For patients fighting acute myeloid leukemia (AML) and myelodysplastic syndromes (MDS), the greatest fear is often not the initial diagnosis, but the relapse. Even after a successful allogeneic hematopoietic cell transplant—the “gold standard” for high-risk cases—the risk of the cancer returning remains a persistent threat. The challenge for physicians has long been a biological catch-22: the most effective targeted therapies often attack the same markers found on both the leukemia cells and the healthy stem cells needed to rebuild the patient’s immune system.
A pioneering first-in-human phase 1/2 trial is attempting to break this deadlock by using CRISPR-Cas9 gene editing to create a “stealth” immune system. By deleting a specific protein marker from donor stem cells before they are transplanted, researchers are creating a biological shield that allows doctors to use potent chemotherapy drugs to kill remaining cancer cells without destroying the patient’s new, healthy blood supply.
The approach centers on a protein called CD33, which is widely expressed on AML cells and healthy myeloid progenitor cells. While the drug gemtuzumab ozogamicin (GO), marketed as Mylotarg, is highly effective at targeting CD33 to destroy leukemia, it typically causes severe myelosuppression because it cannot distinguish between a cancer cell and a healthy one. This trial tests a new product called trem-cel—donor stem cells edited via CRISPR to be CD33-deleted—followed by GO maintenance therapy to mop up any residual disease.
As a board-certified physician, I have seen how the toxicity of maintenance therapies can often limit their utility. The brilliance of this design lies in its sequence: first, replace the patient’s bone marrow with cells that are “invisible” to the drug, and then administer the drug with a level of intensity that was previously too dangerous.
The Architecture of a ‘Stealth’ Transplant
The process begins not with the patient, but with a meticulously screened HLA-matched donor. Once a donor is identified, their hematopoietic stem cells (CD34+ cells) are collected via apheresis. In a laboratory setting, researchers use CRISPR-Cas9 technology to target exon 3 of the human CD33 gene. By creating a precise break in the DNA, the Cas9 nuclease effectively “knocks out” the gene, ensuring that the resulting cells do not express the CD33 protein on their surface.
Once edited and verified, these trem-cel cells are infused into the patient following myeloablative conditioning—typically a regimen of busulfan or total body irradiation (TBI) designed to clear out the patient’s existing, diseased marrow. The goal is for these CD33-negative cells to engraft and begin producing a new lineage of white blood cells, red blood cells, and platelets that the drug Mylotarg will simply ignore.
The trial design incorporates a critical safety window. Patients are monitored for neutrophil engraftment, and at approximately day 60 post-transplant, a bone marrow biopsy is performed. Only those who have achieved hematopoietic recovery and remain in complete remission are eligible to begin the maintenance phase of the treatment. This ensures the new “shielded” immune system is stable before the targeted therapy is introduced.
Trial Design and Dose Optimization
The study (ClinicalTrials.gov identifier: NCT04849910) was structured in two primary parts to determine the maximum tolerated dose (MTD) and the recommended phase 2 dose (RP2D) of the maintenance GO therapy. Because this was a first-in-human study, safety was the primary metric, specifically the cumulative incidence of neutrophil engraftment by day 28.
Part 1 utilized a standard “3+3” dose-escalation design, testing GO doses of 0.5, 1.0, and 2.0 mg/m². This allowed the Data Evaluation Committee (DEC) to monitor for dose-limiting toxicities (DLTs) and pharmacokinetic data before moving to the next cohort. Part 2 then expanded on the RP2D to further assess safety and preliminary efficacy.
| Trial Component | Primary Objective | Patient Status | Dosing Strategy |
|---|---|---|---|
| Part 1 (Escalation) | Identify MTD and RP2D | Engrafted & in remission | 0.5 to 2.0 mg/m² (3+3 design) |
| Part 2 (Expansion) | Assess safety/preliminary efficacy | Engrafted & in remission | Fixed RP2D dosing |
| DTC (Treatment Cohort) | Induction/Consolidation safety | Relapsed or MRD+ | Escalated induction cycles |
For patients who unfortunately relapsed or showed evidence of minimal residual disease (MRD+), the trial included a Dose Treatment Cohort (DTC). In this group, GO was administered as an induction and consolidation therapy, providing a secondary line of defense to attempt to bring the patient back into remission.
Measuring Success and Managing Risk
The complexity of this trial required an exhaustive set of endpoints. Beyond simple engraftment, researchers tracked “donor myeloid chimerism”—essentially confirming that the blood cells in the patient’s body were indeed the edited donor cells and not a return of the patient’s own leukemia. They used deep sequencing and flow cytometry to monitor the percentage of CD33 editing over time, ensuring the “stealth” property persisted throughout the treatment.
Risk management was paramount. The trial established strict “stopping rules” to protect participants. If there was an increased incidence of primary or secondary graft failure, severe acute Graft-versus-Host Disease (GvHD), or an elevated rate of sinusoidal obstruction syndrome (SOS/VOD)—a known complication of Mylotarg—the Data and Safety Monitoring Board had the authority to halt dosing.
The study was conducted across an impressive network of institutions, including Memorial Sloan Kettering Cancer Center, Stanford University, and the National Cancer Institute, reflecting the high-stakes nature of this research. By involving multiple centers, the trial sought to ensure that the results were reproducible across different clinical environments.
What Remains Unknown
While the conceptual framework is robust, the trial faced challenges common to early-phase gene therapy. The Part 2 expansion cohort was not fully enrolled due to the early termination of the study, meaning some of the long-term efficacy data remains incomplete. Specifically, the DTC endpoints were not fully assessed as only one patient was enrolled in that specific arm at the time of the data cut.
the long-term biological impact of having a CD33-negative myeloid system is still being studied. While CD33 is a target for therapy, it is a natural part of myeloid biology; understanding whether its absence affects the overall function of the immune system over years, rather than months, is the next great question for the oncology community.
The next critical checkpoint for this research will be the publication of final follow-up data regarding relapse-free survival (RFS) and overall survival (OS) for the evaluable population. These figures will determine if the CRISPR-edited “stealth” approach provides a statistically significant advantage over standard-of-care transplants in preventing AML relapse.
Do you think gene-edited transplants are the future of leukemia treatment? Share your thoughts in the comments or share this story with a colleague.
