CEM improves breast cancer detection rate over low-energy mammography

Contrast-enhanced mammography (CEM) improves breast cancer detection over low-energy mammography in screening settings, according to research published March 11 in Radiology.

A team led by Joao Horvat, MD, from the Memorial Sloan Kettering Cancer Center in New York found that CEM depicted 90% of breast cancers compared with 10% on low-energy mammograms alone and 50% on low-energy mammograms with whole-breast ultrasound.

“CEM is a suitable modality for screening, especially in women with dense breasts,” the Horvat team wrote.

CEM in recent years has caught the eye of breast imagers by generating low-energy images that are comparable to 2D full-field digital mammography. Additionally, recombined images from CEM are comparable to those of MRI, with the advantage of having lower cost, proponents say.

Previous studies suggest that whole-breast ultrasound, a supplemental breast imaging technique, does not help increase cancer detection with screening MRI. Horvat and co-authors investigated whether the same trend goes for CEM. They compared the respective performances of CEM, low-energy mammograms alone, and low-energy imaging supplemented by whole-breast ultrasound.

Their study included data collected between 2014 and 2019 from 468 women with a median age of 54. Separate breast radiologists interpreting the ultrasound images were blinded to the CEM images, and vice versa.

Images depict true-positive findings in a 51-year-old woman with a left breast focal nonmass enhancement. (A) Mediolateral oblique and (B) cranio-caudal contrast-enhanced mammographic (CEM) images show the enhancement (arrows). No correlate was identified on (C) mediolateral-oblique or (D) craniocaudal low-energy images, or whole-breast ultrasound (not shown). Contrast-enhanced breast MRI shows a suspicious enhancement on (E) sagittaland (F) axial subtraction images (arrows), correlating with the finding on CEM images. MRI-guided biopsy (not shown) showed flat epithelial atypia, withsubsequent surgery confirming ductal carcinoma in situ.RSNA

The team identified nine screen-detected cancers diagnosed in eight women. Of these, CEM detected eight cancers, followed by four found via low-energy mammography with supplemental ultrasound and one by low-energy mammography alone. CEM significantly outperformed low-energy mammography alone and low-energy mammography combined with whole-breast ultrasound in terms of cancer detection. However, it also led to more biopsy recommendations.

Comparison between breast imaging modalities
Measure	Low-energy imaging	Low-energy imaging supplemented by ultrasound	CEM	p-value (CEM compared with other modalities)
Cancer detection rate (out of 1,000)	2.1	8.5	17.1	< 0.05
Abnormal interpretation rate	10.3%	13.7%	18.6%	< 0.001
Biopsy recommendation rate (out of 1,000)	15	38.4	42.7	N/A

CEM recommended 20 biopsies, a positive predictive value (PPV) of 45%. Low-energy mammography had a PPV of 14.3% and low-energy mammography supplemented by ultrasound had a PPV of 27.8%.

The team highlighted that CEM caused a higher number of recalls and false-positive biopsy results, though with a higher PPV of biopsies performed.

“Supplemental whole-breast ultrasound may be averted in patients undergoing screening CEM because it does not add value to cancer detection and may increase false-positive recalls and biopsy results,” it wrote.

The study authors also called for future studies to include more diverse population data to find out whether their results are generalizable.

“Additional studies could also incorporate CEM-guided biopsies as part of the imaging algorithm because the availability of CEM-guided biopsy could reduce additional imaging with MRI,” they added.

The full study can be found here.