Gene-Edited Stem Cell Transplants Shield Healthy Cells in Trial for Aggressive Blood Cancers

Thirty adults with some of the hardest-to-treat blood cancers received a stem cell transplant with an unusual feature: the donor cells had been edited with CRISPR to remove a single protein, CD33. A clinical trial led by researchers at Washington University School of Medicine in St. Louis tested whether that change could make follow-up cancer therapy safer. The results appear in Nature Medicine.

The patients had acute myeloid leukemia (AML) or myelodysplastic syndrome (MDS) and were at high risk of relapse. For aggressive blood cancers, a stem cell transplant is often the only treatment with curative potential, yet the cancers frequently come back afterward. The phase 1/2 trial ran at Siteman Cancer Center, which is based at Barnes-Jewish Hospital and WashU Medicine, along with 14 other sites in the U.S. and Canada.

The problem the edit aims to solve is specific. CAR-T cell therapy, an immunotherapy that works against some aggressive blood cancers, has not worked against all of them, including AML and MDS. John F. DiPersio, MD, Ph.D., the study's corresponding author, explains that myeloid cancers are tricky because the proteins CAR-T cells home in on are also present on healthy myeloid cells, including the donor stem cells given in a transplant. The therapy can therefore destroy healthy blood stem cells, which risks toxicity and a dangerous inflammatory cascade. It may also blunt the treatment, because many CAR-T cells attack the wrong targets and leave cancer cells untouched.

The basic concept was first described by Miriam Y. Kim, MD, now an assistant professor of medicine at WashU Medicine. She began the work as a postdoctoral researcher at the University of Pennsylvania and continued it in the DiPersio lab.

CD33 is an appealing protein to delete because it appears only on blood-forming cells, not other tissues, and people born without it have no apparent health problems. If a patient receives CD33-free donor cells, any remaining cells carrying CD33 should in theory be cancer, so a CD33-targeted therapy would spare the healthy donor cells. The edited product is called tremtelectogene empogeditemcel, or trem-cel.

As a proof of concept, the patients received gemtuzumab ozogamicin after transplant. It is not a CAR-T therapy but an engineered antibody that targets CD33 and carries an anti-cancer drug. The FDA has approved it for CD33-positive AML, and it is in trials for CD33-positive MDS. It helps prevent relapse, but its use is limited by liver toxicity and damage to blood cells, including dangerously low white cell, red cell and platelet counts.

All 30 patients achieved engraftment, meaning the transplanted cells settled in the bone marrow and began working, by day 28. Platelet production returned by day 16 on average. Those timelines are comparable to standard transplants. Nineteen patients received at least one cycle of the antibody in a dose-escalation protocol, and researchers established a recommended dose. Blood counts held across all doses, suggesting the edited transplant protected patients from the low counts usually seen with this drug after a standard transplant.

Side effects resembled those of standard transplants: anemia, low platelets, fever, infections and graft-versus-host disease, in which donor cells attack the patient's healthy tissue. Average survival was just over 14 months. Seven patients died during the study, four from cancer progression and three from transplant-related causes: kidney failure, liver toxicity and sepsis.

"We are encouraged by the results," said DiPersio, who directs WashU Medicine's Center for Gene and Cellular Immunotherapy, noting that outcomes looked very similar to standard transplantation. He hopes to combine the approach with CD33-targeted immunotherapies such as CAR-T cells. A single case study published in October 2025 in JCO Precision Oncology points that way: a patient with a highly aggressive AML relapsed after a CD33-deleted transplant, then received CD33-targeted CAR-T cells made from the same donor's T cells. That patient reached complete remission and remains cancer free more than a year later, with normal blood production from cells lacking CD33. One case is not a trial; the next step is pairing the two approaches in larger studies.