May 17, 2022
Article in Auntminnie
Reminder: Tomosynthesis (or 3D mammography) is a radiological imaging technique that reduces the effect of superimposition of breast tissue as it reconstructs a three-dimensional image of the breast from several low-dose X-rays acquired from different projection angles.
This technique was heavily promoted about 10 years ago. Therefore, a review is done after 10 years of hindsight in the media "AuntMinnie.com."
This is a community website for radiologists and professionals in the medical imaging industry. According to this rather collaborative media that connects radiologists, business managers, and industry professionals to "meet, do transactions, research and collaborate," tomosynthesis has clearly disappointed.
Many questions and doubts about the benefit of using this technique have been raised previously: https://pubmed.ncbi.nlm.nih.gov/30816931/
- tomosynthesis does not reduce false alarms
- the additional use of tomosynthesis does not reduce interval cancers
- tomosynthesis would increase overdiagnosis
- the benefits of tomosynthesis are not clear
1° Cancer detection
Digital mammography alone has been compared with digital mammography + tomosynthesis (a higher-radiation combination): matched studies* have shown that the addition of tomosynthesis made it possible to find more cancers: 8.8 per 1000 women compared with 6.4 per 1000. But in other unmatched studies*, the difference was narrower, 5.7 cancers detected per 1000 women versus 4.5.
* Matching consists of setting up pairs (1 case and 1 control) with the same characteristics (e.g., age) to compare the results while avoiding potential confounding factors. The groups are thus "balanced" on these characteristics.
2° Recall rates
What about recall rates? The recall rate refers to false alarms during screening, i.e., suspicions of cancer that will not be confirmed, but only after recalling the patients who will need to have other complementary explorations before deciding on these suspicions. Here again, the data vary according to the study conducted.
Based on the March 2022 study summarized here, repeated breast cancer screening with 3D mammography only modestly decreases the risk of having a false-positive result compared with standard digital mammography.
What can we learn from this study?
The risk of a false-positive result was lower when screening was performed every two years instead of every year and in the case of non-dense breasts and older women.
However, the difference was modest, and the reduction in false positives by using 3D mammography was only 2.4% compared to standard mammography.
3°How effective are synthetic mammography images?
In 2012 an opening was made for 'synthetic imaging,' which records a single radiological acquisition and therefore delivers a single dose of radiation, thus avoiding the over-irradiation caused by 3D mammography**.
But are the synthesized images an effective alternative to digital mammography images? Clinical results of effectiveness tests of synthesized mammographic images are unfortunately mitigated. Overall, the results between synthesized images are equivalent to digital mammography, although the latter has a better resolution.
**Classically, 2D mammography and 3D tomosynthesis acquisitions are used in combination. This results in a significant increase in the X-ray dose delivered. The X-ray doses delivered by combining 2D mammography and tomosynthesis are about twice the dose of 2D mammography alone.
Synthetic 2D tomosynthesis is an alternative, obtained by reconstruction from 3D acquisitions only; it avoids the joint use of 2D mammography and thus reduces the delivered dose.
4° Does tomosynthesis reduce mortality?
Does tomosynthesis result in a reduction in mortality? According to this article in Autminnie.com, a survey of eight studies conducted between 2016 and 2021 investigated whether tomosynthesis reduces rates of interval cancers (cancers not caught by screening because they occur between two mammograms) compared with digital mammography alone. Interval cancers are often very aggressive and occur quickly, thus missed by screening. They are correlated with mortality because their intrinsic aggressiveness endangers the survival of women, often because of their metastatic potential.
It was found that tomosynthesis does not impact the rate of interval cancer.
Ten years after its use, the benefits of tomosynthesis may be much more modest than clinicians initially expected. In conclusion, this technique is finally similar to digital mammography with no proven advantage.
Even if the detection rate of tomosynthesis seems slightly better, the benefit of this technique remains an open question. If this moderate improvement in cancer detection is gained at the cost of increased overdiagnosis, we cannot conclude that the benefit/risk ratio is favorable.
As usual, the major concern is the information provided to women, as tomosynthesis is sometimes performed in radiology offices without the knowledge of the patient who comes for a routine mammogram, who does not benefit from it and is exposed to unnecessary over-irradiation.
Also read: https://www.bmj.com/content/366/bmj.l4506