Abstract Background: CD7 chimeric antigen receptor (CAR) T-cell therapy has recently emerged as a promising strategy to induce remission of relapsed or refractory (R/R) CD7-positive hematologic malignancies. We previously demonstrated that a novel “all-in-one” approach of sequential CD7-directed CAR T-cell therapy followed by haploidentical hematopoietic stem cell transplantation (haplo-HSCT)—without pharmacologic myeloablation or graft-versus-host disease (GVHD) prophylaxis—was safe and effective in 10 R/R patients, with the potential for durable disease remission (NEJM 2024). Here, we present interim results from an ongoing phase I, prospective, single-arm study evaluating the safety, feasibility, and outcomes of this approach in an expanded cohort (ClinicalTrials.gov NCT05827835). Methods: Patients with R/R CD7-positive hematological malignancies, regardless of prior allo-HSCT history, were deemed eligible for enrollment. Lymphodepleting chemotherapy based with fludarabine, cyclophosphamide, and etoposide was administered, followed by infusion of CD7 CAR T cells at a dose of 2×10⁶ cells/kg derived either from previous transplant or newly HLA-haploidentical donors. Upon achieving complete remission with incomplete hematologic recovery (CRi), patients underwent haplo-HSCT in the absence of pharmacologic conditioning or GVHD prophylaxis. The primary objective was to evaluate safety/tolerability, and secondary objectives were clinical outcomes including chimerism, progression-free survival (PFS) and overall survival (OS). Results: As of August 5, 2025, 19 patients were enrolled, of whom 15 (78.9%) had completed at least 3 months of follow-up. Median age was 48.0 (range, 23.8-81.9) years, and 8 patients (42.1%) were male. Diagnoses included acute myeloid leukemia (AML) in 10 patients (52.6%), T-lymphoblastic lymphoma (T-LBL) in 5 (26.3%), and T-cell acute lymphoblastic leukemia (T-ALL) in 4 (21.1%). 13 patients (68.4%) had received 3 or more lines of prior treatment, and 8 (42.1%) had a history of allo-HSCT, including 1 with 2 previous allo-HSCT. One patient had previously received CLL-1 CAR-T therapy. CD7 CAR T-cell products were successfully manufactured for all patients, which were sourced from newly selected haploidentical donors in 15 cases (78.9%) and from prior transplant donors in 4 (21.1%). Donor sources included offspring in 10 cases (52.6%), parents in 6 cases (31.6%), and siblings in 3 cases (15.8%). All patients achieved MRD-negative CRi following CAR-T infusion, and mostly proceeded to haplo-HSCT within the first month of CD7 CAR T-cell infusion, with median interval of 16 (range, 10–32) days. Cytokine release syndrome occurred in 84.2% (n=16) at grades 1–2 and in 15.8% (n=3) at grade 3. Grades 3-4 cytopenia were reported in 100% (n =19). Acute GvHD (aGVHD) after allo-HSCT was observed in 9 (47.4%) patients at grades 1-2 and 3 (15.8%) patients at grade 3, with the cumulative incidence of aGVHD at 3-month after HSCT of 58.6%. All aGVHD events were effectively managed and resolved. Among 17 evaluable patients, 16 (94.1%) achieved full donor chimerism at 1-month post-transplantation, while 1 patient exhibited mixed chimerism and experienced early relapse thereafter. Chronic GVHD occurred in only 1 of 13 evaluable patients. With a median follow-up of 7.1 (range, 1.1 to 19) months after CAR-T infusion, 13 patients remained alive, of whom 11 were disease-free at their last follow-up. CD7-negative relapse was detected in 5 patients (26.3%), with a median interval from CAR-T to relapse of 3.3 (range, 2.0 to 4.4) months. A total of 6 patients died, including 3 from severe infectious complications and 3 from disease relapse. OS and PFS at 12 months after CAR-T were 54.8% (95% CI, 28.6%-100%) and 48.7% (95% CI,25.1%-94.7%), respectively, with the median OS and PFS of respectively 12.3 months and 9.2 months. Conclusion: This interim analyses further supported that this integrative strategy is safe and feasible in R/R CD7-positive malignancies, offering a promising therapeutic option—particularly for patients ineligible for conventional allo-HSCT.
Yu et al. (Mon,) studied this question.