Digital breast tomosynthesis (DBT): recommendations from the Italian College of Breast Radiologists (ICBR) by the Italian Society of Medical Radiology (SIRM) and the Italian Group for Mammography Screening (GISMa)
- 2.8k Downloads
This position paper, issued by ICBR/SIRM and GISMa, summarizes the evidence on DBT and provides recommendations for its use. In the screening setting, DBT in adjunct to digital mammography (DM) increased detection rate by 0.5–2.7‰ and decreased false positives by 0.8–3.6% compared to DM alone in observational and double-testing experimental studies. The reduction in recall rate could be less prominent in those screening programs which already have low recall rates with DM. The increase in radiation exposure associated with DM/DBT protocols has been solved by the introduction of synthetic mammograms (sDM) reconstructed from DBT datasets. Thus, whenever possible, sDM/DBT should be preferred to DM/DBT. However, before introducing DBT as a routine screening tool for average-risk women, we should wait for the results of randomized controlled trials and for a statistically significant and clinically relevant reduction in the interval cancer rate, hopefully associated with a reduction in the advanced cancer rate. Otherwise, a potential for overdiagnosis and overtreatment cannot be excluded. Studies exploring this issue are ongoing. Screening of women at intermediate risk should follow the same recommendations, with particular protocols for women with previous BC history. In high-risk women, if mammography is performed as an adjunct to MRI or in the case of MRI contraindications, sDM/DBT protocols are suggested. Evidence exists in favor of DBT usage in women with clinical symptoms/signs and asymptomatic women with screen-detected findings recalled for work-up. The possibility to perform needle biopsy or localization under DBT guidance should be offered when DBT-only findings need characterization or surgery.
KeywordsBreast cancer Digital breast tomosynthesis Mammography Screening
In the last years, many reports have appeared about digital breast tomosynthesis (DBT) in both the diagnostic and screening setting. In fact, thanks to a pseudo-three-dimensional reconstruction, DBT allows for overcoming some limitations of standard two-dimensional (2D) digital mammography (DM) caused by structural overlapping and resulting into false-negative and false-positive findings . The X-ray tube moves along an angular direction and generates multiple low-dose variably angled projections of the compressed breast. The reconstruction of multiple images of thin slices of the compressed breast is obtained by specialized software [2, 3]. The technical solutions proposed by vendors differ for tube movement (continuous/discontinuous), width of oscillation angle, number of projections, type of detector, and other characteristics, although no substantial effects on diagnostic performance related to these differences have been reported so far.
This position paper, issued by the Italian College of Breast Radiologists by the Italian Society of Medical Radiology (SIRM) and the Italian Group for Mammography Screening (GISMa), summarizes the available evidence on DBT and provides practical recommendations for its use. For the levels of evidence (LoE) reported here, we refer to the definitions given by the Centre for Evidence-Based Medicine, Oxford, United Kingdom .
DBT for first-level screening
Several studies have evaluated the potential of DBT in first-level screening [5, 6, 7, 8, 9, 10, 11, 12, 13, 14]. In particular, three European prospective trials conducted in the context of organized population-based screening programs [6, 7, 8, 9] and five retrospective studies from spontaneous screening settings [10, 11, 12, 13, 14] have shown that DBT combined with DM (or as a stand-alone approach ) allows for a better diagnostic performance than DM alone.
Results of DBT differ according to the study design, the screening interval, and the type of screening setting (organized versus spontaneous). A review  reported that DBT provides an increase in cancer detection rate from 0.5 to 2.7 per thousand screened women and a reduction in false-positive recall rate from 0.8 to 3.6 per 100 screened women. The possibility of using only the mediolateral oblique DBT projection, supported by one study , has the rationale of dose reduction. However, although many experiences have shown the superior diagnostic performance of two-view DBT compared with one-view DBT [16, 17, 18], the availability of synthetic 2D mammograms generated from the DBT dataset (see below) has substantially overcome the problem of dose. A reduction in recall rate of greatly varying magnitude (from 6 to 82%, median 31%) has been found . Notably, this variability depends on the baseline recall rate with standard DM because the higher the baseline DM recall rate, the higher the absolute and relative reduction obtained with DBT. Thus, the advantage of reducing the recall rate could be less prominent in those screening programs which already have low recall rates with standard DM.
