MRI/ultrasound fusion prostate biopsy in diagnosis of prostate cancer


A new paper in this month’s Journal of Urology has suggested that biopsy of lesions identified using a 3.0 T MRI scan and targeted for biopsy using an MRI/ultrasound image fusion technique may be able to significantly improve the probability of detection of clinically relevant prostate cancers in an office setting.

Sonn et al. carried out a retrospective assessment of prostate cancer detection rates among 171 men consecutively evaluated for risk of prostate cancer between March 2010 and September 2011. The men all fell into one of two groups:

  • 106/171 patients (62.0 percent)  were undergoing targeted biopsy to evaluate them as appropriate candidates for active surveillance.
  • 65/171 patients (38.0 percent) were undergoing targeted biopsy because they had a persistently elevated PSA level but one or more negative, conventional, systematic biopsies.

All patients were managed according to the following protocol:

  • Prior to biopsy they were given a multiparametric, 3.0 T MRI.
  • Lesions evident on this MRI were outlined in three dimensions and classified by a specialized uroradiologist according to a scale indicating increasing risk for cancer (from image grade 1 to image grade 5).
  • The stored MRI data were then fused with real-time transrectal ultrasound data to generate a 3D model of the prostate, and
  • Working from this 3D model each patient received both a targeted biopsy of the lesions identified on MRI and a standard 12-core systematic biopsy while under local anesthesia.

Here are the resulting data:

  • The patients’ average (median) PSA level was 4.9 ng/ml at time of biopsy.
  • Their median prostate volume was 48 cm3.
  • On average, the targeted biopsy was 3.0 times more likely to identify prostate cancer than the systematic biopsy (21% vs 7%).
  • Prostate cancer was found in 90/171 patients (52.6 percent) — implying that 81 patients were biopsy negative on both types of biopsy (although only 65 patients were biopsy negative at the start of the study).
  • 34/90 patients in whom prostate cancer was identified (37.8 percent) had a Gleason score ≥ 7 — implying that 56/90 men had a Gleason score of 6.
  • In men with at least one prior negative biopsy, the rate of cancer diagnosis was 37 percent.
  • In men being evaluated for or while on active surveillance, the rate of positive cancer diagnosis was 63 percent.
  • 13/34 patients with a Gleason score ≥ 7 (38.2 percent) had disease detected only on a targeted biopsy.
  • 16/17 patients (93.8 percent) with a targeted image grade of 5 (highest suspicion) on MRI had prostate cancer on biopsy, including 7 patients with  a Gleason score of ≥ 7.

Sonn et al. express the opinion that this study has three key findings:

  • That they were able to accurately target and biopsy lesions observed on MRI using MRI/ultrasound fusion in an office setting.
  • That addition of targeted biopsies to systematic biopsies increased the rate of diagnosis of all cancers and the rate of cancers with a Gleason score of ≥ 7.
  • That the level of suspicion of lesions identified on MRI correlated with actual cancer diagnosis overall and with diagnosis of cancers with a Gleason score of ≥ 7.

It is clear that this research continues to validate the potential of MRI/ultrasound fusion-guided biopsy for the diagnosis of prostate cancer. It is also interesting to note the authors’ comment in the discussion that, “Given the low risk population in our study …, relatively few patients subsequently underwent radical prostatectomy.” Clearly, this statement indicates that an increasingly substantial percentage of patients at some centers is now being managed over time with active surveillance. However, we should be cautious about the over-use of MRI as part of the first-line diagnosis of prostate cancer. In this study, MRI was being used exclusively in the management of candidates for active surveillance, men already on active surveillance, and men with a history of negative biopsies despite a clearly elevated PSA level.

6 Responses

  1. The key point to me is: “On average, the targeted biopsy was 3.0 times more likely to identify prostate cancer than the systematic biopsy (21% vs 7%).”
    In other words, systemic (untargeted) biopsies are not very good, with a high rate of false negatives!

  2. George:

    We have known that systematic biopsies “aren’t very good” for 20+ years. What we don’t know yet is whether this type of fusion biopsy is reproducibly any better. The ability to carry out this type of biopsy is highly dependent on the skills of uroradiologists to read the MRI scan data with accuracy. Good uroradiologists aren’t exactly falling off trees, and require years of training.

  3. You know know this! But the “Best Practices for Urology” doesn’t. They still recommend doing the systematic biopsies in their office, which makes it litigation proof but still wrong. It does not take years of training but a desire to improve their practice. The targets are visible, they don’t think it’s necessary and will not ask for the MRI test outside their office.
    George

  4. Dear George:

    You need to appreciate that there is a giant step between the use of things like MRI/ultrasound fusion in clinical research studies at academic medical centers and the practical application of such technology in the office of the average urologist.

    The fact that something can be done by a sophisticated research team doesn’t always imply that everyone can do it a few months or even years later. From a purely practical point of view, I am dubious that there are anything like enough skilled uroradiologists in America to be able to accurately read the all required MRI scans. I also suspect that most urologists wouldn’t have the skill to plan the MRI/ultrasound fusion biopsy.

    MRI/ultrasound fusion is a complex and sophisticated technique. If I wanted one tomorrow, I know of few centers with the technology, the staff, and the equipment to carry it out with any real degress of confidence. Just because most of us need a car in America doesn’t imply that we all need a new model Lamborghini a couple of months after the first one comes off the production line.

    Finally, I would note that I was very careful to write that this technique “may” be able to significantly improve the probability of detection of prostate cancer in appropriately selected patients. We need more data before this can be stated to be an established fact.

  5. Will there be a greater number of “insignificant” cancers detected with this technique?

  6. Sam:

    I am not sure that we know the answer to this question yet. However, perhaps the more important question is what people do about clinically insignificant prostate cancer when it is identified. If this test help men to have confidence in active surveillance as opposed to invasive treatment, then perhaps that is a good thing.

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 )

Twitter picture

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

Facebook photo

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

Google+ photo

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

Connecting to %s

Follow

Get every new post delivered to your Inbox.

Join 1,156 other followers