Accuracy of contrast-enhanced spectral mammography for estimating residual tumor size after neoadjuvant chemotherapy in patients with breast cancer: a feasibility study.

OBJECTIVE: To assess the feasibility of contrast-enhanced spectral mammography (CESM) of the breast for assessing the size of residual tumors after neoadjuvant chemotherapy (NAC). MATERIALS AND METHODS: In breast cancer patients who underwent NAC between 2011 and 2013, we evaluated residual tumor measurements obtained with CESM and full-field digital mammography (FFDM). We determined the concordance between the methods, as well as their level of agreement with the pathology. Three radiologists analyzed eight CESM and FFDM measurements separately, considering the size of the residual tumor at its largest diameter and correlating it with that determined in the pathological analysis. Interobserver agreement was also evaluated. RESULTS: The sensitivity, specificity, positive predictive value, and negative predictive value were higher for CESM than for FFDM (83.33%, 100%, 100%, and 66% vs. 50%, 50%, 50%, and 25%, respectively). The CESM measurements showed a strong, consistent correlation with the pathological findings (correlation coefficient = 0.76-0.92; intraclass correlation coefficient = 0.692-0.886). The correlation between the FFDM measurements and the pathological findings was not statistically significant, with questionable consistency (intraclass correlation coefficient = 0.488-0.598). Agreement with the pathological findings was narrower for CESM measurements than for FFDM measurements. Interobserver agreement was higher for CESM than for FFDM (0.94 vs. 0.88). CONCLUSION: CESM is a feasible means of evaluating residual tumor size after NAC, showing a good correlation and good agreement with pathological findings. For CESM measurements, the interobserver agreement was excellent.

OBJETIVO: Avaliar a viabilidade da utilização da mamografia espectral com meio de contraste (CESM) na avaliação do tumor residual em mulheres com câncer de mama submetidas a quimioterapia neoadjuvante. MATERIAIS E MÉTODOS: Foi avaliada a concordância entre a mensuração do tumor residual na CESM e na mamografia digital (FFDM) com os dados histopatológicos de mulheres submetidas a quimioterapia neoadjuvante entre 2011 e 2013. Após as exclusões, três radiologistas analisaram oito CESMs e FFDMs separadamente. A maior dimensão do tumor residual foi considerada para comparação com os resultados histopatológicos. Concordância e correlação da CESM e FFDM com resultados histopatológicos e a concordância interobservador foram avaliadas. RESULTADOS: A CESM teve sensibilidade, especificidade e valores preditivos positivos e negativos maiores que a FFDM - 83,33%, 100%, 100% e 66% versus 50%, 50%, 50% e 25%, respectivamente. A CESM teve correlação boa e consistente com os achados histopatológicos (coeficiente de correlação = 0,76-0,92; coeficiente de correlação intraclasse = 0,692-0,886). A correlação entre FFDM e os achados histopatológicos não foi estatisticamente significante, com consistência questionável (coeficiente de correlação intraclasse = 0,488-0,598). A concordância entre as dimensões do estudo histopatológico foi mais estreita com a CESM do que com a FFDM. A concordância interobservador foi maior na CESM (0,94) do que na FFDM (0,88). CONCLUSÃO: A CESM é viável e pode ser utilizada para avaliação de tumor residual após quimioterapia neoadjuvante. A CESM tem boa correlação e concordância com o estudo histopatológico e excelente concordância interobservador.

Accuracy of contrast-enhanced spectral mammography for estimating residual tumor size after neoadjuvant chemotherapy in patients with breast cancer: a feasibility study / Acurácia da mamografia espectral com contraste para seguimento de tumor residual pós-quimioterapia neoadjuvante em pacientes com câncer de mama: um estudo de viabilidade

Abstract Objective: To assess the feasibility of contrast-enhanced spectral mammography (CESM) of the breast for assessing the size of residual tumors after neoadjuvant chemotherapy (NAC). Materials and methods: In breast cancer patients who underwent NAC between 2011 and 2013, we evaluated residual tumor measurements obtained with CESM and full-field digital mammography (FFDM). We determined the concordance between the methods, as well as their level of agreement with the pathology. Three radiologists analyzed eight CESM and FFDM measurements separately, considering the size of the residual tumor at its largest diameter and correlating it with that determined in the pathological analysis. Interobserver agreement was also evaluated. Results: The sensitivity, specificity, positive predictive value, and negative predictive value were higher for CESM than for FFDM (83.33%, 100%, 100%, and 66% vs. 50%, 50%, 50%, and 25%, respectively). The CESM measurements showed a strong, consistent correlation with the pathological findings (correlation coefficient = 0.76-0.92; intraclass correlation coefficient = 0.692-0.886). The correlation between the FFDM measurements and the pathological findings was not statistically significant, with questionable consistency (intraclass correlation coefficient = 0.488-0.598). Agreement with the pathological findings was narrower for CESM measurements than for FFDM measurements. Interobserver agreement was higher for CESM than for FFDM (0.94 vs. 0.88). Conclusion: CESM is a feasible means of evaluating residual tumor size after NAC, showing a good correlation and good agreement with pathological findings. For CESM measurements, the interobserver agreement was excellent.

