the properties of the antibodies they make and the DNA sequences that encode their function. A single antibody discovery project can generate millions of DNA sequences and single-cell measurements, as well as thousands of target-specific antibodies, each characterized by hundreds of data points. Every project generates more data about the antibody immune response. This creates a competitive advantage whereby Celium extracts insights from the data that allows us to accelerate wet lab experimentation with in silico computation in a continuously iterative process. Because our computation is grounded on real world data, the output of Celium is not theoretical predications. We find real molecules that have been optimized by nature.
2
Our business thesis is based on the belief that technological advancement can improve the drug development process and that maximizing the value and impact of our work is best achieved through partnerships. In March 2020, we tested these beliefs as we mobilized our response to the COVID-19 pandemic. Working with our partner Lilly, we were able to progress from initiation of discovery to clinical trials in only 90 days.
As of March 15, 2021, bamlanivimab, the first monoclonal antibody therapy from this collaboration, has been evaluated both alone and together with other antibodies across multiple clinical trials for the treatment and prevention of mild-to-moderate COVID-19. Bamlanivimab alone and together with other antibodies has been used to treat approximately a million high-risk COVID-19 patients.
Lilly has successfully completed a Phase 1 study of bamlanivimab in hospitalized patients with COVID-19 (NCT04411628), a Phase 2/3 study in people recently diagnosed with COVID-19 in the ambulatory setting (BLAZE-1, NCT04427501) and a Phase 3 study of bamlanivimab for the prevention of COVID-19 in residents and staff at long-term care facilities (BLAZE-2, NCT04497987). A Phase 2 study assessing the efficacy and safety of bamlanivimab alone, and bamlanivimab with other neutralizing antibodies versus placebo for the treatment of symptomatic low-risk COVID-19 in the outpatient setting (BLAZE-4. NCT04634409) has completed enrollment. In addition, bamlanivimab is being tested in the National Institutes of Health-led ACTIV-2 study in ambulatory COVID-19 patients. On January 27, 2021 Lilly announced an expansion to the ongoing BLAZE-4 trials to evaluate bamlanivimab together with VIR-7831.
Bamlanivimab received EUA, from the FDA, on November 9, 2020, and bamlanivimab in combination with etesevimab received EUA from the FDA for the treatment of mild-to-moderate COVID-19 in high-risk patients on February 9, 2021. On November 22, 2020, Lilly was granted authorization for bamlanivimab by Health Canada under the Interim Order Respecting the Importation, Sale and Advertising of Drugs for Use in Relation to COVID-19. Similar authorizations for bamlanivimab have been granted in Europe, the Middle East and Africa. Through its COVID-19 Treatment Guidelines, the National Institutes of Health recommended the use of bamlanivimab together with etesevimab to treat early COVID-19 in high-risk patients on February 23, 2021. Bamlanivimab alone and together with etesevimab received a positive opinion from the European Medicines Agency’s Committee for Medicinal Products for Human Use, on March 5, 2021: the harmonized, European Union, or EU, level opinion on the efficacy, quality, and safety of the antibodies can be used by EU member states when making decisions on the possible use of the therapies at a national level prior to market authorization. On September 16, 2021, the EUA for bamlanivimab together with etesevimab was expanded to include post-exposure prophylaxis for COVID-19. On December 3, 2021, the EUA for bamlanivimab with etesevimab was expanded to include high-risk pediatric patients for the treatment of mild to moderate COVID-19 as well as post-exposure prophylaxis. On January 24, 2022, the FDA reissued the EUA for bamlanivimab with etesevimab and restricted their use in geographic regions where exposure is likely to have been to a non-susceptible SARS-CoV-2 variant, based on available information including variant susceptibility to these drugs, and regional variant frequency. Due to the high frequency of the Omicron variant, bamlanivimab together with etesevimab is not currently authorized in any U.S. region, and these drugs may not be administered for treatment or post-exposure prevention of COVID-19 under the EUA until further notice by the FDA.
Under our partnership with Lilly, we are entitled to receive a specified percentage of proceeds that Lilly receives from these sales. In 2020 and 2021, we recognized a total of $198.3 million and $327.3 million, respectively, in royalty payments related to sales of bamlanivimab. As proud as we are to have played a role in the global response to COVID-19, we believe it is only an example of how our technology can accelerate antibody discovery.
Our business has historically been both high growth and capital efficient. While our historical cash flows may not be indicative of future results, we have generated positive operating cash flow cumulatively since our inception in 2012 and in every year since 2018. Our partnership agreements include: (i) payments for technology access and performance of research, (ii) downstream payments in the form of clinical and commercial milestones and (iii) royalties on net sales of any approved therapeutics. We structure our agreements in a way that is designed to align our partners’ economic interests with our own. The vast majority of the economic value of the contracts that we enter into with our partners is in the downstream payments that are payable if certain milestones are met or approved products are sold. For the years ended December 31, 2020 and 2021, our revenue was $233.2 million and $375.2 million, respectively. For the years ended December 31, 2020 and 2021 our net income was $118.4 million and $153.5 million, respectively. As of December 31, 2021, we have entered into agreements for 156 partnered discovery programs, 131 of which include the potential for milestone and royalty payments from our partners. As of December 31, 2021, we had 386 full-time employees in Canada, the United States and Australia, which was comprised of approximately 49% scientists, 32% business professionals, and 19% engineers and data scientists.
Impact of COVID-19
Since the start of the COVID-19 pandemic, we have taken and continue to take steps to protect our employees, business partners, vendors, and contractors. To address the multiple dimensions of the pandemic, we have been continuously monitoring the
3
global situation, executing mitigation activities whenever and wherever required, following guidance from provincial and federal health authorities, and implementing a hybrid return to our facilities for those employees who have been working remotely.
We continue to take the following actions to address the evolving COVID-19 pandemic:
•
We implemented comprehensive COVID-19 guidelines that are shared via our internal communication platform to provide real-time updates company-wide relying on directives from local health authorities. As information is gathered, we adapt accordingly, including work from home requirements.
•
We implemented protocols for employees necessary to carryout Company functions in the office and laboratory facilities including physical distancing, personal protective equipment, signage, erecting barriers between desks and lab benches, and implementing space restrictions for different areas of the facilities.
•
We have limited business travel to essential activities and have robust procedures to control and monitor all office and facility access .
•
We have not been required to stop research activities due to the COVID-19 pandemic. We will continue to adapt and apply new measures as required and as directed by local health authorities.
Our Strategy
Our mission is to improve health with technologies that transform the way that antibody-based therapies are discovered. To achieve this mission, we aim to become the operating system for next generation antibody discovery and to act as an integral part of our partners’ development efforts.
We seek to expand the industry of antibody therapeutics in two ways. First, we believe our technology can solve discovery problems to unlock new opportunities for therapeutic antibody development. Second, by accessing our teams, technologies and facilities, partners can eliminate the extended delays and costs associated with setting up antibody discovery capabilities. Through our partnership business, we aim to enable our partners to start programs without delay and prosecute them at maximum speed.
Our strategy includes:
•
Create more value with our existing partnerships. We have entered into contracts for more than 155 antibody discovery programs. Our partners include large cap pharmaceutical companies, biotechnology companies of all sizes and non-profit and government organizations. Where appropriate, we seek to expand our single-program partnerships to multi-year, multi-target agreements. As of December 31, 2021, over 80% of partnerships with downstream-participation contracts included multiple targets. Through continual expansion of our capabilities and the addition of new technologies we will also look to increase the royalties associated with our partnership deals.
•
Increase the number of partnerships. We will work to forge new partnerships across our target customers, including large cap pharmaceutical companies, biotechnology companies of all sizes and non-profit and government organizations dedicated to drug development. We plan to gain new customers via increased business development activities, technological expansion and superiority over alternative methods. We will also work to increase the number of deals with smaller early-stage biotech firms by offering flexible deal structures and vertical integration from target to antibody drug candidate.
•
Expand our market by delivering a full solution through forward integration. Many of our potential partners, including early stage biotech companies, are seeking a partner with the infrastructure, resources and expertise to execute early stage discovery and preclinical development programs. Building on our existing platforms, we are adding capabilities and infrastructure to support full chemistry, manufacturing and control, or CMC, activities and GMP manufacturing to provide our partners with a full solution from target to investigational new drug application, or IND, submission.
•
Scale our teams and facilities to meet future demand. We are building capacity to support the execution of additional partnerships and expansion of the scope of discovery programs. To achieve this, we are investing in expanding our workforce and our facilities, and increasing efficiency through automation and software solutions.
•
Further our technological differentiation. We believe we have technological differentiation for discovery of antibody drug candidates from natural immune systems. We intend to maintain our leading position through research and development to amplify and add capabilities in areas such as computation, protein engineering, immunization technologies, genetically engineered rodents and cell line selection. As we do so, we will continue to expand and protect our intellectual property estate. We will continue to look for strategic technology acquisitions to broaden and deepen our capabilities and expertise in antibody discovery and development, or that offer opportunities to expand our partnership business into adjacent therapeutic modalities, including vaccine development and cell therapy.
4
•
Leverage synergy of data and computation. We leverage unique data sets and AI to increase the efficiency, speed, and capacity of our discovery programs. In our partnership programs, we maintain rights to large data sets that connect information at the level of single-cell measurements, DNA sequence and protein function. We use this data to create an accelerating flywheel of learning: data generation from our partnership business provides the basis for AI modules that lead to expanded capabilities and faster data generation which supports our partnership business.
We believe our strategy creates a virtuous cycle, as depicted in Figure 2 below, that will drive our position as the centralized operating system for antibody discovery.
Figure 2: Our Business Strategy
Our Key Competitive Strengths
Our industry position and success are based on the following key competitive strengths:
•
Better antibody discovery, from the start. Our ability to perform comprehensive searches of natural immune systems allows us to expand the universe of antibody candidates and increase the probability of finding those candidates with therapeutic properties. This allows us to be more selective, thereby increasing the quality of leads that advance to development, including finding rare antibodies and reducing the time, cost and effort of having to start again or fix molecules that are broken.
•
A full-stack technology, accessible to all. We have built a centralized operating system for therapeutic antibody discovery and development. Our proprietary technology stack leverages and integrates a wide array of state-of-the-art technological tools and expertise in areas including microfluidics, single-cell analysis, high-throughput genomics, machine learning and hyper-scale data science. Our end-to-end solution allows us to rapidly source, search, decode, engineer and ultimately deliver lead antibodies for our partners’ development efforts. We democratize antibody discovery by enabling drug developers of all sizes throughout the world to benefit from our technologies and initiate their programs without delay.
•
An AI platform built on real world data. Our AI platform is built and continually validated with real world data. Unlike AI-only antibody discovery platforms, the output of our process is not theoretical predictions. We apply computation to find real molecules with desired properties. We inform wet lab experimentation with in silico computation, and vice versa, continuously iterating this process for each discovery campaign.
•
A unique combination of hardware, software and wetware. We believe our approach is differentiated based on the integration of proprietary and patented technologies across hardware, software and wetware. Our founders pioneered single-
5
cell antibody screening in nanoliter volumes and developed proprietary microfluidic devices and custom instrumentation to automate and scale screening. Our workflows incorporate proprietary immunization methods, including the Trianni suite of transgenic mice, optimized molecular biology protocols and patented protein engineering technologies. These technologies are bound together with custom software, data science tools and machine learning algorithms, and are operated by high-performing teams.
•
Industry-innovating business model. We have built a high-growth business that applies to a broad universe of partners and directly aligns with their economic interests. We believe this capital-efficient model allows us to build a diversified portfolio of royalty streams that reach into the future therapeutic antibody market. As we focus on improving our antibody discovery platform, we will continue to make critical investments in technology that benefit the entire industry. As of December 31, 2021, we have entered into 156 partnership agreements.
•
The flywheel of data, partnerships and technology. We believe our technology becomes more powerful and more accurate with each program. Data generated through our discovery partnerships provides the basis for training AI modules that yield new insights into antibody responses and that improve the speed, accuracy and efficiency of our technology. This creates a positive feedback loop through which each round of data analysis improves the speed and efficiency of the next round of data generation.
•
Strong brand built on performance and third-party recognition. Our business is built on the success of our partners and the strength of our technology. The strength of our technology has won the support of governments, including $30.6 million from the U.S. Defense Advanced Research Platform, or DARPA, Pandemic Prevention Platform, or P3, program and CAD $175.6 million ($125.6 million) from the Government of Canada’s Strategic Innovation Fund. We have been covered extensively by the media, including top-tier outlets such as CNN, MSNBC, Time Magazine, the Financial Post, The New York Times, Wired and the Wall Street Journal.
