Comparison of Tomosynthesis Plus Digital Mammography and Digital Mammography Alone for Breast Cancer Screening 1

Transcription

1 Note: This copy is for your personal non-commercial use only. To order presentation-ready copies for distribution to your colleagues or clients, contact us at Original Research n Breast Imaging Brian M. Haas, MD Vivek Kalra, MD Jaime Geisel, MD Madhavi Raghu, MD Melissa Durand, MD Liane E. Philpotts, MD Comparison of Tomosynthesis Plus Digital Mammography and Digital Mammography Alone for Breast Cancer Screening 1 Purpose: Materials and Methods: To compare screening recall rates and cancer detection rates of tomosynthesis plus conventional digital mammography to those of conventional digital mammography alone. All patients presenting for screening mammography between October 1, 2011, and September 30, 2012, at four clinical sites were reviewed in this HIPAA-compliant retrospective study, for which the institutional review board granted approval and waived the requirement for informed consent. Patients at sites with digital tomosynthesis were offered screening with digital mammography plus tomosynthesis. Patients at sites without tomosynthesis underwent conventional digital mammography. Recall rates were calculated and stratified according to breast density and patient age. Cancer detection rates were calculated and stratified according to the presence of a risk factor for breast cancer. The Fisher exact test was used to compare the two groups. Multivariate logistic regression was used to assess the effect of screening method, breast density, patient age, and cancer risk on the odds of recall from screening. 1 From the Department of Diagnostic Radiology, Yale University School of Medicine, PO Box , New Haven, CT From the 2012 RSNA Aual Meeting. Received February 5, 2013; revision requested March 26; revision received June 12; accepted June 14; final version accepted June 19. Address correspondence to L.E.P. ( liane.philpotts@yale.edu). q RSNA, 2013 Results: Conclusion: A total of patients presented for screening mammography; 6100 received tomosynthesis. The overall recall rate was 8.4% for patients in the tomosynthesis group and 12.0% for those in the conventional mammography group (P,.01). The addition of tomosynthesis reduced recall rates for all breast density and patient age groups, with significant differences (P,.05) found for scattered fibroglandular, heterogeneously dense, and extremely dense breasts and for patients younger than 40 years, those aged years, those aged years, and those aged years. These findings persisted when multivariate logistic regression was used to control for differences in age, breast density, and elevated risk of breast cancer. The cancer detection rate was 5.7 per 1000 in patients receiving tomosynthesis versus 5.2 per 1000 in patients receiving conventional mammography alone (P =.70). Patients undergoing tomosynthesis plus digital mammography had significantly lower screening recall rates. The greatest reductions were for those younger than 50 years and those with dense breasts. A nonsignificant 9.5% increase in cancer detection was observed in the tomosynthesis group. q RSNA, radiology.rsna.org n Radiology: Volume 269: Number 3 December 2013

2 The use of screen-film mammography in breast cancer screening has been proved to reduce mortality in women older than 40 years (1). In certain patient populations, conventional full-field digital mammography has shown an incremental benefit in the ability to detect cancer and is currently the standard of care for breast cancer screening (2,3). However, conventional mammography is an imperfect modality and has been faulted for its false-positive rate and sensitivity, particularly in women with dense tissue (4). Digital breast tomosynthesis promises to address both of these concerns (5,6). Tomosynthesis allows for the three-dimensional reconstruction of the breast tissue, which can then be viewed as sequential sections through the breast. Tomosynthesis has been shown to increase the conspicuity of many lesions while reducing false-positive findings from summation of overlapping tissues (7). Recent work has evaluated the performance of tomosynthesis in a variety of observer performance studies, which collectively have established that the combination of digital breast tomosynthesis with conventional digital mammography can decrease screening mammography recall rates without having a negative effect on sensitivity in the detection of Advances in Knowledge The addition of two-view tomosynthesis to conventional digital mammography during screening examinations resulted in a 29.7% decrease in recall rates (P,.01). Reductions in recall rates held across women of all breast