Vol. 05 Issue 1, Winter 2000
Charles Garbo, MD, board certified medical oncologist in private practice in Ithaca, New York
Formerly Assistant Professor,University of Massachusetts Medical Center, Worcester, Massachusetts
When assessing the needed therapy for and prognosis of an individual with breast cancer, there are a number of critical issues. The most important relate to the stage of the cancer (size, lymph node involvement and presence or absence of metastatic disease). Also important are characteristics related to the patient's health - including other medical problems, menopausal status, and the patient's own preference for types of therapy. In addition, the efficacy of the various therapies, their toxicity and the amount of evidence for improved outcomes - survival, lack of recurrence, etc. - are important.
There are other important factors which don't influence the stage of cancer which also have an impact. These include grade (how aggressive the cancer appears under the microscope), hormone receptor status (whether the tumor is likely to respond to hormonal therapy), other receptors, and the histology of the breast cancer. This article will focus on the histology.
Since the breast is a glandular structure, almost all malignancies of the breast are glandular cancers, also called adenocarcinomas. The current belief is that all or almost all invasive breast cancers start in a pre-invasive stage, also called in situ. These lesions don't have the ability to spread to lymph nodes or metastasize unless/until they become invasive. There are two very distinct types of in situ carcinomas of the breast - lobular and ductal.
Lobular-carcinoma-in situ (LCIS) is usually an incidental microscopic finding, as it lacks clinical or mammographic signs. It is not truly a malignancy but a marker for increased risk of breast cancer development in both breasts. There is approximately 10 -15% risk of ipsilateral (the same side of the body) and another 10% risk contralateral (opposite side of the body) invasive breast cancer occurring. Despite the fact that the initial lesion is lobular, most of the subsequent cancers are invasive ductal carcinomas (only about 30% are lobular cancers). The fact that most women with LCIS don't go on to develop breast cancer, that the risk is bilateral and that the usual cancers which occur are ductal prove that this is a marker - a risk for breast cancer and not an actual pre-invasive cancer. With LCIS, there is about a 1% chance per year of developing invasive cancer and it should be thought of as a risk factor, just like family history, age, or nulliparity. Occasionally a decision is made for bilateral mastectomy, especially if there is the additive risk factor of a strong family history. There is no role for unilateral mastectomy or wide local excision or radiation therapy.
The other non-invasive breast cancer, ductal-carcinoma-in situ (DCIS) is very different. It is an actual pre-invasive malignancy with risk of invasive cancer at that site. Accordingly, therapy and risks are very different from LCIS. DCIS is almost always discovered as an abnormality on a mammogram, usually as clustered microcalcifications. In the days before the routine use of mammograms, it was a rare diagnosis. Now it accounts for 20% or more of all breast cancers. There has probably been more debate about the treatment of DCIS than any other type of breast cancer. Even after conservative therapy (less than mastectomy) became a standard option for invasive breast cancer, mastectomy remained standard for DCIS. Today, even though no randomized trial of mastectomy compared to lesser therapies exist, the vast majority of women with DCIS don't need and don't have a mastectomy.
DCIS can be cured with mastectomy 98 - 99% of the time. The lymph nodes are only involved about 1% of the time. With conservative therapy there is certainly a higher percentage of recurrences of both DCIS and invasive cancer, however in appropriately selected patients, survival can be 97% at ten years. A number of studies have looked at whether adding other therapies to lumpectomy improves the outcome. The National Surgical Adjuvant Breast Project (NSABP) has performed a number of studies. NSABP B17 showed that adding radiation to conservative surgery lessened the recurrence rates. More recently NSABP B24 has shown that adding Tamoxifen reduces recurrence in both breasts.
There are still times when mastectomy is needed: (1) inability to remove all the DCIS with conservative surgery; (2) very large areas of DCIS or (3) extensive DCIS that is multifocal and involving multiple quadrants of the breast. In addition, DCIS is a heterogeneous group of lesions varying by patterns of growth. Most commonly these are divided into comedo-type DCIS which has prominent necrosis and large cells, and non-comedo-type. The non-comedo-type DCIS includes cribriform, micropapillary, papillary and other less common types. Comedo type appears more aggressive histologically and also has markers of more aggressive disease. It is less commonly receptor-positive; more commonly it has abnormal amounts of DNA (aneuploid) and more often expresses "bad" oncogenes.
Treatment of DCIS needs to be individualized based on the tumor's characteristics, the family history and the patient's preferences.
Before turning to invasive breast cancer I'd like to discuss benign breast conditions - especially hyperplasia or overgrowth of cells. With "mild hyperplasia of the usual type," i.e. without atypical features, there is not an increased risk of breast cancer development. In "moderate or florid hyperplasia of the usual type" there is a slight increased risk which rises in conjunction with a positive family history of breast cancer. Atypical hyperplasias have some of the characteristics of LCIS or DCIS but lack the complete pathologic criteria, either because of small size or involvement of just one duct or lobule. These lesions have a relatively high risk of developing into invasive breast cancer. The table below presents the relative risk of breast cancer for all of the categories discussed to this point.
