ID #1059 Antigen Prescreening Enables Phase I Clinical Trial Enrollment for GPC2-CAR T Cell Therapy in Patients with Relapsed or Refractory CNS Embryonal Tumors
Katherine Ryan, Kyle Dyson, Thien Nguyen, Richard Sleightholm, Alberto Delaidelli, Megan Troxell, Emily Egeler, Fahn Bryant, Barbra Beebe, Mariah Duncan, Poul Sorensen, David Solomon, Crystal Mackall, Sabine HeitzenederAbstract
Background
Relapsed or refractory (RR) pediatric CNS embryonal tumors, including medulloblastoma (MB), embryonal tumor with multilayered rosettes (ETMR), atypical teratoid/rhabdoid tumor (AT/RT), and pineoblastoma, have dismal outcomes with 5-year overall survival ranging from ∼10% in RR-MB to < 1% in ETMR and AT/RT. Glypican-2 (GPC2) is an oncofetal antigen and promising immunotherapeutic target expressed in pediatric cancers including CNS embryonal tumors. Preclinical studies demonstrating potent antitumor activity of GPC2-CAR T cells in MB models prompted initiation of a first-in-human Phase I clinical trial (NCT07087002) evaluating their feasibility and safety using intracerebroventricular administration in children and young adults with RR-CNS embryonal tumors.
Methods
Because antigen density is critical for GPC2-CAR T-cell potency, enrollment is restricted to GPC2-positive disease, predicted to exceed CAR detection thresholds using an H-score of > 100 as surrogate eligibility cut-off. Tumor specimens undergo CLIA-certified GPC2 immunohistochemistry (IHC) and are scored using a standard H-score (0–300) derived from staining intensity (0–3+) and percentage of GPC2-positive tumor cells. In parallel, an advanced scoring method segmenting intratumoral intensity differences (formula = (1 × % weak) + (2 × % moderate) + (3 × % strong)) is performed to assess potential clinical utility.
Results
To date, twelve patients were screened; 75% met eligibility by standard H-score, and two patients have enrolled. Screened tumors include MB (n = 2/median H-score 190), ETMR (n = 4/median 170), pineoblastoma (n = 2/median 200), and AT/RT (n = 3/median 2). Except for AT/RT, all tumors exceeded the enrollment threshold by either scoring method, with advanced H-scores trending higher than standard scores (p = 0.0977).
Conclusion
These findings demonstrate feasibility of GPC2 antigen prescreening using archival tumor tissue to support early-phase trial enrollment, enabling coherent safety and early efficacy assessment of GPC2-CAR T-cells in pediatric CNS Embryonal tumors. Larger cohorts are needed to determine whether advanced scoring provides added benefit in borderline cases.