A WT-1 and PRAME "Fast-DC" Immunotherapy As a Potential Post-Remission Strategy for AML

Patients with acute myeloid leukemia (AML), not eligible for allogeneic hematopoietic stem cell transplantation (AHSCT), have a significant risk of disease relapse and death due to a lack of long-term disease control. To investigate the possibility of relapse prevention after initial chemotherapy, a safety and feasibility, single-center, open label, first in human phase I/II study using dendritic cells (DC) targeting PRAME and WT-1 was recently completed.

Mature autologous DCs were rapidly (3-4 days) generated using full-length PRAME and WT-1 antigens and a maturation cocktail with a TLR 7/8 agonist. Intradermal vaccination of 2.5-5x10⁶ WT-1 and 2.5-5x10⁶ PRAME RNA-loaded DCs was performed at week 1, 2, 3, 4, 6 and then monthly for the remainder of a 2-year study period. Patients with WT1-positive AML, with or without PRAME positivity, had to be in morphological remission (+/- hematological recovery) after induction chemotherapy without being eligible for AHSCT. The study consisted of a phase I part (n=6) to assess early safety followed by a phase II part (n=14) which assessed further safety as well as clinical and immunological effects of vaccination.

A total of 20 eligible patients (5 female and 15 male, ECOG score of 0) with a median age of 59 years (range 24 to 73) were included into the study. Risk groups based on HOVON/SAKK 102 criteria at screening identified 13 patients as good, 5 as intermediate and 2 as poor risk. All patients were positive for WT-1 and 17 for PRAME prior to chemotherapy. The mean time from first diagnosis to first vaccination was 10.1±3.7 months.

Vaccinations were well tolerated, without any withdrawals from the study due to toxicity. No related serious or unexpected adverse events (AEs) were reported and the most common AEs were injection site related, accounting for 28% of all AEs, which were mild and transient in nature (Grade I; CTCAE v4.03). Grade I/II toxicities were experienced by 18/20 patients, whilst 5 (25%) experienced Grade III toxicity unlikely or not related to treatment.

The 2-year survival probability from time of first vaccination was 80% (95% CI: 55-92). Importantly, patients at or above the age of 60 years (n=10) also had an 80% survival rate at 2 years (risk groups: 4 good, 4 intermediate, 2 poor). Nine patients relapsed under vaccination, of whom 4 died thereafter (three due to the underlying disease, one due to GvHD after transplantation) within the 2-year study duration, equating to a probability of progression free survival (PFS) of 55% (95% CI: 31-74). The PFS rate for the elderly patients (≥ 60 years) was 50%, while it was 60% for the younger patient group. Most relapses (5/9) appeared early, within 80 days after first vaccination. Out of the 9 relapsed patients, 6 could be successfully transplanted (104 to 380 days after first vaccination; risk groups: 4 good and 2 poor risk) and 4 were still alive at the end of the 2-year observation period.

The DC manufacturing process reproducibly yielded a high number of DCs expressing the transfected tumor antigens, WT-1 or PRAME, as well as high surface levels of co-stimulatory molecules. After freeze-thawing, DCs were capable of CCL19-directed migration and, upon stimulation, secreted IL-12, but not IL-10. These phenotypic and functional characteristics indicate efficient DC maturation and activation despite the patients´ previous therapeutic regimens. DC properties were comparable irrespective of the patients' clinical outcome.

Immune-monitoring of bulk T cells from vaccinated patients revealed that patients in stable remission had higher levels of HLA-DR-expressing T cells when compared to relapsing patients, both in peripheral blood and bone marrow. Such difference was observed prior to vaccination and throughout the entire treatment period, suggesting that T cell activation may play a role in maintaining remission. Analysis of 23 common AML mutations revealed that 6/9 relapsed patients had at least 1 mutation, with 4 patients having 3-6 mutations. In contrast, only 1 mutation already present at baseline was found in 4/11 patients in remission.

Administration of autologous DC transfected with PRAME- and WT1-RNA is feasible, safe and well tolerated. The 2-year 80% survival rate, particularly in the elderly, and the 55% progression free survival warrant further studies to assess the efficacy of this vaccine approach in improving outcomes in patients with AML.