If both DM and DBT are acquired (separately or with the so-called “combo” modality), the average glandular dose is approximately doubled. In fact, the DBT dose is similar to that of DM [20, 21, 22]. However, considering the dose reduction obtained through the introduction of DM as an alternative to film-screen mammography, the dose of these DM/DBT protocols too remains below the upper limit defined by the “European guidelines for quality assurance in breast cancer screening and diagnosis” published in 2006 .
Of note, radiation exposure would be an issue to take into consideration for a generalized adoption of DM/DBT protocols for population-based mass screening. The solution has come from the synthetic digital mammography (sDM) obtained through specialized algorithms summing and filtering the DBT datasets. Principles and methods have similarities with those generating maximum intensity projections in computed tomography and magnetic resonance imaging (MRI). These sDMs are virtual 2D mammograms obtained from DBT, paying no tradeoff in terms of radiation exposure [24, 25, 26].
Several authors have compared sDM/DBT protocols with DM/DBT protocols demonstrating a similar diagnostic performance [25, 26, 27, 28]. Therefore, unless additional radiation exposure is specifically justified (as in the work-up of suspicious findings detected at DM), a diagnostic mammography examination can be performed using the sDM/DBT approach.
Notably, two studies [29, 30] have reported that 54–57% of additional cancers detected by additional ultrasonography screening after negative DM were detected by DBT. This is a relevant argument in favor of DBT, considering the practical hurdles for a generalized mass screening with ultrasonography in adjunct to DM.
However, in the context of organized population-based screening programs, a simple increase in sensitivity and overall diagnostic performance of a new tool, even if statistically significant and clinically relevant, is not enough, per se, for its generalized adoption. According to the European Council Recommendation on cancer screening , evidence from randomized controlled trials is needed before introducing new screening tools. In particular, considering the pre-existing evidence in favor of screening mammography (recently confirmed by the International Agency for Research on Cancer ), caution is urged due to the possibility that a substantial part of the additional cancers detected by DBT could be overdiagnosed lesions, i.e. indolent malignant lesions that would never surface clinically during the woman’s life . Importantly, the consequence would be an increase in overtreatment, i.e. unnecessary surgery, radiation, and/or medical therapy.
It is worth to note that most of the additional cancers detected with DBT have been reported to be invasive [6, 7, 8, 9, 10, 11, 12, 13, 14]. A large study from the United States  has compared DM/DBT in approximately 174,000 women with DM alone in approximately 281,000 women. The DM/DBT approach was associated with a 29% increase in detection rate and a 15% reduction in recall rate. All the increase in cancer detection was due to invasive cancers (+41%). No increase in the detection of ductal carcinoma in situ, deemed more probably overdiagnosed, was observed. However, we should consider that also invasive cancers can be indolent and overdiagnosed. Thus, no definitive conclusions can be drawn.
Several other issues cannot be ignored when discussing the potential introduction of DBT for screening. They include at least sufficient availability of DBT equipments, management of suspicious findings at DBT alone, and increased reading time [34, 35, 36], which implies the need for more breast radiologists involved as screening readers.
whether the expected increase in cancer detection at first round with DBT is associated or not associated with a reduction in interval cancer incidence;
whether the total incidence of advanced cancers (stage ≥T2) in the whole study period (i.e. considering all cancers detected in the first or second round as well as interval cancers) is lower in the DBT group compared with the control group; and
whether the total incidence of breast cancer in the DBT group exceeds that of the control group.
The attention paid to interval cancer rate is explained by the possibility to use this measure as a proxy of effectiveness in mortality reduction . To our knowledge, only two articles [38, 39] have reported results concerning the interval cancer rate of DBT screening. The study design was a single-center retrospective analysis  and a multicenter  prospective cohort study. Comparing DM alone with DM/DBT screening, a reduction in interval cancer rate was observed in both studies, from 0.7 to 0.5 per thousand screened women  and from 0.6 to 0.46 per thousand screened women , respectively. These variations were not statistically significant. However, it should be noted that neither study was designed and powered for interval cancer analysis, and that both were conducted in a spontaneous screening setting. These heavy limitations prevent us to draw any conclusion to be extrapolated to European organized biennial screening programs.
DBT for work-up of screen-detected suspicious findings, as a first-line diagnostic examination in symptomatic women, for preoperative staging, and targeted evaluation after MRI
Several studies [40, 41, 42, 43, 44, 45, 46, 47, 48] have shown that DBT is at least equivalent to additional DM views (magnification, spot compression, 90° views, etc.), also reducing radiation exposure in both the diagnostic and screening setting. In symptomatic women, diagnostic accuracy is improved by DBT, reducing the number of suspicious findings and of unnecessary biopsies [43, 45, 46]. In the presence of palpable lesions, DBT has been reported to be never worse, and often better, than DM for estimating tumor size . The better diagnostic performance of DBT is confirmed also in the case of invasive lobular cancers, as shown by a recent study . In the preoperative setting, the combination of DM, DBT, and ultrasonography could provide the same information as MRI . Finally, DBT allows for identifying some findings additionally found on preoperative MRI also when they are not visible at targeted ultrasonography, permitting a reduction in the number of MR-guided biopsies .
Studies evaluating inter-reader variability for DM/DBT versus DM alone have reported for each reader an increase in the detection rate and a reduction in the recall rate with an overall increase in the diagnostic performance [53, 54, 55, 56]. In the particular context of spontaneous screening, DBT has also been found to reduce the number of short-time repeat examinations .
All these results allow for recommending DBT for symptomatic women and for work-up of screen-detected suspicious findings.
asymptomatic women at average risk (first level of organized or spontaneous screening);
asymptomatic women at intermediate risk, including women with a previous breast cancer;
symptomatic women and women needing work-up of screen-detected suspicious findings;
asymptomatic women at hereditary/familial high risk;
asymptomatic women at high risk due to previous chest radiotherapy.
Asymptomatic women at average risk (first level of organized or spontaneous screening)
These women should undergo DM starting not before the age of 40, with a screening interval that may vary according to local health authority’s decision. As recently stated by the European Society of Breast Imaging [57, 58], direct DM should be preferred to film-screen mammography due to a reduced radiation exposure  and an at least equivalent diagnostic performance  (LoE A). The preference is also in favor of DM when compared with indirect digital phosphor storage plate (so-called computer radiography) .
in the context of studies approved by an Ethical Committee, with enrollment after informed consent signature by the woman (for randomized controlled trials, refer to the scheme proposed by the Osservatorio Nazionale Screening );
in the well-defined context of centers being part of public population-based screening programs, with previous experience with ethically approved studies concerning at least feasibility of screening DBT, demonstrated through articles published in peer-reviewed journals.
Asymptomatic women at intermediate risk, including those with a previous breast cancer
Up to now, no substantial evidence has been produced in favor or against the use of DBT for screening women at intermediate risk of breast cancer, i.e. those with a lifetime risk from 15 to 19%, estimated using multifactorial models. All these women should undergo screening DM with the same protocols recommended for asymptomatic women at average risk.
For asymptomatic women with a previous history of breast cancer (included in this general category of women at intermediate risk), we refer to the recommendations provided by the GISMa and the Italian College of Breast Radiologists by SIRM . The recent observation of a reduction in indeterminate findings in surveillance after breast cancer treatment  plays in favor of DBT usage (LoE D) in this setting, with a preference for sDM/DBT protocols.
in the context of studies approved by an Ethical Committee, with enrollment after informed consent signature by the woman;
in the well-defined context of centers having previous experience with ethically approved studies concerning DBT, demonstrated through articles published in peer-reviewed journals.
Symptomatic women and women needing work-up of screen-detected suspicious findings
We include in this category both women with suspicious clinical symptoms or signs (asking for a mammogram usually ordered by a general practitioner or a specialist) and asymptomatic women with screen-detected suspicious findings recalled for work-up in the context of spontaneous or organized screening. Women with previous history of breast cancer, if having suspicious clinical symptoms/signs or after a suspicious finding on an imaging study, are also included in this category.
