The five faces of prostate cancer
Prostate cancer can no longer be treated as one disease. Dr. Alastair Lamb, Clinical Lecturer in Urology, Cambridge Biomedical Campus, UK, outlines some recent work identifying genetic subtypes of prostate cancer and how this could alter future care
The question that often arises from the patient is “When will this prostate cancer kill me, doctor?”. The question may take a softer form than this but the point is the same: men want to know whether this new diagnosis they have just received is going to shorten their life; whether it means they can no longer go on the long-planned family trip to Australia next year; or whether they’ll see their first grandchild born.
And really this is a question about risk stratification. The thing is, as often discussed with men in clinic (or with someone at a dinner party), that it doesn’t necessarily matter if you get prostate cancer; half of all men are going to get it if they live beyond the age of 80. No, the problem isn’t having prostate cancer; it’s what type you have. We already divide men into subgroups in prostate cancer which we believe have different outcomes. Currently we do this on the basis of three things: a blood test (the prostate specific antigen or PSA test), the appearance of tissue under the microscope (the Gleason score), and the size and extent of the tumour based on MRI scanning and digital examination of the prostate (the TNM score).
Based on these three things, we divide men into low, intermediate or high risk. (1)But the problem is that we often come across men in the lower risk groups who, contrary to expectations, do very poorly in terms of outcome. Conversely we also have men in the high risk group who actually do unexpectedly well. So, there must be some additional information about the behaviour of such cancers that we are missing with our current approach.
Genetic information in each cancer cell
An important answer to this problem could be the genetic information stored within the prostate tissue itself. Tissue architecture, tumour growth and, to an extent, release of PSA are all downstream indicators of something that is happening in the DNA of the cancer cells within the tumour. DNA is the library code which contains the information that tells a cell what to do.
The machinery within each cell nucleus turns DNA into RNA (Ribonucleic acid) which in turn determines the sequence of amino acids to form a protein which acts out the function of a cell. It’s these ‘molecules’ – DNA, RNA and protein – which are the key to a cell’s behaviour. Protein is notoriously difficult to measure accurately and so DNA and RNA (a surrogate for protein levels) can together be measured to give a molecular impression of current and future cell activity.
The five faces of prostate cancer
In our study recently published in EBioMedicine (a new open-access journal from The Lancet and Cell Press) we looked at prostates from 259 men from Cambridge, UK and Stockholm, Sweden and measured the DNA and RNA in the tissue looking for mutations in the DNA library code and whether RNA levels had been turned up or down. Based on this, we grouped men with prostate cancer into five distinct groups, each with a characteristic genetic fingerprint, coining the name the “five faces of prostate cancer” in Aine McCarthy’s CRUK blog.
Two of these groups had dramatic changes in the number of ‘copies’ of each gene due to mutations in the DNA and these were linked to marked increase or decrease in RNA levels. By contrast, the two groups had very little change compared to normal prostate tissue. A fifth group was somewhere in between. (2)
We were very interested to see that the two groups with dramatic molecular changes did far worse in terms of prostate cancer relapse compared to the other groups. In fact, membership of one of these two groups was a much stronger predictor of poor outcome than current clinical parameters such as PSA, Gleason score or T stage.
The future of prostate cancer care
Future decision-making in prostate cancer could look very different to current practice. Although we don’t currently screen for prostate cancer (thus lagging behind breast, ovarian and gut cancer) “next-generation screening” based on a composite algorithm of PSA-related tests will soon be available (Lamb & Bratt, Lancet Oncology 2015 in press – details to follow). Having identified a population of ‘at-risk’ men, we will then proceed to diagnosis with a biopsy.
We then have to apply the molecular stratification techniques described above to biopsy tissue – the EBioMedicine study looked at prostate removed at prostatectomy which gives larger samples of tissue than biopsy – or even to small amounts of circulating DNA in the blood. An important step here will be to make sure that the right bit of the prostate is sampled and it is likely that high resolution “multi-parametric” MRI scanning will play an important role in selecting the key target in the prostate.
The molecular information received from selectively targeted biopsies, alongside the current clinical parameters will inform a more comprehensive risk stratification which will permit a far more intelligent approach to treatment decision making and subsequent follow-up regimens. The precise form of molecular score is yet to be decided. Two commercial scores are already available (Prolaris and Oncotype Dx) and it is likely that further approaches will soon follow.
For these to be used alongside the trio of PSA, Gleason and TNM staging, much rigorous testing will be needed in larger cohorts of men to prove that they really can provide large-scale clinical benefit, helping men to avoid the side-effects of treatment including incontinence, impotence and bowel dysfunction, and ultimately saving lives.
Alastair Lamb, Clinical Lecturer in Urology, Addenbrooke’s Hospital, Cambridge & Cancer Research UK Cambridge Institute, Cambridge Biomedical Campus, UK.
1. D’Amico, A. V., Whittington, R., Malkowicz, S. B., Schultz, D., Blank, K., Broderick, G. A., Tomaszewski, J. E., Renshaw, A. A., Kaplan, I., Beard, C. J., and Wein, A. (1998). Biochemical outcome after radical prostatectomy, external beam radiation therapy, or interstitial radiation therapy for clinically localized prostate cancer. Jama 280, 969-974.
2. Ross-Adams, H., Lamb, A. D., Dunning, M. J., Halim, S., Lindberg, J., Massie, C. M., Egevad, L. A., Russell, R., Ramos-Montoya, A., Vowler, S. L., et al. (2015). Integration of copy number and transcriptomics provides risk stratification in prostate cancer: A discovery and validation cohort study. EBioMedicine 2, 1133-1144.