New strategies for detecting steroid metabolic disorders-paneling vs profiling

Thursday, 24 May, 2012
  • Angel O.K. Chan


The steroid metabolic disorder 21-hydroxylase deficiency (21OHD)1 is increasingly being included in blood spot–based newborn-screening programs. Screening most commonly relies on 17-hydroxyprogesterone immunoassay, but it gives a high rate of false positives and may not provide differentiation from 2 other types of congenital adrenal hyperplasia—P450 oxidoreductase deficiency (PORD) and 11-hydroxylase deficiency.

Koyama et al. recently reported in Clinical Chemistry reference cutoffs for 4 urinary steroid metabolites and derived ratios for differentiating 21OHD and PORD (1). Although this report was based on samples from 46 institutions, they did not specify whether samples were obtained during second-line screening after a finding of an increased 17-hydroxyprogesterone result in a blood spot screen. It is useful that the authors provided genotypes for all patients positive for one of these disorders, but they stated that they did not include individuals with nonclassic 21OHD, which suggests to us that they selected results for this survey only after genotyping had been completed. The authors thus appear to have missed the opportunity to show how their approach might be integrated into a staged national screening program.

Koyama et al. used GC-MS selected ion monitoring analysis. The method presented presumably aimed to combine speed of analytical preparation with a selection of a small panel of steroid markers. Their use of solvent extraction will have discriminated against polar steroids, many of which are useful in the early neonatal period, whereas their chosen marker, pregnanetriolone, is relatively unimportant. The authors claimed that cutoffs had not previously been presented for their age range (0–180 days), but Shackleton has published a comprehensive set of reference data for a wide spectrum of steroid metabolites and metabolite ratios for diagnosing various major steroid metabolic disorders. These data covered all age groups and included 21OHD and PORD (2). Urinary steroid profiling is an analytical technique that remains unrivaled for the unequivocal identification of nearly all steroid metabolic disorders (3). In the cyclic scan mode, it is an ideal steroid metabolomic tool that still leads to discoveries of novel steroids, even for such seemingly well-studied conditions as 21OHD (4).

Central to any improved mass screening strategy is likely to be liquid chromatography–tandem mass spectrometry (5), which enables paneling of a selected set of steroids so that all forms of congenital adrenal hyperplasia can be screened at once (6). This strategy should in theory also allow detection of nonclassic 21OHD, which is important for ensuring accurate differentiation of individuals with a spectrum of enzyme deficit from healthy individuals and thus anticipating problems that otherwise would present later in childhood, such as early development of sexual hair and advanced bone age.

For an integrated national service, we would therefore recommend liquid chromatography–tandem mass spectrometry paneling be used for blood spot testing, along with the use of data-reduction algorithms to derive likely diagnoses for a range of steroid metabolic disorders. Positive results would then be evaluated by urinary steroid profiling. The spirit of “profiling” is that every metabolite eluted is available for examination and can be taken into consideration when interpreting the data.

In conclusion, urinary steroid profiling is a sophisticated and powerful biochemical analytical tool that can provide extensive pathophysiological information. The metabolite-excretion pattern, concentrations, and ratios are equally important in enabling a correct diagnosis. The interpretation of results must be individualized for the best interests of the patient.

Where can I read this paper?

Full article on Clinical Chemistry. Note: you will need the credentials or your institution to login or you can purchase access.

Published: 2012 Aug;58(8):1262-3. doi: 10.1373/clinchem.2012.189613. Epub 2012 May 24.