Currently there is evidence to support the inclusion of screening for breast and cervix cancer in programs for cancer control. Breast cancer programs have an important impact in women over the age of 50, but increasingly there are suggestions that, at least in the early years, there are no benefits for women age 40 to 49, and even that mortality from breast cancer may be increased. Cervix cancer programs can have a major impact if appropriately organized. There is no justification at present to recommend screening for colorectal cancer. Screening can be expected to have only a minor influence on reduction in cancer mortality by the year 2000. Biomarkers may be of value as indicators of risk or indicators of disease. Some have been proposed for the former, such as mammographic parenchymal patterns and breast cancer risk. There are good theoretical reasons for not incorporating them in screening programs, as the program sensitivity will decrease, even if the program specificity increases. Biomarkers as indicators of disease may be valuable when the marker is linked to the etiologic agent for the disease, providing that valid screening tests can be developed for the marker, or could be a biological indicator of the presence of disease, such as a monoclonal antibody, specific to the cancer. Examples include markers for hepatitis B virus (HBV) infection and liver cancer screening, and potentially in the future, markers for human papillomavirus (HPV) infection and cervix cancer screening. Specific monoclonal antibodies to cancer utilizable in screening are being sought, but are not yet available for use. One example under evaluation is the CA 125 monoclonal for ovarian cancer.