Mutational analysis in UK patients with a clinical diagnosis of familial hypercholesterolaemia: relationship with plasma lipid traits, heart disease risk and utility in relative tracing.

As part of a randomised trial [Genetic Risk Assessment for Familial Hypercholesterolaemia (FH) Trial] of the psychological consequences of DNA-based and non-DNA-based diagnosis of FH, 338 probands with a clinical diagnosis of FH (46% with tendon xanthomas) were recruited. In the DNA-based testing arm (245 probands), using single-strand conformation polymorphism of all exons of the low-density lipoprotein receptor (LDLR) gene, 48 different pathogenic mutations were found in 62 probands (25%), while 7 (2.9%) of the patients had the R3500Q mutation in the apolipoprotein B (APOB) gene. Compared to those with no detected mutation, mean untreated cholesterol levels in those with the APOB mutation were similar, while in those with an LDLR mutation levels were significantly higher (None=9.15+/-1.62 vs LDLR=9.13+/-1.16 vs APOB=10.26+/-2.07 mmol/l p<0.001, respectively). Thirty seven percent of the detected mutations were in exon 3/4 of LDLR, and this group had significantly higher untreated cholesterol than those with other LDLR mutations (11.71+/-2.39 mmol/l vs 9.88+/-2.44 mmol/l, p=0.03), and more evidence of coronary disease compared to those with other LDLR or APOB mutations (36 vs 13% p=0.04). Of the probands with a detected mutation, 54 first-degree relatives were identified, of whom 27 (50%) had a mutation. Of these, 18 had untreated cholesterol above the 95th percentile for their age and gender, but there was overlap with levels in the non-carrier relatives such that 12% of subjects would have been incorrectly diagnosed on lipid levels alone. In the non-DNA-based testing arm (82 probands) only 19 of the 74 relatives identified had untreated cholesterol above the 95th percentile for their age and gender, which was significantly lower (p<0.0005) than the 50% expected for monogenic inheritance. These data confirm the genetic heterogeneity of LDLR mutations in the UK and the deleterious effect of mutations in exon 3 or 4 of LDLR on receptor function, lipids and severity of coronary heart disease. In patients with a clinical diagnosis of FH but no detectable mutation, there is weaker evidence for a monogenic cause compared with relatives of probands with LDLR mutations. This supports the usefulness of DNA testing to confirm diagnosis of FH for the treatment of hyperlipidaemia and for further cascade screening.