IMPaCT Investigator Highlights (Text Version) - Dr. Robert K. Bright

Title: Prostate Cancer Vaccines Targeting Over-Expressed Self-Oncoantigens

Investigator: Robert K. Bright, PhD

My current PCRP grant focuses on the question of vaccinating against prostate cancer by targeting novel over-expressed genes that we’ve discovered in prostate cancer cells.

Many of the immunotherapy targets for prostate cancer, and other cancers, fall under a category of an over-expressed self protein. So what that means is a normal self protein, the tumor cells makes more of it, and just that difference in amount possibly could be targeted with a vaccine.

In this particular case we’re talking about a gene product called D52 which is a new gene family that’s unique, conserved, and non-redundant. it shares 86-percent identity with the mouse orthologue which has also been cloned, that code for small intracellular proteins, so it’s not a target for an antibody response but a prime target for a T-cell type response or killer immune cell type response.

So we moved into preclinical animal models in mice specifically, asking the question--can we immunize mice against this gene product, this protein, even though it’s expressed in mouse normal tissues at fairly high levels. And if we can immunize mice with the mouse protein even though it’s in normal tissues, not cause damage to normal tissues, but actually reject tumor cells then we probably could do that with humans, that would be our hypothesis.

In the first experiment, we looked at immunizing mice just intramuscularly with this protein with some various molecular adjuvants. And what you can see is, if we immunized and didn't--no--no other immune modulations in the animals they didn't--weren't able to reject the tumor, so this is 100-percent tumor bearing, equal to controls. However if we immunized and tried to remove in vivo, or at least down-regulate the presence of TGF Beta-1, which we knew was associated with these tumors, these tumors secret e this. It’s been shown that tumor cells as they grow in a nodule throw this soluble protein out and it interferes with the immune response as it comes, and it renders it ineffective.

TRAMP cells in particular secrete a significant amount of this, and indeed that did interfere, because we could use an antibody that targets TGF Beta and then bring the tumor takedown to about 40-percent. So this is about 60-percent protection which is actually quite good in this tumor model particularly with a brand new target that’s very self in nature. But we wanted to do better so we shifted to a DNA based vaccine and we got it up to about 70-percent protection, right off the bat without any other mod ulation of TGF Beta or anything else. So there’s something that’s very potent about DNA vaccines in this model.

Interestingly, if we use the human gene in the mouse the immune response is even more potent even though the response being generated against the human is actually attacking the mouse gene product in the mouse tumor cells. So it’s what we call xenogeneic immune response. And we’re playing on the hope that this 15, 20-percent difference in amino acid content--mouse to human is enough of a foreign property that you’ll get a more potent immune response against the human in the mouse or conversely the mouse i n the human--if we got that far at clinical trials, but yet can still recognize the mouse protein in mouse tumors and then theoretically the human protein in human tumors.

We also show you it’s durable because survivors here we re-challenged with a heavy dose of tumor cells almost a year later and it still had an immune response that remembered those tumor cells and killed those tumor cells in at least 40 percent of the mice. Still we would like to get to 100-percent, so we’re still pushing that direction, and we are getting closer.

We decided what if we combine the DNA and the protein--not at the same time but in what’s called the prime boost scenario in heterologous because the vehicles are different--DNA based versus soluble protein based. So they’re different biochemical properties.

So we started with DNA, primed the immune system with that because it works so well here, and then ended with a boost of protein and you can see that we got tremendous protection out--quite a ways, almost--almost 90-percent throughout the whole cycle. So we’re getting closer.

The other thing that came out of this experiment is in separate groups of mice that we immunized and should have been protected, we also depleted subsets of T-cells. This is an intracellular protein, so we know by the nature of how the immune system works that T-cells are going to be important in killing these tumor cells. And we wanted to know is it helper subset or the killer subset that--that have the biggest impact? And they answer is they both do, which was a surprise to me, not so much that I didn't think CD4 cells would be involved because we always need help it seems for killer cells to be more potent killers, but I didn't know it would be this great of an effect. So in these mice they have CD8 cells but we took away the help and the CD8 cells couldn’t do what they needed to do. And in these we took away the killers and again it couldn’t do what they needed to do,

So the bottom line there is I think we’ve found a better way to immunize and we know now that we need to make sure the vaccine will employ to the best of its ability both CD4 and CD8 type T-cells in order to completely kill the tumors.

We believe the vaccine is going to be very effective. We also believe when we get a little bit closer to understanding the molecular biology behind why it even exists, there may be other therapeutic approaches that could be applied, along with vaccination, or in lieu of vaccination that might be just as effective or more effective in treating prostate cancer.