National Cancer Institute Home at the National Institutes of Health |
Please wait while this form is being loaded....

Problems in the Diagnosis of Ductal Carcinoma in Situ

Stuart J. Schnitt, MD; Beth Israel Deaconess Medical Center & Harvard Medical School

View PresentationIcon indicating linked file is archived content

The diagnosis of DCIS is not always straightforward. Many studies have illustrated the central problem in diagnosing DCIS, which is that the heterogeneity of the disease allows it to be easily confused with a variety of other diseases, such as atypical ductal hyperplasia (ADH), microinvasive carcinoma, invasive carcinoma, lymphatic invasion, lobular carcinoma in situ (LCIS), and other intraductal lesions.

Given that there are documented, clinically important differences between ADH and fully developed low-grade DCIS, they should be considered distinct with regard to patient management. However, it is exceedingly difficult to differentiate between ADH and low-grade DCIS because there are no qualitative histologic features or biomarkers that permit the reliable distinction between the two diseases in all cases.

Because DCIS may have areas that mimic invasion, such as duct branching, involvement of lobules and benign sclerosing lesions, distortion of involved spaces, and tangential sectioning, among others, DCIS may be over-diagnosed as microinvasive (or even frankly invasive) cancer. Unfortunately, distinction between mimics of invasion and real invasion is not always possible on standard histologic sections. Additionally, because microinvasive foci may be overlooked or not sampled, under-diagnosis may also occur. In aggregate, however, there are no clear differences between DCIS and microinvasive cancer with regard to disease-free or overall survival; therefore, patients with large areas of DCIS with and without demonstrable microinvasion should be managed similarly.

DCIS and LCIS are managed differently -- DCIS is seen as a precursor and treated with complete eradication and margin evaluation, whereas LCIS is seen as a risk factor and observed (with or without tamoxifen) with no margin evaluation. Because there can be overlap in distribution within the ductal-lobular system (i.e., DCIS can involve identifiable lobules and LCIS can involve ducts) and lesions of one may have features more commonly associated with the other, problems can occur when attempting to distinguish the two disease entities. However, e-cadherin immnostaining may be of help in problematic cases.

Currently, the most appropriate management of patients with histologically ambiguous in situ lesions is unknown.

Last Modified: 18 Oct 2013