Sanaz A. Jansen PhD, Presenter: Nothing to Disclose

Suzanne Conzen MD, Abstract Co-Author: Nothing to Disclose

Xiaobing Fan PhD, Abstract Co-Author: Nothing to Disclose

Erica Markiewicz BA, Abstract Co-Author: Nothing to Disclose

Gillian Maclaine Newstead MD, Abstract Co-Author: Consultant, Naviscan Inc Consultant, Three Palm Software LLC Research suppport, Koninklijke Philips Electronics NV Spouse, stockholder, Hologic, Inc

Gregory Stanislaus Karczmar PhD, Abstract Co-Author: Nothing to Disclose

A major goal of oncologic imaging of the breast is to diagnose cancerous lesions at the earliest stage possible, including as pre-invasive ductal carcinoma in situ (DCIS). This goal is balanced by concerns that DCIS may be overdiagnosed as some indirect suggests not all will progress into invasive cancers. The purpose of this study is to use imaging of transgenic mice to gain insights into both the detection and progression of DCIS.

A total of 36 C3(1) SV40 Tag mice were used with IACUC compliance for the following studies targeting murine DCIS: a) Dynamic contrast enhanced MR imaging (DCEMRI): 12 mice were injected with Gd-DTPA and contrast kinetics was modeled in DCIS according to a two compartment model, yielding physiologic parameters Ktrans and ve. b) Progression: Serial MR imaging every 2-3 weeks was performed in 12 mice. Each DCIS lesion was examined to determine if and when it progressed into invasive carcinoma. c) Relaxometry: Tissue magnetic parameters T1, T2, and T2* of DCIS lesions were measured in 12 mice. All experiments used fat-suppressed T1-weighted gradient echo images of inguinal mammary glands for lesion identification.  

(a) DCEMRI demonstrated gadolinium uptake in ROIs placed on single ducts with DCIS, with average Ktrans=0.21±0.14(min-1) and ve=0.40±0.16. (b)DCIS lesions took vastly different progression paths: (i) 9/21 progressed to invasive tumors with an average progression time of 4.56 +/- 1.9 weeks (ii) 5/21 were stable for over 8 weeks, and were identified by a statistical model to represent indolent disease, and (iii) 2/21 lesions regressed. (c) DCIS lesions displayed biexponential T2 and T2*decay, and short T2* components (<~1ms). DCIS lesions were better appreciated on shorter TE images in terms of morphology, size and signal-to-noise ratio.

Results from in vivo MR imaging of early murine mammary cancer imply that gadolinium is present inside ducts with DCIS and that imaging at shorter TE may improve visualization of lesions. In addition, serial MR imaging provides direct evidence that DCIS is a nonobligate precursor to invasive tumors, even in mice that are genetically predisposed to develop invasive carcinoma.

The methods and data here provide a new foundation for using MR imaging in preclinical studies of early breast cancer, from evaluation of preventative/interventional therapies to improving detection.

Jansen, S, Conzen, S, Fan, X, Markiewicz, E, Newstead, G, Karczmar, G, Insights into Early Breast Cancer Detection and Progression Using Magnetic Resonance Imaging of Transgenic Mice.  Radiological Society of North America 2009 Scientific Assembly and Annual Meeting, November 29 - December 4, 2009 ,Chicago IL. http://archive.rsna.org/2009/8006357.html