USC engineers renewable immune progenitors for cancer immunotherapy
A team at the University of Southern California has taken cell therapy in a different direction: instead of harvesting mature immune cells and trying to keep them alive after infusion, the researchers started with immature precursors and built a system that can keep producing new cancer-fighting cells. The work centers on granulocyte-monocyte progenitors, or GMPs, a blood-forming population that normally matures into macrophages and related immune cells. Under carefully controlled laboratory conditions, those progenitors could be expanded extensively, genetically engineered, and then used to generate functional immune defenders after administration in mice.
The strategy addresses two persistent bottlenecks in cancer immunotherapy. Mature engineered cells are difficult to manufacture in large numbers, and once infused they may not persist long enough to maintain pressure on tumor growth. By working one developmental step earlier, the USC group created a renewable starting material that kept its identity while multiplying in culture. In the study published in Cell, the engineered progenitors established themselves in bone marrow and other blood-forming sites, where they continued to produce new immune cells over time.
The biological logic is straightforward. If a therapy can seed a lasting source of replacement cells inside the body, the treatment no longer depends entirely on a single infusion surviving intact in circulation. That shifts the goal from temporary delivery to sustained immune remodeling.
Self-renewal in GMPs challenges a long-held view of blood development
Hematopoietic stem cells have long been viewed as the main source of long-term self-renewal in the blood system. Progenitor cells are generally considered more committed and less able to persist over time. This work complicates that framework. Under the right laboratory conditions, GMPs divided extensively while preserving the molecular features that define them and retaining the capacity to mature into functional immune cells.
Qi-Long Ying, MD, PhD, described the resulting platform as scalable and engineerable, with potential implications not only for cancer therapy but also for stem cell biology. The finding matters because manufacturing cell therapies often begins with a practical limitation rather than a biological one: there simply are not enough cells to work with, especially when the therapeutic target demands repeated expansion, genetic modification, and preservation during storage and transport. GMPs appear to offer a way around that choke point.
A second investigator group at Stanford independently showed that the same progenitors could be maintained for extended periods and genetically engineered, reinforcing the reproducibility of the approach. In translational cell therapy, reproducibility is not a footnote. It is the difference between a promising lab trick and a platform that can be moved toward clinical development.
Why macrophage-based therapy has struggled
Macrophages are attractive anticancer tools because they naturally migrate into tissues and can infiltrate tumors. Unlike engineered T cells, which have achieved their strongest results in certain blood cancers, macrophage-based therapies may have an advantage in solid tumors, where cell access is often a major obstacle. They can enter the tumor microenvironment and participate in local immune signaling while also engulfing harmful material.
Yet mature macrophages bring their own problems. They are difficult to expand in large numbers outside the body, hard to genetically modify efficiently, and vulnerable to injury during freezing and storage. After infusion, they may accumulate in organs such as the lungs and liver rather than distribute broadly, and they can disappear quickly. Those constraints limit both potency and durability.
Starting with GMPs sidesteps several of those weaknesses. Rather than manufacturing terminally differentiated cells that are already close to the end of their usable lifespan, the researchers expanded a renewable precursor pool that could later generate engineered macrophages and related immune cells on demand.
CAR engineering amplified antitumor activity in mice
The next step was to equip these progenitors with a chimeric antigen receptor, or CAR, so that their descendants could recognize a specific molecular marker on cancer cells. CAR technology is already a major force in cellular immunotherapy, but it has been used most successfully in T cells. Applying the same logic to GMPs opens a separate therapeutic lane, one that could blend direct cellular activity with broader immune activation.
Some of the progenitor cells carried an added immune-stimulating signal designed to recruit and activate nearby immune cells, including T cells. In the mouse studies, that extra signal strengthened the antitumor effect. The engineered cells slowed progression of both blood cancers and solid tumors, and the combination of CAR targeting plus immune activation produced a stronger response than CAR engineering alone.
The marrow findings are particularly important. After administration, the progenitors settled into blood-forming niches and generated additional engineered immune cells from within the body. Mature macrophage therapy has often been limited by rapid clearance; a self-renewing progenitor source changes the exposure profile entirely. One infusion can potentially lead to a continuing internal supply of effector cells.
The work also raises the possibility of off-the-shelf cell products. Because the added immune-activating signal remained effective even when donor and recipient immune characteristics did not match, the approach may be compatible with donor-derived manufacturing rather than fully individualized preparation. That could reduce the delays and logistical burden that often slow access to advanced cell therapies.
Early disease models extend the platform beyond oncology
Cancer was not the only setting tested. In a mouse model of chronic granulomatous disease, an inherited immune disorder that weakens the ability of certain cells to kill bacteria and fungi, administration of GMPs restored the animals’ capacity to fight bacterial infection. The result suggests that the same developmental strategy could be adapted for immune deficiency states, and possibly for infectious disease applications if later studies support safety and efficacy.
The broader implication is not simply that one more engineered cell type can attack tumors. It is that the developmental stage of the cell may be as important as the receptor or payload placed on it. Engineering too late means working with cells that are already difficult to expand and unstable after infusion. Engineering earlier preserves flexibility, manufacturing capacity, and the chance for sustained in vivo output.
This remains preclinical work. The mouse data do not establish safety in people, and long-term persistence inside bone marrow raises questions that will need careful evaluation before any human study can move forward. A renewable source of immune cells can become an advantage only if its control is precise enough to avoid unintended effects, excessive persistence, or misdirected immune activity.
Even so, the study places GMPs in a new therapeutic category. They are no longer just developmental intermediates on the way to mature myeloid cells. Under the right conditions, they become a renewable engineering substrate for cellular immunotherapy, with possible use in cancer, inherited immune defects, and future infectious disease applications.