CD7 CAR T-Cell Therapy Plus Allogeneic HSCT Is Safe, Feasible in CD7+ Hematologic Malignancies


Sequential CD7 CAR T-cell therapy plus haploidentical HSCT without GVHD prophylaxis is effective and safe in CD7-positive hematologic malignancies.

He Huang, MD, PhD

He Huang, MD, PhD

Sequential CD7 CAR T-cell therapy and haploidentical hematopoietic stem-cell transplantation (HSCT) without graft-vs-host disease (GVHD) prophylaxis is effective and safe in patients with relapsed/refractory CD7-positive hematologic malignancies who are not eligible for conventional allogeneic HSCT, according to findings from a study that were published in the New England Journal of Medicine.1

This “all-in-one” treatment strategy elicited complete remission (CR) with incomplete hematologic recovery (CRi) with grade 4 pancytopenia after CAR T-cell therapy in all 10 patients with relapsed/refractory CD7-positive lymphoma or leukemia enrolled in 1 of 2 prospective clinical trials between November 2021 and September 2023. Of these, 9 patients had minimal residual disease (MRD)–negative incomplete hematologic recovery, and 1 patient (patient 6) had MRD-positive incomplete hematologic recovery.

“Collectively, our integrated strategy maximized antileukemic efficacy from both persisting CAR T cells and graft-vs-leukemia potential, providing a feasible approach for patients with relapsed or refractory CD7-positive cancers who are ineligible for conventional allogeneic HSCT,” senior study author He Huang, MD, PhD, of the Bone Marrow Transplantation Center at First Affiliated Hospital, Zhejiang University School of Medicine in Hangzhou, China, and coauthors, wrote in the paper.

Nine patients were enrolled in a clinical trial (NCT04599556) evaluating donor-derived CD7 CAR T cells. Due to manufacturing limitations of donor-derived CAR T cells resulting from the COVID-19 pandemic, 1 patient was enrolled in a compassionate-use program (NCT04538599) that provided access to universal CD7 CAR T-cell therapy after the end of the clinical trial.

Patients were deemed eligible for sequential allogeneic CD7 CAR T-cell therapy and haploidentical HSCT if they experienced CRi and severe bone marrow hypocellularity or pancytopenia following CD7 CAR T-cell therapy, had detectable levels of CD7 CAR T cells, and had no history of allogeneic HSCT.

All patients received an intensified lymphodepleting regimen consisting of fludarabine at 30 mg/m2, cyclophosphamide at 300 mg/m2, and etoposide at 100 mg for 5 consecutive days, followed by an infusion of either haploidentical CD7 CAR T cells at 2 x 106 cells/kg (n = 9) or universal CD7 CAR T cells at 5 x 106 cells/kg (n = 1) on day 0. Patient 1 underwent allogeneic HSCT as salvage therapy for pancytopenia after CAR T-cell therapy, and all other patients underwent allogeneic HSCT as prophylaxis. Patients received no additional GVHD prophylaxis drugs or pharmacologic pre-HSCT conditioning regimens and instead relied on the immunosuppressive effects of lymphodepletion before CAR T-cell therapy and the CD7 CAR T cells. For the 9 patients who received haploidentical CD7 CAR T cells, hematopoietic stem and progenitor cells (HSPCs) were obtained from the same donor. For the 1 patient who received universal CD7 CAR T cells, HSPCs were obtained from a new haploidentical donor.

Seven patients had acute myeloid leukemia (AML), 2 patients had T-cell acute lymphocytic leukemia, and 1 patient had stage IVA T-cell lymphoblastic lymphoma. The median age of patients at enrollment was 56.5 years (range, 13.7-72.5), and patients had received a median of 9.5 courses of therapy (range, 4-15). All patients had bone marrow involvement, patients had a median blast percentage of 36.0% (range, 2%-87%), and 2 patients had extramedullary disease. Patients had a median CD7 expression rate of 93.0% on blast cells (range, 80.7%-97.7%). The median time from diagnosis to CAR T-cell infusion was 13.1 months (range, 4.6-33.7) and the median time from CAR T-cell infusion to HSPC infusion was 19 days (range, 15-89).

