**1. Introduction**

Early detection of breast cancer is crucial to reducing mortality from the disease [1]. Population screening using mammography is the standard approach to early detection. The principal criterion for screening recommendation is patient age, although a family history of breast cancer is also a well-established risk factor for breast cancer. In the last two decades, high mammographic breast density as a risk factor has also gained significant attention in the literature due to the associated increase in breast cancer risk and the resultant reduced sensitivity of screening mammography [2]. Technological advances and the transition from film-screen to digital mammography (DM) reduced the masking effect of breast density and enhanced cancer detection [3]. However, there are still difficulties reading DM images of women with dense breasts [4]. Mammographically dense tissue has been demonstrated to increase radiologists' suspicion during the interpretation of screening mammograms, leadingtohigherrecallratesforwomenwithmammographicallydensebreasts[5].

 Most women with dense breasts recalled for assessment have demonstrated negative or benign outcomes, with false positives ranging between 73% and 97% [6–10]. Unnecessary

**Citation:** Hadadi, I.; Clarke, J.; Rae, W.; McEntee, M.; Vincent, W.; Ekpo, E. Reducing Unnecessary Biopsies Using Digital Breast Tomosynthesis and Ultrasound in Dense and Nondense Breasts. *Curr. Oncol.* **2022**, *29*, 5508–5516. https://doi.org/ 10.3390/curroncol29080435

Received: 21 June 2022 Accepted: 2 August 2022 Published: 4 August 2022

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

recalls increase the cost of screening, may cause psychosocial harm due to false-positive alarms, and can deter women from rescreening [11,12]. Digital breast tomosynthesis (DBT) and ultrasound can mitigate the limitations associated with DM, allowing for a more detailed evaluation of breast tissue and suspicious lesions by minimising superimposition of parenchymal densities [13–15]. In the United Kingdom National Health Service Breast Screening Programme, DBT assessment of women recalled at DM screening led to a 33% reduction in the benign biopsy rate [16], with lower false positives across mammographic features such as mass lesions and asymmetric densities. Some studies have also shown that using ultrasound is accurate in distinguishing between benign and malignant lesions and decreases the number of benign biopsies by 34% to 60% [17,18]. However, it is unclear how DBT and ultrasound compare when assessing women recalled and whether they reduce false-positive biopsies across different breast compositions.

Studies comparing DBT and ultrasound [19,20] have focused on incremental breast cancer detection in mammography-negative dense breasts. In one study [21], the effectiveness of ultrasound screening after the combination of DM and DBT was examined. Despite the widespread concurrent use of DBT and ultrasound in many clinical settings, no study has directly compared their diagnostic efficacy in mammography-recalled women to establish optimised assessment pathways. Importantly, no published work has compared the impact of these tools in reducing unnecessary biopsies of benign lesions, across various breast densities. We hypothesised that DBT and ultrasound result in equal numbers of unnecessary biopsies. Therefore, this study aims to compare the performance of DBT and ultrasound in women recalled for assessment after a positive screening mammogram and to compare the number of unnecessary biopsies in dense and nondense breasts.

## **2. Materials and Methods**

#### *2.1. Study Design and Patient Selection*

This is a data linkage study of women recalled for assessment following a suspicious finding on their screening mammogram. Patients were included if they were recalled for assessment after a screening mammogram, had both DBT and ultrasound assessment, and underwent breast biopsy procedures. All screening mammograms of the recalled women were read by two radiologists who interpreted the images independently from each other. Any discrepancies between these two readers were resolved by a third radiologist.

A five-tier grading system was used to describe the findings on screening mammograms based on BreastScreen Australia's RANZCR breast imaging lesion classification system: grade 1 (normal or no abnormality); grade 2 (benign); grade 3 (indeterminate/equivocal); grade 4 (suspicious); grade 5 (malignant). Each breast was examined using two different mammographic views: the craniocaudal (CC) and the mediolateral oblique (MLO). If more mammography spot views were required, they were obtained. Women whose images were rated 3, 4, or 5 were recalled for assessment using DBT and ultrasound to confirm or rule out breast cancer. A needle biopsy was performed on breast lesions graded 3, 4, or 5 after these imaging assessment tools were assessed. Table 1 shows the baseline characteristics of women recalled for assessment.

Breast density was reported according to the Breast Imaging Reporting and Data System (BI-RADS 5th edition): BI-RADS A: "the breasts are almost entirely fatty"; BI-RADS B: "there are scattered areas of fibroglandular density"; BI-RADS C: "the breasts are heterogeneously dense, which may obscure small masses"; BI-RADS D: "the breasts are extremely dense, which lowers the sensitivity of mammography".

All DBT images used in this study were acquired using Selenia Dimensions (Hologic Inc.). Real time B-mode and colour doppler were performed using an ACU-SON S2000 Ultrasound System (HELX Evolution with Touch Control, Siemens Medical Solutions), equipped with a 12L4 linear array transducer (12–4 MHz). Both DBT and ultrasound assessment results were evaluated according to the RANZCR breast imaging lesion classification scale.


**Table 1.** Characteristics of women recalled for assessment following a suspicious mammography finding.