densities and ages and were statistically significant (P,.05) for all subgroups except those with predominantly fatty breast density and those aged 70 years and older. The greatest reductions in recall rates occurred for patients with dense breasts and those younger than 50 years (,20 examinations needed to prevent one recall). malignancy (8 10). More recently, the Oslo Tomosynthesis Screening Trial is the first published trial to implement screening tomosynthesis on a large scale (11). The Oslo study, which included patients, found a 27% increase in cancer detection for combined mammography plus tomosynthesis compared with mammography alone but only a 15% decrease in falsepositive recall rates. It is unknown how these results will translate to a U.S. screening program. It is also unknown which patients will experience the greatest benefit, if any, from tomosynthesis. We sought to evaluate the performance of breast tomosynthesis in clinical practice by identifying which patients experience the greatest reduction in screening mammography recall rates and evaluating the cancer detection rate. We hypothesized that the greatest reduction in recall rates will be for those patients with dense breasts. Our purpose was to compare screening recall rates and cancer detection rates of combined tomosynthesis plus conventional digital mammography to those of conventional digital mammography alone. Materials and Methods Two authors (M.R. and L.E.P.) are consultants for Hologic. An author without an industry-related conflict of interest (B.M.H.) had control of the data and information submitted for publication. The institutional review board waived the requirement to obtain Implications for Patient Care Tomosynthesis promises to reduce patient anxiety by reducing the number of women who are recalled from screening mammography. The greatest benefit from the reduced recall rates are realized by younger women and those with dense breasts, which could potentially increase compliance with screening mammography in two very sensitive patient populations. informed consent for this Health Insurance Portability and Accountability Act compliant study. All patients presenting for screening mammography in a 1-year period between October 1, 2011, and September 30, 2012, at four clinical sites were identified for review. The four clinical sites included breast imaging clinics located in an academic tertiary care hospital, in two outpatient radiology centers, and in a mobile van based mammography unit. All sites used digital mammography (Selenia; Hologic, Bedford, Mass). The tertiary care hospital site had a digital tomosynthesis unit (Dimensions, Hologic) for the entirety of the study period. One of the two outpatient radiology offices obtained a tomosynthesis unit in May 2012, midway through the study period. At sites where tomosynthesis was available, tomosynthesis was the preferred method of screening. When the machine was available, tomosynthesis was performed in combination with conventional digital mammography in all patients regardless of breast density and at no cost to the patient. Patients were informed before the examination of the use of tomosynthesis and could opt for conventional digital mammography alone. Patients were excluded from receiving tomosynthesis if they had breast implants or large breasts requiring tiled images. Tomosynthesis images were obtained in each breast with both craniocaudal and mediolateral oblique projections. Published online before print /radiol Content code: Radiology 2013; 269: Abbreviations: CI = confidence interval OR = odds ratio Author contributions: Guarantors of integrity of entire study, B.M.H., J.G., M.R., M.D., L.E.P.; study concepts/study design or data acquisition or data analysis/interpretation, all authors; manuscript drafting or manuscript revision for important intellectual content, all authors; manuscript final version approval, all authors; literature research, all authors; clinical studies, B.M.H., J.G., M.D., L.E.P.; statistical analysis, B.M.H., V.K., J.G.; and manuscript editing, all authors Conflicts of interest are listed at the end of this article. Radiology: Volume 269: Number 3 December 2013 n radiology.rsna.org 695

