GlaxoSmithKline’s latest move to reinvigorate R&D is a collaboration with two of the biggest names in CRISPR research to develop new gene editing technologies and use them to identify drug targets.
GlaxoSmithKline plc (LSE:GSK; NYSE:GSK) teamed up with the labs of Jennifer Doudna at the University of California Berkeley and Jonathan Weissman at the University of California San Francisco to form the Laboratory for Genomics Research (LGR), a physical laboratory near UCSF’s Mission Bay campus that will bring together investigators from all three partners.
The lab will receive up to $67 million in funding over five years from GSK and will use CRISPR technologies to interrogate the mechanistic links between genetic mutations and disease biology. The idea is that determining how mutations cause disease will lead to new drug targets. GSK has an option to those targets; details are not disclosed.
CSO and President of R&D Hal Barron said the lab will blend GSK’s human genomics data and drug development expertise with the CRISPR-based functional genomics capabilities of the academics, and layer machine learning on top. The collaborators will focus on immunology, oncology and neuroscience.
“Machine learning can teach us in an innovative way which experiments to do,” from the front end, Weissman said. “On the back end, machine learning is essential for taking the data that we’re getting out of this and extracting the meaning.”
Doudna said: “It really started with a conversation that I had with Hal Barron just over a year ago to talk about how we could take the extraordinary opportunities in functional genomics using the CRISPR technology to probe the underlying genomics of cells; how we could take that and apply it to questions that we all have about fundamental biology, as well as the way those insights will play into drug discovery in the future.”
Synthetic lethality is one focus area for the target search, Barron said.
“Disrupting one gene might not have any impact on a disease or a disease-like phenotype in a cell. Disrupting a second gene might actually have nothing, but when you put the two together, you have a profound impact on the cell,” Barron said. He noted that CRISPR has the ability to test synthetic lethal interactions at scale, which is important when 20,000 genes are interacting with 20,000 other genes to make 400 million data points per experiment.
Uncovering synthetic lethality pairs has not been as straightforward as expected because other mutations and non-genetic alterations can modify the interactions (see “Following PARP, ATR Axis Next”).
Barron said finding synthetic lethal interactions is now done relatively easily in the labs of Doudna and Weissman.
But finding new targets is only one part of LGR’s mission. In parallel, the lab will develop new CRISPR technologies based on the expertise of Doudna and Weissman that will feed into the new target identification projects.
While the partners aren’t sure exactly what CRISPR 2.0 will look like, one goal of the technology advancement is scaling and automation.
Founding LGR is Barron’s latest initiative to boost GSK’s R&D since joining the company in January 2018.
In December, GSK acquired immuno-oncology biotech Tesaro Inc. (NASDAQ:TSRO) for $5.1 billion, which brought the pharma the marketed PARP inhibitor Zejula niraparib and an early stage immuno-oncology pipeline (see “GSK Gains Zejula, Immuno-oncology Pipeline”).
Last July, it partnered with 23andMe Inc. to discover new drug targets and develop therapies for indications including Parkinson’s disease (see “GSK’s Near-term R&D Prospects”).
The LGR deal builds on that collaboration, as well as collaborations with UK Biobank and the Open Targets public-private partnership, by using human genetic data accessed through the partners as the starting points for the target projects.
Through the Open Targets partnership, GSK has participated in other CRISPR screens, including a large-scale study with The Wellcome Sanger Institute that identified and ranked 600 potential new targets by disrupting every gene in over 300 cancer models across 30 different cancer cell types. The paper was published in Nature in April.
Like that study, the new collaboration will use CRISPR as a research tool and not a therapeutic modality. Resulting therapies against the new targets will likely be small molecule or antibody agents.