Detecting and diagnosing prostate cancer

A quarter of all male cancer diagnoses are prostate cancer. Chris Smith sets out to uncover the facts...
18 February 2020

Interview with 

Vincent Gnanapragasam, Michael Eden, Vineetha Thankappannair, & Michael Langford, University of Cambridge


Man waiting


Every year, about 50,000 men are diagnosed with prostate cancer in the UK. That’s about a quarter of all male cancer diagnoses in that period. It sounds alarming - but at the same time, people often also say that this is something you die “with”, rather than “of”. So what is the bottom line? Chris Smith set out to discover how prostate cancer presents in the first place, what it looks like down a microscope, and how we diagnose it. On the way he spoke to experts Vincent Gnanapragasam, Michael Eden, Vineetha Thankappannair, and heard Michael Langford's experience of living with the condition...

Vincent - Everyone is at risk, and they used to say that if you live to 80 then 80% of men will get prostate cancer. And that's probably true. Having said that, certain groups are at higher risk, so particularly men of African descent have a higher risk, or if you've inherited certain gene mutations, in particular the BRCA mutation, but that's only about 1% or so. And we still don't quite understand why some men develop prostate cancer more rapidly than others do. But what we do know, is that when a man presents with prostate cancer, the older they are, the more likely they are to have more advanced disease. So really prostate cancer is a disease that evolves over time. And the real challenge here is to not over-diagnose cancer, which is early, but at the same time, not picking it up when it's too late.

Chris - And that's what we're going to explore as we go through the programme. But first, let's back up very slightly. What is this gland that we're discussing? Well, the prostate is found only in men. It sits below the bladder, surrounding the urethra, the tube that drains urine. Normally the prostate's about the size of a small satsuma and its job is to produce many of the components of seminal fluid. As a man gets older, his prostate naturally enlarges which can obstruct the flow of urine, and although this enlargement usually has nothing to do with the development of cancer, it's nevertheless often the trigger that sends a person to see their GP who then investigates further.

Vineetha - My name is Vineetha Thankappannair . I'm the Macmillan lead nurse for urology and a prostate cancer specialist nurse at Addenbrooke's. Usually, the urinary symptoms are caused by the enlargement of the prostate. All men have their prostate getting enlarged from 40 years of age. When it gets bigger, it causes obstruction to the water pipe because the prostate sits on the bladder and the water pipe goes through it. When it gets enlarged it squeezes the water pipe which causes restriction in passing urine, which then leads to the symptoms like getting up at night or frequency or urgency. things like that. The enlargement of the prostate is the majority of the time due to benign enlargement, but sometimes advanced prostate cancer can cause some of those symptoms as well.

Chris - Which is why if someone has them it's important that they should get checked?

Vineetha - Yes.

Chris - Are there any other symptoms? Because we've dwelled on urinary symptoms. Are there any other ways in which prostate cancer could present in the first instance that the person might not think it's got anything to do with their urinary system?

Vineetha - The other symptom is usually bone pain. And that's the majority of the time when it is an advanced prostate cancer. You know, usually patients get bone pain and they put it down to a stretch or having done something physical to the back or things like that.

Chris - So when the person goes to see the GP with these symptoms, what sorts of things would they expect their GP to have done or checked or have put in place before they would send them to you?

Vineetha - So usually if a patient presented to the GP, the GP should take their history, especially the urinary symptoms and when it started. If it is long term urinary symptoms and getting worse, it's usually because of the enlargement of the prostate. But if it sort of happened in a short period of time, it could be a urinary infection. And so it is important that the GP checks a urine sample. And then if there is no infection, a GP should do the blood test for prostate specific antigen and usually the GP will explain what the prostate specific antigen is, what can make it go high and things like that. So that's what I'm expecting the GP to explain to the patient before they make that referral.

Chris - So significant emphasis is placed on the PSA test, but what actually is prostate specific antigen and how does it highlight the possible presence of prostate cancer?

Vincent - It's a protein secreted by the prostate gland in its natural functions. It just so happens that prostate cancer cells selectively secrete more of it. It's not a perfect marker and its accuracy is maybe about 60 to 70%. But certainly the higher it is, the more likely it indicates prostate cancer.

