BALTIMORE—Having delivered a plenary lecture during last year’s American Society of Gene and Cell Therapy (ASGCT) conference, David Liu, PhD, probably felt obliged to present some new stories after he received the society’s 2024 Outstanding Achievement Award last week. Thankfully, that isn’t a problem in the rapidly moving genome editing space, and certainly not for a group as productive as Liu’s lab at the Broad Institute.
It is not that base and prime editing—groundbreaking technologies developed in the Liu lab from 2016–19—are old news. Liu did devote a few slides to summarizing the clinical progress of both platforms in diseases as wide ranging as spinal muscular atrophy, sickle cell disease, and progeria. There are at least nine base editing clinical trials underway, with three having reported initial findings.
Pride of place goes to Alyssa Tapley, the British teenager who received a CAR-T therapy from Waseem Qasim, MD, and colleagues in London two years ago to treat her advanced T-cell leukemia. In remission, “Alyssa has returned to school and hopes to pursue a career in biomedicine,” Liu told the ASGCT delegates. And the U.S. Food and Drug Administration recently cleared Prime Medicine’s first gene editing therapy for chronic granulomatous disease, less than five years after Andrew Anzalone, MD, PhD, and colleagues unveiled the technology.
But Liu devoted most of his 2024 lecture—“Continuous evolution of genome editing systems for targeted gene-sized DNA integration in mammalian cells”—to describing a pair of independent methods designed to realize another goal of the genome editing field: editing gene-sized DNA fragments measured in kilobases (kb) in human cells without necessitating double-strand DNA breaks.
This approach would be ideal in principle for treating genetic diseases caused by large numbers of different mutations in the same gene. The classic example of this is cystic fibrosis, for which more than 400 discrete mutations have been catalogued in the cystic fibrosis transmembrane regulator (CFTR) gene. And Stargardt disease, an inherited form of vision loss, can result from one of more than 500 mutations in the ABCA4 gene.
“Until a major regulatory streamlining occurs, it’s hard to imagine creating hundreds of different reagents needed to treat certain diseases,” Liu said. The goal would be “to integrate an entire healthy gene or exon into the pathogenic site or a safe harbor locus, without incurring double-strand breaks.”
Two is better than one
The first example builds on the prime editing platform and is based on a method published in 2022 called PASSIGE (prime assisted site-specific integrase gene editing). This enables the integration of large, potentially gene-sized cargos into newly installed landing sites using an integrase enzyme called Bxb1, albeit with modest efficiencies.
As Liu’s lab showed in the early development of base editing, there are ways to overcome limited enzyme efficiencies using a technique developed by Liu, Kevin Esvelt, PhD, and colleagues 13 years ago called PACE (phage-assisted continuous evolution). Liu’s group evolved Bxb1 through hundreds of generations (as will be described in an upcoming paper in Nature Biomedical Engineering), with the most active variants supporting higher levels of gene integration—up to 35%—while still delivering sizeable cargos of up to 10 kb.
Liu also presented some comparative data with a similar method dubbed PASTE, developed two years ago by Omar Abudayyeh, PhD, and Jonathan Gootenberg, PhD, the co-founders of Tome Biosciences. The newly evolved PASSIGE platform (eePASSIGE) performed better than PASTE. (Liu argued that this is a consequence of the fusion of Bxb1 to the prime editor in the PASTE platform.) PASSIGE has been used to engineer integration of half-a-dozen therapeutic genes in various cell lines, including blood clotting factor IX, with a 32% average integration frequency.
A strong CAST
The second approach is the result of a fruitful three-year collaboration with Samuel Sternberg, PhD, at Columbia University, in which PACE once again has had a starring role. Over the past five years, the Sternberg lab has played a leading part in characterizing CRISPR-associated transposases (CASTs). These complexes facilitate integration of large DNA fragments without introducing double-strand DNA breaks in bacteria. But in mammalian cells, wild-type CASTs frustratingly show minimal integrase activity. A drug called ClpX improves efficiency to a degree but has proved to be cytotoxic.
Making the reasonable assumption that transposase catalysis might be the major bottleneck, Liu’s team turned once more to PACE to evolve CASTs in the lab, a process that stretched over thousands of generations. The resulting CASTs yielded integration rates about 200-fold greater than wild-type, resulting in integration efficiencies of 10-20 percent. Moreover, the last set of variants helpfully removed the dependence on ClipX. The optimally evolved CAST (so far) exhibits an integration frequency of up to 30%.
Not surprisingly, several other groups—academic and commercial—are eagerly exploring techniques for gene editing of large inserts. Rahul Kakkar, MD, CEO of Tome Biosciences, gave one of several presentations from the company on preclinical progress in non-human primates targeting the gene corresponding to the one mutated in patients with phenylketonuria. “It is tailor made for our technology,” Kakkar told Brad Loncar of Biotech TV.
Another company making strides in gene-sized programmable editing, or gene writing as it calls it, is Tessera Therapeutics. The company presented several talks last week demonstrating progress in mouse models for sickle cell disease (SCD) and alpha-1 antitrypsin deficiency. The in vivo SCD delivery program shows promise in enabling “true correction to wild type without complex stem cell mobilization and ex vivo cell processing or toxic chemotherapy conditioning,” commented CEO Michael Severino, MD.
Tessera’s chief science officer, Michael Holmes, PhD, a gene editing veteran formerly with Sangamo, said: “We have demonstrated what we believe are industry leading correction levels and the highest fidelity we are aware of across each of our liver rewriting programs,” including phenylketonuria and Wilson’s disease. Holmes called it a testament to “the power and capabilities of our Gene Writing and delivery platforms.”
Doubtless this will be a major theme when ASGCT takes the conference to New Orleans in May 2025.