Continuing Education Column Imaging Findings of Missed reast Cancer: Retrospective nalysis oo -Kyung Han, MD Department of Radiolgy, Sungkyunkwan University School of Medicine E - mail : bkhan@skku.edu J Korean Med ssoc 2008; 51(9): 838-848 bstract Screening mammography has been proved to be an effective tool to detect early breast cancers and to decrease mortality. However, the rate of false-negative mammograms has been reported to be still high as 10~30%. Missed breast cancers are cancers that are visible at previous mammograms only retrospectively and can be classified as three types; interval cancers, subsequent screen-detected cancers, and alternative imaging-detected cancers. In a small group, screen-detected abnormalities recalled for further evaluation may be dismissed due to false negative diagnostic assessment, leading to delays in breast cancer diagnosis. Possible causes for missing include perception errors, interpretation errors, and technical errors. Furthermore, every diagnostic examination has inherent limitations. Perception errors are often attributed to combined multiple factors; peripheral lesions, single view abnormalities, subtle findings, distracting lesions, and dense parenchyma obscuring a lesion. To decrease the false negative rate, radiologists should be alert to take additional mammograms and ultrasonography, and should try to improve the image quality and interpretation techniques comparing with the previous imaging, considering the use of computer-aided detection or double reading. Keywords: reast neoplasms; diagnosis; Missed cancer; Cancer screening - - 838
Imaging Findings of Missed reast Cancer Figure 1. n interval cancer (marked by marker) developing 11 months after previous screening mammogram reported as negative. small spiculated mass which was palpable seen at the time of cancer diagnosis () was not recognized on previous mammographic study (). On retrospective evaluation, a tiny abnormal density was seen () at the periphery of left upper central breast. lesion overlapped with pectoralis muscle caused decreased contrast. Table 1. Classification of Mammographically Missed reast Cancers Symptoms at previous mammograms Symptoms at diagnosis Time interval to diagnosis Interval cancers None Yes Subsequent screendetected cancers lternative imagingdetected cancers None None None None nnual screening: < 1 year iennial screening: < 2 year nnual screening: 1 year iennial screening: 2 year at the time of alternative imaging Table 2. Classification of Interval Cancers Previous mammogram Mammogram at diagnosis Frequency False-negative interval cancers Retrospectively seen seen 25~40% True interval cancers Retrospectively not seen seen 18~63% Occult cancers Retrospectively not seen not seen 8~12% 839
Han K C Figure 2. n interval cancer () (marked by marker) developing 6 months after previous screening mammogram () reported as negative. faint increased density is seen only on previous mediolateral oblique (MLO) view (; right) and it is not apparent even on retrospective evaluation. However, compared with the same view from one more year prior (C), there is an interval change, suggesting of neodensity. Patient s chest wall concavity caused difficult mammographic positioning, especially in craniocaudal (CC) view. Figure 3. n interval cancer due to underestimation of calcifications. screening mammogram () showed a few punctate microcalcifications in clustered distribution (arrow), and was categorized as category 3, probably benign without taking a magnification mammogram. The Patient revisited the same hospital after 16 months and complained of palpable mass. Followup mammogram () shows increasing number and apparent pleomorphism of microcalcifications, prompting a biopsy and diagnosis of microinvasive ductal carcinoma. 840
Imaging Findings of Missed reast Cancer C D Figure 4. Initial MLO () and craniocaudal () mammograms show a nonspecific parenchymal asymmetry without mass or distortion (arrows), which is attributable to physiologic asymmetry (arrow). Follow-up MLO (C) and craniocaudal (D) mammograms obtained 7 months later show slightly prominent density (arrows) and skin thickening. Histologic diagnosis was invasive ductal cancer. 841
Han K Figure 5. mammogram marked by computer-aided detection program. circle and a rectangle mark an abnormal density and an area of microcalcifications, each. - 842
Imaging Findings of Missed reast Cancer Figure 6. Subsequent screen-detected cancer. Initial bilateral craniocaudal and MLO mammograms () show subtle small density (arrow) at the periphery of left breast only on craniocaudal view. Subsequent screening mammograms a year later () clearly show an illdefined mass (arrows). Figure 7. Subsequent screen-detected cancer. Initial mammogram () show several scattered microcalcifications, attributable to benign calcifications. Subsequent screening mammograms a year later () show increasing fine linear calcifications in a wider area. Histologic diagnosis was low-grade ductal carcinoma in situ. Figure 8. lternative imaging-detected cancer. Initial bilateral craniocaudal and MLO mammograms () was initially interpreted as negative. Ultrasonogram taken at the same time () showed an irregular mass at right breast. retrospective evaluation of mammogram show sarchitectural distortion with subtle increased density (arrow) at the corresponding site. 843
