New biomarker for prostate cancer found

First up - prostate cancer. About 50,000 cases of prostate cancer are diagnosed in British men each year. That's more than 10 times the number of cases of cervical cancer, and three times higher than the cervical cancer mortality rate. So why is there a screening programme for cervical cancer but not one for prostate cancer? Part of the reason is that there are some very effective markers for cervical disease, which make it relatively easy to discriminate between healthy people and someone with cancer. That hasn't been the case for prostate disease. Although now, scientists at Johns Hopkins in the US have developed a new antibody that can home in selectively on prostate tissue. And by coupling it to a labelling molecule, it can be used to highlight prostate cancers around the body and even deliver drugs to prostate cells. Daniel Thorek lead the study and spoke to Chris Smith about the discovery...

Daniel - One of the big problems that we have in prostate cancer patient management is how do we accurately detect the disease, is the disease localised to one spot or has it spread to many places? And then once we know what stage of the disease a patient has, what can we learn about that disease in order to best treat it? That's a longstanding problem in the field of prostate cancer because we know that the number of men who have this disease is very, very high. But how we treat these patients optimally is this longstanding question and this project was really aimed towards developing a new tool to detect disease, to characterise it, and to monitor it.

Chris - This is the concept of a biomarker, isn't it? Something you can measure which tells you about the disease, about the disease process and where the disease is.

Daniel - Right. So in prostate cancer we've been very  fortunate because we had a very robust biomarker for quite a while, that is the PSA value, so the prostate specific antigen. This is a protein that is secreted from prostate cells and prostate cancer cells. As men age, we start to see a little bit more of this protein leaking into the blood and, if you have cancer, the number goes up even higher. The problem is using that number can be difficult to guide therapy. So, if the number of the PSA level in the blood is not very high, it can be difficult to discern if a patient has disease or not, or if a patient who has disease is responding to therapy or not.

Trying to improve on that test has been the main goal of this project. What we aim to do was to be able to be able to image individual lesions in the primary prostate to give an idea of the number of lesions and what molecularly was going on in those cells.

Chris - How?

Daniel - So the way we did this was we developed a novel antibody or a new antibody called 11B6. And the antibody targets a very specific portion of a protein produced by prostate cancer cells and by the prostate, and this protein is called HK2. We can then take that antibody, we can radio label it or we can put a fluorescent dye on it and we can inject that into either animals or, hopefully in the near future, into patients.

Chris - Now why did you choose that particular marker in the prostate to make this antibody that should recognise - why go down that path?

Daniel - That's a great question. So finding good biomarkers is really the difficult thing. We know that many cancer cells have aberrant expression of certain genes and, ultimately, of certain proteins. But many cells that are cancerous share all of the markers as healthy normal cells and so being able to select diseased cells specifically is a very difficult challenge. In the case of the prostate though we were a little bit lucky. So HK2, the protein that we're targeting with the 11B6 antibody is very similar to PSA, so it's specifically and only produced in men in the prostate and prostate derived tissues. So the only time you have prostate derived tissues is when you have prostate cancer, so here we have a protein that is specifically expressed by the organ that we're targeting and it's not expressed in any other tissues.

Chris - But that doesn't mean you can discriminate between a healthy cell and a cancer cell, they will both have this marker on because they're both prostate cells?

Daniel - That's absolutely correct. But in the case of prostate cancer we have a specialised case where any prostate tissue should be removed if a patient is at high risk of developing aggressive disease. And so, by using an antibody with an imaging agent tagged onto it, this 11BC antibody, we were able to target both healthy primary prostate tissue and diseased tissue. We get really the full gamut so we're able to really discern both malignantly derived tissue and healthy tissue, both of which we want to be able to characterise and hopefully remove.

Chris - Now will this work, that's the critical question, isn't it? Because you've done this in animals so far, have you got now a way to translate this to human patients to see if a) the antibody is safe, and b) if it does what it says on the tin - will it work in a human the same way it works in your experimental animals?

Daniel - It's always difficult to predict that. We're very confident that, in this case, we may have a really robust imagining agent to detect and characterise cancer in man. What's typically done is we take human cancer cells and put them into immuno-deprived mice, so those are mice that don't really have a functioning immune system which can allow transplanted human disease to grow in them. So that's usually how imagining agents are developed.

What we've done in these studies is we've used those models, but we've also evaluated models that look at disease in the bone. Prostate and breast cancer are very well known to metastasise to the bone, and imagining in the bone is very difficult - our antibody works there. We've also developed genetically engineered mouse models of disease. They have a fully intact immune system but they spontaneously develop cancer and so we can also use the antibody there. And finally, we've already initiated non-human primate studies, so we've done some toxicology in monkeys when we administer the antibody we've not seen any deleterious effects.