TUMOR-MARKER TRENDS IN ASYMPTOMATIC WOMEN AT RISK FOR OVARIAN-CANCER - RELEVANCE FOR OVARIAN-CANCER SCREENING

The purpose of our study was to examine single and serial determinatio ns of five biomarkers in asymptomatic women at risk for epithelial ova rian cancer to define median values and trends in a cancer-free cohort . Women were enrolled in the Gilda Radner Ovarian Cancer Detection pro gram at Cedars-Sinai Medical Center, Los Angeles, California from July 1, 1991, to July 1, 1993. Biomarkers studied include CA-125, LASA, DM -70K, UGP, and HER-2. Biomarker values that were analyzed included tho se obtained on the first patient visit and in a subset of values evalu ated at 6-month follow-ups. A total of 590 women were included in the study; 425 were premenopausal and 165 were postmenopausal. No one in t his study group has developed ovarian cancer within a 12-month follow- up period. Postmenopausal women had significantly lower values for CA- 125 (12 vs 19 U/ml, P = 0.0001) and higher values of LASA (16.6 vs 15. 4 mg/dl, P = 0.0001), DM/70K (11 vs 0 U/ml, P < 0.001), UGP (3.3 vs 0. 3 U/ml, P < 0.001), and HER-2 (12 vs 11 U/ml, P = 0.009) than premenop ausal women. Less than 5% of the women had elevated tumor marker value s on both screens except for CA-125, where 15% of women studied had tw o consecutive visits with elevated values (>35 U/ml). Two women demons trated an exponential rise in CA-125 between visits. Premenopausal wom en differ in median biomarker values compared to postmenopausal women. Elevations in single tumor markers, combinations of markers, and expo nential increases of markers have all produced false-positive results for ovarian cancer within the 12-month follow-up period. Exponential i ncreases produced the smallest false positive rate (n = 2), but furthe r study and follow-up are needed to determine the efficacy of these te sts in ovarian cancer screening. (C) 1995 Academic Press, Inc.