Boston Breast Imaging
Educational Hub
Case 07
(continued)
Answer:
Architectural distortion with enhancement is cancer until proven otherwise.
Architectural distortion (AD) can be difficult to identify on conventional 2D imaging and is more commonly seen on tomosynthesis views. This has presented a new clinical challenge of how to navigate the increased number of AD findings that are seen with tomosynthesis. The accepted management is to perform tomosynthesis-guided biopsy. However, many of the biopsied AD are not cancer (false positives) and represent benign entities such as radial scar/complex sclerosing lesion.
Studies are underway looking at whether CEM can help differentiate benign from malignant AD. Can enhancement on recombined views help identify cases that are more likely to be cancer? As of this writing, the research suggest that architectural distortion WITHOUT enhancement has a very low likelihood of being cancer. While more evidence is definitely needed, this is the start of helping triage these cases. For now, AD WITH enhancement should most certainly be work-up further.
Let's talk about this case: there is an area of AD with associated non-mass enhancement in the upper outer left breast. No ultrasound correlate was seen. What to do?? Tissue sampling really should be performed. Given the AD, this can be sampled with tomosynthesis-guided biopsy. If the AD wasn't clearly seen, it could still be sampled with stereo/tomo guided biopsy given the size of the enhancement. Breast MRI should only reserved when the finding cannot be targeted with standard imaging modalities. Drumroll ---- this was a COMPLEX SCLEROSING LESION with LCIS!
Goh Y, Chan CW, Pillay P, et al. Architecture distortion score (ADS) in malignancy risk stratification of architecture distortion on contrast-enhanced digital mammography. Eur Radiol. 2021 May;31(5):2657-2666.
Patel BK, Naylor ME, Kosiorek HE, et al. Clinical utility of contrast-enhanced spectral mammography as an adjunct for tomosynthesis-detected architectural distortion. Clin Imaging. 2017 Nov-Dec;46:44-52.