Developing Chimeric Antigen Receptor T cells for the Treatment of Acute Myeloid Leukaemia

Abstract

AML is the most common acute leukaemia in adults with a high incidence of relapse following allogeneic HSCT. CAR T cells have shown remarkable outcomes in R/R B-cell leukaemia/lymphoma and MM. However, replicating this success in AML has been challenging chiefly due to heterogeneity of AML and lineage specific antigen expression in AML within and between patients, leading to immune escape and treatment failure.

CD123 is a leading target as it is frequently overexpressed in bulk and LSCs, with relatively lower expression on myeloid cells. Others in our lab have shown that PiggyBac CAR T cells targeting CD123 with natural ligand IL-3 recognition domain are effective in vitro. However, the discovery of CAR T derived lymphomas in treated patients prompted us to seek alternative gene modification systems. In this thesis, I explored 2 main themes, namely (i) to improve the genetic safety of CARIL3 therapy; and (ii) to increase the efficacy of anti-AML CAR T therapy.

Chapter 3 describes the pre-clinical development of CARIL3 using alternative lentiviral and PiggyBat systems with a focus on maintaining efficacy while exploring differences in gene integration patterns.

Chapter 4 looks at TIM-3 as another potential target with the aim of combining with CARIL3 to reduce the risk of antigen escape. The issue of fratricide due to expression of TIM-3 on T cells was overcome by CRISPR/Cas9 KO.

Chapter 5 explores whether the TIM-3 knockout could also improve efficacy of CARIL3, since TIM-3 is studied as a T cell exhaustion marker. The theme of genetic safety is again explored, by using CRISPR-Cas9 directed knock-in into the TIM-3 locus to minimise the risk of random integration.

While further work is required to enable safe and efficacious CAR T cells against AML in the clinic, the work in this thesis presents several proof-of-concept strategies taking significant steps towards realising this goal.