NSABP recently completed its first prevention trial looking at Tamoxifen versus placebo for women at high risk. Those with atypical hyperplasia were among those at highest risk and who had the biggest benefit from Tamoxifen. Benefit has also been shown in LCIS and DCIS.
Among invasive breast cancers (those with the potential to metastasize) the majority are invasive ductal. This includes those tumors that have some characteristics of the more specific types discussed below, but which don't fully meet their criteria. About 30% of ductal carcinomas have some other features. Invasive ductal carcinomas form solid tumors and when large enough are usually palpable. Size and grade are the most important local determinants of prognosis. The more a cancer resembles normal breast tissue, the lower the grade and the better the prognosis. This is certainly true for lymph node (LN) negative tumors and less clear for LN positive tumors. Invasive ductal cancers are found in all age groups and are the typical cancers against which other histologic types are compared.
Other factors having prognostic significance for invasive ductal carcinoma include: blood vessel invasion, invasion of the nerves (perineural invasion), and presence of HER2/neu over expression.
There are several types of invasive cancer which are much less common, have distinctive histologic features and tend to have a better outcome than the invasive ductal carcinomas. Tubular, also called well-differentiated carcinoma, is as the name implies a cancer that closely resembles normal mammary ducts. They almost always have an in situ component. While they constitute only 1 - 2% of breast carcinomas, among small, mammographically detected tumors of 1 cm or less, they constitute 5 - 10%. Because they are smaller and slower growing, they are less likely to involve lymph nodes or to metastasize. Their prognosis is much better than the average breast cancer, even taking their small size into consideration.
Mucinous carcinomas also make up 1 - 2% of breast carcinomas, but more commonly occur in the elderly (over age 75), where more than 5% of carcinomas are mucinous. As the name implies, they secrete mucin and little glandular tissue is seen. They tend to have a better outcome than the typical invasive ductal carcinomas early on but have a high rate of metastases at 10 years or greater.
Medullary carcinoma comprises about 7% of breast carcinomas; it is more common in younger individuals, less than 50 years old. It tends to present with large tumors. Although they tend to grow rapidly and present as large lesions, their prognosis is also better than average. They are less likely to go to axillary lymph nodes and when the nodes aren't involved, the prognosis is especially favorable. This diagnosis has a very rigid set of pathological criteria and if not all the features are present it is called atypical medullary carcinoma. The outcome for this variant is not as good.
Papillary carcinomas make up another 1 - 2%. They are more commonly located near the nipple and frequently present with nipple discharge. They have a frond-like or papillary growth pattern. There are several other even rarer carcinomas which start in the ducts. Details on these can be found in the references.
The other common breast carcinoma is lobular. About 10 - 15% of breast carcinomas are infiltrating lobular. They present much more frequently as a mass and less commonly on mammograms than other types. They don't cause calcification on mammograms. Both grossly and microscopically these tumors don't have well-defined margins. They are more commonly bilateral than other types - either synchronously or metachronously. There are often components of other histological types mixed in. Microscopically these cells are often arranged in a linear fashion.
Infiltrating lobular carcinoma has a different pattern of spread than other histologic types - more commonly going to the lining of the central nervous system (meninges) or to surfaces inside the abdomen. It can also invade many abdominal organs and mimic stomach, uterine or ovarian carcinomas. Overall the prognosis is similar to that of infiltrating ductal carcinoma.
An unusual but poor prognosis type of breast carcinoma is inflammatory carcinoma. The prognosis is poor enough that they are automatically classified as Stage III B (just one step below metastatic disease). Inflammatory carcinoma involves the skin and the lymphatics of the skin. It is usually large, often taking up more than one quadrant of the breast and is very rapidly growing. The skin often has a shriveled appearance called peau d'orange (skin of an orange). With multi-modality treatment in recent years, the prognosis has improved somewhat.
In summary, there are many different types of adenocarcinoma of the breast. Some of these types have clinical significance. The earlier proliferative breast lesions, both hyperplasia and in situ disease all present with increased risk of developing invasive breast cancer. The degree of risk and some of the potential therapies were discussed previously. For more detail on all of the above topics, please refer to the following references.
Fisher, B., J. Costantino, C. Redmond, E. Fisher, R. Margolese, N. Dimitrov, N. Wolmark, D.L. Wickerham, M. Deutsch, L. Ore, E. Mamounas, W. Poller and M. Kavanah. "Lumpectomy compared with lumpectomy and radiation therapy for the treatment of intraductal breast cancer." N. Engl. J. Med. 328 (1993): 1581-1586.
Harris, Jay, Marc E. Lippman and Monica Morrow, eds. Diseases of the Breast. NewYork: Lippincott William & Wilkins, 1995.
Holland, James F., Emil Frei III, Robert Bast Jr., Donald Kufe, Donald Morton and Ralph Weizhselbaum, eds. Cancer Medicine, 4th Edition. Hamilton, Ontario : BC Decker, 1997.
Rose, Paul Peter and Harol Oberman. Tumors of the Mammary Gland, 3rd Series. Washington, D.C.: American Registry of Pathology, 1994.
Tavassoli, Fattaneh A. Pathology of the Breast, 2nd Edition. New York: McGraw-Hill Professional Publishing, 1999.