Novel Healthy Cell Sparing CAR T-cell Therapy Enters Clinical Studies for Solid Tumors

The BASECAMP-1 study is currently enrolling participants to explore the feasibility of manufacturing a novel CAR T-cell therapy for patients with solid tumors that is specific to a tumor antigen but spares healthy tissue by targeting only cells with human leukocyte antigen (HLA) loss of heterozygosity (LOH), which is believed to be a key tumor escape mechanism.

According to a poster presented at the 2022 ASCO Annual Meeting, the phase 0 study plans to enroll those who meet the criteria for treatment and complete leukapheresis for cell therapy manufacturing. If relapse is detected in these patients, they will go on to receive treatment in the phase 1 EVEREST study, which will explore CAR T-cell therapies that are specific to either CEA (A2B530) or mesothelin (A2B694) using the Tmod platform developed by A2 Biotherapeutics, the sponsor of the study.

"BASECAMP-1 is currently enrolling, and this is an exciting clinical trial for our patients," Diane M. Simeone, MD, professor of surgery, director, Pancreatic Cancer Center at NYU Langone, said in a video that accompanied the poster. "CAR T therapies have been challenging to use in patients with solid tumors, despite their efficacy in hematologic malignancies. Part of the issue has been the dose-limiting on-target, off-tumor toxicities, which have been seen in previous CAR T studies for CEA and mesothelin."

The Tmod CAR T-cell therapy was designed to address some of the challenges experienced by prior cell therapies. The therapy is engineered to target cancer cells specifically while sparing healthy tissue, via a dual receptor construct that can be applied to either autologous or allogeneic T cells or natural killer cells. This logic-gated mechanism uses HLA LOH to differentiate between cells, as it is present on normal cells but lost in some tumor cells. In the construct, CEA or mesothelin (MSLN) activates the T cell while a LIR1-based inhibitory receptor specific to HLA-A*02 works as a blocker, allowing the treatment to only function in those with HLA LOH, which has been in observed in approximately 13% of solid tumors.

The third-generation CAR T-cell therapy contains 2 ligand binding domains. The first is for activation and contains either an anti-CEA or anti-MSLN scFv, although any activating antigen could be eligible. This activator has a CD8α hinge and CD28, 4-1BB, and CD3ζ intracellular signaling domains. The blocker consists of an anti-HLA-A*02 scFv with LIR-1 hinge and intracellular signaling domain. Both are co-expressed in a single construct with a cleavable T2A linker and traduced by a single lentivirus transgene.

Preclinical findings supported this mechanism, with the Tmod cells being able to effectively discriminate between normal and tumor cells, based on the HLA status. In an in vitro analysis, 6 days after co-culture, the tumor cells were eradicated while healthy cells remained with a CEA-targeted Tmod CAR T-cell therapy. A CEA CAR T cell control that did not have the blocker molecule eradicated both healthy and tumor cells in vitro. In an in vivo mouse study, both tumor and normal cells were implanted in mice followed by CAR T cell infusion. Similar findings were seen as the in vitro study, with the standard CAR T cells eliminating all cells and with differentiation between HLA-positive and negative cells seen for the Tmod cells.

"The Tmod CAR T cells kill equivalently to standard CAR T cells. The Tmod CAR T cells are very selective in sparing normal cells," said Simeone.

HLA-A*02:01 loss can be detected in approximately 12.6% to 16.3% of patients with advanced solid tumors. In finding from The Cancer Genome Atlas, LOH was found in 25.3% of non-small cell lung cancers (NSCLC), 33.1% of pancreatic cancers, 16.2% of gastroesophageal cancers, and in 9.6% of colorectal cancers. These findings were confirmed using Tempus xT-Onco next-generation sequencing assay, which is being used in the study to detect HLA LOH. In real-world data from 10,867 cases, Tempus identified LOH in 23.1% of NSCLC cases, 20.8% of gastroesophageal, 19.6% of pancreatic, and in 15.6% of colorectal cancers. "The lower rate of pancreatic cancer was perhaps due to the severe desmoplasia in these tumors," Simeone said.

In the BASECAMP-1 study, patients undergoing primary resected are initially screened for HLA-A*02 heterozygous status, with those who are positive are further analyzed for LOH by Tempus NGS. Those who meet these criteria will undergo leukapheresis and Tmod cell therapy manufacturing. The resulting product is cryopreserved, should relapse occur. The study is currently enrolling at NYU Langone Medical Center, Mayo Clinic, MD Anderson Cancer Center, University of California San Diego, and University of California Los Angeles.

"I am happy to report that we've made good progress with this study so far. Thirty-five patients have been screened, of these 13 were found to have the proper HLA subtype and 6 of these have gone tumor analysis by Tempus next-generation sequencing," said Simeone. "On May 11, we identified our first candidate with LOH who will be a candidate for leukapheresis. We expect many more patients to follow."

Reference:

Simeone DM, Hecht JR, Patel SP, et al. BASECAMP-1: Leveraging human leukocyte antigen (HLA) loss of heterozygosity (LOH) in solid tumors by next-generation sequencing (NGS) to identify patients with relapsed solid tumor for future logic-gated Tmod CAR T-cell therapy. J Clin Oncol. 2022;40 (suppl 16; abstr TPS2676).