Q&A: David Liu’s bold vision to make on-demand treatments routine for life-threatening rare genetic diseases

In May 2025, researchers announced that K.J. Muldoon, a baby boy born without the ability to process dietary protein properly, had become the first person to be treated with a customized gene editing therapy. Based on a technology developed by Broad Institute core member David Liu’s laboratory, the treatment is the first in a series of new medicines being tested to treat rare diseases by repairing patients’ particular genetic misspellings. The team that treated K.J. was led by physician-scientists Kiran Musunuru and Rebecca Ahrens-Nicklas at the Children’s Hospital of Philadelphia and U. Penn.

Some of these treatments, like K.J.’s, are built with base editing technology, a gene editing technique developed by Liu’s team in 2016 to directly convert an individual DNA base pair into a different base pair. Others use prime editing, a technique Liu’s group pioneered three years later that can make any kind of small DNA correction. Together, these technologies have entered at least 19 clinical trials, with clinical results from seven of those trials reported so far — all showing that the base edit or prime edit resulted in patient benefit.

Today, K.J. is thriving, and Liu and others in the field are hoping to repeat this success many times over. They plan to harness scientific, medical, regulatory, and manufacturing innovations to enable on-demand genetic treatments like K.J.’s to be produced at scale, making them the standard of care for life-threatening rare genetic diseases.

To learn more, we spoke with Liu, who is the Richard Merkin Professor and director of the Merkin Institute for Transformative Technologies in Healthcare at the Broad, the Thomas Dudley Cabot Professor of the Natural Sciences at Harvard University, and a Howard Hughes Medical Institute (HHMI) investigator.

Tell us about baby K.J. How did the work in your lab contribute to this?

K.J. was born with a severe genetic disease caused by a single-letter gene mutation that prevents the liver from properly removing ammonia from the blood, which causes ammonia to build to toxic, and potentially fatal, levels. About half of the infants with this disease do not survive past infancy, and those who do often suffer long-term brain damage.

The effort to treat him was a massive "village effort" involving many researchers, co-led by Kiran Musunuru and Rebecca Ahrens-Nicklas. My lab's contribution was developing the base editing technology that made it possible to correct K.J.'s mutation, and recommending to Kiran and Rebecca specific components of the base editor that were ultimately chosen to administer to K.J. 

Our development of base editing was made possible by crucial funding from the National Institutes of Health and the resources, support, and expertise of researchers and collaborators at Harvard University and at the Broad.

The teams involved undertook a heroic race against the clock, completing in less than seven months what previously might take seven years. This unprecedented feat required diagnosing K.J.’s specific mutation, creating a mouse model of the disease, determining the optimal base editor, performing extensive safety analyses, working with Danaher to manufacture the therapeutic, conducting toxicity studies, and securing FDA approval for the trial.

The result appears to be a major success. After receiving injections of the base editor, which corrected the single letter misspelling in his liver cells, K.J.'s blood ammonia levels have dropped to the high end of normal for an infant. He can now tolerate protein in his diet and is meeting developmental milestones that patients with this disease typically do not. While doctors are cautious about using the word "cure" so early, it is a triumph for K.J.'s family and for everyone involved.

Have there been recent successes from prime editing, the other gene editing platform that your lab developed?

Yes, in another recent exciting medical milestone, scientists at Prime Medicine (a company I co-founded) announced the first-ever treatment of a human patient with a prime edited medicine, an 18-year old with an immune deficiency caused by a two-letter deletion in his DNA. Using prime editing technology, they inserted the two missing letters into his hematopoietic stem cells and transplanted them back into his bone marrow. The treatment successfully restored the missing function in his immune system, and now a second patient has also been treated, showing even higher (90%) editing efficiency. Together, base editing and prime editing technologies provide powerful tools to bring genetic treatments to more patients.

K.J. wasn’t the first patient to receive base editing therapy. Why is his case so significant, and potentially difficult to repeat?

Indeed, in 2022 Alyssa Tapley became the first human to receive a base-edited therapeutic, treating her life-threatening T-cell leukemia that had not responded to other treatments. Doctors took donor T cells and performed three precise base edits on them, engineering them to attack leukemia cells while leaving healthy cells alone. After an infusion of the edited cells, the treatment rapidly cleared her cancer and she has remained cancer-free for more than three years now. Designed as an off-the-shelf therapy, the treatment also showed promise for other patients with T-cell leukemia in the same trial. 

KJ, on the other hand, carried a unique genetic change that prevented his liver cells from working properly. He needed a personalized, one-of-a-kind therapy. It also needed to be delivered directly to the liver cells in his body, rather than to cells in a lab, and this was achieved by packaging the base editor in a lipid nanoparticle, a delivery system conceptually similar to the one used to deliver COVID vaccines to billions of people. His bespoke treatment resulted from an unprecedented level of coordination and synergy among many different groups, but it’s customized to correct K.J.’s mutation and may never be used to treat another person. This kind of all-out effort is not yet feasible or commercially viable to repeat for the 10 million babies born each year with rare genetic diseases. 

How can we move beyond these heroic one-off efforts, towards making personalized genetic treatments standard of care?

First, it’s important to remember that all of this progress came from public investment in basic science. No one could have predicted that studying repetitive DNA sequences in bacteria would eventually lead to medicine that can rescue a baby from a life-threatening disease — but it did. To realize the full potential of these discoveries, we must continue that investment. Of the 19 base editing and prime editing clinical trials underway now, more are taking place outside the U.S. than within our country, despite the fact that the core editing technologies were developed in the U.S. That’s in part a reflection of the large investment in these technologies, and the lower regulatory barriers to initiate clinical trials, in other countries.

The challenge is that even when the science is successful, that doesn't guarantee an economically viable path to bring these new medicines to the patients who need them. With the right framework and the right support, I believe it will be possible to treat at least 1,000 patients with personalized genetic treatments by 2030. Doing so would not only transform lives, but save billions in health care costs to manage symptoms that can be alleviated by fixing a disease’s root cause. To achieve that goal, we need a new framework for developing and approving drugs. 

What might that framework look like?

Several of us are proposing to create national Centers for Interventional Genetics, FDA-accredited centers of excellence, which would require substantial organization and support. These centers would include five key components. They would have rapid diagnosis and therapeutic development platforms so we can identify and optimize gene editing agents and their delivery vehicles. They would also need chemistry, manufacturing, and controls (CMC) capabilities that can operate swiftly and at a small scale, along with platforms for performing cellular, rodent, non-human primate, or virtual toxicology assessment. Importantly, we’ll need streamlined regulatory pathways for approving clinical trials with only a few patient participants that are founded on the principles of proportionate regulation based on current data and benefit:risk ratios. Finally, we need platforms for sharing these resources and the valuable pre-clinical and clinical data that would result.

Each year 10 million babies are born with one of 10,000 rare genetic disorders, many of which in principle are now treatable with genetic medicines. It should be possible to build a modular system that makes personalized gene editing a scalable infrastructure, an important step towards the ultimate goal of providing each patient receiving a rare genetic disease diagnosis with a treatment made in time to save their life. The insights gained from this effort will catalyze future waves of therapeutic innovation, and also help secure America’s leadership in genetic medicine for the foreseeable future.