Randomized Phase II Study of Proteinase 3-Derived PR1 Peptide Vaccine and GM-CSF with or without PEG-Interferon ALFA-2B to Eradicate Minimal Residual Disease in Chronic Myeloid Leukemia

Abstract

BACKGROUND: Complete molecular eradication of chronic myeloid leukemia (CML) cells occurs only in a minority of patients treated with imatinib (IM). The curative potential of allogeneic transplantation and donor lymphocyte infusions highlights the responsiveness of CML to T-cell-mediated immunity. The proteinase 3-derived PR1 peptide (VLQELNVTV) is a leukemia-associated antigen presented on HLA-A2 to cytotoxic T lymphocytes (CTL) that preferentially kill leukemia over normal hematopoietic progenitors.METHODS: This phase II study was designed to evaluate the anti-leukemic effects of PR1 vaccine (2 mg) with GM-CSF (0.6 mg) and an incomplete Freund's adjuvant (Montanide ISA-51) given subcutaneously to patients with CML on IM in complete cytogenetic response (CCyR) with stable residual molecular disease. Patients were randomized to receive the vaccine alone or in combination with pegylated interferon α (PEG-IFN-α; 0.5 μg/kg) with each vaccination. PR1 vaccination was administered on weeks 0, 3, 6, and 18. Immune responses were assessed by PR1/HLA-A2 tetramer staining and molecular responses by quantitative PCR (qPCR) in peripheral blood before study entry, prior to each vaccination, and every 3 months thereafter.RESULTS: Five of the 20 planned HLA-A2+ patients (3 male) have been accrued. All but patient 4 (b3a2) expressed b2a2 transcripts prior to vaccination. Patients 1 (on IM 600 mg/d for 85 months), 2 (on IM 800 mg/d for 72 months) and 5 (on IM 400 mg/d for 76 months) were randomized to receive vaccine+PEG-IFN-α whereas Patients 3 and 4 (both on IM 400 mg/d, for 43 and 37 months, respectively) received PR1 vaccine alone. BCR-ABL1/ABL1 ratios prior to vaccination were 0.99, 0.79, 0.52, 0.10, and 0.44 respectively. The median follow-up from the first PR1 vaccination is 19 months (range, 4-20). All patients experienced transient mild elevations of BCR-ABL1 transcripts after the first vaccination followed by steady decline in transcript levels in Patient 1 (>1-log), and patients 2, 3, and 5 (<1-log) (Figure 1). Furthermore, patient 1 continues exhibiting an ongoing decrement of transcript levels. BCR-ABL1 transcripts have not decreased in Patient 4. Toxicity was limited to grade 1-2 injection site reactions except for Patient 2, in whom the fourth and last dose of PEG-IFN-α was not administered due to behavioral changes. Peripheral blood lymphocytes were analyzed in all 5 patients with PR1/HLA-A2 tetramer and detailed immunophenotyping (CD8, CCR7, CD45RA). Patients 1, 2 and 3 had immune responses (IR) defined as ≥2-fold increase in PR1/HLA-A2 tetramer+ cells. Patient 4 did not have an IR, but had pre-existing PR1-CTL that decreased transiently 3 weeks after the first vaccination. Patient 5 harbored pre-existing PR1-CTLs as well; follow-up analysis pending. No clear trends regarding memory T cell subsets have been identified. Immunity to control CMV-derived pp65 peptide was evident in all 5 patients but did not change after vaccination. Multiple clones of PR1-CTL could be derived from patients 1, 2 and 3 and studies are underway to analyze their TCR-ab affinity to PR1/HLA-A2.CONCLUSION: PR1 vaccination induces specific immunologic responses and improvement of molecular response in patients with CML in CCyR with stable or rising levels of BCR-ABL1 transcripts receiving imatinib therapy. [Display omitted]