Moving Therapeutic Genome Editing into Global Clinical Trials and Medicine

Introduction

Moving CRISPR-based therapies from discovery to dosing patients in clinical trials and ultimately to approval involves navigating a challenging terrain of highs and lows. In this exclusive interview, conducted as part of the Global Observatory for Genome Editing conference, physician-scientist Kiran Musunuru and genome editor Fyodor Urnov reflect on the past 20 years of their nonclinical and clinical programs in the field, the current landscape of innovation, and what they see on the horizon.

Musunuru is a cardiologist by training and a co-founder of Verve Therapeutics, a biotech company developing gene-editing therapies for cardiovascular disease. Urnov has been working on genome editing for 25 years, including working for more than a decade at Sangamo Therapeutics, during which he and his colleagues coined the term “genome editing.” He is a co-founder of Tune Therapeutics, an epigenome editing biotech company.

The Global Observatory for Genome Editing team caught up with Musunuru and Urnov during a U.S. National Institute of Health meeting for the Somatic Cell Genome Editing Consortium in May 2025.

(This interview has been edited for length and clarity.)

What Are the Major Opportunities and Concerns for the Future of Somatic Cell Gene Editing?

Fyodor Urnov: As someone who has tracked the trajectory of genome editing from 2002 onwards, I marvel at the extraordinary expansion of the toolbox and application space, leading to clinical applications. However, as of today, my concern, bordering on fear, is that the for-profit pipeline of “CRISPR cures” is narrow. It seems that biotech companies are all attempting to work on the same diseases simultaneously. This is not a criticism but a reflection of the current landscape. Companies like Beam Therapeutics, Intellia Therapeutics, and Prime Medicine are focusing on alpha-1 antitrypsin deficiency, while others are working on sickle cell disease (SCD).

I receive countless emails from parents with pictures of their children living with and suffering from genetic diseases. I question what’s going on, especially since the expansion of the toolbox from 2012 to 2016–2017 with the invention of base and then prime editing. The delivery toolbox is expanding, and recent articles in many leading journals showcase this progress. ¹ It is like you’re walking into the world’s greatest food supermarket with an ever-increasing array of ingredients, but everybody’s making a hot dog. This artificial constriction could harm our field.

This situation pushes the responsibility for impactful technology out of the for-profit domain and into the academic and nonprofit sectors and creative public-private partnerships. Recent meetings of the NIH Somatic Cell Gene Editing Consortium ² and other flagship federal programs showcase how much academia can achieve in this space. There is also a significant change afoot in the biomedical enterprise in the United States, originating from both Washington and other areas. These are times of technological blossoming, superimposed on commercial, regulatory, and governmental challenges. Having been part of the team that brought gene editing forward from bench to bedside in 2009, I feel very protective of our “baby,” which is now graduating from college and needs more support to make a big impact. ³

Kiran Musunuru: I couldn’t agree more with Fyodor. His concern is fully justified. We see it happening with PCSK9, ⁴ where Verve Therapeutics and others, including at least two Chinese companies, are working on the same target. Several of these efforts are already in clinical trials. Similarly, if you include RNA editing, there are five or six companies focusing on alpha-1 antitrypsin deficiency.

These targets become attractive because capital is expensive, and companies have to focus down and take their best shot at what appears to be the biggest commercially viable opportunity. Alpha-1 antitrypsin deficiency, though rare, affects around 100,000 people in the United States, making it a plum commercial opportunity. With PCSK9, you’re talking about atherosclerotic cardiovascular disease, the leading cause of death worldwide—a huge market.

We can understand why companies are going there, but it is to the exclusion and detriment of literally thousands of rare and ultra-rare diseases. These diseases are being overlooked entirely. I don’t want to say ignored, since I think companies appreciate that the need is there. But they’re just not going to go there. This puts the onus on the academic sector to fill the gaps and apply validated technologies to tackle these rare scenarios. There is a need, or an opportunity, to apply these same technologies that have now been validated by the companies—the same exact approaches—and tackle all of these rare and ultra-rare and “n-of-one” type scenarios.

Urnov: It’s been 20 years since an article in Nature ⁵ showed that gene editing could repair a point mutation driving a rare immune disorder. The honest-to-goodness truth is that it is probably the lowest-hanging fruit in the realm of gene therapy. It’s a severe combined immune deficiency, where trivial levels of repair are sufficient to be curative. But there is no gene editing trial for that disease. In fact, of the 500 inborn errors of immunity, there is only one trial.

There is no global committee ensuring CRISPR’s maximum footprint. The normal mechanisms in the small molecule or biologics space (e.g., antibodies targeting cancer checkpoints) do not apply here. But the lemonade you get out of this very strange lemon is that we’re returning clinical trial-based innovation from the for-profit sector to the academic and nonprofit sectors. And we in those sectors tend to be more nimble. We can do smaller, more affordable trials because we have academic pricing and not industrial pricing. It’s poignant that Emily Whitehead was dosed at the University of Pennsylvania in 2011–2012, leading to seven approved medicines and nearly 40,000 subjects dosed. ⁶

Editing is now the fastball of biology to gene therapy. We need new ways to develop, derisk, and commercialize these medicines.

Do You Think the Global Landscape Is Different from the U.S.?

