A Blood Dendritic Cell Vaccine for Acute Myeloid Leukemia Expands Anti-Tumour T Cell Responses at Remission

Many acute myeloid leukemia (AML) patients achieve a complete remission (CR) with chemotherapy, but relapse is common. Removal of residual disease remains the greatest challenge. The potential of the immune system to control chemotherapy resistant malignant cells is evident in the success of allogeneic transplantation (aTx), and early results with T cell checkpoint inhibition. However, these non-specific interventions lead to considerable morbidity.

Dendritic cell (DC) vaccination has the potential to generate leukemia-specific autologous immunity with little toxicity. Promising results have been achieved with vaccines developed in vitro from purified monocytes (Mo-DC) delivered to patients with limited residual disease after conventional therapy. To stimulate an effective cytotoxic T cell response to exogenous proteins, DC must be able to cross-present protein antigen. We have demonstrated that pre-formed blood DC (BDC) have superior function to Mo-DC. Both Mo-DC and CD1c⁺ mDC sustained presentation of surface-loading peptide, FMP_58-66, in the context of HLA class I for at least 16 hours. CD1c⁺ mDC presented the peptide at higher density than Mo-DC (p=0.0004). Most notably, CD1c⁺ mDC had a strikingly increased capacity to cross-present FMP_54-74 long peptide as FMP_58-66 bound to HLA-A2 at 16 hours compared with Mo-DC (p=0.0032).

To assess the capacity to isolate and target DC in AML patients. We performed a visualised stochastic neighbour embedding (ViSNE) analysis that allowed us to directly compare healthy and patient BDC populations (HD, n=8, Relapsed refractory, n=6; in CR post-chemotherapy, n=18 or >12 months post-ATx, n=7) by visualising high dimensional cytometry data structures in two dimensions. All orthodox BDC subsets were identified in HD and AML CR patients, but were reduced in active AML. Within 12 weeks following chemotherapy, CD1c⁺ myeloid DC recovered while CD141⁺ DC and plasmacytoid DC showed partial recovery. We saw no significant difference in BDC subset recovery between different chemotherapy regimens used or in patients undergoing ATx.

To determine whether the T cell landscape will permit an active vaccination strategy in AML CR patients, we compared patients receiving non-fludarabine regimens (NFR), fludarabine-containing regimens (FR) with those receiving those post-ATx with age-matched HD. Patients receiving NFR have relatively normal T cell number, phenotype and function at CR, whereas patients who have received FR as salvage therapy have persistent T cell abnormalities including reduced number, altered subset distribution, as well as a failure to expand and increased activation-induced cell death upon ligation of the T cell receptor. A comprehensive analysis of T cell landscape demonstrated increased PD-1 and TIM-3 on CD4⁺ T cells, particularly within the memory CD28⁺CD45RO⁺ population in FR (PD-1 p=0.0005, TIM-3 p=0.001) and ATx (PD-1 p=0.044, TIM-3 p=0.027) patients. In the NFR group there was a trend to increased PD-1 expression by CD4⁺ T cells (p=0.06, Figure 6A). Peripheral blood CD8⁺ T cells showed no significant increase in PD-1 or TIM-3 expression.

We utilised the window of opportunity following induction of CR in NFR or ATx patients to combine active vaccination with BDC in combination with checkpoint inhibition against PD-1 and TIM-3. BDC priming was able to directly enhance functional T cell responses to viral (CMV, Influenza) and the tumour associated Wilms tumour 1 antigen, generating both CD4 and CD8 responses. Checkpoint blockade using anti-PD-1 and anti-TIM-3 antibodies demonstrated additive effects on the enhancement of antigen specific, cytotoxic, interferon gamma producing CD8⁺ T cells.

The generation of capable functional responses in AML requires intact antigen processing and presentation as well functional effector T cells. In AML we have demonstrated that DC function is intact and there is a window of opportunity for active immune intervention in CR following either chemotherapy or ATx. Chemotherapy regimens have a profound impact on the function of effector T cells, notably fludarabine profoundly inhibits the capacity of T cells to respond to antigen priming and induces susceptibility to activation induced cell death. Clinical targeting with active DC vaccination in conjunction PD-1 and TIM-3 blockade may provide a new avenue for targeting residual disease and relapse in poor prognosis patients with AML.