Patient 1 had persistent grade 4 pancytopenia for 3 months after CAR T-cell infusion, which was complicated by candida sepsis followed by an Enterococcus faecalis infection. This patient received a salvage haploidentical HSPC infusion. The 9 other patients in the study had similar conditions but proceeded to haploidentical HSCT within 1 month after CAR T-cell infusion. Mononuclear cells, CD34-positive HSPCs, and CD3-positive T cells were infused at respective median doses of 9.2 x 108/kg (range, 4.8-20.3), 4.8 x 106/kg (range, 3.5-8.4), and 5.5 x 108/kg (range, 2.9-13.7).

Regarding safety, 9 patients experienced cytokine release syndrome (CRS; grade 1, n = 5; grade 2, n = 4), all episodes of which were successfully managed. The median time to CRS onset was 1 day (range, 1-5), and the median duration of CRS was 9.5 days (range, 8-13).

No patients experienced immune effector cell–associated neurotoxicity syndrome. All patients experienced grade 4 pancytopenia with no signs of abatement after medical intervention, and bone marrow assessments indicated severe hypocellularity after CAR T-cell infusion.

Four patients developed GVHD, 1 of whom had CAR T-cell therapy–related GVHD and 3 who had HSCT-related GVHD. Patient 2 experienced grade 2 skin GVHD on day 7 after CAR T-cell infusion, which completely resolved on day 11 after treatment with a glucocorticoid and antipruritic. Patients, 1, 4, and 6 experienced short-term grade 2 acute GVHD following haploidentical HSCT. No patients experienced chronic GVHD.

Five patients experienced any-grade bacterial or fungal infections. Patient 4 experienced a multidrug-resistant bloodstream infection at 3.3 months followed by an intracranial infection; this patient died of septic shock at 3.7 months. Patient 8 died on day 13 after HSCT from encephalitis due to human herpesvirus 6 infection and septic shock due to Staphylococcus haemolyticus. All other infections in other patients were successfully managed with antibiotics. Epstein-Barr virus and cytomegalovirus reactivation was detected in all patients except patient 10.

This study showed successful relief of pancytopenia after allogeneic HSCT. One month after HSCT, 7 of the 9 evaluable patients experienced full donor chimerism. Patient 3 maintained mixed donor chimerism within the first 3 months (quantified at 88.35%) and achieved full donor chimerism 6 months after HSCT. The patients experiencing full donor chimerism had successful hematopoietic recovery within 1 month after HSCT, with a median time to neutrophil engraftment of 11.5 days (range, 8-17) and a median time to platelet engraftment of 12 days (range, 8-29). Patient 5 experienced autologous hematopoietic recovery 1 month after HSCT, which may be attributed to delayed infusion and compromised expansion of CAR T cells. Immune cell recovery was also observed in this study.

On day 28 after allogeneic HSCT, a post-HSCT efficacy analysis in 9 patients showed that all achieved MRD-negative CR. PET-CT scans demonstrated the complete regression of extramedullary lesions in patient 1 at 3 months and patient 10 at 1 month post-HSCT.

At a median follow-up duration of 15.1 months (range, 3.1-24.0) among survivors, 2 patients with AML experienced CD7-negative leukemia relapse in the bone marrow, at 5.6 months and 4.3 months after CAR T-cell therapy in patients 6 and 7, respectively. Patient 7 died from disease progression at 4.8 months after CAR T-cell therapy. At the data cutoff of November 8, 2023, 6 patients continued to be in MRD-negative CR without requiring additional treatment.

The estimated 1-year overall survival rate was 68% (95% CI, 43%-100%), and the estimated 1-year disease-free survival rate was 54% (95% CI, 29%-100%).

All patients exhibited robust in vivo CAR T-cell expansion. The median time to maximum CAR T-cell expansion per quantitative real-time polymerase chain reaction (qPCR) was 16 days (range, 7-20), and the median level of maximum CAR T-cell expansion was 2.9 x 105 copies per μg of DNA (range, 0.1-6.9). The median time to maximum CAR T-cell expansion per flow cytometry was 11.5 days (range, 8-18), and the median level of maximum CAR T-cell expansion was 316.5 cells per µL (range, 155.4-6501.9).

