Rosen's Breast Pathology, 4e

xvii

Introduction

intraepithelial stage, there are substantial differences be- tween organ sites. Among these differences from a clinical standpoint, accessibility is a critical issue when it comes to treatment. Thus, it is possible to directly observe and follow over time the epithelium of organs with external orifices such as the uterine cervix, the urinary bladder and the gas- trointestinal tract. Repeated cytologic samples can be taken from these sites without performing a surgical procedure that removes part of the lesion. These organs, and others such as the prostate gland, can also be monitored with small biopsy samples. As a consequence, it is possible to track preinvasive lesions at these sites in order to distinguish between conditions that can be managed with minimally invasive procedures and those that require more aggressive treatment. On the other hand, despite repeated efforts to improve mammary ductoscopy, virtually all of the complex glandular structure of the breast where carcinomas arise is not acces- sible for direct observation, and current imaging technolo- gies only reveal structural alterations without identifying the precise causes of the changes. Sequential imaging that requires repeated radiation exposure only documents the evolution of structural changes. For example, a mass lesion caused by mammary carcinoma usually has an invasive component, but in some circumstances it may consist entirely of noninvasive, intraepithelial carcinoma (DCIS) when excised and thoroughly examined microscopically. If such a tumor were investigated by needle biopsy and the sample proved to be DCIS, there is currently no method (comparable to culposcopy of the uterine cervix) whereby the nature of the remainder of the tumor could be ascer- tained without removing it entirely. Applying the paradigm of the uterine cervix in the forgoing circumstance would require allowing the mass to remain in the breast and following it with serial imaging studies. If by chance an invasive focus already existed in the mass, this procedure would allow the invasive carcinoma to expand and increase the likelihood of metastatic spread. If left in place, whether indolent or not, the incompletely excised lesion would prob- ably enlarge over time, creating growing anxiety, and yet it might remain entirely noninvasive. This would clearly be an unacceptable way to manage DCIS clinically, and changing the name of the lesion would not solve this conundrum, nor need it change treatment after the diagnosis. Within the lim- its of our current knowledge, the cornerstone of clinical care for intraepithelial ductal carcinoma of the breast (DCIS) is complete excision and thorough histologic examination of the lesion when it is detected. At present, no procedure short of this can provide the full appreciation of the nature and extent of the lesion that is needed to guide treatment. In reviewing the proposal to remove the word carcinoma from the intraepithelial, noninvasive stage of ductal carci- noma, one is reminded of the famed manager of the New York Yankees baseball team, Yogi Berra, who is credited with the remark, “It’s like déjà vu all over again.” In the current context, the effort to delete carcinoma from DCIS

repeats the scenario of the 1980s and 1990s when there was a movement to replace LCIS with names such as lobular neoplasia (LN). The very same reasons were put forth for this change as are now offered for removing the word carci- noma from DCIS: changing the name would reduce patient anxiety and lead to less aggressive therapy. One phase of this effort was to promulgate the concept that LCIS was only a marker of breast carcinoma risk and not a precursor to invasive carcinoma. If this were true, mastectomy, then a frequent treatment option after the diagnosis was made by biopsy, would no longer be appropriate. Research conducted in the past two decades, described in detail in Chapter 31, clearly demonstrated that LCIS is a nonobligate precursor to invasive lobular carcinoma. This means that not every patient with LCIS develops invasive lobular carcinoma in their lifetime, but when it occurs, invasive lobular carcinoma arises from LCIS. It is also well established that patients with LCIS are at increased risk to develop invasive ductal carcinoma that arises independently from the LCIS. In the ensuing years, little progress has been made in distinguishing patients with LCIS who are likely to experience progression to either type of invasive carcinoma from those unlikely to have this outcome. Now, approxi- mately 30 years later, clinical management for LCIS usually consists of medical therapy and clinical observation for a disease that is again widely referred to as LCIS. Thus, in the case of LCIS, concern with “overtreatment” was expressed in proposals to change the name of the dis- ease, but other factors played a much more important role in creating a new treatment paradigm. Most important of these was a better understanding of the clinical course of the disease that resulted from retrospective long-term follow-up studies of untreated patients published in 1978 and sub- sequent prospective studies that are described in detail in Chapter 31. Historically, the movement away from mastec- tomy for the treatment of LCIS coincided with the discovery of hormone receptor proteins, particularly those involved in the response of mammary carcinoma to estrogens and progesterone, and the introduction of selective estrogen receptor modulators (SERMS) such as tamoxifen that have significantly reduced the risk of developing invasive carci- noma in patients with LCIS. While efforts to delete the word “carcinoma” from LCIS may have prompted controversy and drawn attention to the need for changes in therapy, it was better understanding of the clinical and biologic charac- teristics of the disease, accompanied by new, effective forms of treatment that brought about the desired result. The foregoing notwithstanding, the concerns expressed by Drs. Esserman, Thompson, and Reid raise an important question about our understanding of the biology of DCIS. Retrospective, long-term follow-up studies described in detail in Chapter 11 demonstrated that if left untreated after a biopsy, about 40% of patients with low-grade DCIS devel- oped invasive carcinoma. Among the remaining majority of patients, the absence of subsequent invasive carcinoma may have been due to complete removal of the DCIS in the