Apparent Diffusion Coefficient as an MR Imaging Biomarker of Low-Risk Ductal Carcinoma in Situ: A Pilot Study

• Published in: Radiology Vol. 260; no. 2; pp. 364 - 372
• Main Authors: LIMA, Mami, LE BIHAN, Denis, OKUMURA, Ryosuke, OKADA, Tomohisa, FUJIMOTO, Koji, KANAO, Shotaro, TANAKA, Shiro, FUJIMOTO, Masakazu, SAKASHITA, Hiromi, TOGASHI, Kaori
• Format: Journal Article
• Language: English
• Published: Oak Brook, IL Radiological Society of North America 01.08.2011
• Subjects: Adult; Aged; Biological and medical sciences; Biomarkers; Breast Neoplasms - diagnosis; Breast Neoplasms - pathology; Carcinoma, Intraductal, Noninfiltrating - diagnosis; Carcinoma, Intraductal, Noninfiltrating - pathology; Contrast Media; Diffusion Magnetic Resonance Imaging - methods; Female; Gadolinium; Genital system. Mammary gland; Gynecology. Andrology. Obstetrics; Heterocyclic Compounds; Humans; Image Interpretation, Computer-Assisted; Investigative techniques, diagnostic techniques (general aspects); Linear Models; Mammary gland diseases; Medical sciences; Middle Aged; Organometallic Compounds; Radiodiagnosis. Nmr imagery. Nmr spectrometry; Reproducibility of Results; Retrospective Studies; ROC Curve; Tumors; Performance evaluation; Breast disease; Biological marker; Pilot study; Specificity; Predictive value; Adult; Female; Diagnosis; Diffusion coefficient; Woman; Human; Ductal carcinoma; Carcinoma in situ; Exploration; Breast cancer; Malignant tumor; Nuclear magnetic resonance imaging; Mammary gland diseases; Risk factor; Medical imagery; Severity score; Positive predictive value; Technique; Cancer
• ISSN: 0033-8419; 1527-1315
• DOI: 10.1148/radiol.11101892

To evaluate the potential of apparent diffusion coefficients (ADCs) obtained at quantitative diffusion-weighted magnetic resonance (MR) imaging of the breast as a biomarker of low-grade ductal carcinoma in situ (DCIS). This retrospective study was approved by an institutional review board, and the requirement to obtain informed consent was waived. Twenty-two women (age range, 36-75 years; mean age, 56.4 years) with pure DCIS (seven with low-grade DCIS, five with intermediate-grade DCIS, and seven with high-grade DCIS) and three with microinvasion underwent breast MR imaging at 1.5 T between January 2008 and November 2010. MR examinations included contrast material-enhanced (gadoteridol) T1-weighted imaging and diffusion-weighted MR imaging with b values of 0 and 1000 sec/mm(2). ADC maps were generated. The distributions of the ADCs in regions of interest covering the lesions were compared among the three grades by using linear mixed-model analysis, and the discriminatory power of the lesion minimum ADC was determined with receiver operating characteristic analysis. The mean ADC was 1.42 × 10(-3) mm(2)/sec (95% confidence interval [CI]: 1.31 × 10(-3) mm(2)/sec, 1.54 × 10(-3) mm(2)/sec) for low-grade DCIS, 1.23 × 10(-3) mm(2)/sec (95% CI: 1.10 × 10(-3) mm(2)/sec, 1.36 × 10(-3) mm(2)/sec) for intermediate-grade DCIS, 1.19 × 10(-3) mm(2)/sec (95% CI: 1.08 × 10(-3) mm(2)/sec, 1.30 × 10(-3) mm(2)/sec) for high-grade DCIS, and 2.06 × 10(-3) mm(2)/sec (95% CI: 1.94 × 10(-3) mm(2)/sec, 2.18 × 10(-3) mm(2)/sec) for normal breast tissue. The mean ADCs for high- and intermediate-grade DCIS were significantly lower than that for low-grade DCIS (P < .01 and P = .03, respectively), and the mean ADC for low-grade DCIS was significantly lower than that for normal tissue (P < .001). The lesion minimum ADC for low-grade DCIS was also significantly higher than that for high- and intermediate-grade DCIS (P < .01). A threshold of 1.30 × 10(-3) mm(2)/sec for the minimum ADC in the diagnosis of low-grade DCIS had a specificity of 100% (12 of 12 patients; 95% CI: 73.5%, 100%) and a positive predictive value of 100% (four of four patients; 95% CI: 39.8%, 100%). These preliminary results suggest that quantitative diffusion-weighted MR imaging could be used to identify patients with low-grade DCIS with very high specificity. If the results of this study are confirmed, this approach could potentially spare those patients from invasive approaches such as mastectomy or axillary lymph node excision.