3 Examinations were interpreted by eight dedicated breast imaging radiologists with experience varying from 2 to 23 years (J.G., M.R., M.D., and L.E.P. were among these radiologists). All readers were formally trained and certified in tomosynthesis interpretation. Training and certification was achieved through Hologic. One radiologist (L.E.P.) was trained in 2007 as part of the original Hologic tomosynthesis trials. Other radiologists received training and certification through 1-day Hologic training by means of either an on-site course (two radiologists) or a Web-based tutorial (five radiologists). Recall rates (Breast Imaging Reporting and Data System category 0) were calculated for the two groups overall and stratified according to breast density and age. Cancer detection rates for tomosynthesis plus mammography and mammography alone were also calculated by determining outcomes of all diagnostic mammograms and subsequent biopsies performed as a result of a Breast Imaging Reporting and Data System category 0 assessment at the screening examination. Cancer detection rates were stratified according to the presence of either a personal history of breast cancer or a first-degree relative with a history of breast cancer or the absence of either risk factor. The percentage of cancers that were invasive cancers was tabulated for both tomosynthesis plus mammography and mammography alone groups. The number of patients that need to be screened to prevent one recall or to detect one additional cancer was calculated by comparing recall and cancer detection rates between tomosynthesis plus mammography and mammography alone groups. The Fisher exact test was performed to determine the statistical significance in the recall rates between the two screening methods and between the cancer detection rates overall and stratified according to patient risk factors. Multivariate logistic regression analysis was performed to ascertain the difference after controlling for the possible confounding factors, including age, breast density, and either personal or family history of breast cancer. A similar analysis was conducted within age and breast density subgroups to identify those populations that most benefited from tomosynthesis. A two-sided P value of.05 was indicative of a significant difference. Statistical analysis was performed by using Excel v14.3 (Microsoft, Redmond, Wash), SPSS v19.0 (IBM, Armonk, NY), and SAS 9.2 (SAS, Cary, NC). Results During the study period, patients presented for screening mammography; 6100 patients underwent tomosynthesis plus mammography and 7058 underwent conventional digital mammography alone. Although the breast densities and ages of the patients in the tomosynthesis plus mammography and mammography alone groups are similar, patients receiving tomosynthesis were more likely to have a personal history of breast cancer or a family history of breast cancer. Baseline characteristics of patients receiving tomosynthesis plus mammography and those receiving mammography alone are shown in Table 1. Table 1 The overall recall rate was 8.4% (95% confidence interval [CI]: 7.7%, 9.1%) for patients in the tomosynthesis group and 12.0% (95% CI: 11.3%, 12.8%) for those in the conventional imaging alone group (P,.01). Recall rates stratified according to breast density (Table 2) demonstrate reduced recall rates with tomosynthesis plus mammography than with mammography alone for all density groups, with significant differences found for scattered fibroglandular density (7.9% vs 10.6%, respectively; P,.01), heterogeneously dense breasts (10.2% vs 16.7%, P,.01), and extremely dense breasts (6.7% vs 15.6%, P,.01). The number of patients needed to screen to prevent one recall (Table 2) was 37.9 for scattered fibroglandular breast density, 15.2 for heterogeneously dense breasts, and 11.2 for extremely dense breasts. When stratified according to patient age, reduced recall rates were seen with tomosynthesis plus mammography than with mammography alone for all age groups, with significant differences observed in patients younger than 40 years (11.0% vs 25.0%, respectively; Characteristics of Patients Screened with Tomosynthesis Plus Mammography versus Those Screened with Conventional Digital Mammography Alone Parameter Tomosynthesis Plus Mammography (n = 6100) Breast density Predominantly fatty 536 (8.8) 973 (13.8) Scattered fibroglandular 2917 (47.8) 3913 (55.4) Heterogeneously dense 2295 (37.6) 1947 (27.6) Extremely dense 344 (5.6) 211 (3.0) Age (y)*, (2.7) 144 (2.0) (30.6) 1881 (26.7) (33.3) 2311 (32.7) (23.3) 1667 (23.6) (10.1) 1055 (15.0) Personal history of breast cancer 335 (5.5) 197 (2.8) First-degree relative with history of breast cancer 1144 (18.8) 1125 (15.9) Conventional Digital Mammography Alone (n = 7058) Note. Numbers in parentheses are percentages. * The mean patient age was 55.8 years for patients screened with tomosynthesis plus mammography and 57.5 years for those screened with conventional digital mammography alone. 696 radiology.rsna.org n Radiology: Volume 269: Number 3 December 2013

