For the roughly 30,000 Americans living with hemophilia and thousands more with rare enzyme deficiencies, treatment often means a lifetime of intravenous infusions, sometimes multiple times a week. A growing body of research now suggests a radically different possibility: editing blood stem cells so that a patient’s own immune system becomes a self-renewing factory for the very proteins their body cannot make. As of May 2026, the approach has shown clear results in laboratory and mouse models, and multiple research groups are working to close the gap to human trials.
Turning immune cells into drug factories
The core idea exploits a feature the immune system already has. Plasma cells, the mature descendants of B cells, are among the most prolific protein-producing cells in the body. A single plasma cell can pump out thousands of antibody molecules per second, and some plasma cells survive in bone marrow for decades. Researchers reasoned that if they could gene-edit the stem cells that give rise to B cells, the resulting plasma cells would secrete not just natural antibodies but engineered therapeutic proteins.
A study published on PubMed Central demonstrated exactly that. Scientists used gene editing on human hematopoietic stem and progenitor cells (HSPCs), the precursor cells that seed the entire blood and immune system. After transplantation, those edited HSPCs generated B cells that responded to antigen stimulation, expanded, and then matured into plasma cells producing high levels of engineered antibodies and other cargo proteins. The PMC listing for this paper (PMC12871283) indicates a 2025 publication date. The result was a biological assembly line: one edit at the stem-cell level cascading into sustained protein output downstream.
A separate team, publishing in Nature Communications in 2020, took a different route to the same destination. Rather than targeting the B-cell lineage, they edited human HSPCs so that the cells’ red blood cell descendants would express and secrete therapeutic proteins, including clotting factor IX (used to treat hemophilia B) and lysosomal enzymes relevant to storage diseases such as Gaucher and Fabry disease. Functional assays confirmed the proteins were active, and the researchers showed uptake and cross-correction in patient-derived cells, meaning the secreted proteins were absorbed by neighboring deficient cells and restored their function. The edited cells also engrafted successfully in mouse models.
A 2021 review paper on PubMed Central pulled these threads together, framing edited B cells as versatile platforms for producing both antibodies and non-antibody proteins. That review cataloged the gene-editing strategies researchers have used to insert therapeutic cargo into specific genomic locations and assessed the evidence for long-lived plasma cell durability over time. Taken together, these studies establish a clear proof of concept: edited stem cells can populate the immune system with cells that manufacture medically useful proteins on a sustained basis.
The distance still to travel
Proof of concept in mice is not proof of safety or durability in people, and several hard problems remain unsolved.
The most fundamental gap is the absence of any human clinical trial data. Every result published so far comes from cell cultures or humanized mouse models, animals engineered to carry a partial human immune system. The leap from mouse engraftment to reliable, long-term protein production inside a full human body is substantial. Immune environments differ, cell turnover rates differ, and the sheer scale of a human bone marrow compartment introduces variables that mouse studies cannot capture.
One critical translational question is whether stem cells can be edited inside the body, bypassing the costly and complex process of extracting cells, editing them in a laboratory dish, and transplanting them back. Research published in Nature Biotechnology explored this using envelope-engineered virus-like particles to deliver editing tools directly to HSPCs in humanized mice. The results were promising for measuring editing rates across blood cell lineages, but the same work flagged an unresolved concern: the body’s immune response to the delivery particles themselves, which could neutralize the treatment before it reaches its target.
Children pose a particular challenge. Because pediatric immune systems are still developing, a one-time edit delivered in infancy might not persist through decades of immune cell turnover. Whether patients would eventually need a second dose, and whether the immune system would attack the editing machinery on re-exposure, remains an open question with no published answer.
It is also worth noting how this approach fits alongside gene therapies that have already reached patients. The FDA has approved Casgevy for sickle cell disease (the first CRISPR-based therapy), Hemgenix for hemophilia B, and Zynteglo for beta-thalassemia. Those treatments demonstrated that genetic modification of blood cells can work clinically, but each has limitations: Hemgenix uses a viral vector delivered to the liver rather than editing stem cells, and both Casgevy and Zynteglo require harsh bone marrow conditioning. The stem-cell protein factory approach aims to expand the range of treatable diseases and, potentially, reduce the need for repeated dosing, but it has not yet faced the regulatory and safety scrutiny those approved therapies have passed.
Where industry stands
On the commercial side, Editas Medicine disclosed preclinical proof-of-concept work on in vivo HSPC editing in its Form 10-K filing for the fiscal year ended December 31, 2024, submitted to the U.S. Securities and Exchange Commission. The filing describes targeted lipid nanoparticle formulations designed for extrahepatic delivery, meaning delivery to tissues outside the liver, where most current gene therapies concentrate. However, the filing characterizes the work as early-stage preclinical research, and no timeline for regulatory submissions or clinical trials has been made public.
Other academic groups and smaller biotech firms are also pursuing B cells as therapeutic protein platforms, though most remain in preclinical stages. Commercial interest signals that investors see potential, but it does not, on its own, validate the science. The legally binding nature of SEC filings means companies face consequences for material misstatements, which gives those disclosures a degree of reliability, but they describe corporate strategy and milestones rather than peer-reviewed clinical outcomes.
When the first human trial protocol will matter most
The public record currently contains no data from a human clinical trial, no investigational new drug application filed with the FDA or the European Medicines Agency based on this approach, and no long-term safety data on off-target editing effects in human stem cells. The biological mechanism is real and reproducible in controlled settings. The unresolved question is whether it can be made safe enough, durable enough, and manufacturable enough to reach the patients who need it.
For anyone following this space, whether as a patient weighing future options, a clinician tracking the pipeline, or an investor assessing risk, the next concrete milestone will be the first clinical trial protocol registered with a regulatory agency. That filing has not yet been publicly reported. Until it is, the promise of a self-sustaining immune pharmacy remains exactly that: a promise grounded in strong preclinical science, waiting for its first real test in people.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
