Reassessment and Follow-Up Results of BI-RADS Category 3 Lesions Detected on Screening Breast Ultrasound

Transcription

1 Women s Imaging Original Research Chae et al. I-RDS Category 3 Lesions Found on Screening Ultrasound Women s Imaging Original Research Eun Young Chae 1 Joo Hee Cha Hee Jung Shin Woo Jung Choi Hak Hee Kim Chae EY, Cha JH, Shin HJ, Choi WJ, Kim HH Keywords: I-RDS, breast neoplasm, probably benign lesion, ultrasound DOI: /JR Received pril 5, 2015; accepted after revision July 16, ll authors: Department of Radiology, Research Institute of Radiology, san Medical Center, University of Ulsan, College of Medicine, Pungnap-2-dong, Songpa-gu, Seoul , Korea. ddress correspondence to J. H. Cha (jhcha@amc.seoul.kr). This article is available for credit. JR 2016; 206: X/16/ merican Roentgen Ray Society Reassessment and Follow-Up Results of I-RDS Category 3 Lesions Detected on Screening reast Ultrasound OJECTIVE. The purpose of this study is to determine the frequency and the malignancy rate of I-RDS category 3 lesions detected on screening breast ultrasound and to reassess whether they satisfied the requirements of the merican College of Radiology Imaging Network (CRIN) 6666 protocol. MTERILS ND METHODS. Of 28,796 asymptomatic women who underwent screening mammography during 2 years, 12,187 underwent additional ultrasound as part of the screening. Patients for whom I-RDS category 3 lesions were seen on the ultrasound were selected. We reviewed the initial ultrasound showing I-RDS category 3 lesions and mammograms. We also investigated the clinical outcome of these lesions using the reference standard of a combination of pathologic analysis and follow-up for at least 24 months. RESULTS. The frequency of I-RDS category 3 lesions detected on screening ultrasound was 14.6% (1783/12,187). Of the 1164 patients with a follow-up duration of at least 24 months or whose lesions were biopsied, eight were eventually proven to have malignancy (0.7%). The malignancy rate was 2.2% (4/184) for patients with abnormal mammograms and 0.4% (4/980) for those with normal mammograms. When the CRIN 6666 protocols were strictly applied, 225 (19.3%) lesions were retrospectively recategorized as I-RDS category 4 (n = 12) or category 2 (n = 213). ll detected malignancies were early breast cancers with no lymph node metastasis. CONCLUSION. lthough the frequency of ultrasound I-RDS category 3 lesions is considerably high (14.6%), the malignancy rate is very low (0.7%), especially in patients with a normal mammogram. Therefore, with I-RDS category 3 assessment, careful evaluation is required to avoid unnecessary short-interval follow-up or biopsy. ecause there is an increasing demand for improved breast cancer detection [1] and the sensitivity of mammography may be as low as 30 48% in women with dense breast tissue [2, 3], breast ultrasound is an appealing and useful screening tool. It is widely available, is well tolerated by most patients, requires no radiation or contrast medium, and is free of adverse effects [4]. Many previous studies have shown that supplemental screening breast ultrasound provides an increased cancer detection yield of cancers per 1000 women [3, 5 8]. However, the addition of screening breast ultrasound to mammography also causes an increase in false-positive biopsy rates and a decrease in specificity. erg et al. [5] reported that the positive predictive value of breast biopsy decreased from 22.6% for mammography alone to 11.2% for mammography plus screening ultrasound. n assessment of I-RDS category 3 lesions (i.e., probably benign) has been proposed for specific imaging findings, such as circumscribed, oval, and solid masses [9]. I-RDS category 3 lesions have a less than 2% likelihood of malignancy, according to most diagnostic and screening ultrasound series [1, 10 12]. ppropriate use of I-RDS category 3 is clinically helpful in reducing the number of false-positive biopsies while maintaining an acceptable cancer detection rate. The final assessment of I-RDS category 3 is audited as positive at the screening breast ultrasound. For all lesions assessed as probably benign on ultrasound, the surveillance protocol is identical to that used for mammography and involves follow-up examinations at 6, 12, and 24 months. However, the published literature supporting this recommendation is not as convincing as that for mammography. 666 JR:206, March 2016

2 I-RDS Category 3 Lesions Found on Screening Ultrasound 12,187 Women with screening breast ultrasound Fig. 1 Flowchart of I-RDS category 3 lesions in this study. longest dimension along with a measurement orthogonal to it Patients with I-RDS category 3 lesions 1164 Patients with follow-up for at least 24 months or biopsy 980 With normal mammograms 4 Cancers detected 4 Cancers detected Therefore, the purpose of our study is to determine the frequency and the malignancy rate of I-RDS category 3 lesions detected on screening breast ultrasound and to reassess whether they satisfied the requirements of the merican College of Radiology Imaging Network (CRIN) 6666 protocol. Materials and Methods Study Population Our institutional review board approved this retrospective study, and the requirement for informed consent was waived. total of 28,796 women with no signs or symptoms of breast abnormalities (age range, years; mean age, 52 years) underwent screening mammography at our institution between January 2008 and December mong them, 12,187 women underwent self-financed voluntary additional sonography as part of a screening examination. mong those women, we identified those patients whose lesions were classified as I-RDS category 3 on ultrasound, and we included in our study only the patients with pathologic results or clinical follow-up of at least 2 years. 184 With abnormal mammograms Screening Examinations (Mammography and Ultrasound) Mammography was performed using a fullfield digital mammography unit (Senographe DS or Senographe Essential scanner, both from GE Healthcare). Two standard image planes, the mediolateral oblique and the craniocaudal views, were used. Whole-breast ultrasound was performed by one of five board-certified radiologists, with 5 20 years of clinical experience in performing breast ultrasound, using an IU22 unit (Philips Healthcare) equipped with a 50-mm linear array transducer with a bandwidth of 12 5 MHz. t our medical institution, the scanning technique of bilateral whole-breast ultrasound was standardized as follows: it begins with scanning of the right breast using a transverse and sagittal orientation and scanning the inner breast in a supine position and the outer breast in a supine oblique position with the patient s arm raised above her head. The pectoralis muscle had to be seen on all images to ensure that the entire breast was examined. The axilla was also included in the examination. We documented each lesion with an image of its Image Evaluation Sonographic images were retrospectively reviewed by two radiologists with 4 and 14 years, respectively, of clinical experience in breast radiology. Each radiologist was blinded to the other radiologist s readings during the initial review. When there was a discrepancy, the two radiologists reviewed the cases together and reached a consensus. They were also blinded to any clinical or histopathologic information regarding the patients. The radiologists evaluated whether a lesion satisfied the criteria of I-RDS category 3 according to the CRIN 6666 protocol [13]. When a lesion was recategorized as anything other than I-RDS category 3, they provided the reason for reassessment. The interpreting radiologists were fully aware of the CRIN 6666 protocol and were trained to make the final assessment, as outlined here. lesion was classified as I-RDS category 3 when it had the following features: an oval mass (parallel to the skin in orientation) and hypoechoic to fat, with circumscribed borders and no posterior features or minimal posterior enhancement, including multiple bilateral masses with these features if seen only on ultrasound; a hyperechoic mass with central hypoechoic to anechoic components suggesting fat necrosis; a hypoechoic oval mass with homogeneous low-level internal echoes that otherwise met the criteria for simple cysts (i.e., circumscribed acoustic enhancement); a microlobulated or oval mass composed entirely of clustered microcysts with or without layering microcalcifications; probable artifactual posterior shadowing at the interface of fat lobules, without any associated mass, that changes its appearance on changing the angle of insonation; and architec- Fig year-old woman who presented for routine check-up., Initial ultrasound shows 4-mm mass with oval shape, circumscribed margins, and parallel orientation. Mass was categorized as I-RDS category 3 at that time. There was no interval change at 6-month follow-up examination., Twelve months later, lesion size had increased to 11 mm and had microlobulated margins. iopsy was recommended for mass, and invasive ductal carcinoma was confirmed. JR:206, March

