A comprehensive analysis of imaging features and clinical characteristics to differentiate malignant from non-malignant mammographic architectural distortion

Abstract

Background: Mammographic architectural distortion (AD) is usually subtle and has variable presentations and causes, which poses a diagnostic challenge for breast radiologists and consequently a complex decision-making challenge for clinicians and patients. Presently, there is no reliable imaging standard to differentiate between malignant and benign ADs preoperatively. This study aimed to perform a comprehensive analysis of detailed mammographic and ultrasonographic features and clinical characteristics to enhance the diagnostic and differential efficacy for AD lesions. The findings have the potential to boost the diagnostic confidence of breast radiologists when encountering with AD lesions and could be instrumental in refining clinical management strategies for ADs.

Methods: This retrospective study included consecutive female patients with ADs on screening or diagnostic mammography from January 6, 2015, to December 28, 2018. The patient's clinical data, mammographic and ultrasonographic or "second look" ultrasonographic findings, and pathological results were reviewed. The continuous variables were analyzed using the t-test. The categorical variables were assessed using the Chi-square test or two-tailed Fisher's exact test. Logistic regression analyses were conducted to evaluate potential risk factors for pathologically proven malignant ADs. Machine learning model based on multimodal clinical and imaging features was constructed using R software.

Results: Ultimately, 344 patients with 346 AD lesions were enrolled in the study (mean age: 47.40±10.07 years; range, 19-84 years). Of the ADs, 228 were malignant and 118 were non-malignant. Palpable AD on mammography was more likely to indicate malignancy than non-palpable AD (83.43% vs. 49.15%, P<0.001). AD associated with other mammographic findings was more likely to be malignant than pure AD (73.58% vs. 59.36%, P=0.005). Ultrasonography (US) correlates were observed in 345 of these 346 AD lesions. Among these US correlates, 63 (18.26%, 63/345) were detected by "second look" ultrasound. For the US correlates, the mammographic ADs that appeared as non-mass-like hypoechoic areas and masses on US were more likely to be malignant than those that appeared as other abnormalities (P<0.001). The sensitivity, specificity and diagnostic accuracy of the eXtreme Gradient Boosting (XGBoost) model based on clinical and comprehensive imaging features in differentiation of AD lesions in the validation set were 66.46%, 94.23% and 78.9%, respectively, and the AUC was 0.886 (95% confidence interval: 0.825-0.947).

Conclusions: The application of mammograms-guided "second-look" ultrasound could enhance the detection of US correlates, particularly non-mass-like features. The comprehensive analysis based on clinical and multimodal imaging features could be beneficial in improving the diagnostic and differential efficacy for AD lesions detected on mammography and instrumental in refining clinical management strategies for ADs.

Keywords: Breast; architectural distortion (AD); mammography; ultrasound; “second look” ultrasound.

Figure 1

Flowchart of patient selection process. AD, architectural distortion; DBT, digital breast tomosynthesis; n, number of patients.

Figure 2

Superolateral inferomedial oblique spot compression magnification view of a 44-year-old woman showed pure architectural distortion in an upper inner quadrant of the left breast. The surgical histopathology results revealed benign intraductal papilloma with usual ductal hyperplasia and pseudo-invasion.

Figure 3

Craniocaudal (A), mediolateral oblique (B) and spot compression magnification (C) views of a 38-year-old woman showed architectural distortion in the upper outer left breast, associated with fine amorphous calcifications. The insets showed a zoomed-in area of interest. Surgical histopathology revealed ductal carcinoma in situ.

Figure 4

Craniocaudal and mediolateral oblique views of a 54-year-old woman showed architectural distortion in the upper outer right breast, associated with asymmetry. The insets showed the zoomed-in area of interest. Surgical histopathology revealed invasive ductal carcinoma. RMLO, right mediolateral oblique.

Figure 5

The ultrasonographic image showed a non-mass-like hypoechoic area (+) with a score of 1 for the space-occupying effect that corresponded to mammographic finding of Figure 2.

Figure 6

The ultrasonographic image showed a non-mass-like hypoechoic area (white arrows) with a score of 2 for the space-occupying effect that corresponded to mammographic finding of Figure 3.

Figure 7

The sonographic image showed an irregular hypoechoic mass (white arrows) with a score of 3 for the space-occupying effect that corresponded to mammographic finding of Figure 4.

Figure 8

Receiver operating characteristic curves showed the efficacy of mammography, ultrasonography, and combined mammography and ultrasonography to differentiate between malignant and non-malignant mammography-detected architectural distortion. US, ultrasonography; MG, mammography.

Figure 9

Receiver operating characteristic curves showed the efficacy of the XGBoost model based on clinical and comprehensive image features in predicting the malignancy of mammography-detected architectural distortion in the validation set. ROC, receiver operating characteristic; AUC, area under the curve; XGBoost, eXtreme Gradient Boosting.