The Impact of Ageing on CAR T Cell Therapy

Chimeric antigen receptor (CAR) T cell therapy (CTT) is a ground-breaking T cell-based treatment for haematological cancers but response rates to specific cancer types decrease with the average age of diagnosis. Natural ageing can limit T cell activation and drive T cells towards various states of dysfunction, such as terminal differentiation, senescence, and exhaustion. We therefore hypothesise that ageing undermines T cell function, compromising the peripheral blood mononuclear cell (PBMC) starting material used in CAR T cell generation, thereby reducing CTT efficacy. To explore this hypothesis, we first tested whether common markers of T cell ageing (differentiation, exhaustion, senescence and metabolic dysfunction) had predictive value as biomarkers of CTT outcomes, using samples from young and older healthy donors and multiple myeloma (MM) patients. Donor age was the main driver of multiple markers of differentiation, exhaustion (PD1) and senescence (CD57) in CD8 and CD4 T cells. MM drove the expression of specific exhaustion markers (LAG3) and further augmented differentiation. A significant de novo mitochondrial biogenesis defect was observed during CAR T cell generation in samples from older donors and MM patients. In the CAR T cell product, IFN𝛾 production and in vitro cytotoxicity increased with donor age, but no additional functional change was seen in MM patients. When the predictive value of age-related markers was assessed for various CAR T cell outcomes, several markers of T cell naivety predicted CAR T cell yield, killing capacity, quality, and differentiation. In contrast, CD57 expression predicted CAR T cell cytokine production. Followed by evaluating the impact of autologous stem cell transplant (ASCT) on CAR T cell generation, as ASCT has been proposed to drive premature ageing of the T cell compartment. We tested whether age-related biomarkers could predict changes in CAR T cell outcomes before vs after ASCT. In the PBMC starting material, ASCT led to a marked reduction in naïve T (TN) cells and the CD4:CD8 ratio. ASCT did not alter CAR T cell cell yield and transduction efficiency, but it dramatically increased differentiation and elevated in vitro cytotoxicity of the CAR T cell product. Moreover, the reduction in CD4:CD8 ratio seen in the PBMC starting material persisted into the CAR T cell product. Finally, we evaluated which of the current clinical activation protocols (either 𝛼CD3 or 𝛼CD3/𝛼CD28 monoclonal antibody (mAb) stimulation) generated a better CAR T cell product using PBMC samples from older patients. Both advanced age and the 𝛼CD3/𝛼CD28 mAb protocol reduced CAR T cell yield and increased differentiation. Advanced age and the 𝛼CD3/𝛼CD28 mAb protocol also generated CAR T cells with heightened IFN𝛾 production and elevated in vitro cytotoxicity. We generated a novel metric of CD28 abundance that was predictive of CTT outcomes and was a stronger correlate with the 𝛼CD3/𝛼CD28 mAb protocol, suggesting that CD28 abundance has more of an impact on activation strategies that employ CD28 stimulation. Given that T cells are the raw starting material for CTT, it is notable that the functional capacity of a patient’s T cells is not considered explicitly during clinical decision-making for CTT. This thesis has dissected the impact of ageing from the impact of cancer and conventional therapies on CAR T cell generation to help guide decisions for CTT with older, heavily treated patients with haematological malignancy.