Are MRI-targeted biopsies “better” than standard 12-core TRUS-guided biopsies?

We often see recommendations that MRI-guided biopsies may be more accurate that transrectal ultrasound (TRUS)-guided biopsies in the diagnosis of prostate cancer. A new systematic review has attempted to assess the current status of data supporting or refuting this recommendation.

Moore et al. have conducted a systematic review of published data through December 31, 2011. Their goal was to use the available evidence to offer responses to two questions:

  • When MRI was applied to biopsy-naive men, what percentage showed abnormalities suggesting the need for biopsy?
  • When an MRI-targeted biopsy was utilized, what percentage of patients had prostate cancer detected?

The authors report that, based on their review of the available data:

  • The total number of cases available for review was 4,222.
  • Only 50 cases specifically compared MRI-targeted and standard transrectal biopsy results in the same patients.
  • Among all biopsy-naive men, 374/599 (62 percent) had abnormalities visible on MRI that suggested biopsy.
  • Among men who underwent an MRI-targeted biopsy, 248/374 (66 percent) had prostate cancer detected.
  • MRI-targeted and standard biopsies both detected the same number of cases of clinically significant prostate cancer.
    • MRI-targeted biopsy detected 236/555 cases (43 percent).
    • Standard biopsy detected 237/555 cases (43 percent).
  • MRI-targeted and standard biopsies also bothe missed the same number of cases of clinically signifciant prostate cancer.
    • MRI-targeted biopsy missed 13/555 cases (2 percent).
    • Standard biopsy missed 12/555 cases (2 percent).
  • MRI-targeted biopsy required fewer biopsy cores than standard biopsy
    • The average (mean) number of cores taken under MRI-targeted biopsy was 3.8.
    • The average (mean) number of cores taken under standard biopsy was 12.
  • MRI-targeted biopsy avoided diagnosis of clinically insignificant cancer in 53/555 patients (10 percent) among the presenting population.

Moore et al. conclude that:

  • MRI-guided biopsy is capable of detectecting clinically significant prostate cancer in about the same number of men as a 12-core standard biopsy.
  • MRI-guided biopsy achieves this level of accuracy with fewer biopsies in fewer patients, and with a reduction in the number of diagnoses of clinically insignificant cancer.

However, they make the clear point that there is considerable variability in the methodologies used in the studies carried out and reported to date, and that what is really needed — to know whether MRI-targeting of prostate  biopsies is now a sound and viable clinical strategy — is a robust, multi-center clinical trial based on a standardized protocol for the application of MRI-targeted biopsy compared to standard 12-core biopsy.

The other unanswered question (which is not addressed at all in this study) is whether transrectal color Doppler ultrasound-guided biopsy is either “better” than “gray-scale” TRUS-guided biopsy and/or comparable to MRI-targeted biopsy.

11 Responses

  1. With the recent “ruling” by the USPSTF and looming Obamacare, there has now been indroduced an entirely new paradigm to be considered in any discussion of prostate cancer: cost. Did you know that the technical component of doing a prostate biopsy, the ultrasound, is very inexpensive? I can do a prostate biopsy in my office or my ambulatory surgery center quickly and cost effectively. MRIs require a special “coil,” would be done at the hospital, and at considerable cost and inconvenience for the patient and the urologist. So if the difference in making the diagnosis is not striking it, is hard to fathom father government paying for MRIs when it plans not to pay for the $65.00 PSA. Which unfortunately is making articles like this one and the advent of Provenge a moot point. Like Will Rogers said, “Be glad we don’t get all the government we pay for.” JM

  2. Is it worthwhile or common practice to use MRI to look for cancer that has spread outside the prostate? This could be useful when deciding which treatment to choose, such as surgery or radiation therapy.

  3. Hidden tumours located at the top and the anterior of the prostate commonly evade traditional diagnostic procedures, including transrectal ultrasound-guided needle biopsy. When I learn of a man considering active surveillance in view of biopsy indicating insignificant prostate cancer development, or when learning of men whose biopsies are unable to identify any development despite a continuously elevating PSA, I provide the following links:

  4. Dear Dr. McHugh:

    With the greatest respect, cost has been a factor in access to health care for at least 5,000 years that I am aware of. Ask any slave in Egypt in 3,000 BC whether he or she had access to the same level of health care as the pharoahs or the priests.

    As yet, we have no sound data to demonstrate with clarity than an MRI-targeted biopsy is actually better (in 99 cases out of 100) than a standard 12-core biopsy. Futhermore, there is no actual evidence that “the government” (by which I assume you mean Medicare) is not going to pay for appropriate PSA tests. What individual states may be willing to do (through Medicaid) is a very different issue and varies vastly from state to state.

  5. Dear Jim:

    The accuracy of MRI scanning to try to identify whether cancer is localized or locally advanced in an individual patient is improving but is certainly far from ideal. An appropriately conducted MRI, reveiwed by a skilled radiologist, can certainly identify significant extension of a tumor that started within the prostate into or through the capsule of the prostate. Conversely, there is no known form of imaging that can identify the fact that a few cancerous cells have escaped from the prostate into a lymph node (or elsewhere) and started the process of metastasis.

