Cholestane-3β, 5α, 6β-triol: Further insights into the performance of this oxysterol in diagnosis of Niemann-Pick disease type C

In recent years the oxysterol species cholestane-3 beta, 5 alpha, 6 beta-triol (C-triol) has found application as a diagnostic biomarker for Niemann-Pick disease type C. Other studies have described increased C-triol in patients with Niemann-Pick disease type A/B and milder increases in lysosomal acid lipase deficiency (LALD), whereas they note normal C-triol levels in Smith-Lemli-Opitz syndrome (SLOS) and familial hypercholesterolaemia (FH) patients. Herein, we review data collected in our laboratory during method evaluation along with 5 years of routine analysis and present findings which differ from those reported by other groups with respect to LALD, SLOS and FH in particular, whilst providing further evidence regarding the clinical sensitivity and specificity of this biomarker, which are difficult to accurately assess. All of our Wolman disease (severe LALD) patients have demonstrated gross elevations of C-triol at diagnosis, with reduction to normal levels after induction of enzyme replacement therapy. In diagnostic specimens from SLOS patients we observed very low or undetectable C-triol levels whereas in post-therapeutic SLOS patients demonstrated normalised levels; we also describe a homozygous FH patient in which C-triol is significantly elevated. Upon investigation, we found that C-triol was formed artefactually from cholesterol during our sample preparation, i.e. this is a false positive of analytical origin; at present it is unclear whether similar effects occur during sample preparation in other laboratories. Our data demonstrates clinical sensitivity of 100% during routine application to diagnostic specimens; this is in keeping with other estimates, yet in a small proportion of patients diagnosed prior to C-triol measurement, either by Filipin staining of fibroblasts or molecular genetics, we have observed normal C-triol concentrations. Clinical specificity of C-triol alone is 93.4% and 95.3% when performed in conjunction with lysosomal enzymology. These performance statistics are very similar to those achieved with Filipin staining of cultured fibroblasts in the 5 years preceding introduction of C-triol to routine use in our laboratory. It is increasingly apparent to us that although this analyte is a very useful addition to the diagnostic tools available for NPC, with considerable advantages over more invasive and time-consuming methods, the interpretation of results is complex and should be undertaken only in light of clinical details and results of other analyses including enzymology for lysosomal acid lipase and acid sphingomyelinase.