Prostate specific antigen (PSA) is a common blood test used by doctors to assess whether an individual has prostate cancer. It also predicts the risk of developing prostate cancer sometime in the future. It’s a good blood test but not a perfect one and the risk of cancer does increase with increasing levels of PSA in the blood tested.
Doctors have improved the accuracy of the test by also looking at the rate the PSA rises every year, by modifying the interpretation of the results in accordance with the age of the patient, and measuring additional modifications of PSA.
The test is usually performed in conjunction with a digital rectal examination, and if the doctor feels the risk of cancer is sufficiently high, a recommendation may be made to visit a specialist urologist and to consider a prostate biopsy, which is the only way cancer can be formally diagnosed.
Some men won’t benefit from a blood test to screen for prostate cancer – those with less than ten years left to live, for instance, or men under 40 years old (because prostate cancer is extraordinarily rare for them). But for a man in his 40s, a single blood test can help predict the risk of both getting and dying of prostate cancer. And it can help doctors decide how closely he should be monitored.
Trials have shown a reduced likelihood of death from prostate cancer because of PSA testing for men between 50 and 70 years old with a greater than ten-year life expectancy. This is why PSA testing must continue to be available and offered to men in the appropriate age group.
But before I talk about the evidence for PSA screening, let me correct an offensive allegation published in the press last week – that surgeons recommend surgery for commercial gain.
All doctors are patient advocates and would never recommend a course of action unless they firmly believed that it was in their patients' best interests. This is exactly why 42% of men diagnosed in Victoria with low-risk prostate cancer are managed with surveillance. They are monitored and treated only if the disease worsens. This course of action demonstrates that urologists are not in a hurry to operate on those who won’t benefit. It’s also the formal position of our professional society.
Indeed, urologists are only likely to suggest treatment for men with higher-risk prostate cancer where the survival benefit of surgery over observation has been demonstrated.
Evidence for PSA’s effectiveness
Let’s start by looking at Cancer Council Victoria figures. The five-year survival of a man diagnosed with prostate cancer in the late 1980s, when PSA testing was first introduced, was 57%. It’s now 91%, which is a massive improvement. While treatment has also got better during that time, some of the improvement has clearly been the result of PSA-based testing and early detection.
What’s more, a European study of PSA screening (ERSPC) has shown a 31% reduction in the risk of dying from prostate cancer over nine years in men who were tested. In the Swedish arm of this study, which followed-up participants for 14 years, the reduction in the prostate cancer death rate was 44%.
The American PLCO trial didn’t show a difference in survival between screened and control groups. But it had numerous flaws, including the fact that over half the men who were not supposed to be tested actually were. So it’s hardly surprising that a difference between the tested and control groups was not detected.
Urologists often recommend conservative management of prostate cancer. Two studies – the Scandinavian prostate cancer group (SPCG-4) randomized trial and the prostate intervention versus observation trial (PIVOT) – have clearly shown that, over ten years, men with low-risk prostate cancer may not benefit from surgery, but younger men with larger or higher-grade cancers definitely benefit.
Helping decide
In those with an abnormal blood test, a subsequent biopsy will provide valuable information that can help doctors decide if their patient falls into the risk group that benefits from treatment. Without the blood test, the degree of risk will not be known, and men will run the risk of a higher chance of dying from prostate cancer.
Surgery can cause side-effects, such as urinary leakage or erectile dysfunction, in a small number of men, and that has a negative effect on quality of life. But it’s important to note that many men diagnosed with prostate cancer already have pre-existing age-related erectile dysfunction – and many are not bothered by this potential side-effect.
Nonetheless, it’s important that doctors have a frank and open discussion about such side-effects with patients and their partners so that people can make informed decisions about the benefits and harms of their treatment.
And consider this:
Does the public want a faceless committee telling them they can’t have PSA-based testing? A committee that tells them it’s better not to know and to bury their heads in the sand? Men have the right to make decisions about their own lives.
As a male in his fifties, I want the right to be tested, to have a cancer detected early and to make my own decision about whether I feel the risks posed by treatment outweigh the risks of the disease. I don’t want a faceless committee deciding on my behalf that they don’t think the risks of treatment are worth taking, when it has the potential to save my life. I can make that decision for myself with the help of my doctor and medical specialists.
Read the case against PSA testing
John Coochey
Mr
I think a critical issue here is what are the side affects and how common? There seems to be some dispute about this. I worked as a Health Economist for the AIHW some years ago and was virtually ordered to show that minimal intrusion surgery was less effective than traditional TURP surgery for BPH which has similar simptons to prostate cancer and similar side affect. In fact the motive was cost (a counter point to surgeons being accused of seeking financial gain) because although new technologies had more up front costs they had much shorter hospital stays so were actually cheaper. Earlier papers had carelessly put in hospitalization of four days when in fact it was six and a half. When my results were reached I was directed verbally that this would mean more people would want the operation so costs would go up and we should not publish these results. I still have the documentation and they will find them in my papers when I die.
Chris Booker
Research scientist
I remember seeing debates about the utility of PSA testing about a decade ago. I can't believe we're still on this topic. It's obvious that what we need is a different test. As soon as we have a test with better specificity and sensitivity, all these of these issues will largely disappear. Sure, there's no perfect test, but one that's better would be much less controversial, and much more useful, which would make the need for articles like this more or less disappear.
Personally I think there should be no more research funding given to anyone examining things to do with PSA. We already know enough about it - which is that it isn't good enough! All research funding should go towards efforts to find new markers and develop ways to test for these in the clinic.
Jon Hunt
Medical Practitioner
As a GP I have great difficult trying to explain the pros and cons of PSA testing. Here we have an oncologist and a surgeon with opposing views. Clearly it is not clear what you are meant to do. What we need is for these views to be explained simply and concisely. My understanding is that PSA testing has made no difference to mortality, at least in the US which contrasts with what is said above. If it did indeed reduce mortality by 44% what is there confusion? Thats not a bad result. I wonder if the results disagrees with other studies?
Mark Frydenberg
Head of Urology at Monash Medical Centre and Associate Professor of Surgery at Monash University
Jon ,
The main issue is that proponents of testing overestimate the benefits and those who argue against overestimate harms . There is a reduced mortality proven in randomised controlled trials so there is no reason why men should not be offered the test if they wish . They are some harms - but the harms of testing is that men have a blood test , and that some men have a biopsy , which 50% of the time may not detect cancer hence a false positive PSA result . Using derivatives such as free to total ratio , age specific reference ranges etc., help reduce the number of men undergoing biopsies and hopefully the increased use of prostate MRI may reduce this even further . The main issue for me is to avoid treatment for those who don't need it , namely those with low grade cancers , this minimises the harms for the patients significantly
Chris O'Neill
Telecommunications Engineer
"avoid treatment for those who don't need it , namely those with low grade cancers , this minimises the harms for the patients significantly"
It would minimize the harm if treatment for those who don't need it was avoided.
Gavin Moodie
Principal Policy Adviser
Surely no one is arguing for 'a faceless committee telling them they can’t have PSA-based testing'. The most that is being argued is for 'family physicians to stop doing routine screening PSA tests of Australian men unless patients decide to proceed after being told about the latest research and indicate they understand the potential benefits and harms. Indeed, they should be asked to provide informed consent.'
This seems to me just the sort of issue upon which the National Health and Medical Research Council should offer guidance. That guidance should be to patients, to physicians, to courts so that they don't hold physicians liable for not recommending the test, and to governments about funding mass screenings.
For my own part, I prefer to know so I take the test, but I would be very conservative about having any procedure.
Mark Frydenberg
Head of Urology at Monash Medical Centre and Associate Professor of Surgery at Monash University
Gavin , no one is proposing mass population screening ; but it is important to realise that the average GP is very busy and if the US preventative task force recommends against screening , and this is followed by a NHMRC statement , then no one will get informed consent regarding PSA ; it won't be offered at all and any mortality benefit at all will have been lost . I agree patients need to be informed of benefits and harms naturally , but the greatest harms come with treatment rather then the test itself
Richard Hockey
logged in via Facebook
The author probably should have read this https://theconversation.edu.au/understanding-risk-statistics-about-breast-cancer-screening-8602 before using 5 year survival rates to support his argument. It sort ruined his case from then on.
R
Paul Rogers
logged in via Twitter
Correct on every count!
Chris O'Neill
Telecommunications Engineer
Professor Frydenberg: "The five-year survival of a man diagnosed with prostate cancer in the late 1980s, when PSA testing was first introduced, was 57%. It’s now 91%, which is a massive improvement."
That's a misleading way of stating the statistics. PSA testing detects a lot more non-dangerous cancers than previous testing methods. A much better way of determining the success of PSA testing is to see how much improvement there has been in the Prostate cancer death since PSA testing was introduced…
Read morePaul Rogers
logged in via Twitter
It seems to me that the significance of individual cases can be lost in the epidemiology of public health statistics.
For example, a colleague whose father died of aggressive prostate cancer at an early age decided to have a PSA test at around 50, only to be diagnosed with prostate cancer on biopsy with consequent prostatectomy.
