Background
Breast cancer incidence has increased during the past 2 decades [
1]. Intraductal papillary carcinoma (IDPC) is rare, accounting for < 2–5% of breast cancer cases. Although the sample size of malignant ductal breast lesions is very narrow, identify patients with a high likelihood of IDPC ability could develop a more accurate surgical planning and reduce potential risks to patients and the overall health care costs.
Ultrasound (US) because of its cost-effectiveness, non-radiation and high sensitivity is widely used for prediction of breast cancer [
2]. In the assessment of intraductal masses, US is more specific than mammography and magnetic resonance imaging (MRI), and may be considered as the selective manner [
3‐
6]. However, intraductal masses lack the classical ultrasound features of malignancy, such as taller-than-wide shape, and irregular margins (infiltrative, microlobulated). IDPC has some similarities with other intraductal breast masses in imaging and clinical manifestations, especially for those with partially cystic lesions on ultrasound, which is difficult to diagnose clinically. For the breast imaging reporting and data system (BI-RADS) system, intraductal lesions are one of the “special cases”. It is unclear whether all detected by US should be classified as BI-RADS 4a because there is no clear clinical and radiological prediction of malignancy. Whether biopsy is needed for all intraductal masses is still in debate [
7,
8]. Moreover, there are possibilities of false-negative results from biopsy procedures, and patients may experience inadequate resection in the first operation.
To the best of our knowledge, none of the studies has evaluated the US features of IDPC with partially cystic lesions. Except for intraductal papillary carcinoma (IDPC), there are previous papers analysing the ultrasound features of other malignant lesions of intraductal masses (including ductal carcinoma in situ and invasive ductal carcinoma). These ultrasound features include filling the duct more completely and involving more branch ducts [
7,
9]. Herein, we evaluated the US features of IDPC with the pathological results obtained after surgery, aiming to explore an effective ultrasound model of identifying intraductal papillary carcinoma with partially cystic lesions on US preoperatively.
Discussion
In this study, aiming to analyse the ultrasound features of IDPC lesions, we compared the US features between IDPC and intraductal papilloma. A collective model (including microcalcification, multiple lesions, posterior echo enhancement, wide base of solid components and rich colour Doppler flow) was proposed as a reference for patient diagnosis. The collective model could predict malignancy with an AUC of 0.99 (95% CI 0.95–1.00). The collective model had a better net benefit demonstrated by the decision curve.
On the basis of the results of the present study, the partially cystic IDPC showed several differences from the known malignant sonographic features of solid breast nodules [
11]. The findings of this study showed that hypoechogenicity was observed in 37 (92.5%) benign and 17 (81.0%) malignant groups. Therefore, hypoechogenicity was considered unhelpful in the differentiation of partially cystic IDPC from benign IDPC. In the study, microcalcifications and rich colour Doppler flow within the solid component of partly cystic lesions are considered the US features that raise the likelihood of IDPC. 3/21 IDPC had microcalcification on breast sonography, and there was only one microcalcification in the intraductal papilloma group. Twelve of 21 IDPCs had rich colour Doppler flow on breast sonography, and 9/40 with intraductal papilloma had rich colour Doppler flow. Similarly, colour Doppler ultrasound and microcalcifications are helpful for distinguishing between benign and malignant breast lesions [
12,
13]. In BI-RADS Fifth Edition, microcalcification and colour Doppler flow imaging features are also important US characteristics of breast lesions to consider when stratification a nodule [
11]. Since none of the studies has evaluated the US features of IDPC with partially cystic lesions, we found that traditional US features, such as microcalcifications and rich colour Doppler flow, are still helpful.
For diagnosing IDPC, the sensitivity, specificity, PPV, NPV, accuracy and AUC of a wide base of solid components were 90.5%, 97.5%, 95.0%, 95.1%, 95.1% and 0.94 (95% CI 0.86–1.00), respectively, which indicated good diagnostic value. Previous studies showed that the solid component of papilloma usually presents as a focal mass arising from the ductal wall, with a relatively narrow base for attachment [
10,
14]. In our study, the component of intraductal papilloma was identified to have a narrow base arising from the wall. The wide or narrow base configuration means an acute angle or blunt angle, depending on the angle degree between the solid component and the adjacent cyst wall on real-time sonography. A previous study also showed that malignant intraductal masses tended to fill the duct more completely [
7]. In total, 19/21 IDPCs had a wide base of the solid component on US, but only one lesion (1/40) had a wide base in the intraductal papilloma group. This may be due to the intracystic components of carcinomas requiring multiple vascular poles emerging from the base of the papillary projections [
15].
In a previous study, posterior acoustic shadowing of a solid mass was suggestive of invasive carcinoma, but this feature is not a reliable indicator for malignancy [
11,
14]. Unlike in these retrospective studies, the results of this study showed that posterior echo enhancement of a partially cystic mass increased the risk of IDPC. To our knowledge, there has been no other study on the diagnostic efficacy of posterior echo shadowing for partially cystic IDPC. This result may be explained by the fact that compared with the surrounding normal tissue, some tumours have a uniform internal structure. When close to the normal tissue, the boundary of the tissue is obvious, so repeated reflections occur, and the posterior echo is enhanced.
Malignancy rates of complex cystic and solid breast lesions contain 0.3%–50%, as reported in previous studies [
16]. A range of malignant pathologic results has been detected, and the most common malignancies include intraductal carcinoma in situ (DCIS), invasive ductal carcinoma (IDC), intraductal papillary carcinoma (IDPC) [
7,
16]. US appearance of DCIS are nonspecific. Calcified DCIS most commonly manifests as echogenic foci located within a mass or duct, associated with internal microlobulations, or distributed in a branch pattern. Noncalcified DCIS may manifest as a hypoechoic mass with microlobulated margins and no posterior acoustic features, or it may have a “pseudomicrocystic” appearance. Harmonic imaging and coronal reconstruction may improve the detection of noncalcified DCIS [
17]. The typical US appearance of IDC was irregular hypoechoic lesions with microcalcification, and it may have posterior echo attenuation [
18].
There are several limitations to our study. First, different radiologists performed the primary ultrasound examinations, which may lead to interobserver differences. Second, our study included a small number of patients with intraductal papillary carcinoma, and further studies with a larger patient population and prospective inclusion of intraductal masses with subjective criteria are needed to confirm our findings. Finally, only colour Doppler sonography was used to evaluate the vascularity of breast lesions.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.