Breast MRI texture analysis for prediction of BRCA-associated genetic risk.

BACKGROUND: BRCA1/2 deleterious variants account for most of the hereditary breast and ovarian cancer cases. Prediction models and guidelines for the assessment of genetic risk rely heavily on criteria with high variability such as family cancer history. Here we investigated the efficacy of MRI (magnetic resonance imaging) texture features as a predictor for BRCA mutation status. METHODS: A total of 41 female breast cancer individuals at high genetic risk, sixteen with a BRCA1/2 pathogenic variant and twenty five controls were included. From each MRI 4225 computer-extracted voxels were analyzed. Non-imaging features including clinical, family cancer history variables and triple negative receptor status (TNBC) were complementarily used. Lasso-principal component regression (L-PCR) analysis was implemented to compare the predictive performance, assessed as area under the curve (AUC), when imaging features were used, and lasso logistic regression or conventional logistic regression for the remaining analyses. RESULTS: Lasso-selected imaging principal components showed the highest predictive value (AUC 0.86), surpassing family cancer history. Clinical variables comprising age at disease onset and bilateral breast cancer yielded a relatively poor AUC (~ 0.56). Combination of imaging with the non-imaging variables led to an improvement of predictive performance in all analyses, with TNBC along with the imaging components yielding the highest AUC (0.94). Replacing family history variables with imaging components yielded an improvement of classification performance of ~ 4%, suggesting that imaging compensates the predictive information arising from family cancer structure. CONCLUSIONS: The L-PCR model uncovered evidence for the utility of MRI texture features in distinguishing between BRCA1/2 positive and negative high-risk breast cancer individuals, which may suggest value to diagnostic routine. Integration of computer-extracted texture analysis from MRI modalities in prediction models and inclusion criteria might play a role in reducing false positives or missed cases especially when established risk variables such as family history are missing.

High-risk breast cancer surveillance with MRI: 10-year experience from the German consortium for hereditary breast and ovarian cancer.

PURPOSE: To report on 10 years of high-risk service screening with annual MRI in the German Consortium for Hereditary Breast and Ovarian Cancer (GC-HBOC). METHODS: A cohort of 4,573 high-risk, previously unaffected women (954 BRCA1 carriers, 598 BRCA2 carriers, 3021 BRCA1/2 non-carriers) participating in the GC-HBOC surveillance program was prospectively followed. Screening outcomes for 14,142 screening rounds with MRI between 2006 and 2015 were analyzed and stratified by risk group, type of screening round, and age. RESULTS: A total of 221 primary breast cancers (185 invasive, 36 in situ) were diagnosed within 12 months of an annual screening round with MRI. Of all cancers, 84.5% (174/206, 15 unknown) were stage 0 or I. In BRCA1 carriers, 16.9% (10/59, 5 unknown) of all incident cancers (screen-detected and interval cancers combined) and in BRCA2 carriers 12.5% (3/24, 4 unknown) were stage IIA or higher, compared to only 4.8% (2/42, 2 unknown) in high-risk BRCA1/2 non-carriers. Program sensitivity was 89.6% (95% CI 84.9-93.0) with no significant differences in sensitivity between risk groups or by age. Specificity was significantly lower in the first screening round (84.6%, 95% CI 83.6-85.7) than in subsequent screening rounds (91.1%, 95% CI 90.6-91.7), p < 0.001. Cancer detection rates (CDRs) and as a result positive predictive values were strongly dependent on type of screening round, risk group and patient age. CDRs ranged from 43.5‰ (95% CI 29.8-62.9) for the first screening round in BRCA2 carriers to 2.9‰ (95% CI 1.3-6.3) for subsequent screening rounds in high-risk non-carriers in the age group 30 to 39 years. CONCLUSIONS: High-risk screening with MRI was successfully implemented in the GC-HBOC with high sensitivity and specificity. Risk prediction and inclusion criteria in high-risk non-carriers need to be adjusted to improve CDRs and thus screening efficacy in these patients.