MRI as a means of screening for prostate cancer? Probably not!


A report on the HealthDay web site related to a presentation about the role of MRI scanning in the diagnosis of localized prostate cancer seems to demonstrate a significant division of views about the value of standardized use of mpMRI in the diagnosis of prostate cancer.

As explained on the HealthDay web site, in a paper presented by Alberts et al. at the annual meeting of the European Association of Urology (EAU) in London, it was again suggested that the use of mpMRI scanning prior to initiation of a biopsy “might greatly reduce overdiagnosis and overtreatment of prostate cancer in older men.”

The Dutch research team had enrolled a total of 335 men, all aged 71 and older, who had elevated serum PSA levels after a standard blood draw. The investigators then divided these patients into two groups:

  • Group A comprised 177 men who were immediately given a six-core, systematic biopsy.
  • Group B comprised 158 men who were first given an MRI scan and then were also given a 12-core, systematic biopsy.
  • Men in Group B also had additional biopsy cores taken from areas of the prostate that looked potentially suspicious for prostate cancer when the MRI scan was examined after the initial systematic 12-core biopsy.

Based on the data that they collected, Alberts et al. claim that:

  • In most cases, the biopsies were able to detect serious (i.e., clinically significant) prostate cancer among the patients in Group A and Group B.
  • 70 percent of the patients in the study would not have needed a biopsy at all if the MRI had been given prior to the biopsies (because there was no sign of suspicious areas for cancer on their MRI scans).
  • While MRI scanning is more expensive than PSA testing, it would save money in the long run (just as mammography screening has “paid off” for women in screening for breast cancer).
  • MRI can reduce over-diagnosis of prostate cancer by 50 percent, and unnecessary biopsies by 70 percent in men over 70.

Now it has to be said that there seem to be some problems with this argument (at least in the eyes of your sitemaster). And we can start with the idea that mass mammography has really “paid off” as a mechanism for screening for breast cancer. Indeed, there are now plenty of people who think that mass mammography has actually been a cause of massive over-diagnosis and over-treatment for breast cancer!

On the other hand, your sitemaster would truly like to believe that a combination of PSA testing and appropriately used MRI scanning could massively reduce the number of unnecessary biopsies carried out every year in  men who are at best at hypothetical risk for clinically significant prostate cancer.

The solution to us seems to be fairly obvious. Men who want to get a PSA test should be able to have one. If their PSA level shows risk for prostate cancer they should be able to make a choice:

  • Come back and get another PSA test in 3 or 6 months time to see if anything has changed.
  • Get an immediate 12-core systematic biopsy of the prostate
  • Get an immediate mpMRI and then, after the results of that mpMRI are available, make the decision about having an MRI/TRUS-guided prostate biopsy (comprising a systematic 12-core biopsy and additional biopsies of suspicious areas of the prostate)

Asked for his opinion on Dr. Alberts’ presentation, HealthDay reports that Dr. Anthony D’Amico was clearly unimpressed:

There is not enough data to say MRI is a home run, and there is not enough data to say it is cost-effective,

Dr. Anthony D’Amico is quoted as saying.

Data from other institutions shows that MRI finds only 80 percent of severe cancers and misses 50 percent of the other high-grade cancers. So having a negative MRI doesn’t mean that you don’t have aggressive prostate cancer.

We understand that a larger trial has now been started in Europe, which will randomly assign up to 40,000 men to either MRI screening at various PSA levels or to no screening. But the value of this trial is going to be in the details (which we do not know as yet).

At the other end of the scale, HealthDay reports Dr. D’Amico’s opinion that

the only way to know for sure if MRI effectively screens for prostate cancer is to scan thousands of patients and remove their prostates to analyze the type of cancer.

But this would guarantee that thousands of men would be unnecessarily treated surgically for a prostate cancer that they may not even have, so that’s probably not even an ethical strategy any more.

What it seems we need here is a modicum of collaboration between those who seriously believe that the first test necessary for many men at risk for truly localized prostate cancer (i.e., men with a PSA of ≥ 3 and < 50 ng/ml and no other indication of risk for prostate cancer that is not confined to the prostate) and those who believe that all such men should immediately have a biopsy so that we minimize the risk of “missing” any case of prostate cancer.

If  we randomized 40,000 such men into four groups (see below), then in seems to your sitemaster that we might be able to solve the problem. The three groups would be the following:

  • A group of ~ 13,333 men who would be given an immediate 12-core biopsy
  • A group of ~26,667 men, all of whom would be given an immediate mpMRI followed by
    • An immediate 12-core biopsy with additional biopsy cores of suspicious areas (in all men with suspicious areas were visible on their MRI) or
    • An immediate 12-core biopsy (in half of all those who had no suspicious area on MRI) or
    • No immediate biopsy (in the other half of all those who had no suspicious area on MRI)

The real question is not whether PSA screening can be replaced by MRI screening. That’s a horrible idea. The real question is whether we can use MRI testing to cut the number of unnecessary biopsies (because you can always get another PSA test and another MRI in a couple of years’ time as a form of active risk surveillance).

We need to get a degree of realism here. Even today, we know that we cannot find every single case of prostate cancer (let alone clinically significant prostate cancer) on the basis of an initial PSA test and a biopsy. We also won’t be able to find every case of prostate cancer (or clinically significant prostate cancer) on the basis of a PSA test, and MRI, and a biopsy. The question is how do we get better at finding every case of clinically significant prostate cancer that we can while minimizing the risks associated with the over-treatment of forms of prostate cancer that don’t need treatment.

One Response

  1. The idea of the patient making choices between repeat PSA tests, biopsy or mpMRI is difficult. What have they to go on?

    A better idea is that the medical profession makes up its mind about the value of mpMRIs!

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: