P2.04-65 Peptide-Based Cancer Vaccine Shortened the Overall Survival of a Large Portion, but Not All, of Advanced Cancer Patients

Abstract

We have been promoting clinical trials of peptide-based cancer vaccination for various cancers, mainly at Kurume University School of Medicine. Objective: To determine biomarkers predictive of overall survival (OS) in advanced cancer patients treated with a peptide-based cancer vaccine. Patients: The samples from two randomized, double-blind, placebo-controlled, phase III trials of personalized peptide vaccine (PPV) for advanced prostate cancer patients (n=306) and recurrent glioblastoma (n=88) patients, those from one single arm phase II trial of PPV for various types of advanced cancer patients (n=2588), and those from one randomized placebo-controlled non-personalized phase II trial (n=51) for advanced prostate cancer patients were provided for this retrospective biomarker study. No significant differences in clinical benefit (overall survival, OS) were found between the patients receiving PPV and those receiving placebo in each of the two randomized, double-blind, placebo-controlled, phase III trials. The neutrophil or lymphocyte proportion in advanced prostate cancer patients prior to study entry was a biomarker discriminating the PPV patients (70%) who showed significantly shorter OS relative to placebo patients from the remaining PPV patients who showed significantly longer OS relative to placebo patients. The C-C motif chemokine 2 (CCL2) level prior to the study entry in recurrent glioblastoma patients was the other biomarker discriminating the PPV patients (40%) who showed significantly shorter OS relative to placebo patients from the remaining PPV patients who showed significantly longer OS relative to placebo patients. The neutrophil or lymphocyte proportion prior to the phase II study entry (n=2588) was also a biomarker discriminating the PPV patients (60%) with significantly shorter OS from the remaining PPV patients entered in the single arm phase II study in all the advanced cancer patients other than gastric cancer or glioblastoma patients. Notably, the median OS for the 250 of 399 lung cancer patients (62 %) who met either or both the cutoffs of neutrophils ³64% and lymphocytes <26% was significantly shorter than that of the remaining 149 patients (38%) (9.5 month,95% confidence interval: 7.6-11.9 vs 20.0 month,13.0-25.3, p<0.001). The CCL2 level, however, prior to the study entry was not a biomarker in these lung cancer patients receiving PPV. This biomarker was also predictive of OS in a randomized placebo-controlled non-personalized phase II trial of peptide-based vaccine. Peptide-based cancer vaccine shortened the OS of a large portion, but not all, of advanced cancer patients with various types of cancer. Prospective clinical studies of peptide-based cancer vaccines using the newly defined prognostic markers may be warranted.