RESUMEN
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.
Asunto(s)
Proteína BRCA1/genética , Proteína BRCA2/genética , Síndrome de Cáncer de Mama y Ovario Hereditario/diagnóstico por imagen , Interpretación de Imagen Radiográfica Asistida por Computador/métodos , Neoplasias de la Mama Triple Negativas/diagnóstico por imagen , Adulto , Estudios de Casos y Controles , Femenino , Variación Genética , Síndrome de Cáncer de Mama y Ovario Hereditario/genética , Humanos , Imagen por Resonancia Magnética , Persona de Mediana Edad , Proyectos Piloto , Valor Predictivo de las Pruebas , Análisis de Regresión , Medición de Riesgo , Neoplasias de la Mama Triple Negativas/genéticaRESUMEN
BACKGROUND: Adequate patient selection is the key to successful lung volume reduction in patients with pulmonary emphysema. Computed tomography (CT) enables a reliable detection of pulmonary emphysema and allows an accurate quantification of the severity. Our goal was to investigate the usefulness and reliability of color-coded (CC) CT images in classification of emphysema and preoperative lung volume reduction planning. METHODS: Fifty patients undergoing lung volume reduction surgery at our institution between September 2015 and February 2016 were retrospectively investigated. Three readers visually assessed the amount and distribution patterns of pulmonary emphysema on axial, multi-planar and CC CT images using the Goddard scoring system and a surgically oriented grading system (bilateral markedly heterogenous, bilateral intermediately heterogenous, bilateral homogenous and unilateral heterogenous emphysema). Observer dependency was investigated by using Fleiss' kappa (κ) and the intraclass correlation coefficient (ICC). Results were compared to quantitative results from densitometry measurements and lung perfusion scintigraphy by using Spearman correlation. Recommendations for lung volume reduction sites based on emphysema amount and distribution of all readers were compared to removal sites from the surgical reports. RESULTS: Inter-rater agreement for emphysema distribution rating was substantial for CC images (κ=0.70; 95% CI, 0.64-0.80) and significantly better compared to axial and multiplanar images (P≤0.001). The inter-rater agreement for recommended segment removal was moderate for CC images (κ=0.56; 95% CI, 0.49-0.63) and significantly better compared to axial and multiplanar images (P<0.001). Visual emphysema rating correlated significantly with measurements from densitometry and perfusion scintigraphy in the upper and lower lung zones in all image types. CONCLUSIONS: CC CT images allow a precise, less observer-dependent evaluation of distribution of pulmonary emphysema and resection recommendation compared to axial and multiplanar CT images and might therefore be useful in lung volume resection surgery planning.