A First-in-Human Study of Autologous T Lymphocytes with Antibody-Dependent Cell Cytotoxicity (ADCC) in Patients with B-Cell Non-Hodgkin Lymphoma (NHL)

Recent trials in patients with B-cell malignancies have demonstrated the clinical potential of chimeric antigen receptor (CAR)-directed T cells. We developed a chimeric receptor (antibody-coupled T-cell receptor, ACTR) consisting of high affinity CD16 V158, CD8α transmembrane domain, 4-1BB and CD3ζ (Kudo et al. Cancer Res 2014). Expression of ACTR conferred ADCC capacity to T cells. In vitro and in immunodeficient mice, ACTR T cells specifically killed B-NHL and chronic lymphocytic leukemia cells tagged with the anti-CD20 antibody rituximab. Trastuzumab induced ACTR-mediated T cell killing of breast and gastric cancer cells, and anti-GD2 of neuroblastoma and osteosarcoma cells. Thus, infusion of ACTR T cells might significantly augment the anti-tumor potential of immunotherapeutic antibodies, and expand options for cell therapy of cancer.

To assess the safety of this strategy and explore its efficacy, we began a pilot study in patients with relapsed/refractory CD20+ B-NHL (ATTCK20; ClinicalTrials.gov No. NCT02315118). In this first-in-human study, ACTR expression is achieved by mRNA electroporation and, hence, is transient. In Phase 1a, 3 subjects receive one dose of autologous ACTR T cells (0.5 x 10⁶/kg) 2 days after rituximab (500 mg/m²); a second dose of ACTR T cells at 0.5 x 10⁷/kg 28 days later is allowed if there is no toxicity and evidence of response. Interleukin-2 (IL-2) is given subcutaneously at 1 million IU/m²/d x 3 during the first week after infusion. Phase 1b plans up to 4 infusions of ACTR T cells (0.5 x 10⁷/kg) at 10 day intervals, each preceded by rituximab at 375 mg/m²; IL-2 (as in Phase1a) is given after the first 2 infusions only.

Six subjects have been enrolled to date (3 in Phase 1a, 3 in Phase 1b) and received a total of 14 ACTR T cell infusions. All had advanced CD20+ B-NHL (2 diffuse large B cell lymphoma, 2 primary mediastinal B cell lymphoma, 1 mantle cell lymphoma, and 1 mediastinal grey zone lymphoma) and had failed multiple lines of therapy. As a lymphodepleting regimen prior to the first ACTR T cell infusion, 5 of the 6 patients received fludarabine (25 mg/m²/d x 5) plus cyclophosphamide (60 mg/kg/d x 1, n=3; 900 mg/m²/d x 1, n=2); the remaining patient received bendamustine (90 mg/m²/d x 2).

After leukapheresis, T cells were cultured under cGMP conditions with anti-CD3 and -CD28 antibodies, and IL-2. After 10 days, cells were electroporated with ACTR mRNA using the MaxCyte GT system. The next day, median ACTR expression in T cells, as shown by flow cytometry, was 95.8% (range, 82.8%-99.3%; n=14). With Rituximab, this produced a median cytotoxicity of 77.5% (67.6%-97.6%; n=14) against CD20+ Daudi cells at a 4:1 effector : target ratio in 4 hours. As expected, ACTR expression in vitro remained high for the first 3-4 days, declining thereafter.

We met or exceeded the estimated minimum target levels of CD3+ ACTR+ cells in 13 of the 14 infused cell products. CD3+ ACTR+ cells (mostly CD8+) were detectable in peripheral blood 1 day after infusion and reached their peak on day 3-4 when they represented 0.43-11.20 (median, 1.27) cells per µL. They became undetectable after day 6, reflecting the kinetics of ACTR expression in vitro.

As of July 2016, 5 of the 6 patients are evaluable, with a median follow-up of 60 weeks (range, 6-83) after the last ACTR-T cell infusion; the remaining patient has received 3 of the 4 scheduled cycles. The main toxicities were grade 3-4 cytopenias: these were observed in all patients and likely due to the lymphodepleting regimen. There was no non-hematologic toxicity, nor evidence of cytokine release syndrome. In Phase 1a, 2 patients had progressive disease (PD) and 1 had stable disease (SD) on CT scan following the first and second infusions, with PET-CT showing reduced FDG avidity and central areas of photopenia suggestive of necrosis. Of the 2 Phase 1b patients evaluable for response, 1 had PD after the third infusion and was taken off study; the other had SD 1 month after 4 ACTR-T infusions; the 2nd month assessment was interpreted as PD.

The initial results of this ongoing trial using autologous T cells with transient ACTR expression are encouraging and warrant additional studies with more potent formulations of this technology. To this end, a clinical trial for patients with relapsed/refractory CD20+ B-NHL in which ACTR is permanently expressed in autologous T cells via viral vector transduction is planned (ClinicalTrials.gov No. NCT02776813).