Chris - When you say the accuracy is 60 to 70%, what does that mean? Accuracy for what?

Vincent - The accuracy that it predicts the presence of a prostate cancer. Now it's very important to remember that the bigger your prostate is, the more PSA you produce. So in fact, if you naturally have a large prostate, you're going to have a high PSA.

Chris - How does it get from the prostate into the blood?

Vincent - It's secreted in the bloodstream from the prostate gland.

Chris - So a person who has a tumor growing, the reason it's higher is what? Because more of it's leaking into the blood or because the cancer is actively making it?

Vincent - Yes, exactly right. The type of cells that produce prostate cancer are known as the epithelial cells and the cells of origin from prostate cancer tend to arise from this compartment. And that's why they tend to produce more PSA.

Chris - So the GP gets a PSA test result, presumably they don't just act on a one off. If they saw a value that might be of concern, they'd repeat it before they did anything else?

Vincent - You would hope so. And the reason that's important is that the PSA can be affected by other things. Sexual activity and ejaculation will elevate the PSA reflecting the function of the prostates and ejaculatory gland that produces secretory substances. And the message that we try to give out is always to repeat the PSA because a single high level may not be indicative. Having said that, we would always take any PSA elevation seriously for, at least, assessment.

Chris - So how do we interpret the results of a PSA test then, given that it can be raised for many reasons other than just cancer?

Vincent - It's a challenging issue. And so this idea of age referenced PSAs have come about. Now that means that you have brackets of age where the PSA is allowed to be higher. So it's got to do with the natural enlargement of the prostate and an assumption that if you're in a higher age bracket, you've got a bigger prostate. So these age-referenced PSAs are used as guides in screening programmes where they look for prostate cancer. They've set baseline cutoffs like for example, a level of 3. But of course the PSA can vary from 0.1 up to 4, 5 or even 6 or 7 or 8 or 10 depending on the size of the prostate and not have prostate cancer. So interpretation is a difficult thing, but primarily to do with what you would expect to be out of the norm for someone whose prostate is naturally enlarging.

Chris - And if that is the case, under what circumstances then should someone be referred to a specialist prostate clinic?

Vineetha - 2 elevated PSA, 4 weeks apart. That is the referral criteria. Or if it is one high PSA and if the examination of the prostate is abnormal, then the GP should refer to secondary care on a 2 week wait basis. So that's a cancer pathway basis.

Chris - In other words, it's cancer until proven otherwise and that person's coming your way or at least to someone like you?

Vineetha - Yes, that's correct. So usually at the majority of hospitals there is a triaging system where either the doctor or the specialist nurses look at the referrals every day and then are directing people to the right pathway.

Vincent - And that system will look at this individual and check things like, you know, make sure there's no other reasons like a urine infection, which may have caused this. But if it's looking as though the PSA is elevated for age, classically now, the patient will be invited for an MRI scan first to see if we can pick up anything or see anything.

Chris - MRI stands for magnetic resonance imaging. It's a noninvasive scanning technique that uses a powerful magnetic field to enable doctors to see the body's internal structure in three dimensions, meaning they can measure the size and shape of the prostate with great accuracy and even see inside it and other tissues.

Vincent - What the MRI can do is it gives you a value of how big the prostate is. As I mentioned to you, that if your prostate is large and you have your PSA raised, it could just be that's normal for you. So in fact correcting the PSA for the size of the prostate, something we call the PSA density, is actually a much better test. And here's an interesting thing, if we were able to transfer that to primary care then actually we might be able to give GPs more confidence about referring or not referring. And that's something we're looking into.

Chris - What proportion of prostate cancers are detectable straight away on an MRI and what fraction do you subsequently think, well it looks negative, you'd do a biopsy and then find they do actually still nevertheless have a prostate cancer?

