Clinical BCR-ABL Peptide Vaccination in Chronic Myeloid Leukaemia: Results of the EPIC Study.

Chronic myeloid leukaemia (CML) is characterised by the BCR-ABL oncoprotein. The amino-acid sequences spanning the junctional region are completely leukaemia-specific. In vitro pulsing of antigen presenting cells elicits immune response to CML cells. Vaccination of CML patients with peptides from this junctional region could therefore elicit/augment immune responses to CML cells. In our Evaluation of Peptide Immunisation in CML (EPIC) study, the patient’s entry requirements were as follows:

1. first chronic phase of CML,

2. expression of e14a2 (b3a2) BCR-ABL transcript, and

3. prior treatment with imatinib daily (at least 400mg) at a stable dose for at least 6 months.

Each patient received intradermally a cocktail of 3 BCR-ABL peptides:

1. a 9-mer spanning the e14a2 region,

2. the same 9-mer linked to PADRE (a 15-mer non-natural peptide shown to activate CD4+ T cells), and

3. a 13-mer consensus e14a2 junctional peptide linked to PADRE.

Peptides were administered at either 100 (5 patients), 300 (5 patients), 600 (5 patients), or 1000μg (4 patients) with sargramostim on 6 occasions over 2 months. Immune responses to the vaccine were monitored by IFN-γ and IL-5 ELISPOT assays on peripheral blood mononuclear cells. Molecular responses were assessed by quantitative real-time PCR of BCR-ABL mRNA. At entry no patient showed a detectable immune response to PADRE, but all 19 patients had detectable CD4+ T cells responses within 3 months of commencing vaccination. This indicated that the vaccination protocol was capable of stimulating T cell responses in all 19 patients. Immune responses to the 9-mer BCR-ABL junctional peptide used in the vaccine were detected in 11/19 patients, and demonstrated to be CD8+ T cells by cytokine analysis in flow cytometry. BCR-ABL immune responses were also assessed against a longer 18-mer peptide spanning the whole e14a2 junctional region. CD4+ T cells specific for this 18-mer peptide were detected in 14/19 patients. Interestingly, immunophenotyping indicated that these BCR-ABL-specific T cells were of a memory phenotype (CD45RO+). Serial molecular responses were available for at least 12 months on all cases. Of the 6 patients not in major cytogenic response (MCR) at baseline, molecular improvement was only observed in one case. However 12/13 patients in at least MCR at baseline had at least a 1 log fall in BCR–ABL transcripts, though this occurred several months after completing vaccination. Moreover, vaccination improved the fall in BCR–ABL transcripts in patients who had received imatinib for more than 12 months. These data show that peptide immunisation in CML can elicit anti-BCR-ABL peptide responses in CD4+ and CD8+ T cells. It also demonstrates that BCR-ABL peptide vaccination may improve control of CML, especially in patients responding well to imatinib.