Trial Information

Phase I/II Study of Human GM-CSF Gene-Transduced Irradiated Prostate Allogeneic Cancer Cell Vaccines (GVAX®) in Advanced Prostate Cancer Patients Made Lymphopenic and Infused With Autologous Peripheral Blood Mononuclear Cells

Other investigators have focused on tumor cell-based vaccine strategies using either
allogeneic or autologous tumor cells. Vaccines derived from allogeneic melanoma tumor cell
lines are associated with no clinically significant toxicity when given alone.

Most tumor cell line vaccines are lethally irradiated, but not genetically modified.
However, a newer generation of tumor vaccines that have been genetically modified to enhance
tumor-associated antigen presentation to patient T cells have reached early clinical trials.
Based on extensive preclinical studies and clinical allogeneic prostate cancer vaccine
data, a strong impression exists that ex vivo GM-CSF gene transfer may make human prostate
cancer cell lines more potent in inducing clinically relevant antitumor immune responses in
men with micrometastatic prostate cancer.

In the early clinical studies of irradiated GM-CSF-secreting autologous cancer cell vaccine
treatment of advanced kidney cancer and advanced melanoma, immune responses elicited by
vaccination appeared dependent on both the vaccine cell dose administered and the level of
GM-CSF secretion by the vaccine cells. A significant fraction of the nearly 50,000 men each
year who suffer from life-threatening prostate cancer will not be candidates for autologous
tumor vaccine treatments simply because an insufficient number of autologous prostatic
carcinoma cells will be available after surgery. These men may benefit from an allogeneic
GM-CSF-gene-transduced prostate cancer vaccine.

Theoretically, autologous prostate cancer cells may be the best source of prostate cancer
antigens for eliciting therapeutically useful immune responses. However, evidence has
accumulated to suggest that allogeneic prostatic carcinoma cells might also serve as useful
sources of prostate cancer antigens for prostate cancer vaccine construction. One
theoretical concern about the use of allogeneic prostate cancer cells to prime antigen
specific T-cell immunity is whether cytolytic T lymphocytes (CTLs) restricted to the HLA
molecules of the patient's tumors can be primed by vaccine cells expressing different
(allogeneic) HLA molecules. Recent work has shown that the GM-CSF gene transduced tumor
cells do not directly prime MHC class I restricted CTLs but are broken down by the
recipient's antigen-presenting cells, which process and present these antigens on their HLA
molecules to prime T cells. In fact, tumor cells that do not express any MHC class I
molecules are nevertheless capable of generating systemic antitumor immunity that is
comparable to their MHC class I positive counterparts.

Phase I and II studies of the allogeneic prostate cancer vaccine have been performed in
patients with metastatic prostate cancer. One is a Phase I trial of CG1940 alone and one is
a phase I/II trial of CG1940 and CG8711 using the AAV GM-CSF vaccine cells. These trials
are currently in progress to evaluate safety and time to progression by PSA and bone scan.
In these trials the vaccines are administered at dose levels of 5 x 10^7, 1 x 10^8, 2 x
10^8, or 3 x 10^8 cells every 2 or 4 weeks for 3 to 12 vaccinations and a 5 x 10^8 cell
prime vaccination followed by a 3 x 10^8 cell boost vaccination every 2 weeks for a total of
13 vaccinations. Preliminary analysis of the 12 patients enrolled in the phase I trial
shows one patient with stable disease. Eighty patients have been enrolled in the phase II
trial. Sixty-five of these patients in the phase II trial have been followed for a median
of 4 months. One of these patients had a partial PSA response, and 10 maintained stable
PSAs. Follow-up bone scans were obtained in 33 patients. One patient showed improvement in
metastatic lesions on his bone scan, and 12 patients maintained stable disease on bone
scans. ICTP was assayed at Treatment 1, Treatment 4, and at the First Follow-Up Visit in
fifteen patients. The ICTP level decreased in five patients (33%) of whom two had
undetectable levels after vaccine treatment. This normalization of osteoclast activity
suggests a novel mechanism of immunotherapy in metastatic prostate cancer.

As of November 29, 2004, 188 patients have received the prostate GVAX vaccine. In summary,
these phase I and II trials showed that CG1940 and CG8711 has an excellent safety profile.
As of January 24, 2005, a phase III trial of prostate GVAX is open at more than 40 medical
centers in the USA.