MRI for the size assessment of pure ductal carcinoma in situ (DCIS): a prospective study of 33 patients

Eur J Radiol. 2011 Mar;77(3):462-7. doi: 10.1016/j.ejrad.2009.09.003. Epub 2009 Nov 6.

Abstract

Objective: To determine whether MRI assesses the size of ductal carcinomas in situ (DCIS) more accurately than mammography, using the histopathological dimension of the surgical specimen as the reference measurement.

Materials and methods: This single-center prospective study conducted from March 2007 to July 2008 at the Antoine-Lacassagne Cancer Treatment Center (Nice, France) included 33 patients with a histologically proven DCIS by needle biopsy, who all underwent clinical examination, mammography, and MRI interpreted by an experienced radiologist. All patients underwent surgery at our institution. The greatest dimensions of the DCIS determined by the two imaging modalities were compared with the histopathological dimension ascertained on the surgical specimen. The study was approved by the local Ethical Research Committee and was authorized by the French National Health Agency (AFSSAPS).

Results: The mean age of the 33 patients was 59.7 years (± 10.3). Three patients had a palpable mass at clinical breast examination; 82% underwent conservative surgical therapy rather than radical breast surgery (mastectomy); 6% required repeat surgery. MRI detected 97% of the lesions. Non-mass-like enhancement was noted for 78% of the patients. In over 50% of the cases, distribution of the DCIS was ductal or segmental and the kinetic enhancement curve was persistent. Lesion size was correctly estimated (± 5 mm), under-estimated (<5mm), or over-estimated (>5mm), respectively, by MRI in 60%, 19% and 21% of cases and by mammography in 38%, 31% and 31% (p = 0.05). Mean lesion size was 25.6mm at histopathology, 28.1mm at MRI, and 27.2mm on mammography (nonsignificant difference). The correlation coefficient between histopathological measurement and MRI was 0.831 versus 0.674 between histopathology and mammography. The correlation coefficient increased with the nuclear grade of the DCIS on mammography; this coefficient also increased as the mammographic breast density decreased.

Conclusion: MRI appears to assess the size of DCIS better than mammography by limiting the number of under- and over-estimations compared to histopathology findings.

MeSH terms

  • Breast Neoplasms / pathology*
  • Carcinoma, Intraductal, Noninfiltrating / pathology*
  • Female
  • Humans
  • Magnetic Resonance Imaging / methods*
  • Middle Aged
  • Prospective Studies
  • Reproducibility of Results
  • Sensitivity and Specificity