Asymptomatic women at hereditary/familial high risk
Asymptomatic women at high risk due to previous chest radiotherapy
Needle biopsy under DBT guidance
DBT can show doubtful/suspicious findings without any clinical correlate and even undetectable on DM, sDM, or ultrasonography [5, 6, 7, 8]. This may occur in both the screening and diagnostic settings. In those cases, when the DBT finding is neither detectable at targeted ultrasonography nor at DM review, a needle biopsy (and, when necessary, also presurgical localization) should be performed under DBT guidance. Importantly, DBT guidance offers important advantages in terms of shorter procedure duration and reduced radiation exposure [77, 78] (LoE B).
As a consequence, centers offering DBT should also offer DBT-guided interventions for DBT findings not otherwise identifiable. These interventions can be performed at the same center where DBT was done or at another center functionally connected with the first one. At present, in fact, few centers are equipped with the device for DBT-guided interventions.
Evidence available for DBT allows to recommend its usage for all cases of symptomatic women and women needing work-up of screen-detected suspicious findings (considering both spontaneous and organized screening). In these settings, when available, sDM/DBT protocols should be preferred for symptomatic women.
Breast cancer screening using the current mammography technology has been demonstrated to be effective in reducing mortality from the disease. From this perspective, completing the shift from analog film-screen mammography and computed radiography systems, that are still in use, to direct digital mammography systems ranks first in the order of priorities.
A generalized adoption of DBT as a first-level screening tool should wait for a specific evidence, in particular for a statistically significant and clinically relevant reduction in interval cancer rate (hopefully associated with a reduction in advanced cancer rates). When this evidence will be available, both asymptomatic women at average and intermediate risk (including those with a previous breast cancer history) will be allowed to be screened with DBT on a routine basis. Accurate data collection will be necessary for many years in order to assess the overall impact of DBT technology in terms of mortality reduction and other efficacy/effectiveness indicators. For high-risk women, when a mammogram is indicated, a sDM/DBT protocol should be preferred.
The already existing evidence, which has been built with a non-negligible contribution of Italian breast radiologists, plays in favor of DBT. A trend for making DBT the mammography of the next future can be outlined. However, even in the United States, where FDA approved the use of some DBT devices in some cases for screening, the use of DBT is still quite limited. A recent survey  among the members of the Society of Breast Imaging reported that of 670 responders, only 200 (30%) use DBT although 62% of non-users have planned to equip themselves with DBT.
Studies and researches are still needed for a deeper evaluation of DBT, a relevant innovation in breast imaging. In particular, the risk of an increased overdiagnosis and also several organizational issues (including increased reading time) suggest to wait for a more conclusive evidence before adopting DBT in population-based screening, a position shared with other experts and medical or health authority bodies worldwide [15, 57, 58, 80, 81].
Compliance with ethical standards
Conflict of interest
The authors declare no funding and no conflict of interest for this article.
This article does not contain any studies with human participants or animals performed by any of the authors.