3 Chae et al. Fig year-old woman who presented for routine check-up. and, Transverse () and longitudinal () ultrasound examinations show approximately 12-mm hypoechoic mass that was initially assessed as I-RDS category 3. When merican College of Radiology Imaging Network 6666 protocols were strictly applied, lesion was reassessed as I-RDS category 4 because of microlobulated margins (arrows, ). Ultrasound-guided core needle biopsy was performed at patient s request, and papillary neoplasm was confirmed. tural distortion thought to be due to postsurgical scarring. If a lesion showed one or more suspicious findings, it was reclassified as a I-RDS category 4 or 5 lesion. Those findings included an irregular shape, noncircumscribed margin (microlobulated, indistinct, angular, or spiculated), posterior acoustic shadowing, an intraductal mass, nonparallel orientation (i.e., taller than wide), microcalcifications within a mass, duct extension, echogenic halo, or complex cystic and solid echogenicity. If a lesion showed the following features, it was reclassified as a I-RDS category 2 lesion: simple cyst, duct ectasia without intraductal mass, intramammary lymph node, intensely and uniformly hyperechoic mass, circumscribed mass that clearly contains fat, macrocalcification, siliconomas, or accessory breast tissue. We also retrospectively reviewed the mammograms according to the fifth edition of the I-RDS lexicon [9]. If the mammogram indicated I-RDS category 1 or 2, it was classified as a normal mammogram. If the mammography indicated I-RDS category 0, 3, 4, or 5, it was classified as an abnormal mammogram. During the retrospective review of mammograms and ultrasound, the interpreting radiologists reviewed each examination masked to the results of the other. Statistical nalysis The reference standard was a combination of the pathologic results and clinical follow-up results after 2 years. Statistical calculations were performed using commercial software (SS, version 9.3, SS Institute). p value less than 0.05 was considered to indicate statistical significance. The two-sample t test was used to compare the means of the parametric variables, and the Fisher exact test was used to compare the nonparametric variables. Results Frequency and Malignancy Rate of I-RDS Category 3 Lesions mong 12,187 consecutive women who underwent voluntary self-financed screening ultrasound (age range, years; mean age, 53.8 years), 1783 were classified as having I-RDS category 3 lesions using ultrasound (Fig. 1). The patients age range was years, with a mean age of 51.2 years. Therefore, the frequency of I-RDS category 3 lesions was 14.6% (1783/12,187). mong these patients, a total of 1164 who had pathologic results available (n = 63) or clinical follow-up (n = 1101) for at least 2 years after their ultrasound examinations comprised our study population. One thousand seventeen women had a single I-RDS category 3 lesion, and 147 (12.6%) had multiple lesions. Seven women with I-RDS category 3 lesions that showed change on follow-up (n = 3) or biopsy (n = 4) also had multiple benign masses. Fifty-five lesions were diagnosed as benign by ultrasound-guided biopsy (n = 44), fine-needle TLE 1: ssociation etween the Lesion Size and the Lesion Type Size Malignant (n = 8) enign (n = 1156) p Mean ± SD (mm) 6.75 ± ± Size group (no. of lesions) < 10 mm mm aspiration (n = 7), or surgical excision (n = 4); these included 24 fibroadenomas, eight nonproliferative breast lesions, six proliferative breast lesions without atypia, four fibroadenomatous mastopathies, four cysts, three intraductal papillomas, one hamartoma, one nodular adenosis, one fibrocystic change, one duct ectasia with periductal