Can Ovarian RADS MRI Identify Ovarian Cancer in Indeterminate Masses?
Posted: Monday, March 9, 2020
Ovarian-Adnexal Reporting Data System Magnetic Resonance Imaging (O-RADS MRI) was accurate in stratifying the risk of malignancy in adnexal masses, with a specificity of 0.91 and a sensitivity of 0.93. According to Andrea G. Rockall, MRCP, FRCR, of the Imperial College London, and colleagues, this diagnostic tool will be evaluated in a trial at 18 hospitals in the UK. These study findings were published in JAMA Network Open.
“Our tool has the potential to help triage patients who are low risk, so they can have less invasive treatment options, as well as identifying high-risk patients, so they can receive treatment at an earlier stage and have a better chance of long-term survival,” said Dr. Rockall in an Imperial College press release.
This study enrolled 1,340 women, who underwent routine pelvic MRI to look for features in cysts that could not be identified during typical ultrasound. MRI scores were prospectively applied by two readers onsite and one reader who was masked to clinical and ultrasonographic data. Readers characterized up to three masses per patient according to a standard lexicon and assigned a score. Masses were given a score from 1 to 5, with scores of 1 to 3 signifying no mass or benign and scores of 4 to 5 signifying high risk for malignancy.
Of the 1,194 women evaluated, 1,130 (94.6%) had a pelvic mass on MRI. There were 203 patients (18%) who had at least one malignant adnexal or nonadnexal pelvic mass. The positive likelihood ratios of a mass being malignant were 0.01 for a score of 2, 0.27 for a score of 3, 4.42 for a score of 4, and 38.81 for a score of 5. Of the women with pelvic masses identified on MRI, 580 (51.3%) underwent surgery. A total of 244 women (67.4%) had benign lesions and a score of up to 3. In addition, the MRI score was able to reclassify the mass origin as nonadnexal with a sensitivity of 0.99 and a specificity of 0.78.
Disclosure: For full author disclosures, visit https://jamanetwork.com/