- 4.Centre for Evidence-Based Medicine, Oxford, UK. www.cebm.net. Accessed 1 June 2016
- 7.Skaane P, Bandos AI, Gullien R et al (2013) 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. Eur Radiol 23:2061–2071CrossRefPubMedPubMedCentralGoogle Scholar
- 12.McCarthy AM, Kontos D, Synnestvedt M et al (2014) Screening outcomes following implementation of digital breast tomosynthesis in a general-population screening program. J Natl Cancer Inst 106(11). doi: 10.1093/jnci/dju316
- 16.Svahn T, Andersson I, Chakraborty D et al (2010) The diagnostic accuracy of dual-view digital mammography, single-view breast tomosynthesis and a dual-view combination of breast tomosynthesis and digital mammography in a free-response observer performance study. Radiat Prot Dosim 139:113–117CrossRefGoogle Scholar
- 23.Perry N, Broeders M, de Wolf C et al (2006) European guidelines for quality assurance in breast screening and diagnosis. Fourth Edition. www.euref.org/downloads?download=24:european-guidelines-for-quality. Accessed 7 Jan 2017
- 31.European Council Recommendation on cancer screening, December 2, 2003 (2003/878/EC). https://ec.europa.eu/jrc/sites/jrcsh/files/2_December_2003%20cancer%20screening.pdf. Accessed 22 Feb 2017
- 37.Osservatorio Nazionale Screening. Trial tomosintesi. http://www.osservatorionazionalescreening.it/sites/default/files/allegati/documento%20trial%20tomosintesi_con%20controlli%20qualit%C3%A0.pdf#overlay-context=node/59. Accessed 19 July 2016
- 41.Gilbert F, Tucker L, Gillan M et al (2015) The TOMMY trial: a comparison of tomosynthesis with digital mammography in the UK NHS Breast Screening Programme—a multicentre retrospective reading study comparing the diagnostic performance of digital breast tomosynthesis and digital mammography with digital mammography alone. Health Technol Assess 19:1–136CrossRefPubMedPubMedCentralGoogle Scholar
- 48.Whelehan P, Heywang-Köbrunner SH, Vinnicombe SJ et al (2017) Clinical performance of Siemens digital breast tomosynthesis versus standard supplementary mammography for the assessment of screen-detected soft-tissue abnormalities: a multi-reader study. Clin Radiol 72:95.e9–95.e15CrossRefGoogle Scholar
- 57.Sardanelli F, Aase HS, Álvarez M et al (2016) Position paper on screening for breast cancer by the European Society of Breast Imaging (EUSOBI) and 30 national breast radiology bodies from Austria, Belgium, Bosnia and Herzegovina, Bulgaria, Croatia, Czech Republic, Denmark, Estonia, Finland, France, Germany, Greece, Hungary, Iceland, Ireland, Italy, Israel, Lithuania, Moldova, The Netherlands, Norway, Poland, Portugal, Romania, Serbia, Slovakia, Spain, Sweden, Switzerland and Turkey. Eur Radiol. doi: 10.1007/s00330-016-4612-z
- 58.Sardanelli F, Fallenberg EM, Clauser P, European Society of Breast Imaging (EUSOBI), with language review by Europa Donna–The European Breast Cancer Coalition et al (2017) Mammography: an update of the EUSOBI recommendations on information for women. Insights Imaging 8:11–18Google Scholar
- 62.Distante V, Frigerio A, Naldoni C et al (2006) Gruppo Italiano per lo Screening Mammografico. Documento di consenso GISMA in merito all’estensione dello screening mammografico di popolazione alla fascia di età 40–49 e 70–74. http://win.gisma.it/atti/bolognaconsenso05/documento_consenso.pdf. Accessed 7 Jan 2017
- 63.Giordano L, Giorgi D, Frigerio A, Gruppo Italiano per lo Screening Mammografico et al (2006) Process indicators and standards for the evaluation of breast cancer screening programmes. Epidemiol Prev 30(2 Suppl 1):5–9, 11–47Google Scholar
- 64.Bucchi L, Belli P, Benelli E et al (2016) Recommendations for breast imaging follow-up of women with a previous history of breast cancer: position paper from the Italian Group for Mammography Screening (GISMa) and the Italian College of Breast Radiologists (ICBR) by SIRM. 49. Radiol Med 121:891–896CrossRefPubMedPubMedCentralGoogle Scholar
- 68.Sardanelli F, Podo F, Santoro F et al (2011) High Breast Cancer Risk Italian 1 (HIBCRIT-1) Study. Multicenter surveillance of women at high genetic breast cancer risk using mammography, ultrasonography, and contrast-enhanced magnetic resonance imaging (the high breast cancer risk italian 1 study): final results. Invest Radiol 46:94–105CrossRefPubMedGoogle Scholar
- 81.Melnikow J, Fenton JJ, Miglioretti D, Whitlock EP, Weyrich MS (2016) Screening for breast cancer with digital breast tomosynthesis [Internet]. Agency for Healthcare Research and Quality (US), RockvilleGoogle Scholar
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.