Updated results from first-in-human phase 1 dose-escalation trial of TAK-102, a GPC3-targeted armored CAR T cells, in patients with advanced solid tumors.

2543 Background: Glypican 3 (GPC3) is a member of the glypican family of heparan sulfate proteoglycans that are attached to the cell surface by a glycosylphosphatidylinositol anchor. High GPC3 expression rates are reported in numerous cancer types with a high unmet medical need, including hepatocellular carcinoma (HCC). TAK-102 is a GPC3-targeted autologous chimeric antigen receptor T cell (CAR-T) immunotherapy armored with IL-7 and CCL19. The addition of IL-7 and CCL19 to the construct was designed to support the expansion of memory subsets and persistence of CAR-T cells. We hypothesized TAK-102 would help overcome the challenges associated with an immunosuppressive tumor microenvironment that limit the activity of nonarmored CAR-T therapies in solid tumors. Methods: The first-in-human, Phase 1 dose escalation study evaluated TAK-102 in patients (pts) with GPC3-expressing solid tumors who were refractory or intolerant to standard treatments. TAK-102 was administered via a single infusion to 3 dose cohorts after lymphodepleting chemotherapy (LDC; consisting of fludarabine and cyclophosphamide): dose level (DL) 1 (starting cohort), 1x10 7 CAR+ cells/body; DL2, 1x10 8 CAR+ cells/body; DL3, 5x10 8 CAR+ cells/body. Objectives included evaluation of safety, dose-limiting toxicity (DLT), recommended phase 2 dose, cellular kinetic (CK) parameters, tumor markers, cytokine/chemokine and antitumor activity based on RECIST 1.1. Results: Eleven pts were enrolled and infused TAK-102 (DL1: 3 pts, DL2: 3 pts, DL3: 5 pts): 1 gastric neuroendocrine carcinoma, 2 liposarcoma, 8 HCC. Five pts achieved stable disease (SD), HCC (GPC3 H-Score: 36) . In patients achieving SD, the greatest reduction in tumor size was 26.4%. No DLTs or neurotoxicity were observed. Six pts experienced cytokine release syndrome (Grade1: 5 pts, Grade2: 1 pt); all cases were manageable. AFP was measured as a tumor marker for HCC. Among 8 pts with HCC, 4 had SD after treatment with TAK-102 and 3 showed a decrease or stabilization of AFP levels corresponding to their clinical status. CK profiles for pts were assessed by flow cytometry and qPCR-based assays. Overall, there was improvement in TAK-102 exposure (Cmax, AUC) when escalating from DL1 to DL2. There was a slight decrease in Tmax from DL2 to DL3. Homeostatic cytokine (IL-7) spiked post-LDC and showed no further increase after TAK-102 infusion across all the DLs. There was dose-dependency observed in peak CCL19, IFN-γ and IL-6 levels. Conclusions: TAK-102, an armored CAR-T, is well tolerated and has a manageable safety profile with some early signs of anti-tumor activity. For CK, TAK-102 exposure (Cmax, AUC) showed improvement from DL1 to DL2, and slight decrease in Tmax from DL2 to DL3. Dose-dependency was observed in peak CCL19, IFN-γ and IL-6 levels, which may point towards increased signal of activity from DL1 to DL3. Clinical trial information: NCT04405778 .