About Our Research

Proteins made by the breast duct cells and shed into the breast fluid can be measured. We believe that proteins made by cancerous cells are different than normal cells, and these differences can be detected and used to screen for breast cancer.

In the first figure, we use a technique called mass spectrometry to measure the proteins in the breast fluid of women with or without breast cancer. The technique allows us to measure many proteins at once, since they show up based on their size.

In the second figure, we measued a growth factor, basic fibroblast growth factor, in the breast fluid. This concentration of this growth factor was shown in a small study to be associated with breast cancer.

Prostate-specific antigen (PSA) and insulin-like growth factor binding protein type 3 (IGFBP-3) are compounds made in the breast and we have found them to be associated with breast cancer. Notice how the odds of breast cancer (Odds Ratio) decreases as the PSA value (log PSA) increases. This suggests that high PSA values in the breast fluid are good, whereas low values are worrisome. IGFBP-3 values are the opposite. Low values appear to be good, and high values worrisome. It is noteworthy that very low PSA values in premenopausal women are more worrisome than in postmenopausal women. We have found that female hormones tend to increase PSA values, so when premenopausal women have very low values, the likelihood that something is wrong increases.

Thousands of women undergo mastectomy because we are concerned that the breast contains more cancer after biopsy, based on x-rays, physical exam, and/or prior biopsy results. Unfortunately, the mastectomy specimen often does not contain more cancer. Our hope was that NAF might help us to identify a group of women who did not need mastectomy. We developed an algorithm suggesting that women with normal NAF cytology with a biopsy showing tumor close to but not at the margin, where the biopsy was close (<4 cm) to the breast nipple had a low chance of having cancer in the breast. These studies are encouraging, but need to be confirmed before they are put into general practice.

Figure 1. Proteins in NAF from women with breast cancer (top 4 rows) or women without breast cancer (bottom 4 rows). Certain proteins (e.g. 15940) are present only in women with cancer.

Figure 2. bFGF levels were measured in women with (Yes) or without (No) breast cancer. Levels are higher in women with cancer.

Figure 3. PSA and IGFBP-3 levels in NAF were measured in women of all risk categories, ranging from normal risk to known breast cancer. The odds of a woman having breast cancer increased as the PSA value decreased. The steepness of the slope was greater for premenopausal women. The odds of a woman having breast cancer increased as the IGFBP-3 value increased.

Figure 4. Suggested decision making plan to consider which women may be candidates to avoid mastectomy, despite the presence of a biopsy showing breast cancer with tumor at or near the margin.

Top