chronic inflammation, one case of fluid with macrophages, and one case of chronic inflammation with foreign body reaction. During the follow-up period, eight of the 1164 patients eventually proved to have malignant lesions, with an overall malignancy rate of 0.7% (Fig. 2). Of the 1164 patients with follow-up for at least 24 months or a biopsy, 980 had normal mammograms and 184 had abnormal mammograms. One hundred eighty-four abnormal mammograms included 99 masses, 54 calcifications, 29 asymmetries, and two architectural distortions. The malignancy rate was 2.2% (4/184) for patients with an abnormal mammogram and 0.4% (4/980) for those with a normal mammogram. Therefore, the malignancy rate for patients with abnormal mammograms was higher than for those with normal mammograms, and the differences were statistically significant (p = ). In additional analysis of the lesion size, the mean (± SD) diameter of malignant lesions was 6.75 ± 2.76 mm (range, 3 11 mm), whereas that of benign lesions was 6.99 ± 3.93 mm (range, 2 36 mm), thus indicating that there was no statistically significant difference in the size of malignant and benign lesions (Table 1). Two of eight malignant lesions (25%) and 225 of 1156 benign lesions (19.3%) were 10 mm or larger in diameter; thus, there was no statistically significant difference. 668 JR:206, March 2016

4 I-RDS Category 3 Lesions Found on Screening Ultrasound TLE 2: Reasons for the Upgrade to Category 4 and the Diagnosis Lesion No. Reasons for Upgrade Diagnosis 1 Margin not circumscribed, indistinct, not parallel Invasive ductal carcinoma 2 Margin not circumscribed, indistinct posterior shadowing Invasive ductal carcinoma 3 Complex cystic and solid Intraductal papilloma 4 Margin not circumscribed, microlobulated Intraductal papilloma 5 Irregular shape Intraductal papilloma 6 Margin not circumscribed, microlobulated Fibroadenoma 7 Intraductal mass Nonproliferative breast change 8 Margin not circumscribed, microlobulated Probably benign on follow-up 9 Margin not circumscribed, microlobulated Probably benign on follow-up 10 Irregular shape Probably benign on follow-up 11 Irregular shape Probably benign on follow-up 12 Irregular shape Probably benign on follow-up Reassessment of Lesions When the CRIN 6666 protocol was strictly applied on retrospective review of the lesions, 225 lesions (19.3%) initially assessed as probably benign were reassessed as other categories. Twelve lesions were upgraded to I-RDS category 4 (Table 2 and Figs. 3 and 4), and the most common reason for the reassessment was the margin (six of 12 lesions). Other reasons for the upgrade included an irregular shape (n = 4), complex cystic and solid echogenicity (n = 1), and intraductal mass (n = 1). mong the 12 lesions, because two were proven to be malignant, the malignancy rate of the reassessed I-RDS category 4 lesions was 16.7%. The remaining 213 lesions were downgraded to I-RDS category 2 (Table 3 and Fig. 5), and no malignancy was detected. The most common reason for downgrading was reassessment of an anechoic lesion (n = 178; i.e., the findings of a simple cyst). Duct ectasia without an intraductal mass was also a common feature (n = 25). TLE 3: Reasons for Downgrade to Category 2 Reasons for Downgrade No. of Lesions nechoic lesion (i.e., simple cyst) 178 Duct ectasia without intraductal mass 25 Involuting fibroadenoma 3 Intramammary lymph node 3 ccessory breast tissue 1 Macrocalcification 1 Postsurgical scar 1 Lipoma 1 Total 213 Detected Cancers Of the eight malignancies initially classified as I-RDS category 3, six (75%) were invasive (i.e., five were invasive ductal carcinoma and one was invasive tubular carcinoma). Except for one patient who chose to undergo treatment at another hospital, all of the invasive cancers in our patients were early breast cancers of the T1 stage, and the mean diameter of the cancers was 9 mm (range, 2 16 mm). No axillary lymph node metastasis was found in any patient, and the remaining two patients had ductal carcinoma in situ (11 and 25 mm in diameter). Discussion Radiologists commonly encounter probably benign lesions during daily breast ultrasound examinations. In this study, we found that the frequency of I-RDS category 3 assessment was 14.6% among 12,187 