Radiology: Breast MR could inform personalized DCIS treatment

A model incorporating dynamic contrast material-enhanced (DCE) and diffusion-weighted (DW) MR features could help identify women with high-risk ductal carcinoma in situ (DCIS), which may provide information to facilitate lesion-specific targeted therapies, according to a study published in the May issue of Radiology.  

The incidence of DCIS has risen in the last few decades and detection is associated with benefits and potential harms, including unnecessary surgery and radiation treatment for women with disease that would not have progressed to invasive cancer. “Thus, the identification of highly specific imaging biomarkers that could allow more individualized therapies that accurately match lesion-specific DCIS biology could have substantial benefits for patient morbidity,” wrote Habib Rahbar, MD, from the department of radiology at University of Washington, Seattle Cancer Care Alliance, and colleagues.

Rahbar and colleagues aimed to develop a model incorporating DCE and DW MR features to differentiate high-nuclear-grade DCIS, which correlates with a higher risk of progression to invasive disease and local recurrence, from non-high-nuclear-grade DCIS.

The researchers focused on 55 distinct pure DCIS lesions in 52 women who underwent breast MR imaging from Oct. 7, 2005, to June 7, 2008.

Four fellowship-trained radiologists prospectively interpreted DCE MR images, and a breast imaging fellow retrospectively analyzed DW image data.

High-nuclear-grade lesions exhibited a lower contrast-to-noise ratio than non-high-nuclear-grade lesions on DW images with a b of 600 sec/mm², wrote Rahbar et al. Mean contrast-to-noise ratio for high-nuclear-grade lesions was 0.71. For non-high-nuclear-grade lesions the mean contrast-to-noise ratio was 2.9. This difference might be explained by higher fluid content in non-high-nuclear-grade lesions, according to the researchers.

In addition, there were significant differences in mean contrast-to-noise ratio between high-nuclear-grade and non-high-nuclear-grade DCIS on DW images with a b of 0 sec/mm² at 0.07 and 0.92, respectively.

On DCE images, high-nuclear-grade lesions had a larger mean maximum lesion size, averaging 15 mm larger in size than non-high-nuclear-grade lesions.

“The ability to accurately predict in vivo DCIS grade has multiple clinical implications,” wrote Rahbar and colleagues. An effective predictive model might identify which lesions are likely to be upgraded after surgical excision and allow for more accurate pretreatment assessment and planning.

A combination of imaging, histopathologic and clinical data could facilitate a model for more individualized therapy for DCIS, continued the researchers. Specifically, a model might identify non-high-nuclear-grade lesions that have a low likelihood of recurrence without radiation therapy and thus decrease morbidity by identifying patients in whom radiation therapy is unnecessary.