Use of linkage disequilibrium data in prenatal diagnosis of cystic fibrosis.

Parents at risk of bearing a child with cystic fibrosis, and who have no living affected child, often use prenatal diagnosis based on microvillar enzyme assay in second-trimester amniotic fluid samples. If enzyme levels are abnormal and the pregnancy is terminated, it is possible in principle to use the fetal tissues to establish the phase relationship of linked DNA markers for a subsequent first-trimester prenatal diagnosis. However, the probability of a fetus being affected after an abnormal microvillar enzyme test may be no greater than 80%. The strong linkage disequilibrium between haplotypes at the D7S23 locus and the cystic fibrosis gene may be used to increase this probability. If fetal tissues are homozygous for the 6.6-kb band defined by pKM.19 and PstI and also homozygous for the 2.1-kb band with pXV-2c and TaqI, the chance of being affected increases from 80% to between 95% and 97%. We regard this as being sufficiently certain for use in phase determination.