4 Table 2 Recall Rates for Patients Screened with Tomosynthesis Plus Mammography versus Those Screened with Conventional Digital Mammography Alone Parameter Tomosynthesis Plus Mammography Recall Rate (%) Conventional Digital Mammography Alone P Value Percentage Reduction in Recall Rate* No. of Tomosynthesis Plus Mammography Examinations Needed to Prevent One Recall Overall , (19.1, 36.5) 28.0 Breast density Predominantly fatty (27.8, 54.5) 46.3 Scattered fibroglandular , (12.5, 35.7) 37.9 Heterogeneously dense , (29.1, 48.2) 15.2 Extremely dense , (29.2, 74.2) 11.2 Age (y), , (25.7, 73.7) , (24.2, 45.7) , (12.7, 44.6) (12.3, 44.6) (221.3, 41.0) 82.6 * Numbers in parentheses are 95% CIs. P,.01), patients aged years (10.4% vs 16.3%, P,.01), patients aged years (7.6% vs 10.6%, P,.01), and patients aged years (7.4% vs 10.7%, P =.01). The number of patients needed to screen to prevent one recall was 7.2 for patients younger than 40 years, 17.2 for patients aged years, 33.7 for patients aged years, and 31.0 for patients aged years. The differences in recall rates for patients older than 70 years and for patients with predominantly fatty breast density were not statistically significant. Simple and multivariate logistic regression analysis (Table 3) found that younger age and greater breast density were significantly associated with higher risk of recall. The adjusted odds ratio (OR) (after controlling for age, breast density, and risk factor for breast cancer) of recall for tomosynthesis plus mammography versus conventional digital mammography alone was 0.62 (95% CI: 0.55, 0.70; P,.0001), indicating a 38% reduction in odds of recall for tomosynthesis comparing to conventional mammography after adjusting for age, breast density, and risk factor for breast cancer. Age and breast density were independent predictors of the risk of recall. Table 3 Association of Screening Methods, Age, Breast Density, and History of Breast Cancer with Risk of Recall at Screening Mammography Parameter Unadjusted OR* P Value Adjusted OR P Value Screening method,.0001,.0001 Tomosynthesis plus mammography 0.68 (0.60, 0. 76) 0.62 (0.55, 0.70) Conventional digital mammography alone Age (y),.0001,.0001, (1.88, 3.76) 2.59 (1.83, 3.69) (1.57, 2.37) 1.84 (1.49, 2.27) (1.03, 1.56) 1.26 (1.01, 1.55) (1.01, 1.57) 1.31 (1.05, 1.63) Breast density,.0001,.0001 Extremely dense 1.63 (1.16, 2.31) 1.56 (1.10, 2.21) Heterogeneously dense 2.21 (1.77, 2.76) 2.17 (1.73, 2.73) Scattered fibroglandular 1.52 (1.22, 1.89) 1.53 (1.23, 1.91) Predominantly fatty Breast cancer risk factor Present 0.75 (0.55, 1.03) 0.98 (0.85, 1.13) Absent * Unadjusted OR refers to simple logistic regression not adjusting for confounding variables. Numbers in parentheses are 95% CIs. Adjusted OR refers to multivariate logistic regression analysis on the association of screen methods with the risk of recall after adjusting for age, breast density, and presence of a risk factor for breast cancer. Numbers in parentheses are 95% CIs. Patients younger than 40 years had the greatest reduction in odds of recall from tomosynthesis plus mammography screening, with an adjusted OR of 0.36 (95% CI: 0.19, 0.67; P =.001; Figure). Reductions in odds of recall were also significant for patients aged years (adjusted OR: 0.56; 95% CI: 0.46, 0.69; P,.001), patients aged years (adjusted OR: 0.67; 95% Radiology: Volume 269: Number 3 December 2013 n radiology.rsna.org 697

5 Adjusted ORs and 95% CIs of recall for digital breast tomosynthesis plus mammography versus conventional digital mammography alone according to age and breast density. CI: 0.54, 0.83; P,.001), and patients aged years (adjusted OR: 0.67; 95% CI: 0.52, 0.86; P =.002) but not among patients older than 70 years (adjusted OR: 0.78; 95% CI: 0.53, 1.17; P =.23).When multivariate regression analysis was performed among breast density subgroups, reductions in odds of recall appeared to be greatest for patients with extremely dense breasts (adjusted OR: 0.40; 95% CI: 0.22, 0.70; P =.002). Reductions in odds of recall were also significant for patients with heterogeneously dense breasts (adjusted OR: 0.55; 95% CI: 0.46, 0.66; P,.001) and those with scattered fibroglandular breast density (adjusted OR: 0.72; 95% CI: 0.61, 0.85; P,.001) but not among patients with predominantly fatty breasts (adjusted OR: 0.65; 95% CI: 0.41, 1.04; P =.07). Cancer detection rates were similar between the two groups (Table 4). In sum, 35 cancers were found in patients receiving tomosynthesis, for an overall cancer detection rate of 5.7 per In patients receiving mammography alone, there were 37 cancers, for a detection rate of 5.2 per 1000 (P =.70). This corresponds to a 9.5% increase in the cancer detection rate with tomosynthesis or a need to screen 2018 women with tomosynthesis to detect one additional cancer beyond what would have been detected with mammography alone. To ascertain the effect of