Urnov: Immuneel in India is developing indigenous CAR-T cell therapy, ⁷ which costs $40,000–50,000 per dose compared with over $500,000 in the United States. I’m optimistic that global efforts will provide a larger record of safety and efficacy for CRISPR. The first in vivo dosing for hemoglobinopathy happened in China, while companies in the United States are approaching similar goals. My hope is that the learnings from all of that will provide some momentum and also, frankly, some enabling capabilities.

We need to leverage each other’s learnings to achieve regulatory convergence. I worry that we will end up with multiple jurisdictions, each with a different barrier to entry for a CRISPR therapy. I think that will be a loss for the field. The potential risk to patients from Cas9 is the same in China or in India or New Zealand or Philadelphia or San Francisco or London, so we must avoid fragmentation that could slow us down. This is innovative technology, and it requires regulatory proportionality and regulatory nimbleness.

Musunuru: This is a little bit at odds with what Fyodor just said, but I worry that insufficient regulatory rigor could lead to bad outcomes. Some countries are moving quickly, perhaps too quickly. There is something to be said about investigator-initiated clinical trials and being able to innovate quickly because the discretion is left more in the physician’s hands than in the regulators’ hands. But investigator-initiated clinical trials require that individual investigators truly understand the technology and deploy it responsibly.

I worry that the pace at which some of these things are going forward into clinical trials is outpacing our full understanding of the benefits and risks. So, that’s the kind of thing that keeps me up at night. All it takes is one very highly publicized and negative event to take the entire field down. I’m very conscious of this, since I come from the University of Pennsylvania, where what happened at the turn of the century (with Jesse Gelsinger) took the wind out of the sails of the field of gene therapy for a very long time.

Urnov: I second Kiran’s concerns. My entry into gene editing was marked by severe adverse events (SAEs) in retroviral gene therapy for immune deficiency. I have vivid memories of how regulatory authorities poured liquid nitrogen over that and related approaches. But we learned about safety and efficacy from clinical trials. At this point, we have >15,000 subjects who have been dosed with lentivirus with pretty good track records. We must balance the need for more subjects dosed against thoughtful risk management. Furthermore, the more subjects are dosed with medicines that have been developed thoughtfully, the more we learn.

But then, I have a vivid memory of when we at Sangamo were starting to do gene editing therapy for SCD in 2010–11. A prominent transplanter who treats patients living with sickle cell said to me, “look, sickle is an indication where you cannot be a cowboy.” You have to understand what your responsibility is before the patient community and the tragic history of that community through the trajectory of healthcare over the past two centuries.

Broadly speaking, the community of families living with genetic diseases is highly motivated, but their approach to benefit-risk can be different. A parent caring for a child with a severe neurodevelopmental disorder once told me, “I don’t care about your off-targets. Some things are worse than death.” We must balance their motivation with the responsible deployment of genetic therapies. I wish I had some sort of magic formula where you plug in all the risks and all the parameters, like a polynomial, and it says this is acceptable risk. But I don’t think that exists.

Musunuru: There is the danger that we overpromise. For neurodevelopmental disorders, for most patients who come to our attention, sadly it’s too late, and meaningful improvement is unlikely. While delivery technology is improving, getting into the brain remains a big challenge, and the developmental windows for effective interventions may have passed. We want to be sympathetic to pleas from parents on behalf of their children, but right now the potential risks greatly outweigh the possibility of benefit. But you have to balance that against the fact that if we’re not ever willing to try anything, we’ll never learn about and fully understand the risks of these therapies.

Urnov: I’m working with a family whose child has a neurodevelopmental disease. We face the reality of irreversible developmental changes. We’re building a pretty solid base editor. But then we sit down with the clinicians and talk about the number of seizures, and you ask yourself in the cold light of reality, what’s the likelihood of benefit, especially if irreversible developmental changes happened in utero? Different developers will also have different thresholds for risk.

The case of a patient who died from an SAE on an adenoviral vector-based epigenome therapy for Duchenne muscular dystrophy ⁸ highlights the need for dispassionate judgment. Clearly, that patient should never have been dosed. That product was developed through the major effort of the patient’s brother, who led a heroic effort to essentially start a biotech. I’m not saying that the mode in which they did this trial predisposed them for the adverse event. But I would ask to what extent their judgment was clouded by how close they were to what was going on. We have colleagues who are in similar positions, and it’s hard to know.

Does Therapeutic Misconception Apply Here? Is That Concept Sufficient to Address These Sorts of Situations?

Musunuru: Therapeutic misconception is the idea that participants in research studies expect therapeutic benefit, conflicting with the study’s intent to gain new knowledge. Whether it’s the patient themselves giving consent or parents, there is a tendency, understandably, for them to think that the intended outcome of the research study, or at least the desired outcome, is therapeutic benefit. This is a significant concern in rare and ultra-rare scenarios where safety and efficacy are often assessed simultaneously. Our approach is not to try to dose for a cure in the first dose. Rather, we assess safety at a low dose, and then if the low dose proves to be safe, we can re-dose and try higher doses until we see a therapeutic effect. For this approach, I prefer saying ‘dose to effect’ over using curative terminology.

Urnov: We must be blunt about the complexity of disease indications and the need for thoughtful risk management. The patient community’s motivation must be balanced with the responsible deployment of genetic therapies. We need to learn how to dose safely, re-dose to effect in cases where this is prudent and possible, and then learn and iterate until we have deployable therapies that transform lives.