consecutive women who underwent physicianperformed bilateral screening breast ultrasound (i.e., approximately one of seven patients who underwent ultrasound). Hooley et al. [1] and arr et al. [12] reported that I-RDS category 3 lesions were detected at a rate of 20% (187 of 935 and 519 of 2662 participants, respectively) on screening ultrasound in a patient population with an elevated risk. The percentage of I-RDS cat- Fig year-old woman who presented for routine check-up., Ultrasound shows 3-mm hypoechoic mass (between calipers), initially categorized as I-RDS category 3. On further review, lesion was assessed as having illdefined margin and taller-than-wide orientation, which resulted in its reassessment as I-RDS category 4. Short-interval follow-up was recommended for mass, but patient did not return., Twenty-four months later, lesion showed increase in size, irregular shape, and microlobulated margins. iopsy was recommended for mass, and invasive ductal carcinoma was confirmed. JR:206, March

5 Chae et al. Fig year-old woman who presented for routine check-up., Initial ultrasound shows bilateral symmetric focal duct ectasia in subareolar areas of both breasts. It was categorized as I-RDS category 3 at that time. On review, lesion was reassessed as I-RDS category 2 because there was no evidence of intraductal mass. Patient had no symptoms., t time of 2-year follow-up ultrasound, lesion was still unchanged. egory 3 lesions, including those detected in our study, is higher on ultrasound than on mammography, which reported an approximately 1 2% rate of I-RDS category 3 lesion detection [14, 15]. Despite the advantages of ultrasound as a screening examination, one of the major concerns regarding screening breast ultrasound is the risk of unnecessary tests or biopsies for women who do not have cancer (i.e., false-positives) [16]. For auditing purposes, the current I-RDS reclassifies category 3 assessment as audit positive for all screening modalities (mammography, ultrasound, and MRI) and requires additional nonstandard images, but as audit negative for diagnostics. Previously, I-RDS category 3 assessment was audit negative [17]. This change may explain the large percentage of I-RDS category 3 lesions in the present study, which reports ultrasound examinations performed between January 2008 and December 2009, several years before publication of the fifth edition. In our study, nearly 20% of the lesions initially assessed as I-RDS category 3 were recategorized as other categories when the CRIN 6666 protocol was strictly applied on retrospective review. The most common reason for the upgrade to category 4 was the noncircumscribed margin of the lesion (six of 12 lesions; 50%); this finding was similar to that of a previously published report [18]. The malignancy rate of reassessed I-RDS category 4 lesions was 16.7% (two of 12 lesions). In clinical practice, a radiologist may overlook subtly suspicious ultrasound features. No malignancy was found in 213 reassessed I-RDS category 2 lesions. mong 213 reassessed I-RDS category 2 lesions in this study, 178 lesions were downgraded because they were anechoic, consistent with simple cysts. For some lesions, ultrasound image quality might provide an explanation (Fig. 6). s with all imaging modalities, the detection and diagnosis for ultrasound depends on the image quality factors, such as focal zone setting, FOV, gray-scale gain, or transducer frequency. If a lesion is appropriately downgraded as I-RDS category 2, the patient could be followed by routine screening, instead of additional short-term follow-up. mong 1164 patients in our study, 12.6% of patients had multiple lesions. ccording to the current suggestions by I-RDS [9], multiple bilateral circumscribed masses at ultrasound, at least three overall and one in each breast, could be assessed as benign, with a recommendation for routine screening. If those lesions are assessed as I-RDS category 2 at the outset, the positive predictive value could increase considerably. In this study, seven women with I-RDS category 3 lesions that showed change on follow-up or biopsy also have multiple benign masses. When multiple masses are present, each should be evaluated separately, and the 670 JR:206, March 2016