For how many of these men is it going to be too late if PSA test are not recommended? As I asked on the other forum: "do we wait for blood in the urine before we do anything"?
Chris O'Neill
Telecommunications Engineer
"For how many of these men is it going to be too late if PSA test are not recommended?"
I think the PSA antagonists may be leaving a vacuum in implying that PSA tests are no better than randomly selecting men for biopsies as far as the ultimate objective is concerned. But their point is based on the fact that the correlation between PSA level and dangerous cancer is very low and the fact that a lot of Prostatectomies (48 in one study, even more in others) are usually needed to make a difference…
Read morePaul Rogers
logged in via Twitter
"Perhaps the logic of PSA test antagonism is that since it achieves a lower success rate than pre-emptive Prostatectomy, and since pre-emptive Prostatectomy is unthinkable then PSA test-based decisions should also be unthinkable. "
This logic may apply at the population level, but it seems to me that it fails at the level of individual clinical management. The PSA test is (or should be) only one factor in assessing prostate cancer risk. Digital rectal exam, familial history, age, perhaps obesity and general fitness should probably also be considered.
In this setting, PSA is only one diagnostic tool in evaluating risk and treatment. Perhaps PSA alone should not be used to rush into invasive testing or surgery, but I can't see that it is not a useful element in the mix.
Chris O'Neill
Telecommunications Engineer
"This logic may apply at the population level, but it seems to me that it fails at the level of individual clinical management."
So are you saying that in the case of your colleague's father, pre-emptive Prostatectomies for every man at the age of 45 would have failed him? That would have been very unlikely.
"Perhaps PSA alone should not be used to rush into invasive testing or surgery,"
That wouldn't have helped your colleague's father, would it?
"but I can't see that it is not a useful element in the mix."
I'm not saying that it's less useful than doing nothing. It might be a tiny bit more useful than nothing. I'm saying that it's much less useful than Prostatectomy for every man at the age of 45. To me, that means the current strategy is a failure. Why not give it up and just remove every man's prostate at the age of 45?
Paul Rogers
logged in via Twitter
As you have agreed: ". . . pre-emptive prostatectomy at age 45 is unthinkable".
That being so, why raise it as a comparison? Further, comparing pre-emptive prostatectomy with PSA testing is (1) not a valid comparison because pre-emptive prostatectomy is, obviously, grossly invasive, and (2) the comparison should be with total risk assessment, which includes PSA among other factors, but is not limited to PSA. I describe that scenario in a previous post.
Further, although PSA screening may not have saved my colleague's father (but done early enough, very well could have), one can reasonably conjecture that it probably did save my colleague because of early detection of an aggressive form of prostate cancer.
Chris O'Neill
Telecommunications Engineer
"That being so, why raise it as a comparison?"
You seem to have missed the logic which is that the current strategy which involves using PSA tests performs worse than something that is unthinkable. Therefore the current strategy with its PSA tests should also be unthinkable.
Until you get that fundamental point there is no value in considering anything more detailed.
Like you, I used to believe that the current strategy was worthwhile until I found out the risk and success statistics involved…
Read morePaul Rogers
logged in via Twitter
Chris, I would be interested in the mortality stats you quote and the rigour of the study from which they arise.
For comparison: Eur J Cancer. 2010 Jan;46(2):377-83. Prostate cancer mortality in screen and clinically detected prostate cancer: estimating the screening benefit. van Leeuwen PJ, Connolly D, Gavin A, et al.
CONCLUSIONS: A relative reduction in Pca metastasis of 53% and Pca mortality of 37% was observed in the intervention population after 8.5 years of observation.
http://www.ncbi.nlm.nih.gov/pubmed
(Cohort: 11,970; control: 133,287)
I understand this is only one study.
Paul Rogers
logged in via Twitter
"Until you get that fundamental point there is no value in considering anything more detailed."
I agree, because that fundamental point is not fundamental, nor relevant.
Chris O'Neill
Telecommunications Engineer
"that fundamental point is not fundamental, nor relevant."
Thank you for your erroneous assertion. By the way, you haven't answered the question: The current strategy is likely doing worse than the "unthinkable". So why aren't we doing the "unthinkable"?
Perhaps I should extend it to: Why aren't we doing the "unthinkable" or why do we keep doing something that performs worse than the "unthinkable"?
Paul Rogers
logged in via Twitter
1. It's not 'unthinkable', which is self-evident. But it is unachievable.
2. I don't agree that you have presented sufficient evidence that the current strategy (or at least a 'best practice' version) is worse than the "unachievable".
QED?
