This article is more than 5 years old. Information may no longer be current.
MRI-based prediction model may reduce unnecessary biopsies in prostate cancer
Click Here to Manage Email Alerts
Including MRI-derived parameters in a prediction model reduced the number of unnecessary biopsies while still detecting a high rate of clinically significant prostate cancers among men, according to findings published in JAMA Oncology.
“Although prostate biopsy is generally considered safe, there has been an increase in biopsy-related septic complications owing to a rising prevalence of fluoroquinolone-resistant bacterial infections,” Baris Turkbey, MD, associate research physician at the Molecular Imaging Program of the NCI, and colleagues wrote. “Thus, it would be desirable to reduce the biopsy rate in men who ultimately prove to have benign conditions or low-grade disease. ... MRI of the prostate in conjunction with MRI-transrectal ultrasound [TRUS] fusion-guided biopsy could also serve as a biomarker to avoid biopsy in low-risk patients.”
The researchers performed a cohort study of 651 patients who underwent MRI, MRI-TRUS fusion-guided biopsy and 12-core systematic biopsy in one session. Of the patients, 400 from a single institution formed a development cohort used to create a prediction model, and 251 from two institutions formed a validation cohort.
The prediction model included clinical variables and parameters derived from MRIs. Risk for clinically significant prostate cancer served as the main outcome.
Nearly half of patients in the development cohort (n = 193; 48.3%; mean age at biopsy, 64.3 years) had clinically significant prostate cancer — defined as Gleason score of at least 3 + 4 — as well as more than one-third of the validation cohort (n = 96; 38.2%; mean age at biopsy, 64.9 years).
When the researchers applied the model to the validation cohort, the area under the curve rose from 64% to 84% compared with the model used at baseline (P < .001).
Compared with the baseline model, the MRI-derived model had a lower false-positive rate when set at a risk threshold of 20% (46%; 95% CI, 32-66 vs. 92%; 95% CI, 70-100).
A slight decrease occurred in the true-positive rate (89%; 95% CI, 85-96 vs. 99%; 95% CI, 89-100). However, 18 of 100 biopsies would not have been necessary with the MRI-derived model, without missing any cases of clinically significant prostate cancer.
“Our MRI-based risk calculator incorporating prostate volume and Prostate Imaging-Reporting and Data System version 2 score can be used to reduce the number of unnecessary prostate biopsies [among] patients who are unlikely to harbor clinically significant prostate cancer while capturing most of the patients with clinically significant prostate cancer,” the researchers wrote. “The successful validation in two independent external cohorts justifies its use in other external centers for prospective validation.” – by Andy Polhamus
Disclosures: Turkbey reports no relevant financial disclosures. Please see the full study for a list of all other authors’ relevant financial disclosures.
Perspective
Back to Top
This manuscript dovetails with findings from the PROMIS trial, which demonstrated deliverable and considerable clinical benefits to multiparametric MRI imaging prior to prostate biopsy. These data demonstrate that by adding multiparametric MRI to a clinically actionable predictive model, more than 35% of men would have been spared an unnecessary prostate biopsy without significant compromise to the rate of detection of clinically significant prostate cancer.
The need to reduce unnecessary biopsies and to minimize overdiagnosis/overtreatment of low-risk prostate cancer without undertreatment of men who are destined to be harmed by the disease are at the core of the ongoing prostate cancer screening debate. This technology significantly moves the proverbial needle toward higher quality and more personalized care.
In view of these mounting data, it is my hope that insurers like Aetna and Cigna that largely refuse to reimburse patients for multiparametric MRI imaging in the preprostate biopsy setting finally address this major gap in their coverage policies.
Reference:
Ahmed HU, et al. Lancet. 2017;doi:10.1016/S0140-6736(16)32401-1.
Click Here to Manage Email Alerts