6 I-RDS Category 3 Lesions Found on Screening Ultrasound Fig year-old woman who presented for routine check-up., Initial ultrasound shows 7-mm mass with round shape and circumscribed margin. It was categorized as I-RDS category 3 at that time. On review, lesion was reassessed as I-RDS category 2 because reviewer considered it as cyst. t initial ultrasound image, breast tissue occupied only 50% of FOV, and focal zone (marked by line and whiskers on right) is set too widely., t time of 2-year follow-up ultrasound, lesion (between calipers) was still unchanged. Its anechogenicity and posterior enhancement were depicted clearly with proper settings of focal zone and FOV. I-RDS assessment for the case should reflect the highest level of suspicion for any single lesion. Fewer than 1% of ultrasound I-RDS category 3 lesions were malignant in our study (eight of 1164 participants; 0.7%). This is similar to findings of previous studies [3, 12] and is within the I-RDS guidance ( 2%). When the criteria are limited to invasive breast cancer, the frequency of invasive cancer among ultrasound I-RDS category 3 lesions decreased to 0.5% (six of 1164 participants). Considering the purpose of the screening examination, a few malignancies detected during follow-up and with a minimal effect on the prognosis are unavoidable. In our study, the average diameter of the detected invasive cancers initially classified as I-RDS category 3 was 9 mm (range, 2 16 mm), and no axillary lymph node metastasis was found. In this study, the malignancy rate was significantly higher in probably benign lesions detected by an abnormal mammogram than with a normal mammogram. Therefore, an integrated interpretation of ultrasound with mammography can be useful for assessing these probably benign lesions. dditional analysis for determining lesion size did not influence the malignancy rate of these lesions, because six of the eight malignancies (75%) were less than 10 mm in diameter. The frequency of ultrasound I-RDS category 3 assessment is considerably high (i.e., 14.6% in this study), and the cost of additional follow-up studies in these probably benign lesions can be substantial [19]. However, the malignancy rate of ultrasound I-RDS category 3 lesions is low, and all of the cancers detected in our study were early node-negative (T1N0M0) cancers. lthough the current surveillance guidelines of ultrasound I-RDS category 3 lesions includes a 6-month followup, there are diverse opinions among radiologists regarding the proper timing of follow-up for these probably benign lesions [12]. Extending the follow-up interval from the current guideline to 1 year may be appropriate, especially after a normal mammogram, because the number of unnecessary examinations would thereby substantially decrease and the cost-effectiveness of screening breast ultrasound would be increased with improved specificity. further prospective randomized controlled study will be needed to validate the yearly follow-up of I-RDS category 3 lesions detected on ultrasound. dditional research for shear wave or strain elastography and -mode ultrasound would be appropriate to determine if it is appropriate to accurately up- or downgrade lesions between I-RDS categories 3 and 4a. lthough any lesions assessed as category 4b, 4c, or 5 by morphologic criteria cannot be downgraded, elastography may be helpful in confirming an assessment based on morphologic features. Our study has several limitations. ecause it was retrospective, we evaluated only the static ultrasound images, and real-time evaluation was not possible. However, we enrolled consecutive patients with a relatively large number of I-RDS category 3 lesions. In addition, the I-RDS recommendations for category 3 lesions were not always strictly followed because 34% of the patients (n = 619) did not complete their follow-up for at least 24 months. nother limitation of our study is that it involved voluntary screening in a single medical institution, rather than populationbased screening. ecause we did not consider risk factors such as a mutation in RC genes or a personal or family history of breast cancer, women with an increased risk for breast cancer might have been included. In conclusion, the frequency of ultrasound I-RDS