•
Robust IP portfolio including foundation patents. Our patent portfolio reflects our innovative position and end-to-end capabilities in antibody discovery, including microfluidic single-cell screening and cell culture, single-cell genomics, antibody RepSeq and bispecific antibody engineering. Our Chief Executive Officer and Chief Operating Officer are named inventors of the microfluidic single-cell screening and cell culture patents exclusively in-licensed from the University of British Columbia, or UBC. Through the acquisition of Lineage, we control some of the earliest filed RepSeq patents that we are aware of. We leverage the advanced computational and experimental protein engineering methodologies disclosed and claimed in the OrthoMab patent portfolio to generate bispecific antibodies from any two antibody sequences. In addition, we have filed non-provisional applications in the US with counterparts in numerous jurisdictions around the world related to bamlanivimab, and other COVID-19 antibodies, which we have exclusively licensed to our partner Lilly.
•
Founder-led team, custom-built for interdisciplinary technology development. We believe investments in teams and technology are the surest path to a better future and that the frontier of technology lies at the intersection of engineering, biology and data science. Our founder-led team, backed by investors from the life sciences and tech sectors, has been custom-built for technology development at the interface of genomics, microfluidics, computation, biologics, protein engineering and translational sciences. As of December 31, 2021, we employed approximately 400 people, over two-thirds of whom are scientists, engineers, and data scientists.
Industry Background
Nature’s database of antibodies
Antibodies are the body’s solution for fighting infection and disease. Antibodies are Y-shaped proteins, made by the immune system, that circulate in the blood. Their function is to specifically recognize foreign targets (viruses, bacteria, proteins or cancer cells), bind to them and then eliminate them. The repertoire of antibodies made during an immune response is extensive and encodes essential information about our health, our history of disease and our protection against future illness.
Unlike approximately 20,000 genes that are hard-coded in the human genome, antibodies are created de novo in each individual immune cell through a process of the random shuffling of DNA fragments. For each antibody, this random shuffling creates two separate genes, referred to as the heavy chain and the light chain, that assemble to form a complete antibody molecule. Taken together, there are approximately 2.9 million different combinations of heavy and light chain genes. The complexity of this diversity is augmented by additional random DNA insertions and edits. This results in over 100 trillion different possible antibody molecules, roughly 100 billion times the number of hard-coded genes in our genome. This diversity is astonishingly large. To put it into perspective, if all the possible antibody variable sequences were typed back-to-back in 12-point Arial font, the resulting string of letters would extend from the earth to the sun over 1,000 times.
At any given moment, each of us is making approximately one billion different antibodies, an infinitesimal fraction of the possible diversity of antibodies. Each antibody is made by a single immune cell. When provoked by infection or disease, these cells
6
quickly divide and mutate their antibodies to create an expanded family, or lineage, of cells having closely related antibodies. Cells producing antibodies that best bind to the target get a selective advantage and divide faster: those that do not, are eliminated. Through this selection process, immune systems generate large families of optimized antibodies that can bind almost any target tightly and with exquisite precision.
Natural antibodies created by the immune system benefit from the processes of selection, quality assurance and optimization that have evolved over 500 million years. As a result, naturally-produced antibodies generally have superior properties for drug development. As compared to antibodies isolated from man-made libraries, they generally bind more tightly and specifically, and have superior properties for manufacturing. Due to their superior drug-like properties, approximately 92% of all approved antibody drugs have been derived from natural immune systems, as illustrated in Figure 3 below.
Figure 3: Sources of Approved Antibody Drugs
(Source: TAB Database of Therapeutic Antibodies, September 27, 2020).
Our Market Opportunity
Antibodies are the fastest growing class of drugs and are used across multiple therapeutic areas, including oncology, inflammation, neurodegeneration and many more. In 2019, antibody-based therapeutics accounted for over $140.0 billion in sales and represented five of the top 10 selling therapeutics worldwide. Antibodies represented 70% of the sales of all biologics, and 36 antibody therapeutics reached blockbuster status with sales higher than $1.0 billion. Mean peak-year sales for currently marketed monoclonal antibody and monoclonal conjugate antibody drugs are estimated at approximately $3.1 billion (median $1.8 billion). As of September 30, 2020, there were over 115 approved antibody therapeutics, with more than 150 in Phase 3 clinical trials. The cycle time for antibody discovery projects to reach Phase 1 clinical trials from target selection has been approximately 5.5 years. Monoclonal antibody and monoclonal conjugate antibody drugs have taken on average approximately 7.5 years to reach market authorization from the start of Phase 1 clinical trials. From 2014 to 2020, the number of Phase I clinical trials beginning to enroll patients for monoclonal antibody and conjugate antibody therapies increased by more than 120%, from 207 to over 460.
The probability for a molecule discovered by us to succeed in reaching Phase 1 clinical trials has been estimated at approximately 37%. Monoclonal antibody and monoclonal conjugate antibody drugs have had an approximately 18.5% likelihood of receiving market authorization from the start of Phase 1 clinical trials. According to a study of over 9,000 clinical development programs at large pharmaceutical companies, approximately 90% of all molecules fail to reach approval. Antibody-based therapeutics have thus achieved a higher rate of clinical success and offer several advantages:
•
Minimal off-target toxicity
•
Long half-life in circulation
•
Ability to stimulate the immune system
•
Low immunogenicity
•
Higher affinity and potency
7
As shown in the Figure 4 below, the antibody-based therapeutics market is expected to reach approximately $260.0 billion in size by 2025, representing a CAGR of approximately 11% for the period from 2019 to 2025. Further, the more nascent cell therapy market is expected to grow from $1.0 billion in 2019 to over $17.0 billion in 2025, reflecting a CAGR of approximately 60%. Opportunities for accelerating growth of the antibody therapeutics market include improved access to traditionally difficult targets (e.g., G protein-coupled receptors, or GPCRs, and ion channels), the emergence of new therapeutic modalities (e.g., bispecifics, chimeric antigen receptor T cells, or CAR-T, cell therapy and antibody conjugates) and the ever-expanding number of companies entering the space. Our partnership business allows us to participate in the future antibody therapeutic market through royalties and milestones on drugs that have been discovered using our platform. As of December 31, 2021, we had 156 discovery programs that are either completed, in progress or under contract with 36 partners ranging from large pharmaceuticals to biotechnology companies.
Figure 4: Total End Market Sales ($billion)
Challenges
Technology has dramatically improved efficiency in nearly all sectors of our economy. Within drug discovery, technological improvements have also been made, but we believe the impact has been limited. For instance, the cost of developing new drugs has roughly doubled every nine years since the 1950s. This retrograde trend, referred to as Eroom’s Law, stands in stark contrast to Moore’s law, which successfully describes the doubling of computational power every two years.
Looked at from any perspective, drug development still:
•
Fails too often. According to a study of over 9,000 clinical development programs at large pharmaceutical companies, approximately 90% of molecules fail to reach approval. Failure rates are even higher when considering programs that do not advance into clinical development. We believe the prevalent use of outdated technologies and fragmented processes by antibody drug developers of all sizes create significant inefficiencies throughout the drug discovery process. When antibodies with suboptimal properties are advanced into preclinical and clinical development, they are less likely to progress through development.
•
Takes too long. Based on a database of antibody drug development programs, the average antibody drug takes between approximately 9.4 and 12 years from initiation to approval. Selecting an antibody with the desired properties for successful preclinical and clinical development is akin to finding a needle in a haystack. We believe the inefficiencies in established antibody discovery methods can contribute up to years of delays and result in suboptimal antibodies being developed.
•
Costs too much. According to a well-publicized study published by Tufts University in 2016, the average cost of drug development was approximately $2.9 billion. A significant portion of this cost relates to the use of suboptimal drug candidates.
8
We believe that these challenges are often attributable to a continued reliance on outdated technologies. This is a watershed moment to redefine drug discovery. The past 10 years have seen unprecedented advances in the tools to measure biology, including genomics, high-throughput imaging and industrial-scale lab automation. These tools generate an avalanche of data that contain the insights needed to more quickly bring drugs to the clinic. At the same time, computational power and AI now make it possible to see relationships within big data sets that could otherwise not be seen.
Deep integration of high-quality data generation and computational tools are needed to solve the following challenges in discovery:
•
Underpowered and fragmented technologies. There are well-validated drug targets that cannot be addressed using conventional methods because of limitations with:
o
Outdated discovery technologies . Antibody discovery workflows primarily use technology invented more than 35 years ago, including hybrid cells, or hybridoma, and display. Hybridoma provides only a tiny window into the database of natural antibodies. Display technologies do not benefit from natural antibody optimization.
o
Fragmented solutions . Recently developed technologies are not integrated into a complete solution and address only a limited number of steps in antibody discovery.
•
Access to technology. Most companies are unable to access the novel technologies needed to address every step of the antibody discovery process:
o
High entry barrier . The time and capital-intensive nature of building internal antibody discovery capabilities limits the number of firms that can effectively participate in drug development.
o
Siloed technology access . New technologies are often held within companies for internal use and not available broadly.
•
Developing technologies are entrenched in theoretical solutions . The application of machine learning approaches can be limited by predictions that are difficult or impossible to transfer into experimentally validated results.
Our Platform
Our platform is an operating system designed to support many antibody modalities; unlock new targets; increase the speed to clinical development for our partners and increase the potential clinical and commercial success for our partners.
Our full-stack, AI-powered technology sources, searches, decodes and analyzes antibody responses with the ultimate goal of engineering new antibody drug candidates for our partners. Our platform incorporates and integrates modern technology tools from engineering, microfluidics, single-cell analysis, high-throughput genomics, machine learning and hyper-scale data science. We have internally developed, in-licensed or acquired our technologies. We deploy our platform to help our partners in their efforts to identify antibodies with better potency and developability.
We believe our approach of integrating modern hardware, software and wetware is unique. We have pioneered nanoliter volume single-cell antibody screening methods using microfluidics. Our workflows incorporate proprietary immunization methods, including the Trianni suite of transgenic mice, optimized molecular biology protocols and patented protein engineering technologies. The aggregation of these technologies, coupled with our proprietary processes and team, allows us to provide a differentiated offering to our partners. Figure 5 below represents how our technologies are integrated into one stack.
9
Figure 5: Our Technology Stack with Workflow Summary
The computational engine of our platform, Celium, combines software, AI and visualization tools to mine, organize, compute and interactively explore the immense multidimensional data sets that we produce in each antibody discovery campaign. Unlike many AI-based antibody discovery approaches, Celium is continually improved with real world data. We iteratively inform wetlab experimentation with in silico computation, and vice versa. The output of our process is not theoretical predictions. We discover real molecules that have been optimized by nature.
Our platform is an operating system that is designed to provide the following benefits:
•
Support many antibody modalities. Our technology is capable of performing discovery from a variety of species and using a wide array of selection criteria. We believe this expands the starting diversity and increases the likelihood of finding antibodies with specific properties needed for new antibody therapeutic modalities: discovery of binding domains for a variety of antibody formats; discovery of internalizing antibodies for antibody-drug or antibody-siRNA conjugates; single-chain camelid antibodies for use in CAR-T and protein engineering; and antibody pairs that have specific affinity and binding epitope recognition for bispecific antibody applications.
•
Unlock new targets. Our technology is capable of performing deep searches of antibody responses using cell-based assays that preserve the native conformation of traditionally difficult membrane protein targets. We believe this capability, combined with our proprietary immunization methods, provides a unique advantage in the discovery of rare antibodies that modulate the function of GPCRs and ion channels, two large and well-validated families of drug targets for which discovery using traditional techniques has been extremely difficult or intractable. Our proprietary Trianni All-Epitope mouse line (currently in validation) is engineered to mount robust immune responses against difficult targets that have high homology between rodents and humans, such as some high-value GPCR and ion channel class of targets that are prevalent in many diseases and indications including cancer, neurodegenerative and cardiovascular diseases, and pain.
•
Increase the speed to clinical development for our partners. Our integrated technology stack is designed to accelerate the discovery and pre-clinical development process. We anticipate substantial time can be saved through faster workflows and
10
avoiding unnecessary cyclical efforts. Our technology stack is capable of going from screen to hundreds of antibody sequences in only three days, and from sequences to characterized antibody proteins in less than ten days. Speed may also be achieved by increasing diversity at the start of discovery to maximize the chance that suitable leads are found on the first pass, and by minimizing the requirement of protein engineering. Finally, additional speed in discovery can be achieved by integrating all steps of the process seamlessly.
•
Increase the potential clinical and commercial success for our partners. We aim to increase the probability of clinical and commercial success of our partners. Our technology stack is designed to provide a competitive advantage in speed to the clinic and to identify antibodies with superior biophysical properties.
We believe our competitive advantage is derived from integration of multiple proprietary technologies and a seamless workflow. Table 1 below provides how each aspect of our end-to-end technology stack addresses challenges in antibody therapeutic discovery:
Table 1: Our Platform and Solution
Our Partnership Business
We forge partnerships with large cap pharmaceutical companies, biotechnology companies of all sizes and non-profit and government organizations. Our partners select a target and define the antibody properties needed for therapeutic development. We provide discovery solutions to partners that have a range of discovery capabilities, from the highly enabled to the less enabled. We enable discovery against targets that have traditionally been intractable, and we accelerate programs against less difficult targets.
11
Our Target Market
We provide discovery solutions to partners that have a range of discovery capabilities, from the highly enabled to the less enabled. We accelerate discovery programs spanning a range of difficulty, from traditionally “Intractable discovery problems” to less difficult “Tractable discovery problems”. These categories, taken together, create a two-dimensional grid that segments our market opportunities as shown in Figure 6 below.
Figure 6: Our Target Market Matrix
The four segments depicted in Figure 6 above are as follows:
•
Segment 1. Includes tractable targets at large cap pharmaceutical and large biotech companies that have made significant investments in building internal capabilities in antibody discovery, typically using automated hybridoma methods and/or antibody display technology. We believe that our full technology stack allows us to serve this market by delivering antibody drug candidates faster and of higher quality, and by providing access to more advanced technologies for next-generation biologics.
•
Segment 2. Includes intractable targets at large cap pharmaceutical and large biotech companies. This segment holds untapped potential to generate first-in-class therapies for large unmet medical needs (e.g., pain, autoimmunity, metabolism), but technological barriers have resulted in limited success. We believe our technology provides a strong competitive advantage to find viable antibody drugs candidates for difficult targets, including GPCRs and ion channel targets.
•
Segment 3. Includes tractable targets at growth biotech companies. These companies have comparatively limited discovery capabilities versus their larger peers. We believe that in working with these partners we can provide the following advantages: (i) access to a fully integrated technology stack (ii) accelerate their efforts with timely access to the necessary teams and facilities and (iii) improve the quality of the final antibody leads. In this way, we democratize antibody discovery by providing all of our partners with access to our centralized operating system.
•
Segment 4. Includes intractable targets at growth biotech companies. Presently, there are few partners working in this market segment. We believe that our technology stack can unlock these opportunities, leading to an expanded ecosystem of companies developing antibody therapeutics with the potential to become first-in-class therapies.
Our Partnership Deals
Our deals emphasize participation in the success and upside of future antibody therapeutics. Our partnership agreements include near-term payments for technology access, research and intellectual property rights, downstream payments in the form of clinical and commercial milestones, royalties on net sales, and may include alternative investment opportunities including equity in our business partner and various rights for deeper involvement in moving molecules forward.
As of December 31, 2021, we had 156 discovery programs that were either completed, in progress or under contract, including 131 with the potential for milestone and royalty payments. Our partnership agreements are typically terminable at will with 90 days’
12
notice prior to identification of a target, after which point they may only be terminated for cause. A summary of the recent publicly disclosed partnerships established over the last two years are included in Tables 2 and 3 below.
Table 2: Summary Partnership Agreements with Pharmaceutical & Biotechnology Companies that include downstream participation from 2018 to 2021*
Partner
# of Targets & Duration
Therapeutic Indication or Modality
Date Announced
Everest Medicines
Up to 10 targets, multi-year
Oncology and undisclosed
September 22, 2021
Moderna
Up to 6 targets, multi-year
RNA-encoded antibodies
September 15, 2021
EQRx
Multi-target, multi-year
Oncology and immunology (initially)
August 4, 2021
Tachyon
Single target
Oncology
August 3, 2021
Undisclosed biotech
Up to 4 targets, multi-year
Undisclosed
June 30, 2021
*
Angios
Multi-target, multi-year
Ophthalmology
May 6, 2021
Undisclosed biotech
Multi-target, multi-year
Oncology
May 6, 2021
*
Empirico
Up to 5 targets, multi-year
Undisclosed
April 14, 2021
Gilead Sciences
8 targets, multi-year
Undisclosed
April 1, 2021
Abdera Therapeutics
9 targets, multi-year
Oncology
January 14, 2021
Invetx
Multi-target, multi-year
Animal Health
November 19, 2020
Kodiak Sciences
Multi-target, multi-year
Ophthalmology
October 29, 2020
IGM Biosciences
Multi-target, multi-year
Oncology and immunology
September 24, 2020
Undisclosed
Single target
Bispecific
June 3, 2020
*
Eli Lilly
Up to 9 targets, multi-year
COVID-19 program and additional indications
May 22, 2020
*
Regeneron Pharmaceuticals
Multi-target, multi-year
Multiple undisclosed
March 16, 2020
*
Invetx
Multi-target, multi-year
Animal health
February 23, 2020
Undisclosed
Multi-target, multi-year
Cell therapy
September 25, 2019
*
Gilead Sciences
Single target
Infectious disease
June 13, 2019
Denali Therapeutics
8 targets, multi-year
Neurological diseases
February 28, 2019
Novartis
Up to 10 targets, multi-year
Undisclosed
February 14, 2019
Autolus Therapeutics
Single target
Cell therapy (CAR-T)
November 29, 2018
Denali Therapeutics
Single target
Neurological diseases
June 12, 2018
Undisclosed mid-cap biopharma
Undisclosed
Undisclosed
January 25, 2018
Teva Pharmaceutical Industries
Single target
Membrane protein
June 13, 2017
Pfizer
Multi-target, multi-year
Membrane protein
January 5, 2017
Undisclosed global biotech
Multi-target, multi-year
Undisclosed
November 4, 2016
Kodiak Sciences
Single target
Ophthalmology
August 24, 2016
Teva Pharmaceutical Industries
Undisclosed
Undisclosed
February 2, 2016
* Effective date of agreement
Table 3: Downstream Participation
Most of the programs with our partners will generate milestone payments to us should our partners reach certain preclinical, clinical, regulatory, and commercial milestones. In addition, programs that generate molecules that become commercial products may generate royalty payments to us on the net sales of those products . We also have other downstream economic participation including equity/equity-like positions and options to co- invest. The following table represents the potential maximum value of our milestones by development stage and our range of royalty payments :
Milestones (in billions) 1
Royalty on net sales, 5th to 95th percentile range 2
Preclinical
$0.05
2015-2019 contracts
0-4%
Clinical
$0.75
2020-2021 contracts
2.5-8.4%
Regulatory
$1.08
Commercial
$2.72
Other downstream participation
Total
$4.60
Equity/equity-like positions
Options to co-invest
1 All Programs under Contract, not probability adjusted
2 Includes range of royalty (and equivalent) rates of each contract, considering step-downs, if any
13
Reflecting the increasing value that we create for our partners, and our ability to capture that value, are the range and progression of the royalty rates we are entitled to under our agreements. Half of royalty rates associated with the 98 new partner programs under contract in 2020 and 2021 range from 3% to 5% of net sales. At the top end, royalty rates reach over 8% while very few programs have rates below 2.5%.
Further, the total milestones for our portfolio of programs under contract and not adjusted for the probability of success is $4.6 billion. In addition, our equity and equity-like positions in new companies and options to co-invest, particularly with early-stage firms, represent higher value programs within the overall portfolio.
Table 4: Summary of Partnership Agreements with Non-Profit & Government Organizations from 2018 to 2021
Non-Profit & Government
Summary
Therapeutic Indication or
Modality
Date Announced
Government of Canada
• CAD $175.6 million ($125.6 million) over multiple years
• Funding for technology and manufacturing infrastructure for antibody therapies against future pandemic threats
• COVID-19
• Infectious disease/ pandemic response
May 3, 2020
Bill & Melinda Gates
Foundation
• $4.8 million over two years
• Follow-up to successful 2017 project on tuberculosis
• Infectious disease, including HIV and malaria
March 14, 2019
Genome BC and Genome Canada
• CAD $3.0 million
• Genome Applications Partnership Program (GAAP) with UBC
• Duchenne muscular dystrophy
• Fibrosis
August 16, 2018
DARPA
• Up to $30.6 million over four years
• Establish rapid pandemic response platform
• Pandemic response
March 13, 2018
CQDM and Brain Canada
• CAD $0.8 million
• Function-modifying antibodies against GPCR target
January 27, 2018
In March 2020, we entered into a multi-year strategic Research Collaboration and License Agreement, or RCLA, with Lilly on the discovery of antibodies for up to nine Lilly-selected therapeutic targets, including COVID-19. Under the terms of the RCLA, Lilly has the right to develop and commercialize therapeutic products resulting from our collaboration. As part of the RCLA, we received an upfront payment of $25.0 million and are eligible to receive research payments for non-COVID-19 targets. We are also eligible to receive pre-clinical, clinical, and product approval milestones and tiered royalties on future sales for all Lilly-selected targets. We are entitled to receive an aggregate of up to $29.0 million of milestone payments under the terms of the RCLA. As of December 31, 2020 and 2021, we have received $15.0 million and $7.0 million, respectively, for clinical and other milestones related to bamlanivimab. For non-COVID-19 targets, we are eligible to receive royalties in the low single digits based on net sales; whereas for COVID-19, we continue to be eligible to receive royalties in the low- to mid-teens for aggregate sales below $125.0 million and mid-teens to mid-twenties on aggregate sales above $125.0 million. We have recorded royalty payments related to sales of bamlanivimab of $198.3 million and $327.3 million as of December 31, 2020 and 2021, respectively.
Our Technology
Therapeutic antibody discovery has a myriad of challenges. Antibodies that are suitable candidates for therapeutic development must engage a target specifically, induce the desired therapeutic function and also have physical properties that make them suitable for manufacturing and formulation as drug products. Only a small fraction of the antibodies in any given immune response will satisfy all
14
these requirements. Even for those that do, only a small subset will be optimal for drug development. Therefore, a broad and deep search of the database of natural antibodies is needed to expand the universe of quality drug candidates and increase the likelihood of success.
We have built a technology stack with five key capabilities that we believe solve the discovery problem and are critical to the success of any therapeutic program:
1.
Source. Generate and access a high-quality universe of antibodies for any target;
2.
Search. Explore this diversity with sufficient throughput and specificity to isolate the rare cells that make antibodies with the desired properties;
3.
Find. Decode the genetic sequences of selected antibodies and expand diversity with related sequences present in the immune response;
4.
Analyze. Collect data on the relevant therapeutic properties of each selected antibody to generate large multidimensional data packages, and apply computation to select the most promising leads;
5.
Engineer. Use broader information of antibody responses and molecular engineering to optimize and reformat lead antibodies for development.
Our technology stack achieves these functionalities by leveraging state-of-the-art methods from microfluidics, single-cell analysis, high-throughput imaging, AI, robotics, genomics and protein engineering, all complemented by custom software and data visualization capabilities.
Source: Immunization
The first step in a therapeutic antibody discovery program is to generate the source of antibodies that will define the search space for discovery. There are two competing paradigms: generate synthetic antibody diversity in man-made libraries, known as the display methods, or generate antibody diversity in vivo by immunization of animals. A brief description of these methods, along with their challenges, is as follows:
•
Synthetic antibodies. Display methods encompass a set of synthetic discovery approaches that are based on man-made collections, also referred to as libraries, of human antibody genes. These methods, in use for more than 25 years, have resulted in a small fraction of clinically approved antibody therapeutics. The main disadvantages of display libraries are (i) low binding affinity of antibodies from the initial selection step, necessitating protein engineering steps to increase affinity, (ii) difficulty in manufacturing leads due to poor expression and developability and (iii) difficulty in performing selections on high-value cell-surface targets. We believe these shortcomings have contributed to the relatively low success rate of display technologies in the clinic.
•
Natural antibodies. In the same way that seasonal vaccination is used to generate antibodies that protect the population against flu, antibody responses against a drug target can also be generated through the immunization of animals. Antibodies generated in this way benefit from the natural process of selection (enriched for antibodies that bind the target), quality assurance (antibodies can be expressed and are not sticky, making them more developable into a drug candidate) and affinity maturation (the natural process whereby antibodies are optimized within the body to bind tightly to the target). As a result, antibodies generated through immunization are generally of higher quality and have accounted for the large majority of all approved therapeutics. However, some of the challenges faced with many immunization-based approaches are (i) difficulty in raising a strong immune response against poorly immunogenic targets, (ii) traditional screening methods that restrict immunization and discovery to rodents and (iii) the need to convert antibodies discovered in rodents to human antibodies (humanization) before they can be used.
Our Approach to Sourcing Diverse Antibodies. We believe natural immune sources are a superior search space for therapeutic antibodies. To solve the challenges of sourcing antibodies by immunization, we have assembled multi-species screening capabilities, genetically engineered mouse technology and optimized immunization technologies that are capable of generating diverse and high-quality sources of natural antibodies.
•
Any source, any tissue. Our single-cell screening and RepSeq capabilities enable discovery from any host species and any immune tissue, bypassing the restriction to rodent species of traditional screening approaches. To date, we have successfully applied our discovery technology to search immune responses of numerous species, including humans, mice, rats, rabbits, dogs, cats, llamas, alpacas and cows. In addition to greatly expanding our search space, the versatility of our platform also means we can access antibody responses of species with unique properties. For instance, we have completed multiple programs focused on isolating single-chain antibodies, or VHH antibodies, that are made by camelids (e.g., llamas or alpacas). VHH antibodies derived from camelids are of high interest due to their small size, ability to bind to novel epitopes
15
and the ease with which they can be engineered to produce next generation therapeutics, including multi-specific antibodies and cell therapies. Finally, as recently demonstrated in our work on COVID-19, our approach also allows for a deep search of antibody responses from the large and complex immune responses of human donors, which have been naturally exposed to a pathogen.
•
Human antibodies from rodents . The ability to source fully human antibodies from rodents provides our partners the added benefit of bypassing the need to humanize sequences identified from non-humanized animals. We acquired the Trianni humanized rodent platform in November 2020, which includes a suite of genetically engineered transgenic mice that express human variable antibody genes. Currently, the flagship Trianni mouse is available for discovery projects. The flagship Trianni mouse was generated with proprietary in silico design of antibody genes that resulted in a novel antibody gene structure at the heavy, lambda and kappa mouse antibody loci to maximize antibody diversity. This proprietary antibody design contains the human antibody repertoire, plus the natural constant regions of mouse antibody genes and the natural regulatory elements from the mouse genome, making these human antibodies optimized for expression and maturation by the mouse immune system. We believe this feature helps maximize the response to immunization, the diversity of human antibodies, and enable proper affinity maturation in the mouse to isolate quality human antibodies as therapeutic candidates in antibody discovery campaigns. In addition to the flagship Trianni mouse, we acquired a suite of additional transgenic humanized rodent lines currently being validated and available for discovery projects in the near future, as indicated below. These lines aim to provide partners with the following benefits:
▪
Generate multispecifics and access difficult targets with the Heavy-Chain Only, or HCO, Mouse: The HCO Mouse expresses antibodies comprised of only heavy chains, without the accompanying light chains found in conventional IgG antibodies. This smaller HCO antibody can access additional target sites that the larger conventional IgG molecules cannot, due to less steric constraints, thereby expanding the target space. HCO antibodies can also serve as the basis for combining targeting arms for bispecific and multispecific antibodies without the complexity of correct heavy and light chain pairing. In addition, this mouse line provides the opportunity for generating the fully human VHH antibodies described above, without the expense of immunizing large camelids, and with the benefit of starting with fully human sequences that do not need to be engineered to be more human-like. The HCO Mouse was also generated with the proprietary Trianni in silico design that helps maximize immune response, antibody diversity and enable natural antibody maturation.
▪
Break immune tolerance with the All-Epitope Mouse: The All-Epitope Mouse is engineered to overcome immune tolerance to antigens that have high-homology to mouse proteins, to allow generation of robust immune responses against highly-conserved, high-value drug targets such as GPCRs and ion channels that are prevalent in multiple diseases and indications, including cardiovascular diseases, cancer, neurological diseases, inflammation and pain (see the subsection below titled “—Our Technology in Action”).
▪
Target new epitopes with the DD Mouse: The DD Mouse provides additional flexibility to discover antibodies with long CDR3 regions (the region of the antibody that makes primary contact with the target) allowing access of “hidden” or recessed areas of targets that are not available for contact by conventional IgG molecules with typical CDR3 lengths).
▪
Maximize efficiency with the Eazysort Mouse: The Eazysort Mouse is engineered to allow up-front enrichment of immune cells that recognize the target, helping maximize the efficiency of searching for the right antibodies by focusing the subsequent single cell screening to antibodies of interest.
These proprietary transgenic mice, combined with our platform technologies in immunization, single-cell screening, immune repertoire profiling and protein engineering, provide a flexible and synergistic advantage for rapid, next-generation discovery and development of fully human antibodies for drug development across a very broad target space. Notably, the Trianni platform and suite of transgenic rodents, together with expertise from Trianni personnel joining our research and development program, is a foundation to develop additional novel next-generation animals to expand our platform.
In addition to the Trianni transgenic mouse technologies, we also have in-licensed the ATX-Gx humanized immunocompetent transgenic mouse platform from Alloy Therapeutics. Like the flagship Trianni mouse, the ATX-Gx mouse is genetically engineered to express human antibody genes. We believe that by performing immunizations on both the Trianni mouse and the ATX-Gx mouse in parallel, we are able to expand the diversity of human antibody responses, thereby creating a larger search space for antibody discovery. We believe this will allow us to isolate more and higher quality candidate antibodies. We expect to continue to use both the Trianni mouse and the ATX-Gx mouse in our partnered programs, along with the Trianni next generation mice for the most demanding therapeutic discovery projects.
•
Optimized immunization methods. We have developed a suite of immunization technologies that have been optimized (i) for use with our screening platforms (ii) to address the challenge of immunizing each species, and (iii) to address challenging
16
targets. Specifically, we have developed immunization methods that, when coupled with our deep screening capabilities, allow for antibody generation against targets that are 100% identical to the host species. This is particularly valuable for the generation of antibodies with desired species cross-reactivity (e.g., mouse, cynomolgus monkey—important animal models for testing antibodies in preclinical studies—and human), or for targeting highly conserved protein epitopes. Importantly, these strategies negate the perceived advantage of display platforms for bypassing tolerance, the process by which natural antibody responses are suppressed against self-similar targets. For further explanation, see the case study titled “Overcoming Tolerance with Proprietary Immunization and Deep Search Technologies” in “—Our Technology in Action.” We have also developed a suite of genetic immunization methods for targets that cannot be easily expressed or purified as soluble antigens. This is particularly valuable for discovery against high-value membrane protein targets including GPCRs and ion channels.
Search: Microfluidic Single-Cell Screening
Searching natural immune systems is fundamentally a single cell problem. One mL of blood contains approximately 1 million white blood cells, of which approximately 1% are single antibody secreting cells, or B cells, that make antibodies. Of these, only a fraction is likely to bind to a target of interest, and of these binders, only a very small fraction is likely to have properties that make it suitable as a therapeutic. To effectively search the natural immune system for antibodies requires a technology that can scan through millions of antibody-producing cells and make high-resolution measurements to assess the properties of their unique antibodies. B cells are microscopic, having a diameter of approximately 10 microns (about 1/10 of the width of a human hair) and generate only a minute amount of antibody. When analyzed in the volume of conventional screening formats and conventional labware such as a 96-well plate, this small amount of antibody is too dilute, making it essentially undetectable.
It is because conventional methods lack the sensitivity to analyze individual cells that traditional discovery approaches require that each cell be “grown” into a larger population. However, since B cells generally cannot be grown into these larger populations in the lab, the classic approach to this problem is the “hybridoma method”, which is literally “fusing” a special type of cancer cell with immune cells obtained from the spleen of an immunized rodent (typically a mouse or a rat). These hybridomas inherit the immortal properties of the cancer cell line (i .e., can be grown) and continue to secrete a single type of antibody. Although this approach has been the workhorse of antibody discovery for decades, it has major limitations: (i) it is generally limited to rodents, (ii) it is slow (taking weeks to achieve sufficient cells to test the secreted antibodies) and (iii) it loses more than 99% of the available antibody diversity since the fusion process has extremely low efficiency, typically between 0.1% and 1%.
Our Approach to Search Natural Antibodies . To solve these challenges, our founders pioneered and developed a nano-liter volume single-cell microfluidic screening technology that provides the sensitivity, resolution and scalability needed to perform a deep search of any antibody response.
We believe our screening approach provides a unique combination of speed, throughput and versatility in searching antibody responses. Our microfluidic single-cell screening technology consists of custom-made microfluidic devices that integrate 256,000 single-cell analysis chambers on a chip about twice the size of a credit card. At the beginning of a screening experiment, a sample of B cells isolated from an immunized animal is flowed into the device at a concentration selected to result in approximately one single cell per chamber. Once cells are loaded, they are isolated in single-cell analysis chambers, each having a volume of less than one nanoliter. In this small volume, approximately 100,000 times smaller than what is used in a conventional bench-top experiment, a B cell secretes enough antibodies within minutes to be detected by microscopy. Using fluidically-controlled reagent and particle additions, a variety of experimental protocols can be executed so that each single cell is interrogated to determine the properties of the antibody it makes. A screening experiment takes several hours and is performed the same day as immune cells are isolated from immunized rodents. This is significantly faster than the weeks required to establish hybridoma cultures.
Our microfluidic devices are operated using proprietary high-throughput imaging instruments that incorporate custom robotics, fluid control, software systems and AI-based image analysis. Each instrument can run two microfluidic chips in parallel for a total of 512,000 chambers per instrument run. In a two-hour run, each of the 512,000 chambers is imaged up to 10 times, resulting in over five million chamber images, or roughly 700 chamber images per second. A representation of our microfluidic screening devices and assay readouts are showing in Figure 7 below.
17
Figure 7: Our Screening Approach for Antibody Responses
Our AI-based image algorithms perform real-time analysis of these images to identify chambers that contain single cells that make antibodies with the desired properties. As depicted in Figure 8 below, our technology supports a wide array of complex image-based single-cell secretion measurements including multiplexed target binding on up to six targets, affinity-based enrichment, multiplexed cell-binding, ligand blocking assays and a selection of functional assays. We currently have throughput to screen more than four million cells per screening day.
Figure 8: Representations of Exemplary Microfluidic Screening Assays
As compared to other technologies we believe our microfluidic screening platform provides unique combined advantages of:
•
Throughput to screen greater than 500,000 cells per instrument run.
•
Speed to go from immune cells to selected antibody-generating single cells in less than a day.
•
Antibody selections based on a wide array of measurements.
•
Capability to perform selections on both protein and cellular targets.
•
Versatility to search antibody diversity from any species or tissue.
18
Find: Automated Single-Cell Sequencing and RepSeq
The collection and interpretation of antibody sequence data presents several challenges. First, because only a small fraction of antibodies are suitable for therapeutic development, methods that are based on sequencing antibodies from single cells before evaluating their function are extremely inefficient, low-throughput and costly. Second, sequencing antibodies from selected single cells is technically challenging due to the very small quantities of starting material and the large number of possible sequences that need to be captured. Third, although bulk sequencing of antibodies can provide a comprehensive view of immune responses, these methods often lose information on the correct natural pairing of heavy and light protein chains, which together comprise an antibody, and provide no means to assess the functional relevance of each antibody.
Our Approach to Find Natural Antibodies. To solve these challenges, we combine our nano-liter volume microfluidic single-cell screening platform, automated single-cell antibody sequencing and immune repertoire antibody sequencing to enable the deep analysis and functional interpretation of antibody responses.
Automated single-cell sequencing. Our microfluidic single-cell screening technology allows us to evaluate the binding and/or functional properties of antibodies made by millions of single cells at the first step of analysis. For each screening run on each instrument, up to 768 individual cells that exhibit desired properties can be recovered into microplates for the following single-cell sequencing steps. Our proprietary sequencing protocols have been optimized to achieve approximately 90% efficiency in the recovery of high-quality heavy and light chain antibody sequences from single cells and have been adapted for discovery from multiple species. To achieve speed, reproducibility and throughput, our single-cell sequencing pipeline has been implemented using robotic automation and automated bioinformatics software. We currently have throughput to process up to 15,360 single cell samples per week and to recover high-quality, chain-paired sequences from 3,840 single-cell samples in three days.
RepSeq. We believe our RepSeq technology, when coupled with our microfluidic single-cell screening and sequencing technologies, provides unique capabilities for expanding the search of natural antibody responses. RepSeq is based partly on foundational RepSeq patents that we have exclusively licensed from Stanford as part of our acquisition of Lineage in 2017. RepSeq uses high-throughput sequencing to perform near-comprehensive profiling of the repertoire of heavy and light chain antibody genes that are present in a sample. In its highest-throughput implementation, a single RepSeq sequencing run can generate approximately 800 million antibody sequences.
Since these are bulk sequences obtained from mixtures of large numbers of cells, the interpretation of these big data sets has multiple challenges. The first is the loss of information regarding which heavy chain is naturally paired with which light chain. The second is the inability to identify which rare antibody sequences are relevant to the program (e.g., bind to the target or have desired function). We solve both of these problems by annotating RepSeq data with functional sequence data derived from our single-cell microfluidic screening and sequencing technologies. Using the sequences of hundreds to thousands of antibodies with known properties, we are able to search RepSeq data from related samples to identify closely related families of antibodies that can be arranged in a lineage to reconstruct their evolution during the immune response. These expanded family trees provide valuable insights for vaccine research and are sources of ready-made alternative therapeutic candidates in cases where an antibody of interest has one or more suboptimal properties.
We have shown that the combination of single-cell analysis and RepSeq can expand the number of therapeutic antibody candidates by more than 10-fold in a single experiment, increasing the number of candidates from hundreds to thousands. Enriching for sequences of value by isolating target-specific immune cells that go into a RepSeq experiment can amplify the number of therapeutic antibody candidates. The combination of single-cell screening and RepSeq allows deep interrogation of an immune repertoire to find the best antibodies. For further explanation, see the case study titled “Human Immune Profiling” under “—Our Technology in Action.” The relationship between single-cell-derived antibodies and family members discovered using RepSeq data is depicted in Figure 9 below.
19
Figure 9: Antibody Lineage Mapping
Analyze: High Throughput Cloning, Expression and Bioanalytics
Our microfluidic single-cell screening and RepSeq technologies are capable of generating hundreds to thousands of unique antibody candidate sequences from single-cell screening, along with thousands of related antibody sequences from RepSeq. These large antibody sets create formidable challenges for efficiently down-selecting to a small number of the best candidates for development, including (i) the need to produce and handle large numbers of high-quality antibodies by “expressing”, or converting sequences into protein antibodies that can be further analyzed, (ii) the need to perform measurements to characterize each antibody property for target recognition, function, and properties related to suitability for drug development, and (iii) the need for data management and computational tools to organize and understand the resulting data.
Our Approach to Analyzing Natural Antibodies . To address these challenges, we have built a high-throughput antibody generation and characterization pipeline that generates high-dimensional data clouds for each antibody candidate.
Our antibody expression pipeline combines optimized molecular biology protocols, proprietary expression vectors and robotics to enable the rapid cloning, expression and purification of recombinant antibodies using expression systems that are representative of current drug manufacturing standards. When starting from single-cell-derived antibody samples, we are able to generate hundreds of recombinant antibodies within eight days of screening. Our platform supports multiple antibody formats and currently has capacity to generate 960 high-purity antibody samples per week.
These antibodies are then tested across a suite of analytical assays to determine their biophysical properties including their purity, binding properties (e.g., specificity, epitope binning, affinity), thermal stability, expression levels and aggregation state. Expressed antibodies may be further characterized in appropriate functional cell-based assays to assess their potency. Corresponding in silico analysis is performed on each antibody to predict potential development liabilities, biophysical properties and immunogenicity.
We believe that by gathering more high-quality data on more antibodies from the start of the discovery process, we can significantly improve the speed, quality and success of antibody discovery.
20
Celium : Computation and Data Exploration
Our technology stack generates vast and complex data sets. In a single discovery program, our technology stack can produce terabytes of data per screen, including:
•
tens of millions of microscopy images from raw screening data
•
hundreds of millions of DNA sequences from raw single-cell sequencing
•
billions of DNA sequences from raw RepSeq data
•
millions of single-cell antibody secretion measurements
•
hundreds to thousands of unique single-cell-derived antibody sequences
•
tens of thousands of related antibody sequences derived from RepSeq
•
several hundred thousand antibody characterization measurements
•
associated meta data for each experiment
The sheer quantity and complexity of these data sets present formidable challenges. First, without specialized data collection, standardization, and storage solutions, data of this scale quickly becomes unmanageable and unusable. Second, finding hidden relationships in these complex data sets requires sophisticated computational tools that must be customized for the questions being asked and the data types and structures used. Finally, even once data has been reduced to the key properties of hundreds of antibodies, it may include hundreds of thousands of data points, making interpretation difficult or impossible for scientists.
Our Approach to Analyzing Complex Antibody Data. To address these challenges, we have built a computational engine called Celium that integrates data collection, standardization and storage with a suite of computational tools and an interactive visualization interface that allows scientists to quickly explore and interpret complex antibody data sets.
Our technology stack integrates software to automatically standardize the collection, storage, and version control of raw and processed experimental data obtained at every step in the discovery process. Data from every experiment is stored in a central database designed to maintain the relationships that exist between different measurement types, samples, and antibodies. Because we do not rely on third-party data, we are able to maintain strict data quality assurance and standardization. We believe this provides a critical advantage that greatly increases the value of data.
We have built a suite of computational tools for extracting information and uncovering relationships that are hidden within our data. Our data handling and report generation software automates standard analyses, and instantly returns essential information that would otherwise take days of work. We have developed machine learning and AI methods to replace manual data analysis, quality assurance and design steps associated with antibody sequencing, protein engineering and antibody chain-pairing. Similarly, our vast image data sets have allowed us to develop AI-based machine vision tools for real-time processing, enabling single-cell screening at much greater speed and resolution. Finally, we are using machine learning algorithms to explore the relationship between antibody sequence space and important drug-like properties including resistance to aggregation, stability and expressibility. We believe that these approaches will become increasingly powerful and predictive as our data sets grow.
We believe the value of data and computation is greatly amplified by intuitive tools that enable scientists to interactively explore and query complex data sets. Celium achieves this with a dynamic visualization interface that presents each unique antibody as a connection between two unique DNA sequences that encode for the heavy and light chain. Celium presents antibodies as a network of these connections that intuitively represents sequence similarity. Using this visual language, scientists can interactively navigate and filter thousands of antibodies in real time, using hundreds of different data features, as shown in Figure 10. This allows scientists to interactively explore the immune repertoire and set search criteria based on multiple features to find antibodies with the precise characteristics desired for a drug candidate.
21
Figure 10: Celium Detailed View of Heavy and Light Chain Pairing
Antibodies of interest can be further explored to evaluate sequence features and to expand diversity by linking to associated RepSeq data as shown in Figure 11. By integrating RepSeq data, Celium has the power to map antibody lineages to identify lead candidates likely to have improved binding and functional properties. This allows for expansion of diversity around antibodies of interest to identify new drug candidates for testing and development.
22
Figure 11: Celium Chain ID Clonal & Rep Seq Lineage View
We believe Celium is a unique and powerful tool that enables rapid exploration of multiple data sets in hours that would otherwise take many weeks to search. It provides an elegant human interface that allows scientists to quickly explore data and gain insights that inform action.
Engineer: Advanced Computational Protein Engineering Toolkit
Using natural diversity to accelerate engineering. In many cases, antibodies selected for development need to be modified before they are developed as drugs. These modifications are generally made through a combination of computational design and experimental testing, a process known as antibody engineering. Examples of antibody engineering problems include improving how tightly an antibody binds to its target; removing antibody sequences known to be problematic during manufacturing; modifying non-human antibodies to resemble human antibodies; and removing antibody sequences that may induce immune responses in patients.
The central challenge in antibody engineering is that the relationship between DNA sequence modifications and the resulting antibodies is not well understood. Antibody engineering therefore relies heavily on trial and error to identify DNA changes that give the desired phenotypic results. This process is complicated by the fact that optimizing for one property, such as binding affinity, may cause the loss of another desirable property, such as solubility or expression. This challenge is amplified when the source antibodies need major improvements, as is the case for synthetic antibodies that typically require multiple rounds of antibody engineering to improve their binding affinity, biophysical properties, or both.
Our Approach to Engineering Natural Antibodies . We address these challenges by first generating large, diverse and high-quality panels of candidate antibodies early in the discovery process and, second, by applying antibody engineering approaches that are informed by natural antibody responses.
We believe the best approach to protein engineering is to start with a large panel of the best possible candidate antibodies. To achieve this, we apply our microfluidic single-cell screening platform to maximize the diversity of antibodies selected upfront for multiple target-binding properties, followed by thorough characterization of their functional properties and developability. We then apply stringent filtering with the aim of down-selecting to a small number of leads. When performing discovery from humanized rodents, including our proprietary suite of Trianni humanized rodents, or from humans, we have used this approach to generate high affinity antibodies that require minimal engineering prior to development. Starting with a greater diversity of antibodies, which have been pre-selected for desirable properties and that benefit from natural immune responses, can significantly reduce development time and the technical risk of protein engineering.
23
In cases where antibody properties need to be improved, we can use expanded panels of antibody sequences from RepSeq to inform the design and generation of high-confidence optimization candidates. We believe this approach, to use natural antibody variants from the repertoire to help design optimized candidates, is particularly powerful for high-value membrane protein targets such as GPCRs and ion channels (which cannot be optimized by conventional display-based methods). We also believe it can significantly improve the success-rate and reduce the time needed to optimize development leads.
OrthoMab Bispecific Platform . Beyond improving antibody properties, antibody engineering can also build completely new antibody drug formats with novel molecular geometry, binding properties or chemical properties. Of particular interest is the combination of two source antibodies to create a “bispecific” antibody that can simultaneously bind to two targets. Antibodies are normally comprised of two identical heavy chains and two identical light chains, to make a symmetrical “mirror-image” molecule with two identical targeting arms. In contrast, bispecific antibodies are generally comprised of two different heavy chains and light chains, and therefore have targeting arms that recognize two different targets. Because of their unique properties, bispecific antibodies are a rapidly emerging new class of antibody therapy. As of December 31, 2020, following the first approval in 2015, there are now more than 120 molecules in clinical development, including more than 80 in Phase 1. They enable improved and novel therapeutic mechanisms not possible with other modalities. Examples of the application of bispecific antibodies include (i) recruitment of immune cells to help kill cancer cells, (ii) linking together two receptors to activate a signaling pathway, (iii) serve as a protein scaffold to bring proteins together and (iv) modifying the pharmacokinetics and pharmacodynamics of soluble proteins, as depicted in Figure 12 below.
Figure 12: Therapeutic Modalities Using Bispecific Antibodies
To realize this potential, it is important to develop capabilities to:
•
generate a large and diverse panel of starting antibodies that recognize multiple epitopes (locations on their respective targets), with varying binding affinities and binding geometries.
•
identify suitable pairs from this starting panel that bind with the right orientation and the right location on each respective target, and bind with suitable affinity
•
manufacture bispecific antibodies in a scalable and efficient process
•
ensure the resulting bispecific antibody looks as similar as possible to a normal antibody so as not to induce an immune response in the patient.
Our Approach to Engineering Bispecific Antibodies. OrthoMab, addresses these challenges by enabling the combination of any two source antibodies into a bispecific antibody that can be manufactured using conventional expression and purification methods, with minimal liabilities and immunogenicity.
OrthoMab is a clinically validated protein engineering technology that enables the creation of a bispecific antibody from any two source antibodies, each comprised of a unique heavy chain and a unique light chain. The key innovation of OrthoMab is a set of patented DNA mutations that have been computationally designed using molecular structural modeling. These mutations ensure that the four antibody chains pair correctly: the two different heavy chains preferentially associate together, rather than two molecules of the same heavy chain, and each light chain pairs only with its cognate heavy chain as shown in Figure 13 below.
24
Figure 13: Patented OrthoMab DNA Mutations
By using engineered mutations to control chain-pairing, OrthoMab enables manufacturing of bispecifics at high yields and purities using industry-standard processes. Because each side of an OrthoMab bispecific is 99% identical to the source antibody, it lowers the risk of introducing immunogenic epitopes. Finally, in addition to conventional Y-shaped bispecific antibodies, OrthoMab allows for the creation of a wide array of alternative bispecific formats as shown in Figure 14. The OrthoMab technology was patented by scientists at Lilly and the University of North Carolina at Chapel Hill. We acquired non-exclusive rights to use OrthoMab technology from Dualogics, LLC, through an asset purchase agreement in July 2020.
Figure 14: Exemplary Bispecific Formats Achievable with OrthoMab
Our Technology in Action
COVID-19: From Discovery to Clinic in 90 Days
We have been working with DARPA since March 2018 as part of the P3 program to optimize our technology stack to find effective and field-ready therapeutics against pathogenic threats in record time.
Problem. An effective response to a pathogenic threat requires comprehensive deep screening and characterization of a human antibody response at the maximum speed possible, so that pathogen-specific therapeutics can be quickly identified, developed and deployed. Ideally, samples from index patients (patients who are the first to have been confirmed infected with the pandemic virus) who recovered would be made available to deploy our pandemic response platform.
Solution. We developed rapid antibody screening, expression, purification and characterization pipelines to deeply mine human antibody responses. As part of the P3 program, and prior to COVID-19, we pressure tested our technology stack twice in simulated pandemic responses. In late 2018, we demonstrated rapid isolation of hundreds of Middle Eastern Respiratory Syndrome Coronavirus, or MERS-CoV, heavy chain only antibodies, or HcAbs, from infected camelids (a natural host for MERS-CoV) in less than 96 hours from sample receipt. Many of these HcAbs were more potent neutralizers than benchmark antibodies. In early 2019, together with our partners, we discovered influenza-neutralizing antibodies from a single sample from a human donor, and demonstrated that we could deploy our platform from sample receipt to successful testing in animals in 55 working days. Our seven lead antibodies were all 100% protective against a 20-times lethal dose of the 2009 pandemic H1N1 strain of influenza virus in rodents.
Result. We rapidly deployed our pandemic response platform to find a therapeutic antibody against COVID-19 in the spring of 2020 starting from a blood sample obtained from a U.S. patient. We screened approximately 5.8 million single cells to identify over
25
500 unique anti-SARS-CoV-2 antibodies. Each of these antibodies was evaluated computationally and experimentally to identify approximately 500 different properties per antibody which yielded 220,000 data points, which allowed us to filter down to a smaller group of lead candidates. Within 23 days of receiving the sample, we and our partners identified 24 lead antibodies for further development and clinical testing. One antibody drug candidate was selected by our partner Lilly, and the first patients were dosed in the first-ever COVID-19 clinical trial in North America. This was only 90 days from when we received the sample. The antibody, bamlanivimab , has been evaluated both alone and together with another antibody called etesevimab in more than 5,000 patients across multiple clinical trials for the treatment and prevention of COVID-19. Bamlanivimab and together with other antibodies has been used to treat approximately a million high-risk COVID-19 patients. This demonstrates that a rapid discovery-to-clinic timeline is possible with our rapid and high throughput discovery engine.
Unlocking High-value Membrane Proteins (GPCRs and Ion Channels)
Membrane proteins such as GPCRs and ion channels are a validated and highly valuable class of drug targets in multiple prevalent diseases and indications, including cardiovascular diseases, cancer, neurological diseases, inflammation and pain. Membrane proteins are very difficult targets for antibody therapeutics. They are large and complex proteins, often with multiple subunits embedded in the cell membrane with only a small portion exposed outside of the cell. They also exist in families of multiple, closely-related members. The specificity of antibody drugs against membrane proteins would solve the off-target side effects that make many of these targets a barrier for small molecule drugs. However, membrane proteins present major challenges for antibody discovery: (i) they are poorly immunogenic, (ii) they are very difficult to purify and handle, (iii) it is difficult to generate binding or functional assays for them and (iv) they often have high homology with other species that are traditionally used for immunization campaigns (such as rodents), leading to natural tolerance mechanisms inhibiting a good immune response. In addition, legacy screening technologies either do not provide the depth and throughput to effectively search the immune response (hybridoma strategy), or are poorly suited for screening against complex membrane protein targets (display strategy).
Problem. Using conventional screening methods, a partner had failed to discover functional antibodies against a very small epitope of a GPCR protein target with close structural homology to a related protein.
Solution . For this campaign, we tailored our technology stack to include multiplexed live cell screening assays with counter-screens against the closely related homologs in order to identify unique antibodies against the target epitope. The GPCR was expressed in its natural conformation on live cells, bypassing the need for complicated protein purification and handling. The ability to screen against live cells at such high throughput, while simultaneously screening for specificity against closely related homologs, is a powerful application of our platform.
Result. Within three months, we identified hundreds of unique antibodies against the target epitope, including several that were the desired functional antagonists. This number of antibodies from a screening campaign against a GPCR, plus the high frequency of functional candidates, is a significant success, and a lead candidate was subsequently identified by our partner and advanced into IND-enabling studies. This project demonstrates the power of our technology stack to increase the probability of success for antibody discovery campaigns against difficult targets.
Overcoming Tolerance with Proprietary Immunization and Deep Search Technologies
Antibodies produced by immunized animals are subject to immune tolerance, the process by which the body suppresses antibodies that react to self-antigens. This makes it difficult to generate diverse sets of antibodies against human targets that are similar or identical to the analogous proteins expressed by the animal being immunized. The difficulty in generating immune responses against self-similar targets has long been perceived as a drawback of discovery from natural immune repertoires. Our technology and proprietary immunization approaches allow us to generate diverse antibodies against such targets.
Problem. A partner approached us with a human target with 100% homology across mice, rats and humans. In addition, the human target had two related protein homologs, against which any discovered antibodies must discriminate.
Solution. In this campaign we deployed our proprietary and optimized immunization protocols to first generate a robust immune response in rodents. Next, we developed a high throughput single-cell screening strategy that used multiplexed fluorescence detection to find antibodies specific to the target, but that did not bind the two homologous proteins.
Result. We screened four million single cells from the immunized rodents and identified more than 1,900 target-specific antibodies, a hit-frequency that demonstrates a robust immune response breaking tolerance against a 100% homologous target. Of these hits, we identified 428 unique antibody leads with a degree of somatic hypermutation that indicates a mature and directed immune response against a difficult target.
26
Single Chain Antibody Discovery from Camelids
HcAbs are single-chain antibodies lacking light chains, found naturally in camelids such as llamas and alpacas (along with conventional paired heavy and light chain IgG antibodies, at approximately 50% frequency). HcAbs are valuable for various antibody-based therapeutics, such as bispecific antibodies, antibody-drug conjugates and CAR-Ts. In addition, the decreased complexity of having only single chains comprising the antibody molecules means they are more straightforward to produce and manufacture.
Problem. A partner needed to identify HcAbs from immunized llamas against transforming growth factor beta-3, or TGF-ß3, that could also discriminate between closely related homologs, transforming growth factor beta-1, or TGF-ß1, and transforming growth factor beta-2, or TGF-ß2. The partner did not have the screening technology to identify antibodies with such restricted specificity.
Solution. We designed custom reagents to distinguish HcAbs from conventional IgGs in llamas and also designed custom assays to identify HcAbs specific to TGF-ß3 that could differentiate between TGF-ß1 and TGF-ß2. To find these rare antibodies, we deployed our deep screening platform and screened approximately 20 million single cells over four days to deeply mine the immune response. We identified 67 unique HcAbs, which is a <0.001% antigen specific HcAb hit frequency, indicating an exceedingly rare antibody. This illustrates the flexibility of our platform to discover non-conventional antibody modalities, from alternative species, and the depth required to find rare antibodies with very specific properties.
Human Immune Profiling
An important application of immune profiling is devising improved strategies for the prevention and treatment of viral pathogens such as influenza. Influenza is a recurring seasonal epidemic with major pandemic potential. A significant challenge in addressing influenza is that it is constantly changing. Seasonal strains that circulate in a population accumulate mutations that are influenced by the existing population immunity, resulting in the emergence of new strains for which existing vaccines are less effective. Even more concerning, novel strains can sometimes emerge when viruses that normally infect only animals rearrange their genes in a way that allow them to jump to the human population. Functional profiling human immune responses to influenza infection or vaccination may assist in developing antibody or vaccine products that help protect against serious influenza infections.
Problem. Deep sequencing technologies have been applied to broadly survey the diversity of antibody sequences generated during an immune response. However, such technologies do not indicate which antibodies bind to the target of interest, where they bind on the target or which are most protective against infection.
Solution. To identify antibodies against influenza that may be effective in passive immunotherapy treatments (highly specific and potently neutralizing) and vaccine development (recognizes multiple strains of influenza to inform vaccines that generate long-lasting immunity), we screened four million single cells from multiple human donors. We designed a custom multiplexed screening strategy to simultaneously profile antibodies for their binding profiles against four hemagglutinin proteins derived from H1N1, H2N3, H3N2 or H5N1 influenza strains. We then recovered single cells with desired binding properties for sequencing to determine their heavy and light chain sequences. For selected antibodies we then searched RepSeq data for related antibody sequences.
Results. Our influenza screen uncovered 19,920 influenza-specific antibodies, from which we recovered and sequenced 3,646. This resulted in 1,743 unique antibodies grouped within 860 clonal lineages. Through the combination of single-cell sequences and RepSeq data, we were able to construct detailed antibody lineages for virus-specific antibodies, and then to expand the number of selected virus-specific antibodies by approximately 50 times.
Research and Development—Platform Expansions
We have several active research efforts to expand the breadth and depth of our technology stack. In addition to research and development directed to improve the speed, efficiency, throughput and capabilities of our existing technologies, we have initiated the following platform development projects:
GPCR and Ion Channel Targets. We believe our immunization and screening platform provide us with a competitive advantage in the discovery of antibody against traditionally difficult multi-pass transmembrane proteins, including GPCRs and ion channels. To date we have successfully applied our technology to discover panels of hundreds of antibodies against these target classes, with the most advanced of these programs now at late-stage preclinical development. However, we believe further improvements at the Source and Engineer steps of our technology stack would allow us to more fully unlock the potential of these target classes. To this end, in 2019 we started a research and development group, Channel Bio, in Sydney Australia, that is developing technologies specifically for GPCR and ion channel targets. In September 2021 we also acquired TetraGenetics Inc., a biotechnology company with a proprietary platform for generating recombinant human ion channels and other transmembrane proteins.
27
Transgenic rodents. Our recent acquisition of the Trianni platform of humanized rodents, plus the integration of key research and development personnel from Trianni , enables us to generate novel next-generation humanized transgenic rodents to further our platform and offering to partners. The suite of humanized rodents available or in development will serve as the foundation for additional engineering to create novel humanized rodents.
Cell Line Development. Using our microfluidic single-cell screening platform and based on patented clonal selection methods that we have exclusively licensed from UBC, we are developing optimized workflows that we believe may accelerate the development of clonal cell lines with increased productivity in the expression of antibodies.
CMC and GMP Antibody Manufacturing. We are planning a further facilities expansion of approximately 200,000 square feet that will support cell line development, process development and GMP manufacturing of antibody therapeutics. Upon completion of this facility, we expect to be able to support our partners from program initiation to fill-finish. We believe the integration of an optimized manufacturing process with our discovery and protein engineering capabilities will create synergies in speed and efficiency and will allow us to more rapidly test and validate new antibody therapeutic formats, including bispecific antibodies or antibody conjugates. We expect to have completed this facility and to have GMP manufacturing capabilities in commercial use in approximately three to four years. In April 2020, in support of this effort, we received a commitment for up to CAD $175.6 million ($125.6 million) in financing from the Canadian government.
Competition
The market for technologies that enable the discovery and development of therapeutic antibodies, such as ours, is global, characterized by intense competition and subject to significant intellectual property barriers. The solutions and applications offered by our competitors vary in size, breadth and scope, and given the broad promise of antibody therapeutics, we face competition from many different sources, including companies developing single-cell screening technologies, antibody RepSeq and antibody engineering technologies, using a variety of business models, including the development of internal pipelines of therapeutics, technology licensing, and the sale of instruments and devices. We also face competition from integrated contract research organizations that use traditional hybridoma, phage, and yeast display technologies in discovery. Due to the significant interest and growth in antibody therapeutics more broadly, we expect the intensity of this competition to increase.
We are democratizing the industry by providing our partners of all sizes with access to our centralized operating system We seek to deliver a complete solution for our partners by providing uniquely integrated proprietary technologies that address each step in the discovery process, including immune RepSeq, single-cell analysis, AI, and transgenic rodent platforms. Many emerging and established life sciences companies have been built around technologies that focus on one or a limited number of these steps. Examples include:
•
In the field of single-cell screening, we face technical competition from companies that provide access to similar technologies such as Berkeley Lights Inc., or Berkeley Lights, HiFiBio Inc., Ligand Pharmaceuticals Inc. and Sphere Fluidics Ltd.
•
In antibody RepSeq, we face technical competition from companies that provide access to similar technologies such as 10X Genomics Inc., Adaptive Biotechnologies Corp., Atreca Inc. and Distributed Bio Inc. (acquired by Charles River Laboratories in 2021)
•
In bispecific antibody engineering, we face technical competition primarily from companies that provide access to similar technologies such as Abbvie Inc., Genmab A/S, Merus N.V. and Zymeworks Inc.
•
In discovery using genetically engineered rodents, we face technical competition from companies that provide access to similar technologies such as Ablexis LLC, Crescendo Biologics Ltd., Harbour Antibodies BV, Kymab Ltd., Ligand Pharmaceuticals Inc. and RenBio Inc.
We also face direct business competition from companies that provide antibody discovery services using technologies such as hybridoma and display. Companies with discovery business models that include downstream payments include Adimab LLC, Distributed Bio Inc. (acquired by Charles River Laboratories in 2021) and WuXi Biologics Inc. In addition, we compete with a variety of fee-for-service contract research organizations that provide services, in most cases using legacy technologies, that compete with one or more steps in our technology stack.
For a discussion of the risks we face relating to competition, see “Risk Factors—Risks Related to our Business and Strategy—The life sciences technology market is highly competitive, and if we cannot compete successfully with our competitors, we may be unable to increase or sustain our revenue, or achieve and sustain profitability.”
28
Intellectual Property
We strive to protect the proprietary technologies that we believe are important to our business, including seeking and maintaining patent protection intended to cover the compositions of matter of our product candidates, their methods of use, related technology, and other inventions that are important to our business.
