We have assessed the potential role of a test based upon the measurement of serum CA 125 in an ultrasound-based screening program for familial ovarian cancer. A sample of peripheral blood was taken from 1502 self-referred, asymptomatic women whose pedigree showed that at least one close relative had developed the disease. All women in the study underwent one screening by transvaginal ultrasonography (consisting of one or more scans) to detect any persistent lesion and a change in ovarian volume. Women with a positive result were referred for surgery. The concentration of serum CA 125 was measured in all samples at the end of the study. Seven ovarian cancers (4 invasive and 3 of borderline malignancy; 5 FIGO stage Ia, 1 stage IIa, 1 stage III) and 55 benign lesions were detected. We calculated the effect that a prescreening test (based on different threshold values for serum CA 125) would have had on the number of women entering the ultrasound-based screening program, and on the detection rate and false-positive rate of the overall procedure. There was a direct relationship between the number of women referred for ultrasound screening and the detection rate. The use of a threshold value for serum CA 125 > or = 20 U/ml would have meant that 380 women (25.3%) were referred for ultrasonography and 5 out of 7 cancers (71%) would have been detected with a false-positive rate of 1.1%. The odds of a woman with a positive screening result having cancer at surgery would have been about 1:3 (which would improve to about 1:1 if observational indices of color Doppler imaging and a morphological score had been used throughout). We concluded that a prescreening immunochemical test based on the measurement of serum CA 125 (with a threshold value of > or = 20 U/ml) would increase the prior odds for familial ovarian cancer by 2.8, but would lower the overall detection rate by 29% at the prevalence screening.