RevCAR T did not lead to any instances of GVHD or ICANs, and it produced complete remissions in CD123-positive relapsed/refractory AML.
Treatment with a first-in-class, switchable, off-the-shelf allogeneic CAR T-cell therapy (RevCAR T) produced complete remissions and displayed a manageable safety profile in patients with relapsed/refractory, CD123-positive acute myeloid leukemia (AML), according to data from the phase 1a portion of the RevSTAR-123 (AVC-201-01) trial (NCT05949125) presented at the
Findings showed that among response-evaluable patients (n = 12), 4 achieved a response, with all responses concentrated at the two highest dose levels evaluated. In evaluable patients (n = 8) treated at dose level 12 (DL12) and dose level 15 (DL15), 3 achieved a complete remission (CR) or CR with partial hematologic recovery (CRh), including 2 patients who experienced minimal residual disease (MRD)–negative CRs. Moreover, 1 patient achieved a morphologic leukemia-free state (MLFS), and 1 additional patient experienced MRD conversion.
Regarding safety, the maximum tolerated dose (MTD) was not reached, and DL15 represents the recommended phase 2 dose (RP2D). Across all dose levels (n = 17), no instances of graft-vs-host disease (GVHD) or immune effector cell–associated neurotoxicity syndrome (ICANS) were reported. One dose-limiting toxicity (DLT) was reported in a patient who experienced grade 2 immune effector cell–associated hemophagocytic lymphohistiocytosis-like syndrome (IEC-HS); this was successfully managed using the on/off switch mechanism of treatment, anakinra (Kineret), and dexamethasone. Notably, no treatment-related deaths were reported.
“These responses tended to be durable, at least at the higher dose levels,” lead study author Martin Wermke, MD, said in a presentation of the data. Wermke is a senior physician and director of the Trial Management / Early Clinical Trial Unit at the National Center for Tumor Diseases Dresden in Germany.
Although autologous CAR T-cell therapies have been integrated into treatment paradigms for select B-cell malignancies, developing these types of therapies for patients with AML has been challenging due to a lack of clean, optimal antigens to target and intrinsic T-cell dysfunction that complicates the production of effective autologous strategies.
“CD123 is a potentially interesting target antigen in AML because it's expressed by the majority of patients, and not only on more differentiated leukemic blasts, but also on the leukemic stem cells,” Wermke said. “However, it is also expressed on healthy hematopoietic progenitors, so targeting with a conventional CAR T-cell [therapy] might come with the risk of long-lasting aphasia.”
By developing an allogeneic CAR T-cell therapy derived from healthy donor cells, this approach could circumvent the issue of an impaired autologous immune system, Wermke said, noting other genetic modifications to the donor cells.
“We first CRISPR’ed out the T-cell receptor locus to prevent GVHD. We deleted the whole MHC class II locus to prevent host [CD4] T-cell rejection, and we selectively removed the HLA-A molecule to prevent CD8- and NK-cell rejections,” Wermke explained. “That leaves us with HLA-B and -C in place, for which we have to match our donor T cells to the patient’s HLA haplotype.”
He added that RevCAR T is an inert second-generation CAR, which does not recognize any target antigen in the human body. Rather, it carries a universal adapted peptide on its surface, which serves as the target for a R-TM123-targeting module, “which is basically a bispecific antibody,” he said. “This connects RevCAR T to the target antigen CD123 on the AML cell, activates the T cell, and lyses the AML cell.”
RevCAR T is dosed as a single infusion, and R-TM123 is given continuously because it is a small molecule with a short half-life.
“[Continuous infusion], however, also provides us with a safety switch,” Wermke added. “We can switch off the infusion whenever this is necessary due to safety concerns, and we can switch it on again if needed.”
RevSTAR-123 enrolled patients at least 18 years of age with relapsed/refractory or MRD-positive AML who lacked available treatment options.1,2 Patients were required to have CD123 expression on at least 20% of leukemic blasts, an ECOG performance status of 0 or 1, and HLA-B and HLA-C matching with available RevCAR T.1
After 3 days of lymphodepletion with fludarabine and cyclophosphamide, patients received a single infusion of RevCAR T followed by continuous intravenous R-TM123 throughout a 21-day induction cycle, with 12-day consolidation cycles thereafter. The dose-escalation design tested 18 dose-level combinations across 3 RevCAR T cell doses (100 million, 250 million, and 500 million cells) and 6 R-TM123 doses (0.3 to 9.6 mg/day). These ranges included DL12 (R-TM123, 2.4 mg/day; RevCAR T, 500 million cells) and DL15 (4.8 mg/day; 500 million cells)
Safety, DLTs, and determining the MTD/RP2D served as the trial’s primary end points. Secondary end points included clinical activity, pharmacokinetics, and immunogenicity.
The 17 patients treated in phase 1a had a median age of 63 years (range, 41-74). They received a median of 4 prior lines of therapy (range, 1-18), and 65% had undergone prior allogeneic hematopoietic stem cell transplant. Per ELN 2017 classification, 52.9% had adverse-risk disease, and none had favorable-risk disease. Median baseline bone marrow blast percentage was 28% (range, 2%-73%), and the median CD123-positive blast fraction rate was 77% (range, 32%-99%).
At DL15, the median peak expansion was 161,835 copies per μg of genomic DNA (range, 7,240–659,070). The median RevCAR T persistence across all 17 patients was 43 days (range, 0-112; interquartile range, 18-91), and 47% maintained detectable RevCAR T beyond 30 days. At the DL12 and DL15, median persistence was 42 days, and 63% of patients had persistence beyond 30 days.
Cytokine release syndrome (CRS) occurred in 76% of patients, with all instances deemed treatment related. Grade 3 or higher CRS was observed in 1 patient (6%).
Hematologic all-cause adverse effects included thrombocytopenia (any-grade, 29%; grade ≥ 3, 6%), neutropenia (18%; 12%), and anemia (29%; 29%). Treatment-related grade 3 or higher neutropenia occurred in 1 patient (6%). Febrile neutropenia was reported at any grade and grade 3 or higher in 24% of patients, although no instances were deemed related to treatment. The rate of all-grade and grade 3 or higher sepsis was 12% each, with none of these events related to treatment. Fatigue (any-grade/all-cause, 41%; all-grade/treatment related, 29%; all-cause, grade ≥ 3, 12%) and increased gamma-glutamyl transferase levels (any-grade/all-cause, 18%; all-grade/treatment related, 6%; grade ≥ 3/all-cause, 18%; grade ≥ 3/treatment related, 6%) were also reported.
Based on the phase 1a findings, the trial is continuing into phase 1b, which plans to enroll 20 patients with relapsed/refractory AML who will be treated at DL15.
