Discordant and false-negative interpretations at digital breast tomosynthesis in the prospective Oslo Tomosynthesis Screening Trial (OTST) using independent double reading.

OBJECTIVES: To analyze discordant and false-negatives of double reading digital breast tomosynthesis (DBT) versus digital mammography (DM) including reading times in the Oslo Tomosynthesis Screening Trial (OTST), and reclassify these in a retrospective reader study as missed, minimal sign, or true-negatives. METHODS: The prospective OTST comparing double reading DBT vs. DM had paired design with four parallel arms: DM, DM + computer aided detection, DBT + DM, and DBT + synthetic mammography. Eight radiologists interpreted images in batches using a 5-point scale. Reading time was automatically recorded. A retrospective reader study including four radiologists classified screen-detected cancers with at least one false-negative score and screening examinations of interval cancers as negative, non-specific minimal sign, significant minimal sign, and missed; the two latter groups are defined "actionable." Statistics included chi-square, Fisher's exact, McNemar's, and Mann-Whitney U tests. RESULTS: Discordant rate (cancer missed by one reader) for screen-detected cancers was overall comparable (DBT (31% [71/227]) and DM (30% [52/175]), p = .81), significantly lower at DBT for spiculated cancers (DBT, 19% [20/106] vs. DM, 36% [38/106], p = .003), but high (28/49 = 57%, p = 0.001) for DBT-only detected spiculated cancers. Reading time and sensitivity varied among readers. False-negative DBT-only detected spiculated cancers had shorter reading time than true-negatives in 46% (13/28). Retrospective evaluation classified the following DBT exams "actionable": three missed by both readers, 95% (39/41) of discordant cancers detected by both modes, all 30 discordant DBT-only cancers, 25% (13/51) of interval cancers. CONCLUSIONS: Discordant rate was overall comparable for DBT and DM, significantly lower at DBT for spiculated cancers, but high for DBT-only detected spiculated lesions. Most false-negative screen-detected DBT were classified as "actionable." CLINICAL RELEVANCE STATEMENT: Retrospective evaluation of false-negative interpretations from the Oslo Tomosynthesis Screening Trial shows that most discordant and several interval cancers could have been detected at screening. This underlines the potential for modern AI-based reading aids and triage, as high-volume screening is a demanding task. KEY POINTS: • Digital breast tomosynthesis (DBT) screening is more sensitive and has higher specificity compared to digital mammography screening, but high-volume DBT screening is a demanding task which can result in high discordance rate among readers. • Independent double reading DBT screening had overall comparable discordance rate as digital mammography, lower for spiculated masses seen on both modalities, and higher for small spiculated cancer seen only on DBT. • Almost all discordant digital breast tomosynthesis-detected cancers (72 of 74) and 25% (13 of 51) of the interval cancers in the Oslo Tomosynthesis Screening Trial were retrospectively classified as actionable and could have been detected by the readers.

Digital Mammography versus Breast Tomosynthesis: Impact of Breast Density on Diagnostic Performance in Population-based Screening.