Resumo Objetivo: Avaliar a viabilidade da utilização da mamografia espectral com meio de contraste (CESM) na avaliação do tumor residual em mulheres com câncer de mama submetidas a quimioterapia neoadjuvante. Materiais e métodos: Foi avaliada a concordância entre a mensuração do tumor residual na CESM e na mamografia digital (FFDM) com os dados histopatológicos de mulheres submetidas a quimioterapia neoadjuvante entre 2011 e 2013. Após as exclusões, três radiologistas analisaram oito CESMs e FFDMs separadamente. A maior dimensão do tumor residual foi considerada para comparação com os resultados histopatológicos. Concordância e correlação da CESM e FFDM com resultados histopatológicos e a concordância interobservador foram avaliadas. Resultados: A CESM teve sensibilidade, especificidade e valores preditivos positivos e negativos maiores que a FFDM - 83,33%, 100%, 100% e 66% versus 50%, 50%, 50% e 25%, respectivamente. A CESM teve correlação boa e consistente com os achados histopatológicos (coeficiente de correlação = 0,76-0,92; coeficiente de correlação intraclasse = 0,692-0,886). A correlação entre FFDM e os achados histopatológicos não foi estatisticamente significante, com consistência questionável (coeficiente de correlação intraclasse = 0,488-0,598). A concordância entre as dimensões do estudo histopatológico foi mais estreita com a CESM do que com a FFDM. A concordância interobservador foi maior na CESM (0,94) do que na FFDM (0,88). Conclusão: A CESM é viável e pode ser utilizada para avaliação de tumor residual após quimioterapia neoadjuvante. A CESM tem boa correlação e concordância com o estudo histopatológico e excelente concordância interobservador.

Low energy mammogram obtained in contrast-enhanced digital mammography (CEDM) is comparable to routine full-field digital mammography (FFDM).

PURPOSE: Contrast enhanced digital mammography (CEDM) uses low energy and high energy exposures to produce a subtracted contrast image. It is currently performed with a standard full-field digital mammogram (FFDM). The purpose is to determine if the low energy image performed after intravenous iodine injection can replace the standard FFDM. METHODS: And Materials: In an IRB approved HIPAA compatible study, low-energy CEDM images of 170 breasts in 88 women (ages 26-75; mean 50.3) undergoing evaluation for elevated risk or newly diagnosed breast cancer were compared to standard digital mammograms performed within 6 months. Technical parameters including posterior nipple line (PNL) distance, compression thickness, and compression force on the MLO projection were compared. Mammographic findings were compared qualitatively and quantitatively. Mixed linear regression using generalized estimating equation (GEE) method was performed. Intraclass correlation coefficients (ICC) with 95% confidence interval (95%CI) were estimated to assess agreement. RESULTS: No statistical difference was found in the technical parameters compression thickness, PNL distance, compression force (p-values: 0.767, 0.947, 0.089). No difference was found in the measured size of mammographic findings (p-values 0.982-0.988). Grouped calcifications had a mean size/extent of 2.1cm (SD 0.6) in the low-energy contrast images, and a mean size/extent of 2.2 cm (SD 0.6) in the standard digital mammogram images. Masses had a mean size of 1.8 cm (SD 0.2) in both groups. Calcifications were equally visible on both CEDM and FFDM. CONCLUSION: Low energy CEDM images are equivalent to standard FFDM despite the presence of intravenous iodinated contrast. Low energy CEDM images may be used for interpretation in place of the FFDM, thereby reducing patient dose.

Variability of breast density assessment in short-term reimaging with digital mammography.

OBJECTIVE: To evaluate the variability of breast density assessments in short-term reimaging with digital mammography. MATERIALS AND METHODS: In 186 women, short term (mean interval, 27.6 days) serial digital mammograms including CC and MLO views were obtained without any treatment. Mammographic density assessments were performed by three blinded radiologists for Breast Imaging Report and Data System (BI-RADS, grades 1-4) and visual percentage density (PD) estimation, and by one radiologist for computer-aided PD estimation. The variability of assessments was analyzed according to the age, breast density, and mammography types by multivariate logistic regression. RESULTS: In BI-RADS assessments, 29% (161 of 558) of breast density categories were assessed differently after short-term reimaging and the mean absolute difference in PD for CC and MLO view was 7.6% and 8.1% for visual assessments, and 7.4% and 6.4% for computer-aided assessments, respectively. Among all computer-aided assessments, 29% (54 of 186) of CC view and 22% (41 of 186) of MLO view assessments had discrepancy over 10% in PD. Younger age (<50), greater breast density (grades 3 and 4), and different mammography types were significantly associated with the variability. CONCLUSION: Considerable variability in breast density assessments occurred in short-term reimaging with digital mammography, particularly in women with younger age and greater breast density and when examined using different types of mammography.