Our success depends in part on our ability to obtain and maintain intellectual property protection for the components of our technology stack and products arising from the same; to defend and enforce our patents, to preserve the confidentiality of our trade secrets, and to operate without infringing valid and enforceable patents and other proprietary rights of third parties; and to identify new opportunities for intellectual property protection.
As of December 31, 2021, we owned or exclusively licensed over 70 issued or allowed patents and over 70 pending patent applications worldwide, which includes over 30 issued U.S. patents and over 20 pending U.S. patent applications. We own registered trademarks and trademark applications for AbCellera, Channel Bio, Celium, Trianni, the Trianni Mouse and TetraGenetics in the U.S., Canada, Australia and Europe.
Obtaining patent protection is not the only method that we employ to protect our propriety rights. We also utilize other forms of intellectual property protection, including trademark, copyright, internal know how and trade secrets, when those other forms are better suited to protect a particular aspect of our intellectual property. Our belief is that our propriety rights are strengthened by our comprehensive approach to intellectual property protection. It is our policy to require our employees, consultants, outside scientific collaborators, sponsored researchers and other advisors to execute confidentiality and invention assignment agreement upon accepting employment or consulting relationships with us. These agreements provide that all confidential information concerning our business or financial affairs developed or made known to the individual during the course of the individual’s relationship with us is to be kept confidential and not disclosed to third parties except in specific circumstances. In the case of employees, the agreements provide that all inventions conceived by the individual, and which are related to our current or planned business or research and development or made during normal working hours, on our premises or using our equipment or proprietary information, are our exclusive property. We are diligent in taking precautions that our proprietary information is not released to third parties through the use of security measures. Our trade secrets encompass certain reagent compositions and concentrations, nucleic acid vector sequences and immunization protocols.
Data Rights
Our product to partners is data on the composition of matter of antibodies and their properties. We enter into contracts that allow us rights to use the data that we generate for the purpose of improving our technology stack and fueling machine-learning algorithms. We maintain strict firewall protocols so target-specific data derived from a client cannot be used to inform the discovery on another project by a different client.
Patent Portfolio
We have developed an expansive patent portfolio with claims related to multiple aspects of our technology stack, beginning with our first patent applications exclusively licensed from UBC, in 2013. We continuously assess new ways to improve our technology platform through license or acquisition of third-party patent portfolios, as was the case with our acquisitions of Lineage in 2017 and the OrthoMab platform from Dualogics in 2020, our acquisition of Trianni Inc. in 2020, our acquisition of TetraGenetics, Inc. in 2021, and our license agreement with Alloy Therapeutics in 2020.
Our patent prosecution strategy encompasses the pursuit of protection for our technology stack and tangentially related methods.
UBC License
In December 2013, we executed a license agreement with UBC, or the UBC License, to gain a worldwide, exclusive license to certain patents, or the UBC Patents, patented at UBC by Dr. Hansen and his team for the later of 20 years from the start date of the UBC License, or the expiry date of the last patent licensed under the UBC License. Under the terms of the UBC License, we have the right to sublicense a subset of the UBC Patents and a worldwide, exclusive license to UBC Improvements and/or Joint Improvements on these Patents solely in the antibody field of use. In addition, for a second subset of the UBC Patents, we have a worldwide, exclusive license to use and sublicense solely within the antibody field of use.
Under the terms of the UBC License, we paid a CAD $0.1 million ($0.1 million) initial license fee and pay annual license fees to UBC during the term of the UBC License. We also pay UBC a low single-digit royalty on our revenue and a single-digit royalty of our sublicensing revenue during the term of the UBC License. UBC was also granted a single-digit percent equity position in our company.
29
Under the terms of the UBC License, in consultation with UBC we manage the filing, maintenance and prosecution of the licensed patents and we pay all costs associated with the same while we control all litigation associated with the licensed patents.
UBC may terminate the license under certain circumstances, including in the case of our insolvency, winding up or liquidation, if a court or similar process is levied on the rights under the agreement or on money due to UBC that is not released, if the subject technology becomes subject to a security interest that is not released, if we or any of our directors or officers have materially breached or failed to comply with securities laws, or in the event of certain breaches of, or failure to perform, our obligations under the license or other agreements between us and UBC. Either party may terminate the license for any breach which is not remedied within certain specified time periods.
The UBC Core Patents
The UBC Core Patent license includes a patent family directed toward certain systems, devices and methods for microfluidic cell culture. This patent family includes four issued U.S. patents and one pending U.S. non-provisional patent application. Issued patents from this family are expected to expire in July 2031, absent any disclaimers or extensions available.
The UBC Core Patent license also includes a patent family directed toward systems and methods for assaying binding interactions between a protein produced by a single cell, e.g., an antibody produced by a single B cell, and a second biomolecule (e.g., antigen) in microfluidic chambers and devices. This patent family includes twelve issued U.S. patents and three pending U.S. non-provisional patent applications. Issued patents from this family are expected to expire in July 2031, absent any disclaimers or extensions available.
A patent family directed toward methods for assaying functional properties exhibited by a protein produced by a single cell, e.g., an antibody produced by a single B cell, and a second biomolecule (e.g., antigen) in microfluidic chambers and devices is also included in the UBC Core Patent license. This patent family includes patents issued in the U.S. and Australia and granted in Europe, as well as one pending U.S. non-provisional patent application and seven pending foreign counterpart patent applications. Issued patents from this patent family are expected to expire in March 2034, absent any disclaimers or extensions available.
Lastly, the UBC Core Patent license includes a patent family directed toward methods for determining lymphocyte receptor chain pairs, for example, antibody heavy and light chain pairs. This patent family includes an issued U.S. patent and a granted patent in Europe, as well as one pending U.S. non-provisional patent application and two pending foreign counterpart patent applications. Issued patents from this patent family are expected to expire in May 2035, absent any disclaimers or extensions available.
Lineage
The Lineage patent portfolio complements our single-cell microfluidic intellectual property with downstream methods of sequencing reaction preparation, immune RepSeq and analysis. The immune repertoire patents and applications that we obtained from Lineage form the basis for the sequencing technologies that we currently use in our technology stack.
Stanford License
Through our acquisition of Lineage, we obtained an exclusive license from Stanford University to patents and patent applications directed toward immune RepSeq. Our Stanford license includes one patent family directed toward methods of characterizing an immune repertoire. This patent family includes four issued U.S. patents and one granted patent in Europe, as well as one pending U.S. non-provisional patent application and one pending foreign counterpart patent application. Issued patents from this patent family are expected to expire in May 2031, absent any disclaimers or extensions available. The Stanford license also includes another patent family directed toward methods of characterizing immune response and vaccine selection. This patent family includes two issued U.S. patents and one pending U.S. non-provisional patent application. Issued patents from this patent family are expected to expire in February 2034, absent any disclaimers or extensions available.
Under the terms of the Stanford license, we are required to pay Stanford a yearly license maintenance fee, as well as certain milestone payments in an aggregate amount not to exceed $0.1 million. We are also required to pay Stanford low single-digit royalties on net sales of licensed products as well as a portion of non-royalty sublicensing revenues. The term of the Stanford license runs until the last licensed patent expires, and our obligation to pay royalties will continue so long as there is a valid claim of a licensed patent. Stanford may terminate the agreement governing the license if we are in material default in the provision of any report or payment of any amounts due to Stanford under the agreement, we do not use commercially reasonable efforts to develop or commercialize licensed products, we do not achieve certain diligence milestones, we are in material breach of any provision of the agreement, or if we provide any materially false report to Stanford. We may terminate the agreement at any time upon at least 30 days notice to Stanford.
30
In addition to the Stanford license, the acquisition of Lineage included a patent portfolio comprising four patent families. One patent family is directed toward methods of determining the immune repertoire of a subject. This patent family includes one granted patent in Europe, two pending U.S. non-provisional patent applications, and four pending foreign counterpart patent applications. Issued patents from this patent family are expected to expire in March 2034, absent any disclaimers or extensions available.
Another patent family is directed toward tagging target oligonucleotides. This patent family includes two issued U.S. patents, one issued patent in China, and two granted patents in Europe. This patent family also includes one pending U.S. non-provisional patent application and one pending foreign counterpart patent application. Issued patents from this patent family are expected to expire in March 2034, absent any disclaimers or extensions available.
An additional patent family is directed toward methods for detection of isotype profiles as signatures for disease. This patent family includes patents issued in Japan and China, as well as a patent granted in Europe. This patent family also includes three pending foreign counterpart patent applications. Issued patents from this patent family are expected to expire in September 2032, absent any disclaimers or extensions available.
Lastly, the Lineage patent portfolio includes a patent family directed toward compositions and methods for analyzing heterogeneous samples. This patent family includes a granted patent in Europe and an issued patent in Hong Kong. This family also includes one pending U.S. non-provisional patent application and three pending foreign counterpart applications. Issued patents from this patent family are expected to expire in September 2032, absent any disclaimers or extensions available.
OrthoMab
As part of our agreement to purchase certain assets from Dualogics related to its OrthoMab bispecific antibody platform, we were assigned Dualogics’ interests and rights to that certain Exclusive License Agreement between Dualogics and the University of North Carolina at Chapel Hill, effective February 22, 2019, or the UNC Agreement. Under the UNC Agreement, we have an exclusive license to UNC’s rights under three patent families.
One patent family is directed toward methods of producing a fragment, antigen binding (Fab). This patent family includes three issued U.S. patents and one patent granted in Europe. Issued patents from this patent family are expected to expire in March 2034, absent any disclaimers or extensions available.
Another patent family is directed toward IgG bispecific antibodies and processes for preparation. This patent family includes one issued U.S. patent, one pending U.S. non-provisional patent application, and one foreign counterpart patent application. Any patents that issue from this patent family are expected to expire in January 2036, absent any disclaimers or extensions available.
The last patent family is directed toward methods for producing Fabs and IgG bispecific antibodies. This patent family includes one pending U.S. and one pending foreign counterpart patent applications. Any patents that issue from this patent family are expected to expire in December 2037, absent any disclaimers or extensions available.
Under the terms of the OrthoMab asset purchase, we granted Dualogics a sublicense under the three patent families to develop, market, sell and otherwise commercialize its existing programs related to the OrthoMab technology.
Under the terms of the UNC Agreement, we are required to pay UNC an annual license maintenance fee, low single-digit royalties on net sales of clinically approved and other products as well as sublicense fees. The term of the license and our obligation to pay royalties runs until the last licensed patent expires. UNC may terminate the agreement governing the license if there is a material breach by us of the agreement and we fail to cure such breach, which breaches include but are not limited to our failure to deliver payment to UNC when due, to provide progress reports, to meet or achieve performance milestones or to possess and maintain insurance, or the execution of a sublicense that complies with the terms of the agreement. We may terminate the agreement at any time upon at least 60 days notice to UNC.
Trianni
Through our acquisition of Trianni Inc., we acquired all existing intellectual property including issued patents and pending applications worldwide relating to the flagship Trianni mouse and new platforms in development. We also acquired Trianni’s trademarks including the terms “Trianni” and “Trianni Mouse” that have been issued in the United States and various other jurisdictions worldwide.
The Trianni intellectual property portfolio includes issued patents and pending applications in the U.S. and certain jurisdictions around the world.
31
In one patent family, the patents are directed to transgenic animals and methods of use. This patent family includes eight issued patents including in the U.S., Australia, the Russian Federation, Europe, India, and Japan. There are three pending applications, one in the U.S., one in Canada and one in Japan. Patents issuing from this family are expected to expire in July 2031, absent any disclaimers or extensions available.
Another patent family is directed to enhanced production of immunoglobulins. This patent family includes seven pending applications including one in the U.S and six in pending foreign counterparts including Australia, Canada, Europe, Israel, Japan and Korea. Any patents that issue from this family are expected to expire in February 2037, absent any disclaimers or extensions available.
Another patent family is also directed to enhanced production of immunoglobulins. This patent family includes eight pending applications including one in the U.S. and six in pending foreign counterparts including Australia, Canada, Europe, Israel, Japan, China and Korea. Any patents that issue from this family are expected to expire in August 2035, absent any disclaimers or extensions available.
Another patent family is directed to enhanced immunoglobulin diversity. This patent family includes one issued patent in the U.S. and two pending applications including one in the U.S. and one in Europe. Issued patents from this family are expected to expire in December 2035, absent any disclaimers or extensions available.
Another patent family is directed to transgenic mammals that express canine-based immunoglobulins. This patent family contains one issued U.S. patent and one pending application in the U.S. Issued patents from this family are expected to expire in July 2031, absent any disclaimers or extensions available.
Another patent family is directed to transgenic mammals that express bovine-based immunoglobulins. This patent family contains one issued U.S. patent. Issued patents from this family are expected to expire in July 2031, absent any disclaimers or extensions available.
Another patent family is directed to enhanced production of immunoglobulins. This patent family includes one pending application in the U.S. Any patents that issue from this family are expected to expire in August 2035, absent any disclaimers or extensions available.
Another patent family is directed to transgenic mammals that express canine-based immunoglobulins. This patent family contains two pending applications, one in the U.S. and one PCT application. Issued patents from this family are expected to expire in July 2039, absent any disclaimers or extensions available.
Another patent family is directed to transgenic mammals that express bovine-based immunoglobulins. This patent family contains one pending PCT application. Issued patents from this family are expected to expire in July 2039, absent any disclaimers or extensions available.
Another patent family is directed to single chain VH and heavy chain antibodies. This patent family contains seven pending applications including one in the U.S. and Canada, Australia, China, Europe, Israel and Japan. Issued patents from this family are expected to expire in July 2037, absent any disclaimers or extensions available.
Another patent family is directed to long germline DH gene and long HCDR3 antibodies. This patent family contains two pending applications including one in the U.S. one in Europe. Issued patents from this family are expected to expire in October 2037, absent any disclaimers or extensions available.
Another patent family is directed to transgenic mammals and methods of use. This patent family contains two pending applications including one in the U.S. and one in Europe. Issued patents from this family are expected to expire in August 2039, absent any disclaimers or extensions available.
AbCellera
We also aim to continue developing our product portfolio. We currently own several recently filed pending U.S. non-provisional patent applications directed toward methods for high throughput screening of multispecific antibody libraries and anti-coronavirus antibodies and methods of use.
The term of individual patents depends upon the legal term of the patents in the countries in which they are obtained. In the countries in which we file, the patent term is 20 years from the earliest non-provisional filing date, subject to any disclaimers or
32
extensions. The term of a patent in the United States can be adjusted due to any failure of the United States Patent and Trademark Office following certain statutory and regulation deadlines for issuing a patent.
In the United States, the patent term of a patent that covers an FDA-approved drug may also be eligible for patent term extension, which permits patent term restoration as compensation for a portion of the patent term lost during the FDA regulatory review process. The Hatch-Waxman Act permits a patent term extension of up to five years beyond the original expiration of the patent. The protection provided by a patent varies from country to country, and is dependent on the type of patent granted, the scope of the patent claims, and the legal remedies available in a given country.
For a discussion of the risks we face relating to intellectual property, see “Risk Factors—Risks Related to our Intellectual Property—If we are unable to obtain and maintain sufficient intellectual property protection for our technology, including our platform and Celium, our proprietary antibody visualization software, or if the scope of the intellectual property protection obtained is not sufficiently broad, our competitors could develop and commercialize technologies or a platform similar or identical to ours, and our ability to successfully sell our data packages may be impaired.”
Commercial
Our partnership agreements commonly include: (i) payments for technology access and performance of research; (ii) downstream payments in the form of clinical and commercial milestones and (iii) royalties on net sales of therapeutics. We also participate in agreements with alternative investment opportunities including equity in our business partner and various rights for deeper involvement in moving molecules forward. We structure our agreements in a way that directly aligns our partners’ economic interest with our own. We believe a long-term value of our business will be driven by downstream milestone and royalty payments.
We forge partnerships with drug developers of all sizes, from large cap pharmaceutical to small biotechnology companies. Our partners are predominantly based in the United States and Europe. As of December 31, 2021, we had a total of 36 partners for whom we were conducting antibody discovery activities. For the year ended December 31, 2020, three of our partners accounted for 35%, 25%, and 14% of our research fees revenue and eight partners accounted for the remaining 26% of research fees revenue. For the year ended December 31, 2020, we recognized our first milestone and royalty revenue streams, totaling $213.3 million, exclusively from our partnership with Lilly. For the year ended December 31, 2021, four of our partners accounted for 29%, 17%, 11%, and 10% of research fees revenue and twelve partners accounted for the remaining 33% of research fees revenue. For the year ended December 31, 2021, we recognized milestone revenue of $7.0 million and royalty revenue of $327.3 million, exclusively from our partnership with Lilly. Our partnership with Lilly constituted one of the partnerships that generated 10% or more of our consolidated revenues during the one or more periods described above. With respect to the other partners, we do not believe the loss of any one or more of such partners would have a material adverse effect on us and our subsidiaries taken as a whole.
Over the past year we have grown our workforce by 87%, moving from 206 to 386 full-time employees. As of December 31, 2021, we had 386 full-time employees in Canada, the United States and Australia, which was comprised of approximately 49% scientists, 32% business professionals, and 19% engineers and data scientists.
Our strategy involves:
•
creating more value with our existing partnerships by seeking to expand our single-program partnerships to multi-year, multi-target agreements
•
working to forge new partnerships across our target customers, including large pharmaceutical companies, biotechnology companies of all sizes and non-profit and government organizations dedicated to drug development
•
adding capabilities and infrastructure to support full CMC activities and GMP manufacturing to provide our partners with a full solution from target to IND submission
•
investing in expanding our workforce and our facilities, and increasing efficiency through automation and software solutions
•
maintaining our competitive advantage by undertaking extensive research and development to amplify and add capabilities in areas such as computation, protein engineering, immunization technologies, genetically engineered rodents and cell line selection
•
leveraging proprietary data sets obtained from our partnership programs and AI to increase the efficiency, speed, and capacity of our discovery programs
Our sales and marketing team is growing in proportion to the needs of the business and has been complemented with research and development staff attending a variety of scientific conferences, which has helped increase the business development pipeline. We
33
plan to further expand our commercial sales, marketing and business development teams, increase our presence globally and increase marketing activities to drive awareness and adoption of our platform.
Employees and Human Capital Resources
As of December 31, 2021, we had 386 full-time employees in Canada, the United States and Australia, which was comprised of approximately 49% scientists, 32% business professionals, and 19% engineers and data scientists. None of our employees are represented by a labor union or covered under a collective bargaining agreement. As of December 31, 2021, 343 of our employees were employed in Canada, 25 were employed in Australia and 18 were employed in the United States. We consider our relationship with our employees to be excellent.
Our human capital resources objectives include, as applicable, identifying, recruiting, retaining, incentivizing and integrating our existing and additional employees. The principal purposes of our equity incentive plans are to attract, retain and motivate selected employees, consultants, and directors through the granting of stock-based compensation awards.
Government Regulation
Our focus is on the discovery of antibodies that our partners use to improve the speed and success of their antibody discovery efforts; however, we ourselves are not currently involved in antibody discovery, do not manufacture any products and do not conduct any clinical trials. As such, while we are subject to a number of regulations, such as those governing our laboratory facilities as well as regulations that apply to businesses in the private sector generally, we are not subject to many of the types of regulations that ordinarily apply to companies in the life sciences, biotechnology and pharmaceutical sectors and industries. However, we believe that the long-term success of our business depends, in part, on our partners’ ability to successfully develop and sell products using the antibodies that we discover. The regulations that govern our pharmaceutical and biotechnology partners are those we therefore believe have the most significant impact on our business.
Government authorities in the United States, at the federal, state and local level, and in the European Union and other countries and jurisdictions, extensively regulate, among other things, the research, development, testing, manufacturing, quality control, approval, labeling, packaging, storage, record-keeping, promotion, advertising, distribution, post-approval monitoring and reporting, marketing and export and import of pharmaceutical products, including biological products such as those that our partners develop. The processes for obtaining marketing approvals in the United States and in foreign countries and jurisdictions, along with subsequent compliance with applicable statutes and regulations and other regulatory authorities, require the expenditure of substantial time and financial resources.
Our partners will be subject to a variety of regulations in applicable jurisdictions governing, among other things, clinical studies and any commercial sales and distribution of their products. Whether or not our partners obtain FDA or EU approval for a product, they must obtain the requisite approvals from regulatory authorities in foreign countries prior to the commencement of clinical studies or marketing of the product in those countries. The requirements and process governing the conduct of clinical studies, product licensing, coverage, pricing and reimbursement vary from country to country.
One such regulatory authority that is applicable to certain of our partners is the United States Secretary of Health and Human Services’ authority to authorize unapproved medical products, to be marketed in the context of an actual or potential emergency that has been designated by government officials. The COVID-19 pandemic has been designated such a national emergency. After an emergency has been announced, the Secretary of Health and Human Services may authorize the issuance of, and the FDA Commissioner may issue EUAs, for the use of specific products based on criteria established by statute, including that the product at issue may be effective in diagnosing, treating, or preventing serious or life-threatening diseases when there are no adequate, approved, and available alternatives. An EUA is subject to additional conditions and restrictions and is product-specific. An EUA terminates when the emergency determination underlying the EUA terminates. An EUA is not a long-term alternative to obtaining FDA approval, licensure, or clearance for a product. The FDA may revoke an EUA where it is determined that the underlying health emergency no longer exists or warrants such authorization, so it is not possible to predict how long an EUA may remain in place.
Similar to the United States, Canada has developed a mechanism to authorize unapproved medical products to be marketed in the context of certain emergencies. In particular, under the Food and Drugs Act (Canada), the federal Minister of Health may make an Interim Order if the Minister believes that immediate action is required to deal with a significant risk to health, safety or the environment. The Minister has made various Interim Orders in the context of the COVID- 19 pandemic. These Interim Orders provide the Minister with the authority to permit the sale of a COVID-19 drug in Canada via multiple new mechanisms, including authorizing a COVID-19 indication for a new drug with a modified set of application requirements with the potential for additional terms and conditions, as well as the possibility of authorizing a drug based on certain elements already being authorized by a foreign regulatory authority. Each Interim Order is valid for no longer than a one-year term and an authorization for importation and sale issued under an Interim Order is only valid for as long as the Interim Order is in effect. As is the case with EUAs in the United States, authorizations
34
issued under an Interim Order are not a long-term alternative to obtaining Health Canada licensure for a product. Health Canada is currently considering various options to minimize disruptions for the ongoing authorization of drugs upon the expiry of an Interim Order with the intent to implement transition as needed.
Additional Regulation
In addition to the foregoing, provincial, state and federal U.S. and Canadian laws regarding environmental protection and hazardous substances affect our business. These and other laws govern our use, handling and disposal of various biological, chemical and radioactive substances used in, and wastes generated by, our operations. If our operations result in contamination of the environment or expose individuals to hazardous substances, we could be liable for damages and governmental fines. We believe that we are in material compliance with applicable environmental laws and that continued compliance therewith will not have a material adverse effect on our business. We cannot predict, however, how changes in these laws may affect our future operations.
Anti-Corruption Laws
We are subject to the U.S. Foreign Corrupt Practices Act of 1977, as amended, or the FCPA, the U.S. domestic bribery statute contained in 18 U.S.C. § 201, the U.S. Travel Act, the USA PATRIOT Act, the Canadian Corruption of Foreign Public Officials Act and possibly other state and national anti-bribery and anti-money laundering laws in countries in which we conduct activities, such as the UK Bribery Act 2010 and the UK Proceeds of Crime Act 2002, collectively, Anti-Corruption Laws. Among other matters, such Anti-Corruption Laws prohibit corporations and individuals from directly or indirectly paying, offering to pay or authorizing the payment of money or anything of value to any foreign government official, government staff member, political party or political candidate, or certain other persons, in order to obtain, retain or direct business, regulatory approvals or some other advantage in an improper manner. We can also be held liable for the acts of our third party agents under the FCPA, the Canadian Corruption of Foreign Public Officials Act, the UK Bribery Act 2010 and possibly other Anti-Corruption Laws. In the healthcare sector, anti-corruption risk can also arise in the context of improper interactions with doctors, key opinion leaders and other healthcare professionals who work for state-affiliated hospitals, research institutions or other organizations.
Available Information
Our website address is www.abcellera.com. Our Annual Reports on Form 10-K, Quarterly Reports on Form 10-Q, Current Reports on Form 8-K, including exhibits, proxy and information statements and amendments to those reports filed or furnished pursuant to Sections 13(a), 14, and 15(d) of the Securities Exchange Act of 1934, as amended, or the Exchange Act, are available through the “Investors” portion of our website free of charge as soon as reasonably practicable after we electronically file such material with, or furnish it to, the SEC. Information on our website is not part of this Annual Report on Form 10-K or any of our other filings with the SEC unless specifically incorporated herein by reference. In addition, our filings with the SEC may be accessed through the SEC’s Interactive Data Electronic Applications system at www.sec.gov. All statements made in any of our filings with the SEC or documents available on our website, including all forward-looking statements or information, are made as of the date of the document in which the statement is included, and we do not assume or undertake any obligation to update any of those statements or documents unless we are required to do so by law.
Our code of conduct, corporate governance guidelines and the charters of our Audit Committee, Compensation Committee and Nominating and Corporate Governance Committee are available through the “Investors” portion of our website.
35
Text extracted from the filing as submitted to EDGAR. Formatting, tables and exhibits are simplified for reading; the original document is authoritative for anything you rely on.