the increased prevalence of patients with either a personal history of breast cancer Table 4 Cancer Detection Rates according to Risk for Breast Cancer and Screening Method Parameter Tomosynthesis Plus Mammography or a first-degree relative with breast cancer in the tomosynthesis plus mammography group, cancer detection rates were calculated for both high-risk and non-high-risk groups (Table 4). Among patients at increased risk for breast cancer, the cancer detection rate is 8.6 per 1000 patients in the tomosynthesis group and 7.9 per 1000 patients in the conventional mammography group (P =.83). For patients not at increased risk for breast cancer, the cancer detection rate is 5.1 per 1000 patients in the tomosynthesis group and 4.5 per 1000 patients in the conventional mammography group (P =.65). Of cancers in the tomosynthesis plus mammography group, 24 of 35 cancers (69%) were infiltrating carcinoma, compared with 25 of 37 cancers (68%) in the mammography alone group (P =.93). Discussion Conventional Digital Mammography Alone No. of cancers All patients Increased risk Baseline risk No. of patients All patients Increased risk Baseline risk Cancer detection rate All patients 5.7 per per Increased risk 8.6 per per Baseline risk 5.1 per per P Value Note. Patients at increased risk are those with either a personal history of breast cancer or a first-degree relative with breast cancer. Patients at baseline risk are those with neither of these risk factors. In our nonrandomized retrospective examination, recall rates were significantly lower in patients who underwent tomosynthesis plus conventional mammography than in those who underwent conventional mammography alone. These reductions in recall rates were observed 698 radiology.rsna.org n Radiology: Volume 269: Number 3 December 2013

6 across all breast densities and patient ages. Statistical significance was observed for all subgroups except patients with predominantly fatty breast density and patients older than 70 years. These results persisted when multivariate logistic regression was used to control for differences in breast density, age, and presence of a risk factor for breast cancer. Although all breast density and age subgroups benefitted from the addition of tomosynthesis to conventional digital mammography, the patients receiving the greatest benefit were women with dense breasts and those younger than 50 years. Cancer detection rates were similar between the two groups. These results are in keeping with those from previous studies that showed that tomosynthesis has the potential to reduce screening recall rates. Our study showed a 30% reduction in recall rate overall (8.4% compared with 12.0%). Poplack et al (8) demonstrated that tomosynthesis can mitigate patient recalls for pseudolesions caused by summation of overlapping tissue, with a reduction in recall rates of up to 40%. However, the study by Poplack et al only examined the effect of the addition of tomosynthesis on patients who had a conventional mammogram for which further imaging or diagnostic evaluation was necessary; it did not evaluate whether tomosynthesis used in clinical practice would help identify patients for recall who would otherwise not have been recalled with mammography alone. Therefore, the 40% reduction in recall rates identified by Poplack et al may be higher than what could actually be achieved in clinical practice. Two subsequent studies contained both negative and positive screening examinations in an observer performance study design. Gur et al (9) demonstrated a 30% reduction in recall rates with the addition of tomosynthesis to conventional mammography. Meanwhile, Rafferty et al (10) performed a receiver operating characteristic analysis and demonstrated an increase in the area under the curve of approximately 7% with the addition of two-view tomosynthesis. Our results are consistent with data from these earlier studies. Our results are also consistent with those of the other large tomosynthesis screening study, the Oslo Tomosynthesis Screening Trial (11). The Oslo trial showed a more modest reduction in recall rate (15%) and a higher increase in cancer detection (27%). The baseline recall rate for mammography alone in our study is higher than that in the Oslo study. Therefore, with higher recalls from screening, it follows that our program should catch more cancers in the mammography alone group (ie, increased sensitivity) at the cost of reduced specificity. The addition of tomosynthesis allowed us to realize improvements in specificity without observing large increases in sensitivity, which is likely already relatively high. The value of decreasing screening mammography recall rates is immense. In particular, the greatest