A randomized trial of personalized peptide vaccine (PPV) plus low-dose estramustine (EMP) versus full-dose EMP in patients with hormone-refractory prostate cancer

3007 Background: Personalized selection of the right peptides for each patient could be a novel peptide-based immunotherapy for boosting anti-cancer immunity in many patients (pts). This randomized study evaluated the anti-tumor effect and safety of PPV plus a low-dose EMP compared with full dose EMP for patients with hormone-refractory prostate cancer (HRPC). Methods: This was a randomized (1:1), open labeled, cross-over study in pts with HRPC. Pts were randomized to arm A; PPV plus low-dose EMP (280 mg/day) or arm B; full dose EMP (560 mg/day) according to age and PSA levels. In arm A, prevaccination plasma were measured for their IgG levels for each of the 14 or 12 candidate peptides which can induce HLA-A2 or A24-restricted CTL activity against cancer cells followed by biweekly subcutaneous administration of the top four peptides (3mg each) showing the strongest IgG responses. Disease progression (PD) was defined as three consecutive and 125% increase from baseline PSA levels at least two weeks apart or objective PD by RECIST criteria. After PD, pts were treated with the opposite regime. The primary endpoint was progression-free survival (PFS), and the secondary endpoints were overall survival and toxicity. The planned sample size was 80. Results: A total of 54 pts from 4 institutions were enrolled between June 2006 and December 2008. The accural into arms A and B was 27 and 27 pts, respectively. The main pts characteristics are (arm A/B): median age 71/69 years, EOCG performance status 0/1 96%/4% and 100%/0%, HLA A2/A24/A2A24 40%/32%/28% and 54%/27%/19%, median PSA 27/25 ng/ml, and metastatic HRPC 96%/85%. All pts were evaluable for their response at the time of interim analysis. The personalized peptide vaccination was well tolerated with no major adverse effects. Increased levels of IgG responses to the vaccinated peptides were observed in 20 of 23 (87%) patients tested. The median PFS time was 246 days in the arm A group and 85 days in the arm B, respectively. The PFS time in the arm A was statistically longer than that in the arm B (log-rank test: p = 0.0007, hazard ratio: 0.27, 95%CI: 0.12 to 0.615). Conclusions: PPV plus low-dose EMP was associated with improvement in PSA-based PFS compared to full-dose EMP alone. No significant financial relationships to disclose.