Perspective from Alexander B. Sevrukov, MD
Source: Parsian S. Radiology. 2012; doi: 10.1148/radiol.12112672.
November 19, 2012
3 min read

Diffusion-weighted imaging shows promise identifying false-positive lesions

Perspective from Alexander B. Sevrukov, MD
Source: Parsian S. Radiology. 2012; doi: 10.1148/radiol.12112672.
You've successfully added to your alerts. You will receive an email when new content is published.

Click Here to Manage Email Alerts

We were unable to process your request. Please try again later. If you continue to have this issue please contact

Diffusion-weighted imaging successfully identified as negative/benign nearly half of nonmalignant breast lesions that had been incorrectly identified as positive by MRI, according to results of a retrospective study.

Savannah C. Partridge, PhD, research associate professor in the department of radiology at the University of Washington Seattle Cancer Care Alliance, and colleagues reviewed diffusion-weighted imaging characteristics of nonmalignant lesion subtypes that had been classified as category 4 or 5 false positives using MRI.

The study included 175 breast lesions that were suspected of malignancy and recommended for biopsy using conventional breast MRI, but were found to be benign on the biopsy.

The lesions under analysis were taken from 165 women. Apparent diffusion coefficients from diffusion-weighted imaging (b = 0, 600 sec/mm2) were calculated for each sample, according to the results. Researchers then conducted a comparison between subtypes with an apparent diffusion coefficient threshold of  1.81 × 10−3 mm2/sec.

Forty-six percent of lesions yielded coefficients greater than the prespecified threshold. Certain lesion subtypes were more likely to have coefficients above the threshold, including:

  • fibroadenoma ([1.94 ± 0.38 {standard deviation}] × 10−3 mm2/sec; n = 30)
  • focal fibrosis ([1.84 ± 0.48] × 10−3 mm2/sec; n = 19)
  • normal tissue ([1.81 ± 0.47] × 10−3 mm2/sec; n = 13)
  • apocrine metaplasia ([2.01 ± 0.38] × 10−3 mm2/sec; n = 13)
  • usual ductal hyperplasia ([1.83 ± 0.49] × 10−3 mm2/sec; n = 12)
  • inflammation ([1.95 ± 0.46] × 10−3 mm2/sec; n = 10).

“We have previously found that breast cancers typically exhibit lower apparent diffusion coefficient values than benign lesions on diffusion-weighted imaging,” Partridge told HemOnc Today. “We defined a useful diagnostic apparent diffusion coefficient threshold or cut-off to best differentiate benign and malignant lesions without missing any cancers.” 

The most common lesion subtype to have a coefficient below the threshold was atypical ductal hyperplasia ([1.48 ± 0.36] × 10−3 mm2/sec; n = 23).

The lowest mean coefficient among nonmalignant lesions was reported for lymph nodes  ([1.28 ± 0.23] × 10−3 mm2/sec; n = 4).

The most common lesions with apparent diffusion coefficients below the threshold were atypical ductal hyperplasia and lobular neoplasia. These lesions had significantly lower coefficients than for other benign lesions (P,.0001).

“Assessing [apparent diffusion coefficients] along with dynamic contrast-enhanced MR imaging features may decrease the number of avoidable false-positive findings at breast MR imaging and reduce the number of preventable biopsies,” the researchers concluded.

If the apparent diffusion coefficient threshold had been added to the conventional MRI assessment so that only suspicious lesions with apparent diffusion coefficient values below the cut-off were biopsied, 46% of the benign lesions could have avoided biopsy, Partridge said.

“They exhibited high apparent diffusion coefficient values, above the diagnostic threshold value,” Partridge said. “Essentially it is additional criteria radiologists can use to decide whether or not a lesion they see on MRI is suspicious for malignancy to help reduce unnecessary biopsies.”

Disclosure: The study was supported by a grant from NIH.