benefits are seen in younger patients and those with dense breasts, precisely the population in which conventional mammography is most limited. Decreased recall rates can be directly translated into less patient anxiety and decreased costs for additional diagnostic examinations (12 14). As well, for patients who are recalled, tomosynthesis performed during screening has the potential to abbreviate the subsequent diagnostic examination; tomosynthesis recently has been shown to be comparable to spot compression views for the characterization of masses and may obviate the need for some specialized mammographic views (15). This could further improve the patient mammography experience. There currently are certain downsides to the use of tomosynthesis during screening mammography. Most notably, the tomosynthesis views approximately double the patient dose compared to that with mammography alone (5). Current estimates place the lifetime attributable risk of a radiationinduced fatal cancer from a single digital breast tomosynthesis acquisition performed at age 40 years at per examinations (16). This concern may shortly be obviated because the U.S. Food and Drug Administration has approved technology that allows for the creation of a synthesized two-dimensional mammogram from the tomosynthesis data set (17). This technology has the potential to eliminate the need for acquisition of the conventional mammograms in addition to the tomosynthesis images. However, it is unknown how this technology will perform in clinical practice and whether a separate two-dimensional acquisition can be omitted. The similar cancer detection rates between combined tomosynthesis plus mammography and mammography alone is encouraging. This study was underpowered to demonstrate small differences in the cancer detection rate. A much larger number of patients would be necessary to achieve such power. However, we did observe a nonsignificant 9.5% increase in cancer detection. It is paramount to note that the addition of tomosynthesis did not adversely decrease cancer detection rates by falsely dismissing cancerous findings. Continued follow-up of patients in this study and examination of cancers detected with the two modalities will be important to assess and is the subject of ongoing investigation. The cancer detection rate in the group receiving conventional digital mammography alone is slightly confounded by the possible use of tomosynthesis during the subsequent diagnostic work-up of patients recalled from screening. Of the four study sites, diagnostic work-ups are performed at the academic teaching hospital and the one outpatient center that also has tomosynthesis, and so any patient presenting for diagnostic mammography might undergo tomosynthesis during that workup. The use of tomosynthesis later during a diagnostic examination in the patients who initially underwent only mammography might have identified cancers that otherwise would not have been detected or sampled for biopsy; this could have increased the cancer detection rate above what it otherwise would have been in the mammography alone group. Therefore, the difference in the cancer detection rates between the tomosynthesis and mammography alone groups could actually be greater than what was observed. Radiology: Volume 269: Number 3 December 2013 n radiology.rsna.org 699

7 There are several directions for further investigation. A larger patient series or multi-institutional trial to more accurately assess the tomosynthesis cancer detection rate during screening examinations would be valuable. Larger numbers could strengthen the differences in recall rates in the individual breast density or age groups, possibly achieving statistical significance for those groups that are underpowered in our study specifically the women with predominantly fatty breast density and those aged 70 years and older. As well, little is known about how screening tomosynthesis and screening wholebreast ultrasonography (US) compare; such an investigation would be particularly valuable considering that our data support the value of tomosynthesis in women with dense breasts, which is precisely the population in which screening breast US is gaining popularity (18). The limitations of our study are that data were collected in a retrospective maer and patients were not randomized. Also, only patients at the sites offering tomosynthesis had the potential to be included in the tomosynthesis plus mammography group. Differences in the patient populations presenting to each of these sites could have biased the study results. Patients receiving tomosynthesis plus mammography were more likely to have had breast cancer themselves or to have a first-degree relative with breast