Cancer ResearchResearch PaperPaywall

CAR T Cell Therapy Targets Rare Solid Tumors via GPNMB Antigen

A new CAR T cell product called GCAR1 shows early clinical promise against rare, treatment-resistant solid tumors driven by gene fusions.

Sunday, July 5, 2026 3 views
Published in Nat Cancer
A laboratory technician in gloves examining a vial of clear cell therapy product against a lit background in a clinical immunology lab

Summary

Researchers at the University of Calgary developed GCAR1, a CAR T cell therapy targeting GPNMB, a protein highly expressed in two rare, hard-to-treat cancers: alveolar soft-part sarcoma (ASPS) and translocation renal cell carcinoma. Both cancers are driven by MiT/TFE-family gene fusions that cause uniform, stable GPNMB expression on tumor cell surfaces — making it an attractive target. In lab models and patient-derived organoids, GCAR1 showed strong anti-tumor activity. In a first-in-human clinical trial, one patient with relapsed metastatic ASPS achieved stable disease for three months with resolution of multiple non-target lesions and no serious safety concerns. The CAR T cells expanded in the bloodstream and remained detectable for a month. Combining GCAR1 with immune checkpoint blockade improved results in animal models, suggesting a promising combination approach.

Detailed Summary

CAR T cell therapy has revolutionized treatment for blood cancers, but extending its success to solid tumors has proven difficult — largely because safe, uniformly expressed surface targets are rare. This study addresses that challenge by identifying GPNMB (glycoprotein NMB) as an ideal target in solid tumors driven by MiT/TFE-family oncogenic gene fusions.

The researchers focused on two rare cancers: alveolar soft-part sarcoma (ASPS) and translocation renal cell carcinoma. Both are defined by chromosomal gene fusions that activate MiT/TFE transcription factors, which in turn drive high, homogeneous, and stable GPNMB expression on tumor cell surfaces. This consistency — absent in most solid tumor antigens — makes GPNMB an unusually reliable target.

The team developed GCAR1, a GPNMB-directed CAR T cell product, and tested it across multiple models: patient-matched cancer cells, patient-derived organoids, and mouse xenograft models. In all settings, GCAR1 demonstrated potent and selective anti-tumor activity. The therapy was then advanced to a first-in-human, open-label clinical trial (NCT07104682). A post hoc interim analysis of the first treated participant — a patient with relapsed, refractory, metastatic ASPS — showed stable disease for up to three months, resolution of multiple non-target lesions, and a favorable tolerability profile. GCAR1 T cells expanded as a polyclonal population in peripheral blood and remained detectable for one month.

Spatial transcriptomics analysis of a treatment-resistant lesion revealed immunosuppressive microenvironmental niches that may blunt CAR T activity. Encouragingly, combining GCAR1 with immune checkpoint blockade in xenograft models produced synergistic tumor control, pointing toward a rational combination strategy.

This work establishes proof of concept for targeting oncogenic gene fusion-driven surface antigens with CAR T therapy. If validated in larger trials, it could open a new therapeutic paradigm for a broader class of fusion-driven solid tumors that currently have very limited treatment options.

Key Findings

  • GPNMB is highly and uniformly expressed in ASPS and translocation renal cell carcinoma due to MiT/TFE gene fusions.
  • GCAR1 CAR T cells showed potent activity in patient-derived cells, organoids, and xenograft tumor models.
  • First-in-human use of GCAR1 achieved stable disease for 3 months with resolution of multiple lesions and good tolerability.
  • GCAR1 T cells expanded as a polyclonal population and persisted in peripheral blood for at least one month.
  • Combining GCAR1 with immune checkpoint blockade synergistically improved tumor control in preclinical models.

Methodology

The study combined preclinical work (cell lines, patient-derived organoids, and mouse xenograft models) with a first-in-human open-label individual-participant clinical trial (NCT07104682). Spatial transcriptomics was used to characterize the tumor immune microenvironment of a treatment-resistant lesion. The clinical data reported represent a post hoc interim analysis of a single patient.

Study Limitations

This summary is based on the abstract only, as the full paper is not open access. Clinical evidence is limited to a single patient in a first-in-human trial, making efficacy conclusions very preliminary. Longer follow-up, larger cohorts, and randomized data are needed before clinical conclusions can be drawn.

Enjoyed this summary?

Get the latest longevity research delivered to your inbox every week.

Enter your email to subscribe: