In this issue of ONCOLOGY, Drs Oppong and King provide an organized and comprehensive summary of guidelines for the clinical management of patients with lobular carcinoma in situ (LCIS). Although these approaches address breast cancer risk, it is noteworthy that management strategies for LCIS have not evolved from any unifying concept of LCIS as a disease—nor can they be targeted, as we are not able to predict with any certainty the disease course and outcome for individual women who carry the diagnosis. Rather, the authors present a set of interventions—logical though they are—that have necessarily been formulated piecemeal from diverse, fragmentary clinical and pathological observations. We should bear in mind that even the diagnosis of lobular neoplasia (LN; ie, LCIS and/or atypical lobular hyperplasia [ALH]) is accidental; it is usually an incidental finding on breast biopsies that were performed to target other lesions. LN has no distinctive clinical presentation or imaging features. The prevalence of LN therefore likely exceeds incidence. Thus, it might fairly be said that the clinical management of LCIS modifies an outcome that is driven at least as much by what we don’t see as by what we do see.
This awkward state of affairs is hardly surprising given the complexities of this disease process. Although we now know that all forms of LN are unified by functionally significant loss of E-cadherin gene expression, the histological distinction between ALH and LCIS is often problematic and based largely on subtle differences in the degree to which lobules are involved. Second, as demonstrated by the observed intervals between lobular neoplasia diagnosis and subsequent breast cancers, we know that the natural history of disease progression is quite variable but may be decades long, depending at least in part on whether there is occult malignancy present in unsampled breast tissue at the time of LN diagnosis. Finally, we know that lobular neoplasia is characterized not only by cytological heterogeneity but also by variable yet frequent multicentricity; however, we have no standard methods for quantifying either cytological type or volume of disease. The latter is relevant to the degree that cohort studies reveal strong associations between disease extent and patient risk.[1,2]
Most troubling from a pathogenetic standpoint is the apparently chimeric nature of LN as both a marker of constitutional (ie, bilateral) risk as well as a direct precursor for some invasive carcinomas. In our experience with ALH, the role of risk marker appears to be more prevalent. Specifically, we studied 175 women with ALH and followed them for later breast cancers. Breast cancer developed later in 34, and only 3 of these cancers were invasive lobular cancers. On the other hand, Page et al have reported the disproportionate ipsilateral cancer risk in patients with LN. Recent molecular studies, noted by King, have documented the role of direct precursor in some cases; these studies highlight the need for improved subclassification of LN in order to facilitate the individualization of local as well as systemic interventions. Collectively, the above considerations imply that definitive characterization of LCIS as a clinicopathological entity would be a daunting task and would likely require hundreds of patients, decades of follow-up, exhaustive tissue sampling, and comprehensive genomic analyses. We applaud Oppong and King for their recommendation of prospective studies, particularly with respect to surgical excision following core biopsy diagnosis and the assessment of margin status in these samples, since both these areas are of immediate clinical import and are lacking in sufficient data.
Is it possible, given the complexities reviewed above, to even propose a coherent explanation for the role of LN in breast carcinogenesis? We believe it would be useful to try, if only to provide a framework for the eventual testing of hypotheses. The following is an attempt to develop a preliminary basis for understanding:
1. LCIS is characterized by clonal genetic alterations and is therefore a neoplasm in the conventional sense of the term. However, while LN is a clue that other more aggressive lesions may be present in unsampled tissue, it may in some cases represent a self-limited process that lacks the cellular-genetic events necessary for neoplastic progression.
2. LN is a component and early manifestation of a definable subset of breast neoplasia having a distinctive gene expression signature and characterized by as yet undefined dysregulation of hormone receptors. In terms of genetic pathogenesis, recent studies demonstrate that LN and atypical ductal hyperplasia (ADH) have a greater degree of similarity than do ADH and high-grade ductal carcinoma in situ (DCIS).[4,5] Hence, the traditional distinction between “ductal” and “lobular” may have less relevance than that between low- and high-grade early precursor lesions. Although mortality data are limited, the few studies published do not suggest that LN constitutes a short-term risk for breast cancer death.
3. Unfortunately, to our knowledge there is presently no reliable clinical test to predict cancer risk for individual women with ALH or LCIS.[7,8] We believe that multiparameter models using number of foci, age, and other histological data may eventually refine our ability to predict subsequent breast cancer events. Eventual discovery of cellular or genetic changes might also better define progression potential. The well-documented outcome differences between pathologic LN subsets (ie, minimal ALH, ALH with ductal involvement, LCIS, etc) suggest that such advances are within reach.
Obviously the microscopic nature of LCIS, in addition to its complete lack of clinical detectability, represents a formidable barrier to the full elucidation of its role in breast cancer pathogenesis. Fortunately, the results of chemoprevention studies demonstrate that effective management is nonetheless possible without a detailed understanding of LCIS biology and pathogenesis. For those of us who study early events in breast cancer pathogenesis, clinical progress in the management of women with LN should emphasize not merely better prediction of which patients will develop breast cancers, but also improved ability to detect clinically occult markers of increased risk, including but by no means restricted to LN.
Financial Disclosure: The authors have no significant financial interest or other relationship with the manufacturers of any products or providers of any service mentioned in this article.
1. Degnim AC, Visscher DW, Berman HK, et al. Stratification of breast cancer risk in women with atypia: a Mayo cohort study. J Clin Onc. 2007;25:2671-7.
2. Page DL, Kidd TE, Dupont WD, et al. Lobular neoplasia of the breast: higher risk for subsequent invasive cancer predicted by more extensive disease. Hum Pathol. 1991;22: 1229-32.
3. Page DL, Schuyler PE, Dupont WD, et al. Atypical lobular hyperplasia as a unilateral predictor of breast cancer risk: a retrospective cohort study. Lancet. 2003;361:125-9.
4. Simpson PT, Reis-Filho JS, Gale T. Molecular Evolution of breast cancer. J Pathol. 2005;205:248-54.
5. Abdel-Fatah TMA, Powe DG, Hodi Z, et al. High frequency of co-existence of columnar cell lesions, lobular neoplasia and low grade ductal carcinoma in situ with invasive tubular carcinoma and invasive lobular carcinoma. Am J Surg Pathol. 2007;13:417-26.
6. Fisher ER, Land SR, Fisher B, et al. Pathologic findings from the National Surgical Adjuvant Breast and Bowel Project: 12-year observations concerning lobular carcinoma in situ. Cancer. 2004;100:238-44.
7. Pankratz VS, Hartmann LC, Degnim AC, et al. Assessment of the accuracy of the Gail model in women with atypical hyperplasia. J Clin Oncol. 2008;26:5374-9.
8. Boughey JC, Hartmann LC, Anderson SS, et al. Evaluation of the Tyrer-Cuzick (International Breast Cancer Intervention Study) model for breast cancer risk prediction in women with atypical hyperplasia. J Clin Oncol. 2010;28:3591-6.