Introduction: Relapsed/refractory (r/r) peripheral T cell lymphomas (PTCL) are highly aggressive tumors associated with very poor prognosis. Unlike B-cell lymphomas, PTCL has not benefited from advances in cellular immunotherapies. Pan T-cell depletion is highly toxic and there are few or no T-cell lymphoma-specific antigen targets that discriminate malignant from normal T cells. We recently described a targeting strategy based on the mutually exclusive expression of T cell receptor beta-chain constant domains 1 and 2 (TRBC1 and TRBC2) (Maciocia, PM. et al, Nat Med 2017) which can spare a proportion of the normal T cell compartment. Here we present first in human clinical findings of AUTO4, a TRBC1 directed autologous CAR T cell therapy for patients with TRBC1-positive PTCL.

Methods: This ongoing phase 1/2 study (NCT03590574) consists of a dose escalation and an expansion phase. Here we report the findings of the dose escalation. Biopsies from patients with r/r PTCL were screened for TRBC1-positivity using next-generation sequencing. Four flat dose levels of AUTO4 were explored: 25 x 106, 75 x 106, 225 x 106, and 450 x 106 CAR T cells were administered as a single dose. Patients received lymphodepletion with fludarabine (30mg/m2 for 4 days) and cyclophosphamide (500mg/m2 for 2 days) (Flu/Cy) prior to AUTO4 infusion. Primary endpoints were incidence of Grade 3 to 5 toxicity occurring within 60 days of AUTO4 infusion and the frequency of dose limiting toxicities within 28 days of AUTO4 infusion (DLT period). Overall response (CR+PR) rate post AUTO4 infusion by PET-CT (Lugano 2014 criteria) was a secondary endpoint.

Results: Tumor biopsies from n=73 patients were screened for TRBC1, 36% were TRBC1+. Thirteen patients were apheresed and enrolled, 10 patients have been dosed with AUTO4. The median age was 55 years (range 34 to 63), the median number of prior lines was 3 (range 1-5). The PTCL subtypes treated were PCTL NOS (n=5), AITL (n=4), and CD30+ ALCL (n=1). Three patients had prior SCT. 70% of patients received bridging therapy. After lymphodepletion with Flu/Cy, 3 patients received 25 × 106 CAR T cells, 2 patients received 75 × 106 CAR T cells, 1 patient received 225 × 106 CAR T cells and 4 patients received 450 × 106 CAR T cells. The most common treatment-emergent adverse events were cytopenias (anemia and neutropenia). Any grade CRS was observed in 4/10 patients, 1 patient developed Grade 3 CRS, all CRS events occurred at the 450×106 dose level. None of the patients experienced any neurotoxicity/ICANS, and no dose limiting toxicities (DLT) were observed. Serious treatment-emergent adverse events (SAE) Grade ≥3 within 60 days post AUTO4 infusion, regardless of relationship to AUTO4, were observed in 3/10 patients. No grade 3 or higher infections and infestations were observed. The median (range) number of CD3+ T-cells/µl in blood prior to lymphodepletion and at the end of the DLT period (Day 28) was 204 (94-689 and 123 (19-458), respectively. Nine patients were evaluable for response at Month 1: n=5 were in complete metabolic response (CMR) by PET-CT, though one patient was in CMR after bridging at the time of lymphodepletion, 1 patient achieved a PR, and 3 patients did not respond. One patient was not evaluable at Month 1 by PET-CT due to COVID19 infection. Three of the 4 patients at the 450x106 cell dose achieved a CMR at Month 1. With longer follow-up, 2/4 patients at the 450x106 cell dose maintained a CMR at 6 and 9 months, respectively. No CAR T expansion was detected by PCR in peripheral blood, though CAR T cells were detected in an on-treatment lymph node biopsy of a patient who then achieved CMR.

Conclusion: AUTO4 treatment was well tolerated with no DLT. Early responses are encouraging, although no CAR T cell expansion was seen in peripheral blood which may translate to responses that are not durable. To address this, optimisation of the AUTO4 manufacturing process has been performed, resulting in a product with a more naïve phenotype. The study is ongoing, with additional patients due to be treated to define the recommended phase 2 dose using the new manufacturing process. Longer follow-up and additional patients will be presented at the conference.

Cwynarski:Roche, Takeda, KITE/Gilead, Incyte: Speakers Bureau; BeiGene: Research Funding; Roche, Celgene/BMS, Takeda, KITE: Other: Travel to scientific congress; Roche, Takeda, Celgene/BMS, Atara, Gilead/KITE, Janssen, Incyte: Consultancy. Iacoboni:NOVARTIS, KITE/GILEAD, BMS/CELGENE, ASTRAZENECA, ROCHE, ABBVIE, JANSSEN, MILTENYI: Honoraria; NOVARTIS, KITE/GILEAD, BMS/CELGENE: Consultancy. Shang:Autolus Inc: Current Employment. Zhang:Autolus Inc: Current Employment. Basilico:Autolus Ltd: Current Employment. Huber:Autolus Ltd: Current Employment. Vinson:Autolus Ltd: Current Employment. Lilova:Autolus Ltd: Current Employment. Brugger:Autolus Ltd: Current Employment. Pule:Autolus Ltd: Current Employment.

Author notes


Asterisk with author names denotes non-ASH members.

Sign in via your Institution