Tumor markers for cancer detection. II.

Updated results of a prospective study assessing the value of tumor marker determinations in a supposedly healthy population (2,000) for identification of a group at risk for cancer are reported. With observation periods varying from 1 to 6 years (mean 3.5 years), repeated determinations by RIA were routinely carried out for CEA, AFP, beta-HCG, beta 2-M, ferritin, and, more recently, beta 1-SP. Preliminary data on TPA, CA 12-5, and CA 19-9 were also obtained. A comparative study of methods for CEA determination using monoclonal and polyclonal antibodies revealed that preference should be given to polyclonal antibodies. In the group considered to be "at risk" (ie, having at least one abnormal marker value) (N = 481), the cancer detection rate was 29 per 1,000 against 3.2 per 1,000 in the normal group (N = 1,519). These figures were significant, even if the number of malignancies detected was small (N = 27). By associating general tumor markers such as CEA, TPA, and CA 19-9 with site-specific markers such as PAP and CA 12-5, it seemed that marker determinations played a useful role in risk assessment in cancer detection programs.