Vincent - So these are very complex questions and in fact MRI is part of the diagnostic pathway now, but many issues still remain because in the end, unlike PSA, which is a blood test you measure in a lab, the MRI interpretation is down to the experience of the radiologist. And if they are very good radiologists, their ability to see things is much better. But if they're not, then it may not be so good. So everything is always a balance of risk and benefit. With regards to the MRI, they have come up with a scoring system known as the PI-RAD system, and this PI-RAD system gives a score of between 1 and 5 in terms of how certain they are that there might be something. And if you have a lesion score, say 4 or 5, that's pretty sure that it looks abnormal. That doesn't mean it's cancer, but it looks abnormal and the chances are that it will be a cancer when you biopsy it. Scores of 1 and 2 are almost negative. And so in that case, it's probably unlikely that they can see anything. It doesn't mean there's no cancer there, but it's unlikely. Where the situation gets a bit murkier is around the level of 3 because this is a bit of uncertainty. Now in general terms, if you were to biopsy men across that scale of 1 to 5, your likelihood of finding prostate cancer gets higher with each level. And the other thing it does, is it tells you where these abnormal areas are so you can actually target your biopsy to those areas.

Chris - MRI scans can therefore help us to highlight areas of interest and predict the risk of finding a cancer. But the diagnostic gold standard is a trans-rectal biopsy. This is where, guided by an ultrasound probe introduced into the rectum, small pieces of tissue are taken from the prostate, including from areas that are judged to be both normal and those that might be cancerous.

Vineetha - Trans-rectal biopsies are done in outpatients in our clinic. And we have special machine that sort of help us to outline the MRI and then target is outlined. And then ultrasound helps the consultant or the doctor to target those areas. So it's targeted biopsy, which is done through the back passage, if that's what the patient needs.

Chris - So you use the scan to inform - combined with ultrasound on that day - where you want to sample. So you make sure you basically are sampling the bits that you spotted in the scan and were interesting.

Vineetha - Yeah, so we do targeted biopsy, which means we take three or four samples from the target, which is the suspicious area within the prostate. And we do standard biopsies, like six from the left side of the prostate and six from the right side of the prostate. The reason for that is the majority of the time prostate cancer develops in the peripheral area. So to make sure we are not missing anything, we do target plus standard.

Chris - What's actually involved in doing that biopsy?

Vineetha - What they are expected to do is lie on their left hand side and then usually a probe will go through the back passage. The probe itself is slightly bigger than the thumb, and that is all fitted with equipment which guides the needle. Before the biopsy we numb the prostate gland to make sure we can take the sample pain-free. Patients will feel a scratch, like you're at the dentist but it's the wrong end. After numbing the prostate, they take the sample. What I have heard from patients is that they feel a heavy touch-like feeling when they fire the biopsy gun, that can be uncomfortable.

Chris - So it's basically firing a needle through the gland on a trajectory that you've chosen to sample the right bits. And as it fires the needle through it's grabbing a core, as it were, of that tissue and pulling it back for you to then send to the pathologist.

Vineetha - Yes, that is right.

Chris - What side effects will the patients experience afterwards? Or what do you warn people, "when you go home after this, this is what to expect"?

Vineetha - The main side effect is bleeding through back passage and blood in urine. Usually we tell the patient, make sure they drink plenty to flush their system. The other side effect is blood in semen, which can last sometimes four to five weeks. And it doesn't present as bright red blood in the semen, it's usually blackish, old blood. We do say "don't get alarmed, it's normal to see this." The other side effect is infection, so there's a risk of urine infection and there is a risk of septicaemia, which is infection in the bloodstream. So we do give antibiotics for three days post procedure.

Chris - Increasingly though, practices are changing and it's now possible to make better use of the scan data to achieve even more accurate sampling from the prostate.