    Thus, the answer to your question is, “It all depends” on the degree of disease progression, whether the right type of MRI scan is used, and whether the radiologist is sufficiently skilled and experienced.

    The use of MRI scans to assist in clinical decision-making about the treatment of individual patients is certainly increasing. Whether an MRI scan is always justified and whether most MRI scans actually make any difference (let alone an accurate difference) to the clinical recommendations for treatment in individual cases is a much harder question to answer.

    Bone scans and CT scans are routinely given to many patients who meet the D’Amico criteria for low-risk prostate cancer. There is not a single guideline that endorses such use of bone scans or CT scans, and there is a good deal of literaure actually advising strongly against such behavior. However, many urologists recommend such scans. Why do they do this? Well in some cases it is most certainly for self-protection (legally). In others it may be because their practice group owns the relevant technology and can actually make money. In many cases it may simply be patient pressure.

    It would be just as easy for us to waste a great deal of time, money, and human resources over-using MRI scans. When is an an MRI actually appropriate and necessary in an individual patient? That’s a very hard question to answer today. It would be an easier question to answer if the cost of an MRI scan in America was similar to the cost in somewhere like Japan (i.e., about one tenth of the cost in the USA).

  6. Chuck:

    The approriate use of MRI targeting and other methods to either ensure that (a) a man considering active surveillance really does have a high probability of low-risk disease or that (b) a man with two or more negative biopsies and a rising PSA does not have cancer are quite separate issues from the one being addressed by Moore et al., which is whether MRI-targeted biopsies should be used to help diagnose prostate cancer in previously undiagnosed and biopsy-naive patients.

    As Dr. McHugh has accurately observed, there are major cost implications (in the USA and elesewhere) associated with the use of MRI-targeting of prostate biopsies if this was to be used routinely.

  7. The lesson provided by this study is that the average number of cores by MRI is 4 to yield the same outcome with standard TRUS 12 core. So what was the “clinically significant” definition for the four MRI cores? The 12-core standard takes 12 cores in order to get the clinical standard staging regardless of whether the TRUS targets a suspicious location or simply samples within the pattern. Perhaps the TRUS would yield the same result as the MRI if only suspicious sites were targeted. The message to ponder is whether the current staging is the best option if the standard calls for three times the number of necessary cores. Each core has its own full set of risk factors and as such should be applied sparingly. What am I missing here?

  8. Dear Terence:

    I am not at all sure what you are implying in your comment.

    Clinical staging is not based on biopsy cores; it is based on other factors entirely. Furthermore, one of the problems of gray-scale TRUS is that it is very inaccurate as a method of telling what areas need to be biopsied to find cancer. (This has been proven in carefully conducted clinical trials, and that was part of the reason for the development of the standardized, 12-core, TRUS-guided biopsy.)

    The point that this paper is trying to make is that MRI-guided biopsies may not need to use a “standardized” biopsy process at all because the clinically significant areas that need to be biopsied all appear to be much more likely to be visible under MRI, and can therefore be biopsied selectively. (This is not possible under gray-scale TRUS.)

    I think this hypothesis still needs to be proven in much larger studies, but the hypothesis is not unreasonable. It would, however, eliminate the utility of nomograms like those developed by Kattan and others that have been extremely useful in projecting the outcomes of different types of prostate cancer treatment. And then there is the cost of MRI-guided biopsies, which could rapidly become outrageous.

  9. Reply to Sitemaster

    Thank you for your response. The current staging standard includes: grade, % of core, % of cores, and PSA. So the staging standard (e.g., T1c = 3 + 3, PSA < 10, < 5% tumor) is intrinsically based in part on the core numbers. If the number of cores is un-necessary to determine clinical significance then what is the "clinical standard" that is being applied to the MRI results since you cannot get a Txx staging from MRI directed results? It suggests that some new MRI standard might surplant the current system of staging which lends itself to excessive coring.

  10. Dear Terry:

    I am sorry, but I think we are talking at cross-purposes. The clinical stage as defined by the TNM system is based solely on the apparent presence of the cancer inside or beyond the prostate and nlotg on biopsy data; the grade is based on the Gleason score, which is only available after biopsy. Even if you are referring to an older system known as the AJCC staging system, the amount of cancer in a specific core is not part of any formal staging system that I am aware of, so I have no idea what you are referring to when you state that “The current staging standard includes … % of cores.”

    The number of cores taken and the number of those cores that are postive for cancer is used in nomograms like the Kattan nomograms to determine prognostic significance (along with the patient’s age, PSA level, clinical stage, and Gleason score).

  11. Reply to Sitemaster:

    My mistake was to confuse biopsy core percentage of cancer tissue with TURP tissue percentage of cancer as well as percentage of lobe involvement. The Prostate Cancer Canada pamphlet I have simply refers to “prostate tissue sample” in the T stage definitions, which I assumed to be biopsy tissue not TURP. So now I know “what I missed”. Thanks for taking the interest and providing the appropriate links.


Leave a Reply

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

You are commenting using your 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: