Within a week of receiving a blood sample from one of the first U.S. patients to recover from COVID-19, AbCellera, a Canadian therapeutic antibody discovery company, put its rapid pandemic response platform to work, identifying more than 500 unique fully human antibody sequences, the largest panel of anti-SARS-CoV-2 antibodies reported to date.
Eli Lilly will partner with Vancouver, BC-based AbCellera to co-develop the most promising of those 500+ antibodies, through a collaboration whose value was not disclosed. The companies will carry out their co-development work after AbCellera taps into the expertise of another partner—the Dale and Betty Bumpers Vaccine Research Center of the NIH’s National Institute of Allergy and Infectious Diseases (NIAID).
The Center will assess those antibodies, and find those with the greatest potential against COVID-19 by identifying antibodies that bind the pandemic strain of SARS-CoV-2 the best.
“We are about to find out very quickly which ones of those 500 will be effective neutralizers through our partnership at the Vaccine Research Center,” Ester Falconer, PhD, AbCellera’s Head of Research & Development, told GEN. “We will have that data very quickly, and that will help us down-select a number to develop together with our partners at Lilly for further manufacturing and further testing.”
AbCellera and Lilly committed to equally share initial development costs towards a treatment, after which Lilly has agreed to oversee all further development, manufacturing and distribution. If successful, Lilly said, it will work with global regulators to bring a treatment to patients
Daniel Skovronsky, MD, PhD, Lilly’s chief scientific officer and president of Lilly Research Laboratories, added in a statement: “We’ve partnered with AbCellera because we’re impressed with the speed and quality of their efforts. We are moving at top speed to create a potential treatment to help patients.”
“While typically a new therapeutic antibody program might take years to get in the clinic, our goal with AbCellera is to be testing potential new therapies in patients within the next four months,” Skovronsky added.
Appearing on CNBC’s “Mad Money,” Lilly chairman and CEO David A. Ricks told host Jim Cramer: “We hope to be in a clinical trial this summer.”
DARPA P3 participation
AbCellera developed its pandemic response platform over the past three years through its participation in the Defense Advanced Research Projects Agency (DARPA) Pandemic Prevention Program, also called P3. The four-year, $30 million program was created to establish a robust technology platform for pandemic response capable of developing field-ready medical countermeasures within 60 days of isolation of an unknown viral pathogen, in partnership with NIAID Vaccine Research Center and Ichor Medical Systems.
“Presently, our platform is the world’s best for antibody discovery. And Lilly, a worldwide, well-known pharma company, is one of the top in the world at manufacturing and distribution, so the synergy aligns because AbCellera can find the best antibody,” says Falconer, “but if there’s no one to take those antibodies and help develop them, manufacture them, get them through clinical trials, they’re not going to be useful to the people who need them most.”
AbCellera’s platform can take immune cells from essentially any source with a natural immune system, and isolate the B cells that make antibodies with the therapeutic properties of interest. The cells are loaded into microfluidic devices, which comprise over 150,000 individual reaction chambers, each one of which has a closed volume of about 1 nanoliter.
“In a single day of screening, we can screen through millions of immune cells, and find the best antibodies that will then go on to be developed as therapeutics,” Falconer said. “A single nanoliter volume effectively shrinks down a petri dish millions of times, and in a matter of an hour, a single antibody secreting cell can build up enough concentration in that very tiny volume so that we can actually perform a number of tests for antibody properties.”
Those tests include, individually or in combination:
- Finding antibodies that bind a target expressed in its native conformation on a cell surface.
- Finding antibodies that bind a specific antigen, and counter-screening against unwanted interactions.
- Screening for antibodies that are cross-reactive for different antigens or various specific epitopes.
- Enriching for antibodies with high affinity for a single target or against multiple targets.
- Screening for antibodies that bind a target, differentiating between receptor-ligand interactions such as ligand blocking.
From months to days
That process, according to AbCellera, is combined with high-throughput imaging, fluidic control over the reagents, and machine learning algorithms to isolate the chambers that have the cells that make an antibody of interest—all in one day, versus the three-month timeframe of traditional hybridoma technologies. The sequences of those antibodies can be isolated and verified in about five days.
Further downstream, AbCellera applies single-cell sequencing and its proprietary antibody visualization software, called Celium™: “It lets us look at properties of thousands of antibodies in a highly dynamic and visual way, and that really helps us down-select the best antibodies to move forward for further development,” Falconer said.
AbCellera’s antibody platform earned the Vancouver, BC-based company a spot earlier this month on Fast Company’s 2020 edition of its annual list of The World’s 50 Most Innovative Companies.
Through DARPA’s P3 program, AbCellera optimized its antibody platform, performing numerous capability demonstrations. Last year, AbCellera simulated a pandemic response scenario using the pandemic H1N1 California 2009 strain of influenza.
“We went and performed an end-to-end discovery program starting with a single donor sample, and obtained hundreds of unique antibodies that recognize that pandemic strain of influenza, tested them for propensity to neutralize the virus through our partners at the Vaccine Research Center, and developed a number to take forward for testing in an animal challenge model,” Falconer said.
“We showed that all of the seven leads that we had chosen to test in a mouse model prevent influenza, even when 20 times the lethal dose was delivered to the mouse. All seven of these antibody leads were fully protective,” Falconer added. “So we knew form our previous work that from a single donor sample and in a very rapid timeframe, our platform can discover fully protective antibodies that were able to neutralize the virus and prevent disease. And that is the strategy and the approach that we’re taking for this pandemic virus strain as well.”