URINARY MULTIPLE MARKER SCREENING FOR DOWNS-SYNDROME

We have examined the possibility of using multiple markers in maternal urine rather than serum in order to screen for Down's syndrome. Urine samples were available from 36 cases (24 Down's syndrome, five Edward s' syndrome, three Turner's syndrome, one Klinefelter's syndrome, one triploidy, one triple-X, one twin discordant for Down's syndrome) and 294 controls, including three twins. Three markers were tested: the be ta-core fragment of human chorionic gonadotrophin (hCG), total oestrog en (tE) and the free a subunit of hCG. Levels were corrected for creat inine excretion and expressed as multiples of the gestation-specific m edian (MOM) level from the singleton controls, The median value for th e singleton Down's syndrome cases was 6.02, 0.74, and 1.08 MOM for bet a-core-hCG, tE, and alpha-hCG, respectively. The increases in beta-cor e-hCG and the reduction in tE levels were highly significant (P<0.0001 and 0.005, respectively; Wilcoxon rank sum test) but the increase in free alpha-hCG was not (P=0.40). On the basis of a mathematical model, the expected detection rate for a 5 per cent false-positive rate was 79.6 per cent for beta-core-hCG alone, which increased to 82.3 per cen t when combined with tE. Aneuploidies other than Down's syndrome were characterized by low levels of tE and either low or high beta-core-hCG .