Clinical implications of abnormal mammograms

In a mature screening population, two or three breast cancers a year are diagnosed per 1,000 screening mammograms performed. To make the diagnosis, however, a fair number of biopsies must be performed. The best evidence suggests that approximately one-third of patients undergoing a biopsy for a mammogram- detected lesion will have breast cancer. In a very mature, well-audited mammography practice, that number may be as high as 40 percent.

In the United States the average is about 20 percent. We analyzed western New York claims data by surgeons, radiologists and medical centers and found the rate of breast cancer detected by mammogram ranged from seven to 40 percent (2.1). Interestingly, rates of biopsy-confirmed breast cancer were not correlated with the volume of mammograms performed.

Another issue is the potential overuse of biopsy for mammograms that are socalled “probably benign,” or BIRADS-3 mammograms. Abnormal mammograms are a source of enormous consternation and emotional distress for women, and the medical system isn’t well set up to provide counseling and support for those women. They go to a mammography center, have an abnormal mammogram, are set up for a stereotactic needle biopsy and are sent to a surgeon whom they’ve never met. No support is available for them.

We need better mechanisms to inform patients about what we’re looking for in mammograms and what the appropriate evaluation would be for them. Patients need to understand that time is not of the essence, and they can take time to ensure that they are appropriately evaluated and receive all the opinions they need.

We also need better mechanisms to reassure women when we recommend short-term follow-up as opposed to biopsy. My office is filled every week with women who have been told they need a short-term follow-up and are petrified that they might have breast cancer. They ask, “Why would I wait six months to find out if I’ve got breast cancer?” And yet, if we performed a biopsy on every woman who had a BIRADS-3 mammogram, they’d be lined up out the door and would undergo a lot of unnecessary morbidity.

Radiotherapy in the management of patients with DCIS

We have an unresolved dilemma between the high-level evidence from randomized clinical trials and anecdotal evidence, suggesting that large subsets of patients with in situ carcinoma do not need radiation therapy. The very best evidence suggests radiation therapy does not affect survival. A high percentage of local failures are in situ and most that are invasive are effectively treated. A very small percentage of individuals with in situ carcinoma will ultimately die of breast cancer whether we give them radiation therapy after breast-conserving surgery or not.

The NCCN guidelines recommend radiation therapy for almost all women with DCIS. The guidelines state the use of radiation therapy may be considered optional for individuals with lesions that are less than one-half centimeter and low grade, but I think many surgeons and many radiation oncologists extend that size limit well above one centimeter.

Sequencing and switching hormonal therapy

The aromatase inhibitors appear to be equivalent or even more effective than tamoxifen up front. My only hesitation is the lack of long-term follow-up in patients receiving aromatase inhibitors for a significant period of time.

I think many, if not most, oncologists have switched their practice and are exclusively using aromatase inhibitors in their postmenopausal patients. I think many of them are also using aromatase inhibitors quite broadly after five years of tamoxifen. Many of us in the NCCN and certainly in my cancer center are a little more skeptical about that approach, particularly in women in whom breast cancer is detected very early.

Many of my patients who had very small tumors are on hormonal therapy alone, so my practice is skewed to patients with node-negative, ER-positive, T1B breast cancer who are now in their mid-seventies. Should they be switched to letrozole after five years of tamoxifen? Their risk of recurrence after five years with an eight-millimeter Grade II cancer is one to three percent.

Letrozole may reduce that from three to 2.7 percent with a therapy that may impact bone health and result in fractures, and certainly will entail a cost to patients. Many of my older patients do not have prescription coverage. I do not think the blanket use of these drugs after five years of tamoxifen is necessarily the right answer.

In patients at higher risk, I’d be more likely to switch from tamoxifen to an aromatase inhibitor. These women have a substantial risk of distant recurrence after five years. In women with very high-risk disease, I have recommended continuing tamoxifen after completing five years. There’s a theoretical chance that we might select resistant cells, but I think many medical oncologists have left people on tamoxifen in that situation. Currently, I would not hesitate to put women at higher risk on an aromatase inhibitor.

Contraindications to sentinel lymph node biopsy

I do not perform a needle biopsy for patients with an obviously positive node, rather I perform an axillary lymph node dissection. I can’t remember a patient in whom I dissected axillary lymph nodes and they turned out to have negative nodes. I have not performed sentinel node biopsy with neoadjuvant chemotherapy because we are generally doing neoadjuvant chemotherapy for large breast cancers with high rates of node involvement. When one truly has locally advanced breast cancer, I think you’re beyond the issue of cosmetics and concerns about lymphedema. Locally advanced breast cancer is a horrible, life-threatening and mutilating disease. Local failures in breast cancer are disasters that occur in only a small minority of our patients, so I’ve not regretted performing axillary node dissections in patients with locally advanced breast cancer.

This year we’ve modified the NCCN guidelines to allow consideration of sentinel node biopsy before neoadjuvant therapy and then omission of axillary node dissection. One of the premises behind that change is that neoadjuvant chemotherapy and hormonal therapy are being utilized with smaller cancers than in the past. It wouldn’t seem right to deny a patient a sentinel node biopsy because we put them on a neoadjuvant hormonal trial.

Select publications

Dr Edge is Chair of the Department of Breast and Soft Tissue Surgery at the Roswell Park Cancer Institute and Professor of Surgery at the State University of New York at Buffalo in Buffalo, New York.

Home · Search