INTRODUCTION

In just May 2025, six-month-old “Baby KJ” became the world’s first patient to receive a personalized CRISPR gene editing therapy, which was fast tracked to be developed for him after his CPS1 deficiency diagnosis, a life-threatening genetic disorder that caused ammonia to build up in his bloodstream. Children’s Hos­pital of Philadelphia and the University of Pennsylvania worked closely with several companies to create the CRISPR therapy, which corrected the defective gene.

Aldevron, a global leader in the production of DNA, RNA and proteins, was one of the companies that contributed to this effort, and we believe the success of Baby KJ provides a model that could be replicated many times over to treat people with ultra-rare genetic diseases. Creating more of these “N of 1” ther­apies will be far from simple, however.

The main challenge of personalized cell and gene therapy is not scaling it up, but rather scaling it out – establishing efficient workflows that researchers in any setting can use to create ther­apies for small patient populations, often on compressed time­lines. Right now, the biopharmaceutical industry is set up to manufacture products for millions of patients, and attempting to adapt those workflows to N-of-1 development would be time-consuming and costly.

Optimizing CRISPR development to serve patients who need rapid, personalized therapies for ultra-rare diseases will require novel approaches to everything from how editing tools are deliv­ered into the body to how CRISPR therapies can be manufactured at extremely small volumes. By working together, academic sci­entists and biopharmaceutical developers are continuing to in­novate novel solutions to these challenges.

HOW MRNA IS ADVANCING CRISPR

One key to the success of KJ’s CRISPR therapy was the de­velopment of a novel guide RNA sequence and an mRNA-en­coded base editor, designed in collaboration with Integrated DNA Technologies (IDT), which, like Aldevron, is part of the Danaher company family. Delivering CRISPR gene-editing tools in mRNA form helps to improve efficiency and lowers the risk of off-target editing.¹

The ability to quickly develop mRNA-encoded CRISPR without compromising safety represents a major challenge. One method for accelerating this process is by limiting the size of the quality-control panel. For example, researchers could consider starting with 10 candidates, and select the top three from those. They could then test for purity and efficacy, and then move on.

In KJ’s case, the development team coordinated with the US FDA from the beginning, facilitating the simultaneous completion of several steps of the process. In the first three months, the most precise gene-editing approach was established. Months four through six were spent manufacturing the CRISPR batch for toxi­cology studies, performing safety studies in mice and nonhuman primates and manufacturing the batch that would be brought to the clinic.² Future development of N of 1 therapies will require similar close coordination between regulators and developers.

INNOVATIONS IN CRISPR DELIVERY

The most commonly used vehicle for delivering DNA into the body is adeno-associated virus (AAV). The problem is, AAVs are limited in the size of the cargo they can carry and their ability to target cells beyond the liver. AAVs can also raise the risk of im­munogenic reactions.³

Several promising alternatives to AAV delivery are under development. They in­clude lipid nanoparticles (LNPs), which have a well-documented safety profile due to their use in COVID-19 mRNA vaccines. LNPs have been demonstrated to protect mRNA constructs from degradation and to aid in tropism, or the delivery of CRISPR tools to the intended targets.⁴ Baby KJ’s mRNA CRISPR treatment was encapsulated in liver-directed LNPs.

Other alternatives for CRISPR delivery that are starting to gain traction include extracellular vesicles and virus-like parti­cles. What’s promising about these op­tions is that they can carry larger cargos than AAVs typically can, and they can tar­get cells beyond the liver.⁵

CRISPR components are typically transported into target cells using plasmid DNA backbones, which are effective but potentially cytotoxic. Novel plasmid DNA backbones are smaller than traditional plasmids, and they are free of antibiotic markers that have traditionally been used in plasmid DNA. These attributes lower the risk of toxicities and transgene silencing. In a 2022 study, Genentech compared the efficiency of three DNA donor templates for generating CD8 T cells using CRISPR. They reported that the smallest of the plas­mid vectors generated twice the number of edited cells as did a traditional plasmid and three times as much as a linear dou­ble-stranded DNA donor template.⁶

IMPROVING EFFICIENCY & TARGETING

Since the earliest development of CRISPR, the enzyme Cas9 has been the preferred enzyme for cutting DNA. But it’s not always the best choice because it is often inefficient and it can cause off-target editing. There are now several alternative enzymes that can reduce errors and im­prove efficiency. For example, dCas9 fu­sions are deactivated mutants of the enzyme that can improve targeting. Also available are non-Cas9 nucleases such as Cas12a, which can often target genomic sites that may not be reachable with Cas9.

In the effort to help more patients like KJ, CRISPR developers are searching for other ways to make the technology more adaptable to a wide range of genetic dis­orders. In some cases, that may require not just removing faulty genes, but also in­serting therapeutic genes. The most widely used insertion strategy is homology-di­rected repair (HDR), which is often ineffi­cient. One promising alternative is microhomology-mediated end joining (MMEJ), which involves using programma­ble nucleases to create double-stranded DNA breaks and then inserting the genes at those sites. In studies, MMEJ has been shown to be up to three times more effi­cient than HDR. Another advantage of this process is that it is active in all phases of the cell cycle. That could provide more op­portunities for introducing therapeutic genes into patients’ genomes, potentially widening the universe of diseases that can be targeted with CRISPR.

EMBRACING AUTOMATION

The ability to industrialize N of 1 CRISPR development to help tens of thou­sands of patients with ultra-rare diseases will require several innovations on the process side. Basically, everything the in­dustry does now must be scaled down to accommodate smaller batch sizes. Au­tomation will help us get there.

One innovation that could make it easier to replicate the process that led to KJ’s cure can be described as “RNA in a box”—the ability to use one tool to move quickly from a digital sequence to the final CRISPR product. It would be a true plug-and-play approach: With the use of a sin­gle machine, researchers could put in DNA and enzymes, hit “play,” and receive an RNA drug packaged in a nanoparticle. This isn’t a reality yet, but several compa­nies are working on technology to realize this vision.

To manufacture N of 1 therapies effi­ciently, developers need to reduce reaction volumes to move from producing several hundred milligrams to under 100. This is a challenge, because normally a drug is advanced from the reaction stage into a purification system, which can cause a lot of the volume to be lost. When the final volume is several hundred milligrams, los­ing a small amount is generally not prob­lematic. But with smaller batch sizes, manufacturers can’t afford to lose any amount of product. Reducing the complex­ity of the flow path will prevent product loss. Innovations on the equipment side to accommodate small batch sizes will be critical.

THE EVOLVING ROLE OF AI IN CRISPR DEVELOPMENT

Artificial intelligence could improve every stage of N of 1 CRISPR development, further enhancing efficiencies. KJ’s devel­opment team didn’t rely on AI to complete the CRISPR therapy in six months, but it’s easy to see how the technology could com­press timelines even more. Designing per­sonalized CRISPR therapies for robust expression and stability with low off-target editing risk takes many rounds of in silico and in vivo experimentation. AI-enabled tools can help researchers optimize these characteristics faster. AI can also help de­velopers quickly determine which manu­facturing parameters will result in high rates of purity and yield.

Today, CRISPR still requires largely manual processes that can be time-con­suming and expensive. The closer we can get to full automation of all steps, from de­sign through manufacturing, the faster we’ll be able to develop N of 1 therapies. We’ll also drive down costs.

Most scientists, myself included, keep our emotions in check. But because it’s rare to see one’s work contribute directly to significant milestones, I couldn’t help but feel satisfied by KJ’s positive outcome. Most importantly, I am encouraged that this experience will light a path towards further advances in CRISPR therapeutics. From this experience, scientists in acade­mia and industry can apply lessons learned to future innovations, so there can be millions more KJs that can be cured with CRISPR.

REFERENCES

1. Front. Bioeng. Biotechnol., 21 November 2022. Sec. Cell and Gene Therapy. Volume 10 – 2022. https://doi.org/10.3389/fbioe.2022.942325.
2. N Engl J Med. 2025; 392:2235-2243. Vol. 392 No. 22. Published May 15, 2025. doi: 10.1056/NEJMoa2504747.
3. J Control Release. Feb. 2022. Vol. 342, pp. 345–361. doi: 10.1016/J.JCONREL.2022.01.013.
4. J Control Release. Feb. 2022. Vol. 342, pp. 345–361. doi: 10.1016/J.JCONREL.2022.01.013.
5. https://www.frontiersin.org/journals/bioengineering-and-biotechnology/articles/10.3389/fbioe.2022.895713/full.
6. J Exp Med. 2022 Apr 22;219(5):e20211530. doi: 10.1084/jem.20211530

Dr. Venkata Indurthi is Chief Scientific Officer of Aldevron, a Danaher Life Sciences company that works with CRISPR innovators from early research through clinical use, providing custom nucleases, next-generation plasmid vectors, RNP complexing and analytic services, and more. He has been a member of the Aldevron team since he earned his PhD in Pharmaceutical Sciences from North Dakota State University, Fargo, ND, in 2016.