Vincent - And so we are now starting to do much more what we call transperineal biopsies. And this is where we still have a probe in the back passage, but the needles themselves pass through the perenium, the space between the rectum and the scrotum. And there are many ways to do that. The classical ways to do it are under general anaesthetic in theatre, but that's expensive. But you can get to the front and the size of the prostate much better than you can from the trans-rectal approach. But we and others are starting to do local anaesthetic, transperineal biopsies using new devices that have been developed. And we have one in Cambridge which is going to be very cheap. And we think it will revolutionize how the biopsies are done, because you're doing it through the perenium, but also under local anaesthetic, and for the same time as well. But in essence, all these biopsy methods are designed to do one thing, to take samples from the prostate to make a diagnosis. Now we used to do what's called systematic biopsies, which means basically taking samples from different sectors of the prostate. Now we do that but also targeted biopsies, which means going to the area the MRI has defined and that is the samples we take and then send off to the laboratory for our histopathologist to look at and tell us what they find.

Chris - The biopsy samples are small cylinders of tissue collected at each of the biopsy sites in the prostate gland. These are sent to the laboratory where pathologists like Michael Eden look carefully at each of them for the telltale signs of cancer.

Michael - My role as a consultant histopathologist is to diagnose prostate cancer after a surgeon has taken a sample of tissue from a patient that they suspect has prostate cancer.

Chris - And when you receive these specimens, how do you investigate them?

Michael - We would take each of those bits of tissue. That tissue is cut into very, very thin sections because the cylinder of tissue contains thousands and thousands of cells, but actually down the microscope, what we want to see is one or two cells at a time. So we will cut very, very thin sections and we can then stain it and then look at those cells which have been stained down the microscope.

Chris - So a lot of your diagnosis is literally made by eye. You look down a microscope and you're looking for cells that don't quite look right - or arrangements of cells, how they're organized inside the prostate - that doesn't quite look right.

Michael - That's absolutely correct. So depending on the cancer, we'll be looking at "what do the cells as individuals look like? What are they doing as a group? Are they growing together in a way that looks abnormal?"

Chris - What can you say about the likely behaviour of the cancer, based on the appearance? When you see it, can you give some kind of insights into how likely it is to progress, or grow aggressively, or be a more indolent disease?

Michael - Yeah, there are a number of things that we can look at to try and give an indication to the surgeon and to the patient as to how that cancer will behave. The first thing we try to work out is, what type of cancer are we dealing with? The second would be to assign what's known in histopathology as the grade of the cancer - does it look very, very aggressive? In which case we think that this cancer is going to behave very badly. Or do we think that the grade is very, very low,? In which case we think it's behaving more like normal cells. So it's less worrying. The third thing we look at is, do we have any evidence to show how far the cancer has spread? By that I mean, has it spread into other parts of the body?

Chris - You've got a microscope sitting here. Could you show me what the specimens you look at look like?

Michael - Yeah, absolutely. So I've got a selection of prostate biopsies here. Typically prostate cancers are graded using a Gleason Score and the Gleason Score typically comprises two numbers which are added together. And the numbers can range from one to five. We don't usually see grade one and grade two. So typically we're looking at a number ranging from three to five, added to another number ranging, from three to five. And that gives you a total number, which will range from six which would be the lowest, up to 10 which would be the highest or the worst grade. So what I'll show you now is a prostate cancer, which is a Gleason three plus three which equals six which is the lowest grade. So as you can see here, these sack-like spaces lined by two layers of cells. There is what's called an epithelial layer of cells, which produces secretions in the prostate. And underneath them a myoepithelial layer of cells which are smooth muscle in origin. If we see these smooth muscle layer of cells, I can be fairly confident that this is probably benign.

Chris - So just to describe what I'm seeing here, we've got this thin strip of tissue and there are lots of little holes. This is because the prostate is a gland, isn't it? So it's going to have these structures, which you described as sack-like, which is where the secretions are going to be produced and then put into and then expelled from the prostate. And around those sack-like areas are rows of cells, which have got nice purple centres. Those are the nuclei in the middle of the cells, aren't they? And they're all very nicely organized. And underneath those you're saying, and I can see them clearly, there's another layer of cells. Those are the myoepithelial cells that you're saying if they're there it's usually a sign of good health.

Michael - Correct. That's absolutely right. So my job is to look through systematically, looking to see really, is there anything in the architecture of the cells, how the cells are arranged in relation to one another, that would indicate this might be prostate cancer rather than benign prostate glands? And one thing that does catch my eye, which you get with many years of training, is that areas of prostate cancer architecturally show glands which are quite closely spaced together and they look a bit smaller.

Chris - This bit over here that you've now highlighted - and I see you've drawn with red pen on the slide. So that must've been the hotspot that you noticed - that's very compacted and squeezed together. It's subtle though.

Michael - As you zoom in on them, you'll see the two main features, really. One is that the myoepithelial cell layer is missing from many of these glands, if not all of them. And the nucleus of the cell, in some of them you can see a little dot, which we call a nucleolus, which is quite prominent at this magnification, which you don't see in normal prostate cells. So that, combined with the absence of myoepithelial cells, suggest to me that this is prostate cancer. And then, at that point, if I'm happy it's prostate cancer, then I'd be looking at trying to grade it to see how aggressive it is.

Chris - Preparing the tissue and then waiting while pathologists, like Michael, study it, and report the findings takes a week or two. When the results are ready, the entire clinical team, including the pathologists, radiologists, surgeons, and nurses then meet to discuss each patient, and the best way to manage their case.

Vineetha - It depends on the biopsy type, because if it is a transperineal biopsy, we take more samples. So that takes long, usually two weeks. But if it is a transrectal, standard biopsy, it usually takes 10 days. And then we discuss the case in our multidisciplinary team meeting and we see the patient and give the results.

Chris - And at that stage, what do you say to the patients, or what information can you impart to them, and what's the next step?

Vineetha - So in the result clinic we will have the information about the biopsy result, whether it is a cancer or not. If it is a cancer, what grade of the cancer. And we would have discussed what treatment option is available for patients. So when we see the patient, we will discuss the findings of the biopsy. Is it advanced, or is it locally advanced or early prostate cancer. And we discuss the treatment options with the patient.

Vincent - We can now work out with quite good accuracy what the likely prognosis, meaning to say the outcome is, with or without treatment, and that's what we use together with the patient to work out what to do next. First thing to know is that there is absolutely no evidence that rushing through a diagnosis of prostate cancer has any impact on survival. So it doesn't really matter how long it takes. The key thing is that it is done well. In general terms from the minute somebody presents, to say the hospital for investigations, to get an MRI scan, to be seen to have a biopsy and get a diagnosis. We try to do that within two or three weeks, and then we can make decisions about what to do next.

Chris - Some people listening to this may be wondering why it is that we have screening programs for certain diseases, but we don't have one for this, given that you've said that 80% of men aged 80 may well have prostate cancer, which argues it perhaps would be worth looking for it, so it's therefore a very common condition. Why is there no screening program for prostate cancer?

Vincent - So the problem with screening in prostate cancer is not the fact that it doesn't save lives because it does. The problem is that it picks up a lot of disease which never needs treatment. And that over-diagnosis can also lead to over-treatment because of poor understanding of the natural history of disease. A sort of instinct to treat rather than monitor. And if you look at the statistics as well, it's very hard to justify it. So for example, in prostate cancer terms, the lifetime risk of developing prostate cancer is nearly 10 times the risk of dying of it. You have to find an awful lot of men to save a single life. But the problem with screening is that it's tended to rely on a single test at a single point in time. And the test isn't good. PSA, like I said, is a nice test, but it's just not good enough. And so a lot of our work, a lot of work of others, is actually trying to do two things; define what is the right time point to pick it up. And secondly, trying to get a better test to pick it up.

Chris - Are you having any luck?

Vincent - Yes, there are actually a number of tests out there which are already better than the PSA. The problem is that they cost more as well. We are very interested in actually combining better tests, with a much more evolving way of looking at this. Because the way we see it is that each man's risk changes with time, and how we put the two together is what we are exploring at the moment. But I do think we have the tools to be able to better detect cancers which are going to spread for example, and actually intervene early enough.

Chris - And would that translate then into a saving?

Vincent - That's exactly what the Holy Grail is, because in the end of the day, screening, or early detection for cancer actually has a converse side, which is not picking up things that you don't want to pick up. And that in itself is a goal which I think is worth doing, because if you can reduce the number of people you're going to be looking for something, that means you can get your resources redirected to finding the ones which are important. Most of the time we are so fixated on finding more and more cancers. That's what we want to do. We forget about the fact that there's a lot of people who are investigated who turned out not to have cancers. And that is actually a bigger burden for the health economy, for the individual. And ultimately, if you do diagnose something which is too early, you do condemn someone to be monitored for something which they didn't know about. It's a little bit like saying, if we started to do genetic tests at birth for everyone, what it tells you is a potential that you might get something, rather than you will get something, and then you're going to end up stressed about it for all your life and it may or may not happen.

Chris - Obviously for some people it does happen.

Michael L - My name is Michael Langford. I'm 88 and a half, and I am a semi-retired professor of philosophy. 18 months ago I was diagnosed with prostate cancer, and for a time I was really pretty sick, but a combination of steroids and hormone treatment seems to have made me a lot better. Though, at one point I had to go in just over a year ago for blood transfusions, because my anaemia got very serious, but at the moment, touch wood, I am a lot better.

Chris - How did you first detect something was amiss.

Michael L - Well there's a bit of a history. 23 years ago while I was in Canada, I had a flow problem with my urine, and I had a little operation which made the prostate much smaller, then everything was fine until about 18 months ago when I was diagnosed with a combination of polymyalgia and prostate cancer, which they then discovered had spread. The cancer had spread to my femur and my ribs.

Chris - Did the GP immediately suspect that it might be prostate cancer or was it because, we should be clear, a couple of years ago, I mean you were doing martial arts. You were a very fit and active person.

Michael L - I was until 18 months ago, the doctor diagnosed the polymyalgia and wasn't sure what else was there, and I was sent for more blood tests to Addenbrooke's, and there they found that I had quite severe prostate cancer.

Chris - What tests did you have to make the diagnosis and confirm what it was?

Michael L - I was sent for an MRI scan, and that showed that I had secondary growth in the femur and the ribs.

Chris - You didn't have any biopsies? It wasn't deemed necessary, then?

Michael L - I didn't have biopsies, they were quite sure I had cancer anyway.

Chris - And how did you take that?

Michael L - I was a bit shaken, but being in my eighties, and having friends who've got prostate problems, I knew the old saying, almost all old men die with prostate cancer, but not of it. And I took consolation from that. But a year ago when I was feeling really down, I did begin to wonder whether I had to start rewriting my will and so on. My family were very supportive.

Chris - And what actually then happened to you? So you get the diagnosis, obviously you're saying it did hit you a bit hard, but clearly you've bounced back, because you look much better now.

Michael L - Well, I was the lowest about a year ago, when I had to have blood transfusions because of the anaemia. And then as soon as I started on the hormone treatment, I gradually began to feel better. And apart from one bad week last summer, which I think was an infection of some kind, I've been continually much better. I do have trouble if I walk more than a mile, I get out of breath, and if I go up more than two or three flights of stairs, I get out of breath. So I paced myself and I still teach a bit, and lecture a bit, but I try to take a lot of rest.

Chris - And what have they told you about the prognosis?

Michael L - Well, they were quite honest about this. When I first was diagnosed 18 months ago, one of the oncologists said, the average life expectancy for someone presenting as you did is about three years. But it can vary from one year to 20 years, you know? In other words, it's very, very dependent on how you respond to the treatment.

Chris - And how are they keeping an eye on you?

Michael L - I'm called in to see the oncologist about every two or three months, and I have blood tests about once a month and my anaemia level, I keep an eye on.

Chris - Do they rescan you?

Michael L - I've had two scans. They're very expensive, so understandably, I don't expect them too often. I had one about three weeks ago, which I haven't had the results of yet, but they haven't phoned me up, so I'm hoping they're good.

Chris - You know what they say? Ignorance is bliss, isn't it?

Michael L - Yes. Yes. That's probably true!


Very interesting. Thanks. Sounds like it’s presence is close to certainty, but what about preventive manoeuvres, ie are the any, proven, likely or even quack (shame in me!).

Add a comment