Feasibility of abbreviated magnetic resonance imaging (AB-MRI) screening in women with a personal history (PH) of breast cancer

PURPOSE: To investigate the feasibility of abbreviated magnetic resonance imaging (AB-MRI) in women with a personal history (PH) of breast cancer as a screening tool. MATERIALS AND METHODS: We retrospectively reviewed 1880 screening AB-MRIs in 763 women with a PH of breast cancer (median age, 55 years; range, 23–89 years) between October 2015 and October 2016. The total acquisition times of AB-MRI were 8.3 min and 2.8 min with and without T2-weighted imaging, respectively. The tissue diagnosis or one-year follow-up status was used as the reference standard. The characteristics of tumor recurrences detected on AB-MRI screenings were analyzed. The cancer detection rates (CDRs) and additional CDRs for the 1(st) round and overall rounds of AB-MRI screening were calculated. The recall rate, sensitivity, specificity, positive predictive values for recall (PPV1) and biopsy (PPV3) for the 1(st) round of AB-MRI screening were calculated. The diagnostic performance of the combination of mammography and ultrasonography was compared with that of AB-MRI by receiver operating characteristic (ROC) curve analysis. RESULTS: Fifteen of a total of 21 recurrences were detected on the 1(st) round of AB-MRI screening: 93.3% were node-negative T1 tumors (median tumor size, 1.02 cm; range, 0.1–2 cm) or Tis; 66.7% were high-grade tumors; 8 of these 15 were mammographically and ultrasonographically occult. The CDR and additional CDR for the 1(st) round of AB-MRI screening were 0.019 and 0.010 per woman, respectively. The sensitivity, specificity, recall rate, PPV1 and PPV3 for the 1(st) round of AB-MRI screening were 100%, 96.0%, 14.3%, 13.8% and 58.3%, respectively. For detecting secondary cancer, AB-MRI showed a higher sensitivity and PPV than the combination of mammography and ultrasonography (95.2%, 57.1% vs 47.6%, 38.5%). The area under the ROC curve was higher for AB-MRI (0.966; 95% CI: 0.951–0.978) than the combination of mammography and ultrasonography (0.727; 95% CI: 0.694–0.759) (P<0.0001). CONCLUSION: AB-MRI improved cancer detection with a high specificity, sensitivity and PPV in women with a PH of breast cancer. AB-MRI could be a useful screening tool for detecting secondary cancer considering its high diagnostic performance and short examination time.