Originally published 11/1/2018
It’s astonishing to think of a huge contribution to 3T multiparametric MRI (mpMRI) in the brief three years since we posted the blog below. Talk about a “noteworthy update”! The integration of Artificial Intelligence (AI) into radiology is a huge game changer, and we now devote an entire section of our website to this topic. We have numerous blogs on the application of AI, so hopefully you’ll spend some time exploring our posts.
We want to draw particular attention to a branch of AI called machine learning. When software is specifically “trained” to recognize imaging features, it can “… provide better standardization and consistency in identifying prostate lesions and enhance prostate carcinoma management.”[i] One of the benefits is increased efficiency for busy radiologists who interpret MRI results and write reports. More importantly, however, is the improved accuracy in characterizing elements as prostate cancer that escape perception by the human eye.
From a patient’s viewpoint, pinpointing suspicious lesions that have all the earmarks of cancer means avoiding overdiagnosis and overtreatment of nonaggressive tumors that are amenable to observation. This gives patients peace of mind while preserving their quality of life. If repeat MRI is part of their surveillance, and if it incorporates the same software, any change will be picked up by AI tools. Thus, in addition to the important updates we reported below, the latest news takes mpMRI to a whole new level. At our Center, we are proud to have been early adopters of AI in our imaging services.
I’ve been following current studies and trends on the use of 3Tesla (3T) multiparametric MRI of the prostate, or simply 3T mpMRI. In this context, the term Tesla or T refers to the strength of the magnet. The higher the T, the better the imaging quality—and thankfully, 3T magnets are gradually replacing the 1.5T magnets that were most commonly available.
One of the key principles in prostate cancer (PCa) therapy is the importance of matching treatment to disease. Before mpMRI came along, doctors based treatment recommendations on known clinical factors (age, PSA, Gleason grade, tumor stage). With the advent of visual information from MRI, doctors now had the added ingredient of the location, size, shape and clues as to the aggression of the tumor – a true game-changer when it comes to decision-making!
3 studies provide noteworthy updates
I want to share three newly published studies that have to do with the predictive value of 3T mpMRI before treatment. as examples of the urologic and radiologic recognition that 3T mpMRI is finally gaining to detect PCa before treatment.
- “Pretreatment Multiparametric MRI is Independently Associated with Biochemical Outcome in Men Treated with Radiation Therapy for Prostate Cancer,” is a perfect illustration of how mpMRI before radiation correlates with treatment results. In this paper by Kauffmann et al.[ii], 123 patients with intermediate or high risk PCa underwent radiation. Each had a pre-treatment 3T mpMRI. The scans were analyzed for adverse features (tumor extension beyond the capsule, seminal vesicle invasion, and tumor size > 15mm). The patients were followed for an average of 50 months and tracked for freedom from biochemical failure (FFBF), meaning no rise in PSA. As you might expect, there was a strong link between patients who had adverse features seen on mpMRI and biochemical failure. Because of this, the authors note that “… mpMRI may aid risk stratification beyond clinical risk factors in men treated with radiation therapy…” with the implication that advance knowledge from imaging can assist with individual treatment plans.
- “Which Scores Need a Core? An Evaluation of MR-targeted Biopsy Yield by PIRADS Score Across Different Biopsy Indications,” adds to a growing body of evidence that 3T mpMRI can determine if a needle biopsy is necessary. The study by Sathianathen et al.[iii] involved 225 PCa patients about to have a TRUS biopsy. They were grouped according to their clinical history as either never having had a biopsy, had a previous negative TRUS biopsy, or were on Active Surveillance (AS). All patients had pre-biopsy mpMRI scans which were scored using the PIRADS system. (PIRADS classifies suspicious mpMRI areas on a score of 1-5, where 1 is least suspicious and 5 is greatly suspicious for PCa.) The purpose of the study was to correlate PIRADS scores with actual biopsy findings. According to one news story, “Clinically significant cancers were found in 0% of men with PIRADS scores between 1 and 2, in 8.9% of those with a PIRADS score of 3, in 21.4% of those with a PIRADS score of 4 and in 62.7% of men with a PIRADS score of 5.” The study authors concluded that PCa detection rates are “significantly associated with PIRADS score,” and suggested that men with PIRADS 3 (or below) who have a previous negative biopsy might be able to avoid a rebiopsy at that time.
- “Added Value of Multiparametric Magnetic Resonance Imaging to Clinical Nomograms in Predicting Adverse Pathology in Prostate Cancer” is a study out of the National Institutes of Health[iv], and among its co-authors are Drs. Peter Pinto and Baris Turkbey, both of whom have a strong interest in 3T mpMRI of the prostate. The title raises the question, What’s a clinical nomogram? In my blog on the Partin tables and other nomograms, I wrote that a nomogram is a set of scales that can be used to calculate an unknown value, and when adapted for medicine, they act as a statistical modeling tool. With regard to PCa, it’s based on combined risk factors compared with a global population, and it helps making best-match treatment decisions. While it’s not 100% predictive, it’s better than an educated guess. The “Added Value” study involved over 500 patients who had 3T mpMRI before undergoing radical prostatectomy, so the surgically removed glands were available for physical comparison with pre-operative clinical factors and imaging results. The authors found that the mpMRI scans, in combination with the Partin tables and the Memorial Sloan Kettering nomogram, provided “significant additional predictive ability of adverse pathology at the time of radical prostatectomy. This information can be greatly beneficial to urologists for preoperative planning and for counseling patients regarding the risks of future therapy.”
These three 2018 published papers are samples of the affirmation that 3T mpMRI has gained, and that continues to grow. It has undeniable merit for avoiding unnecessary biopsies and matching treatment with disease. In addition, planning the treatment itself is made more accurate thanks to the detailed portrait the scans offer. In short, 3T mpMRI shows us how to fulfill the wisdom of the ages-old adage that reminds us, “Look before you leap.”
NOTE: This content is solely for purposes of information and does not substitute for diagnostic or medical advice. Talk to your doctor if you are experiencing pelvic pain, or have any other health concerns or questions of a personal medical nature.
[i] Bardis M, Houshyar R, Chang P, Ushinsky A et al. Applications of Artificial Intelligence to Prostate Multiparametric MRI (mpMRI): Current and Emerging Trends. Cancers (Basel). 2020 May; 12(5): 1204.
[ii] Kauffmann G, Arif F, Patel P, Oto A, Liauw S. Pretreatment multiparametric MRI is independently associated with biochemical outcome in men treated with radiation therapy for prostate cancer. Send to
Urol Oncol. 2018 Aug 16. pii: S1078-1439(18)30252-7. doi: 10.1016/j.urolonc.2018.07.004. [Epub ahead of print]
[iii] Sathianathen NJ, Konety BR, Soubra A, Metzger GJ et al. Which scores need a core? An evaluation of MR-targeted biopsy yield by PIRADS score across different biopsy indications. Prostate Cancer Prostatic Dis. 2018 Jul 23. doi: 10.1038/s41391-018-0065-6. [Epub ahead of print]
[iv] Rayn KN, Bloom JB, Gold SA, Hale GR et al. Added Value of Multiparametric Magnetic Resonance Imaging to Clinical Nomograms in Predicting Adverse Pathology in Prostate Cancer. J Urol. 2018 May 29. pii: S0022-5347(18)43271-5. doi: 10.1016/j.juro.2018.05.094. [Epub ahead of print]