Chris O'Neill
Telecommunications Engineer
I got the mortality and diagnosis stats from "Prostate Cancer Research News" which the Cancer Council Vic sent me. They have it on their website: http://www.cancervic.org.au/downloads/cec/prostate-newsletters/Prostate-news-Nov-2008.pdf
"Randomised prostate cancer screening trial: 20 year follow-up" at http://www.bmj.com/content/342/bmj.d1539 cites"Screening and prostate-cancer mortality in a randomized European study. N Engl J Med2009" in its comparisons and quotes from it: "To prevent one death from cancer, 1410 men would need to be screened and 48 treated."
Chris O'Neill
Telecommunications Engineer
Current strategy: 48 treatments per life saved.
"Unachievable" or whatever you want to call it strategy: 23 treatments per life saved (deaths from Prostate cancer in 2010 were 4.4% of the total).
Paul Rogers
logged in via Twitter
"To prevent one death from cancer, 1410 men would need to be screened and 48 treated."
I don't see that as unacceptable, considering that the unthinkable is, well . . . unachievable.
Thanks for the papers, I'll read up, but it occurred to me that in these sorts of compilations, determining (and recording) the cause of death in older men is sometimes fraught. And one would have to consider morbidity and quality of life as well as mortality to make a meaningful comparison.
Chris O'Neill
Telecommunications Engineer
"I don't see that as unacceptable, considering that the unthinkable is, well . . . unachievable."
OK, that works from the point of view that humans do not make rational choices and have to be frightened into taking action. But I'd like to be permitted to make rational choices for myself.
"And one would have to consider morbidity and quality of life as well as mortality to make a meaningful comparison."
I'm glad you brought that up. Where does consideration of incontinence and impotence fit into your "I don't see that (48 treatments per life saved) as unacceptable" statement?
Paul Rogers
logged in via Twitter
"Where does consideration of incontinence and impotence fit into your "I don't see that (48 treatments per life saved) as unacceptable" statement?"
Compared to what? Metastases and chemo? And not all treatments post PSA screen result in impotence and incontinence, short or long term.
"Watchful waiting" is a nice catch phrase, but waiting for what? Evidence of metastases, doubling of PSA, Gleason scores? All of the above, and PSA is no doubt important in the mix, but let's hope the first clue…
Read moreChris O'Neill
Telecommunications Engineer
"Compared to what?"
I'm asking you what consideration of morbidity and quality of life influences your judgement that "I don't see that (48 treatments per life saved) as unacceptable"? What level of morbidity and damage to quality of life would you consider to be unacceptable?
"And not all treatments post PSA screen result in impotence and incontinence, short or long term."
Where did I say that?
You answer very few of my direct questions. One might think you are being evasive. I note that you do not dispute my proposition that only 23 treatments would be required per life saved by pre-emptive treatment compared with 48 treatments per life saved by the current strategy. I also note that you do not dispute that the current strategy works by frightening people into action. What sort of a basis for medical strategy is that? I'm appalled. Is this the way the medical profession works?
Paul Rogers
logged in via Twitter
Well no; I've answered all of your questions in some detail. If you don't like the answers, then so be it.
And as I've agreed here and on the other thread, over-treatment is a problem to address, but it should have nothing to do with PSA screening in best clinical practice.
Chris O'Neill
Telecommunications Engineer
"I've answered all of your questions in some detail."
You didn't provide an ultimate answer to the following. Just evasions. I asked:
"Why aren't we doing the "unthinkable" or why do we keep doing something that performs worse than the "unthinkable"?"
You replied:
"I don't agree that you have presented sufficient evidence that the current strategy (or at least a 'best practice' version) is worse than the "unachievable"."
I provided such evidence that you haven't disputed:
"Current…
Read moreChris O'Neill
Telecommunications Engineer
"For example, a colleague whose father died of aggressive prostate cancer at an early age decided to have a PSA test at around 50, only to be diagnosed with prostate cancer on biopsy with consequent prostatectomy."
It's unfortunate that your father's colleague died from prostate cancer, but it's very unlikely that PSA testing at any time and subsequent treatment would have saved his life. Prostatectomy only saves the lives of men who have a "Goldilocks" cancer: not too aggressive and not too indolent. Too aggressive and it's going to kill you even with the earliest possible detection and treatment. Too indolent and it won't kill you anyway, even if you do nothing. It might help a tiny minority, but for the vast majority of men (at least 98%), the "cures" they are being given are far worse than the disease.
Chris O'Neill
Telecommunications Engineer
Professor Frydenberg:: "Does the public want a faceless committee telling them they can’t have PSA-based testing?"
I don't think that is the PSA antagonists' position. I think their position is more-or-less that PSA testing is not significantly more valuable than random selection for Prostatectomy.