By, Jason Wright, Ph.D., Sr. Director, Molecular Biology & Genomics

I’ve been part of the Homology team since the Company’s launch, when there were a few of us in the lab beginning to translate our technology platform into treatments. I was trained in genetics, but my role at Homology was my first foray into biotechnology drug development. In retrospect, when discussing the role at Homology, I did not ask any practical questions about the company’s operations, business plan or financials. Because of my background in science, I focused instead on understanding the technology and its potential, and I got lucky because the Homology leadership team is fantastic, easily accessible and has a wealth of experience in drug development. It was an amazing experience to be exposed to all of the aspects and functions needed to get a small biotechnology company off the ground.

Off the bat, and most importantly, the leadership team established clear therapeutic targets to which we should apply our technology, which I later came to appreciate were based on a number of considerations, such as market potential, regulatory pathways and the potential to make an impact in a particular disease. These targets allowed us to hit the ground running and have continued to drive our mission to deliver one-time treatments for people with genetic disorders. Here, I am going to reflect on the journey of one program in particular, our in vivo, nuclease-free, gene editing candidate HMI-103 for phenylketonuria (PKU). We recently announced the dosing of the second participant in the Phase 1 pheEDIT clinical trial with this candidate. Having worked on this program from conception through optimization and now seeing HMI-103 dosed in humans is what I am most proud of in my entire career in science. Now, let’s rewind.

The Platform

To provide some background about why I was interested in joining Homology, I should first explain the technology. Homology’s AAVHSC platform is based on the discovery of naturally occurring adeno-associated viruses (AAVs) in human hematopoietic stem cells (HSCs) (Smith, 2014). We’ve shown in preclinical studies that when delivered intravenously (I.V.), AAVHSCs can enter, or transduce, cells of multiple organs implicated in genetic diseases with large unmet needs, including the liver, heart, eye and muscle, as well as cross the blood-brain-barrier to the central nervous system (Ellsworth, 2019). We are deploying our technology in three distinct ways based on the disease that we are targeting: gene editing, gene therapy, and GTx-mAb, or gene therapy to create monoclonal antibodies. Learn more here.

In my opinion, Homology takes a powerful approach utilizing the body’s natural DNA repair process of homologous recombination (HR), without the need for a nuclease, or a DNA-cutting mechanism. This technology allows for the direct correction of loss-of-function mutations (i.e., a genetic variant that results in a deficiency or lack of a needed protein) while minimizing the risk of introducing additional variants to the genome.

Conception

Prior to Homology, I worked in labs studying the genetics of human disease as well as the development of gene editing technologies (e.g., CRISPR and TALENS). These experiences fostered my belief in the importance of directly addressing the underlying cause of genetic disease by employing new gene editing technologies. Once introduced to Homology and AAVHSCs, I quickly realized that there was an opportunity to use this technology to achieve direct correction of genetic disease while minimizing risks attributed to DNA-cutting technologies. The first project that I started to work on was developing a gene editing candidate for the rare disease PKU. PKU is a rare inborn error of metabolism caused by mutations in the PAH gene, which disrupts the normal metabolism of phenylalanine (Phe). Phe comes into the body through dietary protein. In people with PKU, Phe builds up in the brain and blood. If toxic levels of Phe build up, it can cause severe neurological impairment.

Given that we were starting with an academic discovery, I knew that each step of the discovery and research process would provide learnings to propel the program forward. We were learning about our capsids alongside the development of the program. Science usually involves a series of challenges and questions to overcome, and that is why I always like having back-up plans to my back-up plans.

When we started the PKU gene editing development program, we employed a practical and basic design, including homology arms, which are long stretches of DNA that guide the gene to the precise area of the genome, and the PAH gene. Through HR, the PAH gene is designed to integrate directly into the genome, essentially replacing the disease-causing gene with a functional copy.

This was a great place to start from because we learned how best to optimize the product candidate. As we conducted in vitro and in vivo experiments, we advanced our vector design, homology-arm design and packaging techniques. We also added a liver-specific promoter in order to utilize PAH expression from episomes. We learned that the design of the gene editing vector with a promoter had a greater impact on disease endpoints than either a promoter-driven gene therapy without homology arms or a promoter-less gene editing vector. Ultimately, we were able to design a gene editing murine surrogate of HMI-103 that was ten times more potent than the non-gene-editing vector when tested in the murine model of PKU. Given that a disease-relevant human model was not available, a murine surrogate with murine homology arms was necessary to test the candidate in this model.

You can see how this all works here.

There are two important questions to ask for any gene editing modality: 1. Are you making the changes you intend to make? And 2. Are you inadvertently making unwanted mutations to the target (on-target error) or elsewhere in the genome (off-target error)? We developed assays to answer both of those questions using long-read sequencing, and we saw in nonclinical studies that the gene editing candidate specifically integrated at the desired location with no undesired mutations, such as indels or inclusion of viral elements. In addition, we found that the only location in the genome where we saw genome integration was at the intended target itself, which supported that our vector is specific with no increased off-target mutations. We showed this in human and murine genomic models of the target tissue and disease.

By addressing the underlying genetic cause of PKU, HMI-103 has the potential to restore the biochemical pathway so that treated patients may be able to create the PAH enzyme and metabolize Phe.

pheEDIT Trial Phase 1, Dose-Escalation Trial for PKU

Now fast forward seven years – Homology is conducting the first gene editing clinical trial for PKU. Since those early days, we have learned so much about our technology. We prioritized questions to get critical answers faster. I am extremely pleased with the work we did, and I want to do it again. There are greater than 3,000 monogenic diseases in humans and if we can develop a platform that can address the underlying cause of these diseases, we can make a huge difference.