Prostate screening is a complicated mix of opinions, evidence, emotions, and controversy. There are many stakeholders interested in appropriate prostate screening and it is a question we as Australian GPs get asked regularly about.
So what do you do? What should you do? And what do you use to help make the decision with your patient?
Question 1: What is prostate cancer?
Here is a great little article on types of prostate pathology, their anatomical location, and some colourful histology photos – Webpath
Here is a really good and 3D description of the prostate, its function and anatomy – Youtube
Question 2: How common is prostate cancer and what is its mortality rate in Australia?
All statistics are from the Prostate Cancer Foundation of Australia website – Link
It is the second most common cause of cancer death in Australian men (after lung cancer). It is estimated that 267,000 Australian men will be living with prostate cancer by 2017. It has a 5 year survival after diagnosis of 92% in 2014 (59% in 1986!). Men at the age of 75 have a 1 in 8 risk OF BEING DIAGNOSED. By the age of 85, the number increases to 1 in 6. (The Malmo autopsy study showed the prevalence of prostate cancer in men in their 80’s was 44%!).
Question 3: So what is “screening?”
This 10 key points is from the WHO based upon “Wilson’s Criteria” upon which we still plan our screening tests. Let’s compare this for prostate screening.
- The condition should be an important health problem. – Tick
- There should be a treatment for the condition – Tickish – arguably can cause harm and no one best form of treatment
- Facilities for diagnosis and treatment should be available – Tick
- There should be a latent stage of the disease – Tick
- There should be a test or examination for the condition – Tick’ish – Discussed below
- The test should be acceptable to the population – Tick’ish – No one likes a DRE….
- The natural history of the disease should be adequately understood – Tick’ish – Which ones cause problems? Difficult to define which ones will be clinically significant.
- There should be an agreed policy on whom to treat – Hmmm. This one is unclear.
- The total cost of finding a case should be economically balanced in relation to medical expenditure as a whole – Arguable
- Case-finding should be a continuous process, not just a “once and for all” project – This would be the aim
So does prostate screening fit all these criteria? Do you think differently to me about this?
Question 4: “Isn’t prostate screening just straight forward? We have a test for it right?”
We actually have two tests available.
Firstly lets discuss the Digital Rectal Exam (DRE).
Most doctors by the time they get to GP have done many DREs (usually on surgical terms i.e. bowel obstruction etc.) Actual formal learning on performing a DRE to evaluate the prostate can be a bit sporadic and specifically what are we looking for when doing one to evaluate for prostate cancer? – Here is a link to a video on performing an appropriate DRE and some information on what to look for – Youtube Link – Advance to 9mins 42secs. [Warning – this content is explicit and may offend some viewers – Health professional information only]
(By the way – I found several videos on Youtube explaining to patients how to do self examinations of their prostate. I would suggest this is not a good thing to do. Ever. See a medical professional.)
Question 5: How good at we at performing DREs for prostate signs? (Summarised from UpToDate)
A) DRE can help us detect tumors in the posterior and lateral portions of the prostate gland. Unfortunately, only 85% of prostate cancers arise where they can be detected by a finger.
B) NO controlled studies have shown reduction of morbidity OR mortality from prostate cancer when detected by DRE ALONE at ANY age (Krahn et al. 1994) – Link. This is thought to be because when they are detected by DRE alone they are clinically advanced. NB: When used with PSA it IS useful. This is discussed further below.
C) We all have different sized fingers and it is hard to compare findings when training. Subsequently, interrater errors are common in Urologists. It has never been compared in primary care (I would assume possibly worse but have no evidence of this).
D) 2-3% of all men >50yo will have an abnormality on examination. Positive predictive value (positive test actually being cancer) is about 28% in a meta-analysis (link below).
E) Sensitivity and specificity — A meta-analysis of DRE – sensitivity for detecting prostate cancer of 59 percent / specificity of 94 percent (Hoogendam, Buntinx & de Vet (1999)) – Link
F) Interestingly! In one study, up to 25 percent of prostate cancers detected with biopsy after abnormal DRE findings were found in a different area than the palpable abnormality; in other words, the prostate cancer was not directly related to the DRE findings! (JAMA article 1997 – Link)
Question 6: What are general patient views on getting a DRE done?
Couldn’t find any evidence on this!? I would probably suggest that the patient would rather be doing something else. I have no evidence to back this up however…
There are great guidelines on this from the “General Medical Council” in England – Link. Interesting information on what to do if a patient refuses a chaperone also.
The Prostate Specific Antigen is a glycoprotein produced only by the prostate. Its usual role is believed to be in liquefying semen. Subsequently, some can leak into the blood supply and we measure this. Many things can increase its systemic level and these are discussed below. This is discussed really well in this summary article from Australian Family Physician (AFP) – Link
Question 9: What is a ‘free PSA’ level?
PSA in the serum is primarily protein bound and a ‘free PSA’ is the unbound fraction. The theory is that in the patients with a PSA between 4-10 ng/ml it may help differentiate benign causes from prostate cancer (i.e. if patients have a DECREASED proportion of free PSA then they have increased risk of prostate ca. The theory is ‘showers’ of PSA from benign causes will be bound and unbound. In chronic release of PSA bound PSA will accumulate due to its longer half-life compared to free PSA). I could not find any specific guidelines that explain its use. Urologists may use this test more effectively than in primary care (I am probably sending these levels to a urologist anyway). Urologists use values from research including: Free fraction <10% – 56% chance of cancer, Free fraction >25%, 8% chance of cancer, to help guide who they should biopsy.
Australian Prescriber has a great article on this available here – Link
Basically the use of free PSA is ONLY useful in the group between 4-10ng/mL to risk stratify. There is no indication to order it with PSA in the normal range.
You may also hear about cPSA (Complexed PSA). It is another tool to help differentiate but not in wide use. I did not find much evidence yet either (watch this space).
Question 10: So what causes the PSA to go up or down?
Elevated PSA has been associated with prostate cancer. It has also been associated with:
Benign prostatic hypertrophy (BPH), prostatitis (returns to normal over 6-8 weeks), prostatic infarction (sounds unpleasant), postaggressive DRE (also sounds unpleasant!), bicycle riding, and post ejaculation (though no recommendations that patients abstain before test), post biopsy, and patients in acute urinary retention.
Decreased levels of PSA can be associated with 5-alpha-reductase inhibitors (eg finasteride, dutasteride). Some guidelines suggest if a patient has been on these medications for more than 6 months then double the PSA value and use that to determine management.
Question 11: What values do we worry about?
Firstly, PSA is a continuous variable. There is no ‘normal’ level below which prostate cancer can be excluded. Certain factors can be useful to help is to improve the positive predictive value of PSA.
There seems to be two main factors to this: overall value compared to age adjusted ranges, free fraction, and the PSA velocity (For PSA <4 velocity >0.35ng/ml/yr abnormal, PSA >4 velocity >0.75ng/ml/yr abnormal).
For the overall value, the PSA is expected to rise with age, therefore, we have age related reference ranges. Different laboratories will have different reference ranges.
For PSA velocity, Australian Prescriber suggests that BPH causes of elevated PSA lead to a doubling every five years in the early stages. If the PSA concentration is doubling faster (for example every three years or faster), this is more likely to be due to prostate cancer. (Australian Prescriber – Link)
Most guidelines seem to agree that if the patient is above the age expected norms of PSA level then they should be referred to a urologist for discussion of possible further investigation (discussed below).
Question 11a: When to start testing, how often and when to stop?
Baseline testing in a man’s 40’s is often mentioned. Its suggestion is sporadic and not endorsed by the majority of guidelines. Evidence is not generally strong. Here is a prospective trial showing benefits of risk stratification in 40’s but only of 270 men – Link. There is some suggestion that using this early age check can help risk stratify higher risk patients – but this is also unclear. Another commonly quoted Swedish trial that did 3 separate PSA readings over 15 yrs starting at 40 and risk stratified patients based upon this – Link
Most guidelines suggest that if the PSA is below the median value for their age, then they probably only need testing every 5 years. If above the median, yearly or second yearly is most likely the correct time interval.
There is NO role for screening in men with less than 10-15yrs life expectancy. Most guidelines suggest stopping at 70-75.
Question 12: Why should we do a PSA AND a DRE?
Because the two tests when combined together have much better statistical strength than either test alone. This trial was on 117,000 men who were screened between 1993-95 using age-specific reference ranges. This was published in the Prostate journal (1999) – Link
|PSA & DRE||92.2%||31.8%||63.7%|
Here is a good youtube video explaining how a prostate biopsy is done – Link
(I do not endorse any specific urologist – this was just the clearest video that I could find).
Question 13: So if my patient has an elevated PSA and I send them to a Urologist, what does the biopsy result mean?
Most guidelines suggest repeating the PSA if a test is elevated to help define false positive readings.
Cancer.org explains really well how to interpret biopsy results and gleason scoring. Written for patients but very detailed – Link
Question 14: We have tests, we know the problem is prevalent, we know it kills Australian men. Sounds like we should test everyone! Why aren’t we doing it?
Because the numbers don’t add up (yet). It is such a prevalent condition, that is slow growing, and unlikely to kill many men who will die of another cause EVEN if they have prostate cancer. Let’s look at the numbers.
The NHMRC Australian guidelines go through the numbers in a nice clear way – Link.
I will summarise it here:
|1000/1000||Australian men aged 60 with no first degree relatives have a PSA test performed|
|2/1000||Avoid death before 85yo because of testing|
|2/1000||Avoid metastatic prostate cancer before 85yo (same 2? Unclear)|
|87/1000||Have a FALSE positive and have a biopsy|
|28/1000||Have a moderate/major problem from the biopsy requiring healthcare (bleeding and infection are two most common)|
|28/1000||Have prostate cancer diagnosed but would have been asymptomatic for life(Interestingly RACGP Red book states that suicide and CVD increase in this group by 8 and 11 times respectively!)|
|25/1000||Undergo treatment who would have done well without it|
|7-10/25||Of the above treatment group will have persistent impotence and/or urinary incontinence and some will have bowel trouble|
The numbers in America are slightly different but they have a great chart which you can laminate to explain the potential pitfalls to patients: Chart from National Cancer Institute – Link
UpDATED – 21/9/14 – Dr Justin Coleman has a great series of videos backed by National Prescribing Service (NPS) describing some of the statistics behind prostate screening. See it here – Link
UpDATED – 24/9/14 – BroomeDOCS (Dr Casey Parker) Medical educator extraordinaire has put up a 5min podcast on the problems with prostate cancer statistics. If you can’t get your head around it (I am still struggling. I hate statistics… too much thinking!) Then check out this great explanation – Link
Please note I am summarising here. If you want more detail go to the relevant colleges/councils.
Screening for prostate cancer is not recommended unless:
1. The man specifically asks for it; and
2. He is fully counselled on the pros and cons
Australian Health Minsters’ Advisory Council 2010 – No
PSA testing is not suitable for population screening, as the harms outweigh the benefits.
PSA testing may assist in the diagnosis of prostate cancer for men with symptoms; however, levels may be misrepresentative of disease presence/absence. The effectiveness of DRE is limited
PSA test is not recommended for a population-based screening program. Individual men may ask their doctors for testing and doctors may recommend it
PSA test and DRE offered to men aged 55–69 after they are informed of risks and benefits.
Consider testing men aged > 40 with a positive family history
Men can be offered a PSA test and a DRE from age 40 as a baseline measure of risk.
Test annually men with PSA levels above the age-related median. Test less frequently (though they don’t state how often – ?every 5-7 yrs) men with PSA levels below the median
Prostate Cancer Foundation of Australia – 2014 – Yes’ish
PCFA retains its current advice which is that men over age 50, or 40 with a family history of prostate cancer, should talk to their doctor about testing for prostate cancer using the PSA test and DRE as part of their annual health check up.
Mass population screening is not warranted based on current evidence.
Informed (or shared) decision making should be supported and testing should not take place in uninformed men, if at all.
Andrology Australia – They also have a really great summary of all the evidence as a brief print out sheet if you are interested in their current position statement – really interesting read – Link
The Melbourne Consensus guidelines 2013 – Yes maybe kind of?
In men aged 50-69 PSA reduces mortality and metastatic disease (but they don’t then encourage population screening? It’s a bit confusing)
Baseline testing in their 40’s can help to guide future risk of prostate cancer
Question 16: Alot of the councils/colleges recommend that patients are high risk should possibly be investigated. Who are these ‘high risk’ patients?
High Risk patients (Information from AIHW Prostate Cancer – Link)
Males with one first degree relative (2.2-2.8 increased risk). If two first degree relatives (3.3 increased risk).
African descent males are at increased risk
History of an STI, high calcium diets, and processed meat consumption have been shown in some studies (sounds very poor though)
Lower Risk patients – ATSI patients (though they are equally likely to die as population average). Asian patients have a lower risk.
A good calculator which you can use to gauge your own patients risk is from the ERSPC Rotterdam trial and is available here – Link
Important piece of evidence from RACGP is that lower urinary tract symptoms (LUTS) do not appear to have an increased risk of prostate cancer. The most common cause of LUTS is benign prostate enlargement. They state that early prostate cancer never has any symptoms. There is no relationship between LUTS and early prostate cancer. While sorting out their prostatomegaly, urologists will often take this opportunity to screen and this is why they do a PSA.
Question 17: So what ages do I start and stop prostate screening if a patient requests it? How often do I test?
If only it was that easy. The large majority of studies have used variable ages and therefore correlation is difficult. Australian Urological Society suggests an initial PSA at 40yo and, if above the age adjusted median, the patient is placed into a high risk category and gets testing yearly/second yearly.
Other guidelines have suggested yearly from 50yo if the patient wants screening (though 55yo has also been used). Some papers have suggested that if the patient has very low PSA screening may be stretched to every 2-3yrs – and perhaps even 5-7yrs.
The consensus seems to be that testing after the age of 75yo is unlikely to identify cancers which will cause significant morbidity or mortality to the patient so it is generally ceased (What if they are having increasing values and you test them at 75 and it is significantly elevated? Like pap smears, do we have a “must be below a certain number for two years before stopping screening? This is unclear).
Also, if the patients life expectancy is less than 10 years there is not a role for any further testing.
Stop Stop Stop!! Ok information overload. What should I do with this patient in front of me?
John is a 50yo M who presents to your clinic. He has seen the posters about getting a health check now he has turned 50 and wants ‘everything’. You discuss with John what he means by ‘everything’ and he states he’s here for that tune-up now he’s turned 50. Also, his mate had prostate cancer so he wanted to get the blood test for that. What would you do?
So based on the guidelines he is interested in screening (I didn’t bring it up as per the RACGP guidelines). So now I discuss with him that screening involves both a DRE and a PSA. He has no family history of prostate cancer so is a normal risk category. I explain what is involved with the DRE and also talk to him about the PSA. I pull out my chart of patients and explain to John the possible risks with testing. I also explain to him that if he has a high PSA I will need to send him to a urologist for the follow up testing and what this might entail.
I perform the DRE on John after gaining consent and offering a chaperone (he jokingly states he “wants no witnesses” – I document this in the notes – his DRE is normal). His PSA comes back at 7ng/ml. I discuss his high level and ask about cycling/sport, recent ejaculation, or symptoms of UTI/prostatitis (he has none of these). I then repeat his test two weeks later with a urine MCS (to rule out UTI). It again is elevated at 7ng/ml. I refer this gentleman to urologist for discussion of biopsy.
Resources you can use:
Prostate.org.au – Great chart which you can laminate and use to discuss the issues with patients – Link
Andrology Australia – Patient handout on the PSA test – Link
Australian Family Physician – Really nice handout from RACGP stance- Link
Dr Mike Evans – Youtube videos on patient education which are GREAT!! – Link
What is coming?
Cancer.org summarises new tests coming soon such as Progensa (PCA3 level test), test for abnormal gene TMPRSS2:ERG, and new imaging including MRI guided biopsies. This will be an evolving space in the next few years. Hopefully we get a more accurate test soon! – Link
Final sticking points to discuss below:
What do you do with the patient who sees your colleague usually and gets their PSA ‘every year’. Do you re-counsel these patients?
What would you suggest to your father?
Thanks very much to those who made it to the bottom. Hopefully you enjoyed this discussion on another ‘grey’ area of medicine and you have enjoyed this ‘Map’ to what the internet has to offer in regards to education and resource materials. Til next time, the gloves are off! Cheers, Rob.
PS. A final note from an exasperated urology colleague of mine – PLEASE do not check a PSA in a man with a known UTI. It WILL be elevated. This has no prognostic significance for prostate cancer in this setting – nor will it differentiate cystitis from prostatitis – this is a clinical diagnosis (history and prostate exam).