Han K C D Figure 9. lternative imaging-detected cancer. nnually taken right mammograms (, ) in a patient after left mastectomy show no apparent abnormality. Chest CT (C) incidentally detected a mass at medial portion of right breast (arrow). fter verification of abnormal mass with targeted ultrasound (D), a retrospective evaluation of mammograms () showed a subtle increased density (dotted arrow) at right upper posterior breast, which was denser than previous mammographic evaluation (). 844
Imaging Findings of Missed reast Cancer Figure 10. lternative imaging-detected cancer. Mammograms () in a patient with metastatic lymph node at right axilla show no apparent abnormality in breast. reast magnetic resonance imaging before (left) and after (right) contrast administration () detected an enhancing mass (arrow). fter verification of abnormal mass with targeted ultrasound (C), a retrospective evaluation of mammograms () showed a tiny mass (dotted arrow) at right upper posterior breast, which was histologically diagnosed as invasive ductal cancer. C 845
Han K C D Figure 11. Recalled but dismissed cancer. Right unilateral mammograms () in a patient after left mastectomy showed an abnormal density (arrow) only seen on MLO view. Problem solving breast ultrasonographic evaluation did not detect any suspicious lesion. captured sonogram of right upper outer quadrant (; left) show normal fibroglandular tissue and a captured image of right upper inner quadrant (; right) show a benign-appearing mass (), which cannot be accounted for mammographic density. On magnification mammogram (C), which was taken 6 months later an ill-defined mass is clearly seen. Re-evaluation sonographic evaluation (D) identified a 1.5 cm sized malignant mass at right middle outer portion, 9 o clock position. - 846
Imaging Findings of Missed reast Cancer C D Figure 12. Recalled but dismissed cancer. Initial mammogram () shows subtle microcalcifications (arrows) in a whole upper outer quadrant, which were poorly defined due to poor compression and contrast. Initial sonogram of the same quadrant () shows a normal appearing fibroglandular tissue. Six months follow-up mammogram (C) shows pleomorphic microcalcifications (arrows), better seen with improved compression and exposure. Follow-up sonogram of the same site (D) shows multiple microcalcifications (arrows), which was confirmed as ductal carcinoma in situ. 847
Han K 11. Fletcher SW, lack W, Harris RP, Rimer K, Shapiro S. Report of the International Workshop on Screening for reast Cancer. J Natl Cancer Inst 1993; 85: 1644-1656. 12. Harris R, Leininger L. Clinical strategies for breast cancer screening: weighing and using the evidence. nn Intern Med. 1995; 112: 539-547. 13. Kerlikowske K, Grady D, Rubin S, Sandrock C, Ernster VL. Efficacy of screening mammography. meta-analysis. JM 1995; 273: 149-154. 14. Taba r L, Fagerberg CJ, Gad, aldetorp L, Holmberg LH, Gröntoft O, Ljungquist U, Lundström, Månson JC, Eklund G. Reduction in mortality from breast cancer after mass screening with mammography. Randomised trial from the reast Cancer Screening Working Group of the Swedish National oard of Health and Welfare. Lancet 1985; 1: 829-832. 15. Houssami N, Irwig L, Ciatto S. Radiological surveillance of interval breast cancers in screening programmes. Lancet Oncol 2006; 7: 259-265. 16. Warren RM, Young JR, McLean L, Lyons K, Wilson R, Evans, Duffy SW, Warsi IM. Radiology review of the UKCCCR reast Screening Frequency Trial: potential improvements in sensitivity and lead time of radiological signs. Clin Radiol 2003; 58: 128-132. 17. Harvey J, Fajardo LL, Innis C. Preview mammograms on patients with impalpable breast carcinoma: Retrospective vs blind interpretation. JR m J Roentgenol 1993; 161: 1167-1172. 18. Warren R, Duffy S. Interval cancers as an indicator of performance in breast screening. reast Cancer 2000; 7: 9-18. 19. uist, DS, Porter PL, Lehnman C, Taplin SH, White E. Factors contributing to mammography failure in women aged 40-49 years. J Natl Cancer Inst 2004; 96: 1432-1440. 10. renner RJ. False-negative mammograms. Medical, legal, and risk management implications. Radiol Clin North m 2000; 38: 741-757. 11. Han K, Kim JY, Shin JH, Choi YH. Performance of computeraided detection in false negative screening mammograms of breast cancers. J Korean Radiol Soc 2004; 51: 465-472. 12. Soo MS, aker J, Rosen EL. Sonographic detection and sonographically guided biopsy of breast microcalcifications. JR m J Roentgenol 2003; 180: 941-948. 13. Majid S, de Paredes ES, Doherty RD, Sharma NR, Salvador X. Missed reast Carcinoma: Pitfalls and Pearls. Radiographics 2003; 23: 881-895. Peer Reviewers Commentary 848