Among the 6 patients who remained in MRD-negative CR at the data cutoff, 5 with donor engraftment had detectable CAR T cells upon their last assessment. The patient who achieved autologous hematopoietic recovery had no detectable CAR T cells 3 months after CAR T-cell infusion. In the 2 patients with relapsed CD7-negative leukemia, CAR T cells were undetectable by flow cytometry at relapse but were detectable by qPCR.

The median time to eradication of CD7-positive T cells in the peripheral blood was 8.5 days (range, 5-13) and was accompanied by the expansion of CD7-negative/CD3-positive T cells. Of 9 evaluable patients, CD7-positive T cells and natural killer cells were undetectable in 8 patients until the last follow up, excepting patient 5, who had autologous hematopoiesis recovery. In the 2 patients with relapsed CD7-negative leukemia, all normal T cells remained CD7 negative. An assay for transposase-accessible chromatin sequencing on CD7-negative T cells showed reduced chromatin accessibility at the CD7 locus.

“These results suggested that the persisting CAR T cells are able to eliminate CD7-positive cells and that surviving T cells have suppressed expression of CD7 as a consequence of the selective pressure,” the study authors noted.

A one-way mixed-lymphocyte reaction assay was performed on T cells from 3 patients who had over 18 months of follow up to clarify the low incidence of GVHD in the absence of pharmacologic GVHD prophylaxis. This assay demonstrated reduced proliferation after allogeneic cell stimulation vs T cells from corresponding donors, suggesting lower alloreactivity in donor-derived CD7-negative normal T cells after HSCT. A T-cell receptor (TCR) analysis in these 3 patients and their corresponding donors showed increased post-HSCT clonality and decreased diversity of the TCRβ complementary determination region 3 repertoires over time. In patients 1, 2, and 3, the total frequency of the top 50 prevalent TCR clones at 12 months was over 70%, further indicating clonal expansion. These clones had a total frequency of less than 25% in donor T cells.

“Collectively, these data suggest that the long-lasting CD7 CAR T cells can eradicate donor-derived CD7-positive T cells and may contribute to GVHD prophylaxis,” the authors explained. “On the other hand, the recovered T cells still maintained their capability for activation, and their TCR dynamics indicated the potential for a graft-vs-leukemia effect.”

Furthermore, the investigators inferred that observed relapses in the 2 patients with AML were not induced by genetic mutations of CD7 but occurred because of the expansion of pre-existing CD7-negative blasts or transcriptional CD7 expression suppression.

A phase 2 trial (NCT05827835) investigating sequential allogeneic CD7 CAR T-cell therapy and haploidentical HSCT in a larger cohort of patients with relapsed/refractory CD7-positive hematologic malignancies is ongoing.2


  1. Hu Y, Zhang M, Yang T, et al. Sequential CD7 CAR T-cell therapy and allogeneic HSCT without GVHD prophylaxis. New Engl J Med. Published online April 24, 2024. doi:10.1056/NEJMoa2313812
  2. CD7 CAR-T bridging to alloHSCT for R/R CD7+malignant hematologic diseases. Updated April 25, 2023. Accessed April 24, 2024.
Related Videos
Lori A. Leslie, MD
James Ignatz-Hoover, MD, PhD
Changchun Deng, MD, PhD
Sundar Jagannath, MBBS
Guenther Koehne, MD, PhD
Krina K. Patel, MD, MSc
Ibrahim Aldoss, MD
Olalekan O. Oluwole, MBBS, MD, associate professor, medicine, hematology/oncology, Vanderbilt Institute for Infection, Immunology and Inflammation, Vanderbilt University Medical Center
Andrew P. Jallouk, BS, MD, PhD, assistant professor, Vanderbilt University
Elias Jabbour, MD, professor, Department of Leukemia, Division of Cancer Medicine, The University of Texas MD Anderson Cancer Center