BackgroundPrevious studies comparing digital breast tomosynthesis (DBT) to digital mammography (DM) have shown conflicting results regarding breast density and diagnostic performance.PurposeTo compare true-positive and false-positive interpretations in DM versus DBT according to volumetric density, age, and mammographic findings.Materials and MethodsFrom November 2010 to December 2012, 24 301 women aged 50-69 years (mean age, 59.1 years ± 5.7) were prospectively included in the Oslo Tomosynthesis Screening Trial. Participants received same-compression DM and DBT with independent double reading for both DM and DM plus DBT reading modes. Eight experienced radiologists rated the images by using a five-point scale for probability of malignancy. Participants were followed up for 2 years to assess for interval cancers. Breast density was assessed by using automatic volumetric software (scale, 1-4). Differences in true-positive rates, false-positive rates, and mammographic findings were assessed by using confidence intervals (Newcombe paired method) and P values (McNemar and χ2 tests).ResultsThe true-positive rate of DBT was higher than that of DM for density groups (range, 12%-24%; P < .001 for density scores of 2 and 3, and P > .05 for density scores of 1 and 4) and age groups (range, 15%-35%; P < .05 for all age groups), mainly due to the higher number of spiculated masses and architectural distortions found at DBT (P < .001 for density scores of 2 and 3; P < .05 for women aged 55-69 years). The false-positive rate was lower for DBT than for DM in all age groups (range, -0.6% to -1.2%; P < .01) and density groups (range, -0.7 to -1.0%; P < .005) owing to fewer asymmetric densities (P ≤ .001), except for extremely dense breasts (0.1%, P = .82).ConclusionDigital breast tomosynthesis enabled the detection of more cancers in all density and age groups compared with digital mammography, especially cancers classified as spiculated masses and architectural distortions. The improvement in cancer detection rate showed a positive correlation with age. With use of digital breast tomosynthesis, false-positive findings were lower due to fewer asymmetric densities, except in extremely dense breasts.© RSNA, 2019Online supplemental material is available for this article.See also the editorial by Fuchsjäger and Adelsmayr in this issue.

Digital Mammography versus Digital Mammography Plus Tomosynthesis in Breast Cancer Screening: The Oslo Tomosynthesis Screening Trial.

Background Digital breast tomosynthesis (DBT) is replacing digital mammography (DM) in the clinical workflow. Currently, there are limited prospective studies comparing the diagnostic accuracy of both examinations and the role of synthetic mammography (SM) and computer-aided detection (CAD). Purpose To compare the accuracy of DM versus DM + DBT in population-based breast cancer screening. Materials and Methods This prospective study, performed from November 2010 to December 2012, included 24 301 women (mean age, 59.1 years ± 5.7 [standard deviation]) with 281 cancers, of which 51 were interval cancers. Each examination was independently interpreted with four reading modes: DM, DM + CAD, DM + DBT, and SM + DBT. Sensitivity and specificity were compared for DM versus DM + DBT, DM versus DM + CAD, DM + DBT versus SM + DBT, and DM versus DM + DBT at double reading. Reader-adjusted performance characteristics of reading modes were evaluated on the basis of pre-arbitration (initial interpretation) scores. Statistical analysis was based on cluster bootstrap analysis using 10 000 random resamples. Results Sensitivity was 54.1% (152 of 281) for DM and 70.5% (198 of 281) for DM + DBT. Reader-adjusted difference was 12.6% (95% confidence interval [CI]: 5.2%, 19.7%; P = .001). Specificity was 94.2% (false-positive fraction [FPF], 5.8%; 1388 of 24 020) for DM and 95.0% (FPF, 5.0%; 1209/24 020) for DM + DBT, with a reader-adjusted difference in FPF of -1.2% (95% CI: -1.7%, -0.7%; P < .001). Sensitivity was 69.0% (194 of 281) for SM + DBT and 70.5% (198 of 281) for DM + DBT, with a reader-adjusted difference of 1.0% (95% CI: -6.2%, 8.5%; P = .77). Specificity was 95.4% (FPF, 4.6%; 1111 of 24 020) for SM + DBT and 95.0% (FPF, 5.0%;1209 of 24 020) for DM + DBT, with reader-adjusted 95% CIs for FPF of 4.7%, 5.4% and 5.0%, 5.7%, respectively, and a difference of -0.3% (95% CI: -0.8%, 0.2%; P = .23). Differences in sensitivity and specificity with the addition of CAD were small and not significant (P > .2). Conclusion Addition of digital breast tomosynthesis to digital mammography resulted in significant gains in sensitivity and specificity. Synthetic mammography in combination with digital breast tomosynthesis had similar sensitivity and specificity to digital mammography in combination with digital breast tomosynthesis. © RSNA, 2019 See also the editorial by Lång in this issue.

Digital Breast Tomosynthesis and Synthetic 2D Mammography versus Digital Mammography: Evaluation in a Population-based Screening Program.

Purpose To compare the performance of digital breast tomosynthesis (DBT) and two-dimensional synthetic mammography (SM) with that of digital mammography (DM) in a population-based mammographic screening program. Materials and Methods In this prospective cohort study, data from 37 185 women screened with DBT and SM and from 61 742 women screened with DM as part of a population-based screening program in 2014 and 2015 were included. Early performance measures, including recall rate due to abnormal mammographic findings, rate of screen-detected breast cancer, positive predictive value of recall, positive predictive value of needle biopsy, histopathologic type, tumor size, tumor grade, lymph node involvement, hormonal status, Ki-67 level, and human epidermal growth factor receptor 2 status were compared in women who underwent DBT and SM screening and in those who underwent DM screening by using χ2 tests, two-sample unpaired t tests, and tests of proportions. Results Recall rates were 3.4% for DBT and SM screening and 3.3% for DM screening (P = .563). DBT and SM screening showed a significantly higher rate of screen-detected cancer compared with DM screening (9.4 vs 6.1 cancers per 1000 patients screened, respectively; P < .001). The rate of detection of tumors 10 mm or smaller was 3.2 per 1000 patients screened with DBT and SM and 1.8 per 1000 patients screened with DM (P < .001), and the rate of grade 1 tumors was 3.3 per 1000 patients screened with DBT and SM versus 1.4 per 1000 patients screened with DM (P < .001). On the basis of immunohistochemical analyses, rates of lymph node involvement and tumor subtypes did not differ between women who underwent DBT and SM screening and those who underwent DM screening. Conclusion DBT and SM screening increased the detection rate of histologically favorable tumors compared with that attained with DM screening. © RSNA, 2018 Online supplemental material is available for this article.

Performance of breast cancer screening using digital breast tomosynthesis: results from the prospective population-based Oslo Tomosynthesis Screening Trial.

PURPOSE: Digital breast tomosynthesis (DBT) has the potential to overcome limitations of conventional mammography. This study investigated the effects of addition of DBT on interval and detected cancers in population-based screening. METHODS: Oslo Tomosynthesis Screening Trial (OTST) was a prospective, independent double-reading trial inviting women 50-69 years biennially, comparing full-field digital mammography (FFDM) plus DBT with FFDM alone. Performance indicators and characteristics of screen-detected and interval cancers were compared with two previous FFDM rounds. RESULTS: 24,301 consenting women underwent FFDM + DBT screening over a 2-year period. Results were compared with 59,877 FFDM examinations during prior rounds. Addition of DBT resulted in a non-significant increase in sensitivity (76.2%, 378/496, vs. 80.8%, 227/281, p = 0.151) and a significant increase in specificity (96.4%, 57229/59381 vs. 97.5%, 23427/24020, p < .001). Number of recalls per screen-detected cancer decreased from 6.7 (2530/378) to 3.6 (820/227) with DBT (p < .001). Cancer detection per 1000 women screened increased (6.3, 378/59877, vs. 9.3, 227/24301, p < .001). Interval cancer rate per 1000 screens for FFDM + DBT remained similar to previous FFDM rounds (2.1, 51/24301 vs. 2.0, 118/59877, p = 0.734). Interval cancers post-DBT were comparable to prior rounds but significantly different in size, grade, and node status from cancers detected only using DBT. 39.6% (19/48) of interval cancers had positive nodes compared with only 3.9% (2/51) of additional DBT-only-detected cancers. CONCLUSIONS: DBT-supplemented screening resulted in significant increases in screen-detected cancers and specificity. However, no significant change was observed in the rate, size, node status, or grade of interval cancers. ClinicalTrials.gov: NCT01248546.

Average glandular dose in paired digital mammography and digital breast tomosynthesis acquisitions in a population based screening program: effects of measuring breast density, air kerma and beam quality.

The main purpose was to compare average glandular dose (AGD) for same-compression digital mammography (DM) and digital breast tomosynthesis (DBT) acquisitions in a population based screening program, with and without breast density stratification, as determined by automatically calculated breast density (Quantra™). Secondary, to compare AGD estimates based on measured breast density, air kerma and half value layer (HVL) to DICOM metadata based estimates. AGD was estimated for 3819 women participating in the screening trial. All received craniocaudal and mediolateral oblique views of each breasts with paired DM and DBT acquisitions. Exposure parameters were extracted from DICOM metadata. Air kerma and HVL were measured for all beam qualities used to acquire the mammograms. Volumetric breast density was estimated using Quantra™. AGD was estimated using the Dance model. AGD reported directly from the DICOM metadata was also assessed. Mean AGD was 1.74 and 2.10 mGy for DM and DBT, respectively. Mean DBT/DM AGD ratio was 1.24. For fatty breasts: mean AGD was 1.74 and 2.27 mGy for DM and DBT, respectively. For dense breasts: mean AGD was 1.73 and 1.79 mGy, for DM and DBT, respectively. For breasts of similar thickness, dense breasts had higher AGD for DM and similar AGD for DBT. The DBT/DM dose ratio was substantially lower for dense compared to fatty breasts (1.08 versus 1.33). The average c-factor was 1.16. Using previously published polynomials to estimate glandularity from thickness underestimated the c-factor by 5.9% on average. Mean AGD error between estimates based on measurements (air kerma and HVL) versus DICOM header data was 3.8%, but for one mammography unit as high as 7.9%. Mean error of using the AGD value reported in the DICOM header was 10.7 and 13.3%, respectively. Thus, measurement of breast density, radiation dose and beam quality can substantially affect AGD estimates.

Interpretation of automated breast ultrasound (ABUS) with and without knowledge of mammography: a reader performance study.

BACKGROUND: Automated breast ultrasonography (ABUS) has the potential to be an important adjunct to mammography in women with dense breasts. PURPOSE: To compare reader performance and inter-observer variation of ABUS alone and in combination with mammography. MATERIAL AND METHODS: This retrospective study had ethical committee approval. All women gave written informed consent. One hundred and fourteen breasts in 90 women examined by digital mammography and ABUS were interpreted by five radiologists using BI-RADS categories. The 114 breasts included 38 cancers and 76 normal or benign findings. In the first reading session ABUS only was interpreted, and in the second ABUS plus digital mammography. Image interpretations were done without knowledge of clinical or imaging results. A consensus panel analyzed false negative and false positive interpretations. Reading time was recorded for one radiologist. AUC was used for performance measurement, and kappa statistic for inter-observer variability. RESULTS: Mean size for cancers was 16.2 mm; area under the curve (AUC) values for ABUS alone and for combined reading were, respectively: reader A, 0.592-0.744; reader B, 0.740-0.947; reader, C 0.759-0.823; reader D, 0.670-0.688; reader E, 0.904-0.923; and all readers combined 0.730-0.823. The higher AUC for combined reading was statistically significant (P < 0.05) for reader B and for all readers. There was a considerable inter-observer variability. Observer agreement revealed following kappa values for ABUS alone and combined reading, respectively: reader A, 0.22-0.30; reader B, 0.33-0.44; reader C, 0.32-0.39; reader D, 0.07-0.14; and reader E, 0.34-0.43. Shadowing from dense parenchyma was the most common cause of false positive ABUS interpretations. Mean interpretation time for a bilateral normal ABUS examination was 9 min. CONCLUSION: Observer agreement was higher and all radiologists improved diagnostic performance using combined ABUS and mammography interpretation. Combined reading should be standard if ABUS is implemented in screening of women with dense breasts.

Two-view digital breast tomosynthesis screening with synthetically reconstructed projection images: comparison with digital breast tomosynthesis with full-field digital mammographic images.

PURPOSE: To compare the performance of two versions of reconstructed two-dimensional (2D) images in combination with digital breast tomosynthesis (DBT) versus the performance of standard full-field digital mammography (FFDM) plus DBT. MATERIALS AND METHODS: This trial had ethical committee approval, and all participants gave written informed consent. Examinations (n = 24 901) in women between the ages of 50 and 69 years (mean age, 59.2 years) were interpreted prospectively as part of a screening trial that included independent interpretations of FFDM plus DBT and reconstructed 2D images plus DBT. Reconstructed 2D images do not require radiation exposure. Using analyses for binary data that accounted for correlated interpretations and were adjusted for reader-specific volume, two versions (initial and current) of reconstructed 2D images used during trial periods 1 (from November 22, 2010, to December 21, 2011; 12 631 women) and 2 (from January 20, 2012, to December 19, 2012; 12 270 women) were compared in terms of cancer detection and false-positive rates with the corresponding FFDM plus DBT interpretations. RESULTS: Cancer detection rates were 8.0, 7.4, 7.8, and 7.7 per 1000 screening examinations for FFDM plus DBT in period 1, initial reconstructed 2D images plus DBT in period 1, FFDM plus DBT in period 2, and current reconstructed 2D images plus DBT in period 2, respectively. False-positive scores were 5.3%, 4.6%, 4.6%, and 4.5%, respectively. Corresponding reader-adjusted paired comparisons of false-positive scores revealed significant differences for period 1 (P = .012) but not for period 2 (ratio = 0.99; 95% confidence interval: 0.88, 1.11; P = .85). CONCLUSION: The combination of current reconstructed 2D images and DBT performed comparably to FFDM plus DBT and is adequate for routine clinical use when interpreting screening mammograms.

Prospective trial comparing full-field digital mammography (FFDM) versus combined FFDM and tomosynthesis in a population-based screening programme using independent double reading with arbitration.

OBJECTIVES: To compare double readings when interpreting full field digital mammography (2D) and tomosynthesis (3D) during mammographic screening. METHODS: A prospective, Ethical Committee approved screening study is underway. During the first year 12,621 consenting women underwent both 2D and 3D imaging. Each examination was independently interpreted by four radiologists under four reading modes: Arm A-2D; Arm B-2D + CAD; Arm C-2D + 3D; Arm D-synthesised 2D + 3D. Examinations with a positive score by at least one reader were discussed at an arbitration meeting before a final management decision. Paired double reading of 2D (Arm A + B) and 2D + 3D (Arm C + D) were analysed. Performance measures were compared using generalised linear mixed models, accounting for inter-reader performance heterogeneity (P < 0.05). RESULTS: Pre-arbitration false-positive scores were 10.3 % (1,286/12,501) and 8.5 % (1,057/12,501) for 2D and 2D + 3D, respectively (P < 0.001). Recall rates were 2.9 % (365/12,621) and 3.7 % (463/12,621), respectively (P = 0.005). Cancer detection was 7.1 (90/12,621) and 9.4 (119/12,621) per 1,000 examinations, respectively (30 % increase, P < 0.001); positive predictive values (detected cancer patients per 100 recalls) were 24.7 % and 25.5 %, respectively (P = 0.97). Using 2D + 3D, double-reading radiologists detected 27 additional invasive cancers (P < 0.001). CONCLUSION: Double reading of 2D + 3D significantly improves the cancer detection rate in mammography screening. KEY POINTS: • Tomosynthesis-based screening was successfully implemented in a large prospective screening trial. • Double reading of tomosynthesis-based examinations significantly reduced false-positive interpretations. • Double reading of tomosynthesis significantly increased the detection of invasive cancers.

Comparison of digital mammography alone and digital mammography plus tomosynthesis in a population-based screening program.

PURPOSE: To assess cancer detection rates, false-positive rates before arbitration, positive predictive values for women recalled after arbitration, and the type of cancers detected with use of digital mammography alone and combined with tomosynthesis in a large prospective screening trial. MATERIALS AND METHODS: A prospective, reader- and modality-balanced screening study of participants undergoing combined mammography plus tomosynthesis, the results of which were read independently by four different radiologists, is under way. The study was approved by a regional ethics committee, and all participants provided written informed consent. The authors performed a preplanned interim analysis of results from 12,631 examinations interpreted by using mammography alone and mammography plus tomosynthesis from November 22, 2010, to December 31, 2011. Analyses were based on marginal log-linear models for binary data, accounting for correlated interpretations and adjusting for reader-specific performance levels by using a two-sided significance level of .0294. RESULTS: Detection rates, including those for invasive and in situ cancers, were 6.1 per 1000 examinations for mammography alone and 8.0 per 1000 examinations for mammography plus tomosynthesis (27% increase, adjusted for reader; P = .001). False-positive rates before arbitration were 61.1 per 1000 examinations with mammography alone and 53.1 per 1000 examinations with mammography plus tomosynthesis (15% decrease, adjusted for reader; P < .001). After arbitration, positive predictive values for recalled patients with cancers verified later were comparable (29.1% and 28.5%, respectively, with mammography alone and mammography plus tomosynthesis; P = .72). Twenty-five additional invasive cancers were detected with mammography plus tomosynthesis (40% increase, adjusted for reader; P < .001). The mean interpretation time was 45 seconds for mammography alone and 91 seconds for mammography plus tomosynthesis (P < .001). CONCLUSION: The use of mammography plus tomosynthesis in a screening environment resulted in a significantly higher cancer detection rate and enabled the detection of more invasive cancers. Clinical trial registration no. NCT01248546.

Digital breast tomosynthesis (DBT): initial experience in a clinical setting.

BACKGROUND: Digital breast tomosynthesis (DBT) is a promising new technology. Some experimental clinical studies have shown positive results, but the future role and indications of this new technique, whether in a screening or clinical setting, need to be evaluated. PURPOSE: To compare digital mammography and DBT in a side-by-side feature analysis for cancer conspicuity, and to assess whether there is a potential additional value of DBT to standard state-of-the-art conventional imaging work-up with respect to detection of additional malignancies. MATERIAL AND METHODS: The study had ethics committee approval. A total of 129 women underwent 2D digital mammography including supplementary cone-down and magnification views and breast ultrasonography if indicated, as well as digital breast tomosynthesis. The indication for conventional imaging in the clinical setting included a palpable lump in 30 (23%), abnormal mammographic screening findings in 54 (42%), and surveillance in 45 (35%) of the women. The women were examined according to present guidelines, including spot-magnification views, ultrasonography, and needle biopsies, if indicated. The DBT examinations were interpreted several weeks after the conventional imaging without knowledge of the conventional imaging findings. In a later session, three radiologists performed a side-by-side feature analysis for cancer conspicuity in a sample of 50 cases. RESULTS: State-of-the-art conventional imaging resulted in needle biopsy of 45 breasts, of which 20 lesions were benign and a total of 25 cancers were diagnosed. The remaining 84 women were dismissed with a normal/definitely benign finding and without indication for needle biopsy. The subsequent DBT interpretation found suspicious findings in four of these 84 women, and these four women had to be called back for repeated work-up with knowledge of the tomosynthesis findings. These delayed work-ups resulted in two cancers (increasing the cancer detection by 8%) and two false-positive findings. The side-by-side feature analysis showed higher conspicuity scores for tomosynthesis compared to conventional 2D for cancers presenting as spiculated masses and distortions. CONCLUSION: Tomosynthesis is a promising new technique. Our preliminary clinical experience shows that there is a potential for increasing the sensitivity using this new technique, especially for cancers manifesting as spiculated masses and distortions.