Background: Therapeutic options for CRPC pts are limited, but a promising therapeutic modality is the combination of vaccines in concert with immunomodulating adjuvants. The vaccine utilized in this study (designated PSA-TRICOM) is comprised of poxviral vectors which encode a modified agonist epitope of PSA and 3 costimulatory molecules, ICAM, LFA-3 and B7.1. A phase II trial evaluating recombinant vaccinia (rV) and fowlpox (rF) PSA-TRICOM previously demonstrated PSA declines and an objective response (OR) with a significant (p < 0.0003) correlation of clinical benefit and immunologic response (Gulley et al. ASCO 2005). Blockade of a T cell negative costimulatory molecule, CTLA-4, enhances anti-tumor immune responses in preclinical and clinical studies.
 Methods: Pts received a fixed dose of vaccine and GM-CSF, with 4 sequential cohorts receiving escalating doses of an anti-CTLA-4 monoclonal antibody, ipilimumab at 1, 3, 5 and 10 mg/kg. 2 x 10^8 pfu rV-PSA-TRICOM was administered on day 1 with GM-CSF, 100 mcg sc, on days 1-4. Boost vaccinations combined rF-PSA-TRICOM (1 x 10^9 pfu), GM-CSF, and ipilimumab beginning on day 15 and continuing every 28 days thereafter while pts remained on study. Ipilimumab was discontinued after 6 doses or earlier for immune related adverse events (irAE). Restaging scans were performed every 3 months.
 Results: 18 pts with metastatic CRPC have been enrolled. There have been no dose limiting toxicities on study and no > grade 2 toxicities attributed to vaccine. irAE attributed to ipilimumab included colitis (n=6), panhypophysitis (n=3), rash (n=1) and neutropenia (n=1). The median number of cycles with ipilimumab was 3, (range 1-6) with 5 pts receiving only 2 cycles and 3 pts receiving only 1 cycle with ipilimumab due to irAE. Only one pt has received 6 does of ipilimumab.
 The six patients with prior chemotherapy (all treated at the 1 and 3 mg/kg doses) had no evidence of clinical benefit (no decrease in PSA or objective response) with a median time to progression (TTP) of 3.2 months (mean 2.8; range 1.6-3.7).
 Of the 9 pts treated on the 3 and 5 mg/kg dose who had no prior chemotherapy, the median TTP was 5 months (mean 5.4; range 2.9-9.3) with 2 patients still on trial at 5 and 8 months. 5 of these 9 patients had declines in PSA by at least 50% from peak PSA during study. This may be clinically significant as 4 of these pts had radiographically and clinically stable disease for at least 6 months (TTP 6, 7, 8+ and 9.3 months). There was one unconfirmed PR (out of 10 evaluable pts) by RECIST and another patient with a sustained decrease in PSA >95% who remains on study at 8 months.
 3 pts have been enrolled in the 10mg/kg cohort. One patient had a solitary new lesion at L2 on bone scan with stable soft tissue disease and PSA and was taken off study and the other 2 patients remain with stable disease on study at 2.5 and 4 months.
 This combination has activity in pts with chemotherapy naïve metastatic CRPC and is associated with manageable side effects. From data thus far, it is unclear if safety events or responses seen are related to the dose of anti-CTLA4. Further expansion of the 3 mg/kg and 5 mg/kg cohorts for additional toxicity and efficacy data are planned.

AACR-NCI-EORTC International Conference: Molecular Targets and Cancer Therapeutics-- Oct 22-26, 2007; San Francisco, CA