7PD - Preexisting immunity as a potential biomarker for clinical response to AE37 vaccination in breast cancer patients

Date 20 November 2015
Event ESMO Symposium on Immuno-Oncology 2015
Session Poster Discussion session
Topics Breast Cancer
Presenter Nikolaos Pistamaltzian
Citation Annals of Oncology (2015) 26 (suppl_8): 1-4. 10.1093/annonc/mdv513
Authors N.F. Pistamaltzian1, E.A. Anastasopoulou1, P. Tzonis1, J. Kalogeropoulou1, A. Ardavanis2, E.A. Mittendorf3, G.E. Peoples4, C.N. Baxevanis1, M. Papamichail1, S.A. Perez1
  • 1Cancer Immunology And Immunotherapy Center, Saint Savas Cancer Hospital, 11522 - Athens/GR
  • 21st Medical Oncology Clinic, Saint Savas Cancer Hospital, 11522 - Athens/GR
  • 3Department Of Surgical Oncology, MD Anderson Cancer Center, Houston/US
  • 4Cancer Vaccine Development Program, MD Anderson Cancer Center, Houston/US



There is an unmet need for predictive biomarkers in cancer immunotherapy. We have recently reported that preexisting levels of IFNɣ in response to AE36 might serve as a predictive factor for immunological and clinical responses to vaccination with AE37, the Ii-Key hybrid peptide of HER2776-790 (AE36) in prostate cancer patients. The purpose of this study was to investigate the predictive significance of preexisting AE36 immunity, in relation to immunological and clinical response, in breast cancer patients vaccinated with AE37 in a randomized phase II clinical trial.


This trial enrolled node-positive or high risk node-negative patients with any degree of HER2 expression (IHC 1-3+ or FISH > 1.2), rendered disease-free following standard of care therapy. Patients were randomized to receive either AE37 + GM-CSF (VG, n = 55) or GM-CSF alone (CG, n = 44) in 6 monthly intradermal inoculations followed by 4 boosters administered every 6 months. Preexisting immunity was determined using as cutoff the IQR of the values generated by in vitro (IFNɣ ELISPOT) and in vivo (Dermal Reaction-Induration of 1st vaccine, DRI1) assays. Kaplan-Meier and log-rank test were used for survival analysis. P-values ≤0,05 denoted statistical significance.


After a 59 month median follow-up, VG patients exhibited a 41% relative risk reduction (RRR) in recurrence vs CG patients (p = 0.31). RRR was 71% (p = 0.22) in VG patients with preexisting vs non preexisting IFNɣ response, and 84% (p = 0.03) in VG patients with preexisting vs non preexisting DRI1 response. Preexisting immunity (IFNɣ+ and/or DRI1) conferred a 92% RRR in VG patients (p = 0.004). There were no differences in clinicopathologic characteristics between VG patients with or without preexisting immunity.


These data suggest that preexisting immunity might serve as a predictive biomarker for clinical response in high risk breast cancer patients vaccinated with AE37. This may be taken into consideration in the design of a future clinical trial using preexisting immunity as a pre-specified stratification parameter.

Clinical trial identification



G.E. Peoples: partial inventor rights to AE37, and consultant to Antigen Express. All other authors have declared no conflicts of interest.