category 3 assessment is considerably high (14.6%), and these probably benign lesions have a low malignancy rate (0.7% including in situ cancer; 0.5% limited to invasive breast cancer), especially with normal mammograms (0.4%). With the introduction of the breast density notification law, screening breast ultrasound is becoming more widely used. Careful assessment is needed with I-RDS category 3 to avoid unnecessary biopsy or short-interval follow-up, and further prospective validation of the yearly follow-up of I-RDS category 3 lesions would be promising. References 1. Hooley RJ, Greenberg KL, Stackhouse RM, Geisel JL, utler RS, Philpotts LE. Screening US JR:206, March

7 Chae et al. in patients with mammographically dense breasts: initial experience with Connecticut Public ct Radiology 2012; 265: Mandelson MT, Oestreicher N, Porter PL, et al. reast density as a predictor of mammographic detection: comparison of interval- and screen-detected cancers. J Natl Cancer Inst 2000; 92: Kolb TM, Lichy J, Newhouse JH. Comparison of the performance of screening mammography, physical examination, and breast US and evaluation of factors that influence them: an analysis of 27,825 patient evaluations. Radiology 2002; 225: Mendelson E, Tobin CE. Critical pathways in using breast US. RadioGraphics 1995; 15: erg W, lume JD, Cormack J, et al. Combined screening with ultrasound and mammography vs mammography alone in women at elevated risk of breast cancer. JM 2008; 299: uchberger W, Niehoff, Obrist P, DeKoekkoek-Doll P, Dunser M. Clinically and mammographically occult breast lesions: detection and classification with high-resolution sonography. Semin Ultrasound CT MR 2000; 21: Leconte I, Feger C, Galant C, et al. Mammography and subsequent whole-breast sonography of nonpalpable breast cancers: the importance of radiologic breast density. JR 2003; 180: FOR YOUR INFORMTION 8. Chae EY, Kim HH, Cha JH, Shin HJ, Kim H. Evaluation of screening whole-breast sonography as a supplemental tool in conjunction with mammography in women with dense breasts. J Ultrasound Med 2013; 32: D Orsi CJ, Sickles E, Mendelson E, et al. CR I-RDS tlas, reast Imaging Reporting and Data System. Reston, V: merican College of Radiology, Graf O, Helbich TH, Hopf G, Graf C, Sickles E. Probably benign breast masses at US: is follow-up an acceptable alternative to biopsy? Radiology 2007; 244: Raza S, Chikarmane S, Neilsen SS, Zorn LM, irdwell RL. I-RDS 3, 4, and 5 lesions: value of US in management follow-up and outcome. Radiology 2008; 248: arr RG, Zhang Z, Cormack J, Mendelson E, erg W. Probably benign lesions at screening breast US in a population with elevated risk: prevalence and rate of malignancy in the CRIN 6666 trial. Radiology 2013; 269: merican College of Radiology Imaging Network (CRIN). CRIN protocol CRIN website. Published November 30, ccessed October 27, aum JK, Hanna LG, charyya S, et al. Use of I-RDS 3 probably benign category in the merican College of Radiology Imaging Network Digital Mammographic Imaging Screening Trial. Radiology 2011; 260: Kerlikowske K, Smith-indman R, braham L, et al. reast cancer yield for screening mammographic examinations with recommendation for short-interval follow-up. Radiology 2005; 234: rem RF, Lenihan MJ, Lieberman J, Torrente J. Screening breast ultrasound: past, present, and future. JR 2015; 204: Mendelson E, aum JK, erg W, et al. I-RDS: ultrasound, 1st ed. In: D Orsi CJ, Mendelson E, Ikeda DM, et al., eds. reast Imaging Reporting and Data System: CR I-RDS breast imaging atlas. Reston, V: merican College of Radiology, Moon HJ, Kim MJ, Kwak JY, et al. Malignant lesions initially categorized as probably benign breast lesions: retrospective review of ultrasonographic, clinical and pathologic characteristics. Ultrasound Med iol 2010; 36: limoglu E, ayraktar SD, ozkurt S, et al. Followup versus tissue diagnosis in I-RDS category 3 solid breast lesions at US: a cost-consequence analysis. Diagn Interv Radiol 2012; 18:3 10 This article is available for CME and Self-ssessment (S-CME) credit that satisfies Part II requirements for maintenance of certification (MOC). To access the examination for this article, follow the prompts associated with the online version of the article. 672 JR:206, March 2016