When a diagnosis of prostate cancer is made, it is usually based on two things. The first is an elevated PSA (prostate-specific antigen) blood test indicating that a man is at slightly higher risk than a man with a normal PSA. The second requires men with an elevated PSA to have a prostate biopsy. This involves inserting needles into the prostate to take a dozen or so samples of tissue.
The detection of a cancer by the biopsy is a process that is based upon chance. The operator does not know where the tumour is but distributes the needles throughout the prostate in the hope that one or more will hit cancer. The reality is that needles tend to get bunched to the base and back of the prostate, as these parts are easiest to sample.
It is not too hard to see why this strategy currently misses important cancers, misclassifies important cancers as unimportant (due to a “glancing blow”) and finds the many small unimportant cancers that we know are common in men over the age of 50. These unimportant cancers, as the old adage states, are those which men “die with and not from”. Finding them can lead to unnecessary anxiety and unnecessary treatments.
It is customary to blame PSA when the prostate cancer diagnostic pathway is being challenged. However, it looks increasingly likely that our main source of error lies not in PSA, but in the biopsy. Why? Because PSA on its own is a powerful predictor of whether a man will die or not die of prostate cancer. Swedish studies, which looked at blood collected from men who died many years later, showed that a very low PSA was a virtual guarantee a man would not die of prostate cancer.
Information on tumour location is going to be the big step-change that we have been waiting for
In a recently published clinical trial, comparing surgery versus observation in a group of men with prostate cancer, PSA was the only predictor that indicated that men would live longer if they were randomised to surgery. When the findings of the biopsy were added to this, the ability to predict outcomes vanished.
I believe, as do many others, that information on tumour location is going to be the big step-change that we have been waiting for. The best way to get this information is by having a scan of the prostate using magnetic resonance imaging, or MRI, before a biopsy is considered. This is a scan that takes about 30 minutes and does not involve any radiation. Many men will already have had MRIs to investigate back pain or knee and hip disorders.
It may well be that a biopsy can be avoided in men who have a normal MRI. This might save half a million unnecessary biopsies per year in Europe. On the other hand, if there is an abnormal area within the prostate, we might use this information to reduce the number of needles used to biopsy the prostate and also to direct those needles to the area in the prostate that is most likely to represent clinically important cancer.
However, before we use MRI in this way, we need to establish just how good a test it is. To this end, we are running a trial called PROMIS that is open to all men who have been recommended to have a biopsy (and have not had a biopsy before) and can have an MRI (cardiac pacemakers and claustrophobia are the two main reasons why some men cannot have an MRI). The trial offers all men a high quality MRI and a very accurate biopsy of the prostate, which is different to those we described earlier because it “maps” the prostate.
The trial is designed to tell us whether or not we can use MRI to avoid a biopsy in a large number of men and do the biopsy better in the few men that will need to have the prostate sampled. Men taking part in the study will not benefit from either of these possible outcomes of the trial. They will, however, get the most accurate diagnostic testing possible anywhere in the world.
Mark Emberton is professor in interventional oncology at University College London and chief investigator on the PROMIS trial. His clinical and research interests are on diseases of the prostate gland, and he is currently focused on evaluating minimally invasive therapies for localised prostate cancer.