cancer, which could potentially affect both the recall rates and cancer incidence among these patients. Because the recall rates might be expected to be higher in such patients, the true reduction in recall rates with tomosynthesis may in fact be greater than our reported rate. Meanwhile, when the high-risk patients are separated from the remainder of the study population, the increased cancer detection rates persist, suggesting that the trend toward increased cancer detection observed in this study is not substantially based on the higher risk profile of patients in the tomosynthesis plus mammography group. In summary, in our study we found that breast tomosynthesis reduces screening mammography recall rates, particularly for younger women and women with dense breasts, without significant changes in cancer detection. When adopted, tomosynthesis promises to alter the mammography workflow by decreasing screening recall rates and to subsequently improve the mammography experience for many women. Acknowledgment: The authors kindly thank Fangyong Li, MPH, MS, for his assistance in statistical analysis. Disclosures of Conflicts of Interest: B.M.H. No relevant conflicts of interest to disclose. V.K. No relevant conflicts of interest to disclose. J.G. No relevant conflicts of interest to disclose. M.R. Financial activities related to the present article: none to disclose. Financial activities not related to the present article: is a paid consultant for Hologic. M.D. No relevant conflicts of interest to disclose. L.E.P. Financial activities related to the present article: none to disclose. Financial activities not related to the present article: is a paid consultant for Hologic. Other relationships: none to disclose. References 1. Fletcher SW, Black W, Harris R, Rimer BK, Shapiro S. Report of the International Workshop on Screening for Breast Cancer. J Natl Cancer Inst 1993;85(20): Lewin JM, Hendrick RE, D Orsi CJ, et al. Comparison of full-field digital mammography with screen-film mammography for cancer detection: results of 4,945 paired examinations. Radiology 2001;218(3): Pisano ED, Gatsonis C, Hendrick E, et al. Diagnostic performance of digital versus film mammography for breast-cancer screening. N Engl J Med 2005;353(17): [Published correction appears in N Engl J Med 2006;355(17):1840.] 4. Nelson HD, Tyne K, Naik A, et al. Screening for breast cancer: an update for the U.S. Preventive Services Task Force. A Intern Med 2009;151(10): , W237 W Baker JA, Lo JY. Breast tomosynthesis: state-of-the-art and review of the literature. Acad Radiol 2011;18(10): Park JM, Franken EA Jr, Garg M, Fajardo LL, Niklason LT. Breast tomosynthesis: present considerations and future applications. RadioGraphics 2007;27(Suppl 1):S231 S Niklason LT, Christian BT, Niklason LE, et al. Digital tomosynthesis in breast imaging. Radiology 1997;205(2): Poplack SP, Tosteson TD, Kogel CA, Nagy HM. Digital breast tomosynthesis: initial experience in 98 women with abnormal digital screening mammography. AJR Am J Roentgenol 2007;189(3): Gur D, Abrams GS, Chough DM, et al. Digital breast tomosynthesis: observer performance study. AJR Am J Roentgenol 2009; 193(2): Rafferty EA, Park JM, Philpotts LE, et al. Assessing radiologist performance using combined digital mammography and breast tomosynthesis compared with digital mammography alone: results of a multicenter, multireader trial. Radiology 2013;266(1): Skaane P, Bandos AI, Gullien R, et al. Comparison of digital mammography alone and digital mammography plus tomosynthesis in a population-based screening program. Radiology 2013;267(1): Defrank JT, Brewer N. A model of the influence of false-positive mammography screening results on subsequent screening. Health Psychol Rev 2010;4(2): DeFrank JT, Rimer BK, Bowling JM, Earp JA, Breslau ES, Brewer NT. Influence of false-positive mammography results on subsequent screening: do physician recommendations buffer negative effects? J Med Screen 2012;19(1): Brewer NT, Salz T, Lillie SE. Systematic review: the long-term effects of falsepositive mammograms. A Intern Med 2007;146(7): Noroozian M, Hadjiiski L, Rahnama-Moghadam S, et al. Digital breast tomosynthesis is comparable to mammographic spot views for mass characterization. Radiology 2012; 262(1): Hendrick RE. Radiation doses and cancer risks from breast imaging studies. Radiology 2010;257(1): FDA panel gives nod to 2D synthetic mode for Hologic breast tomo. Aunt Miie Web site. sec=ser&sub=def&pag=dis&itemid= Accessed April 15, Berg WA, Blume JD, Cormack JB, et al. Combined screening with ultrasound and mammography vs mammography alone in women at elevated risk of breast cancer. JAMA 2008;299(18): radiology.rsna.org n Radiology: Volume 269: Number 3 December 2013