The CA 15-3 protein is a member of the family of proteins known as mucins, whose normal function is cell protection and lubrication. It plays a role in reducing cell adhesion and is found throughout the body. Elevated levels in breast cancer tissue may be involved in metastasis. CA 15-3 levels can also be elevated in patients with other cancers (lung, colorectal, ovarian, pancreatic) or because of hepatitis or cirrhosis of the liver.
fluid would increase or decrease as the cancer progresses or regresses and could be determined by a simple, reliable, inexpensive assay.
Individual biomarkers currently in clinical and experimental use fall far short of this ideal, however. (See Table 5-1 for a summary of potential issues and limitations of biomarkers for specific events in the development of breast cancer.) Most are synthesized by normal as well as malignant tissues and are only rarely elevated in premalignant or early stage disease. For example, in cancerous breast tissue high levels of the protein CA 15-3 are produced, but usually not until the cancer has reached an advanced stage (Box 5-1).18 Few of the biomarkers in use today are found among all patients with a particular type of cancer, and with the exception of prostate specific antigen, none are organ-specific.18
Absent an ideal biomarker, it is likely that any biomarker-based assay used as a primary screen for breast cancer in normal-risk populations will produce significant numbers of false positives. However, biomarker-based screening may prove to be a practical means of screening women at high risk for breast cancer for premalignant disease and/or occult cancer. Such an assay could detect clusters of proliferating cells at a preclinical stage, as well as cell clusters that may never require treatment. With the discovery of additional or better markers, bioassays may eventually be developed that not only detect the presence of breast cancer or precancer, but also predict clinical course.
As with mammographic screens, the performance of a biomarker assay should increase as additional time points are taken, particularly if the marker(s) reflect disease burden. This is true of existing biomarker assays for prostate, ovarian, and colon cancer. Therefore, although it may be unreasonable to expect that a single assay measurement can replace mammographic screening, multiple measurements taken over time that show a consistent rise in value could be indicative of an enlarging mass. This type of algorithm is likely to be the first implementation for biomarkers in breast cancer screening.
Biomarker assays could also be used to aid the decision-making process