ISSN 1470-3947 (print) | ISSN 1479-6848 (online)

Endocrine Abstracts (2019) 63 P321 | DOI: 10.1530/endoabs.63.P321

Automated free testosterone assay: validation and usual values

Fidéline Bonnet-Serrano1,2, Amina Bouzerara1, Théo Rousselle1, Marie-Claude Menet1,3 & Jean Guibourdenche1,4


1Hormonology Department, Cochin Hospital, Paris, France; 2INSERM U1016, CNRS UMR8104, Université Paris Descartes, Cochin Institute, Paris, France; 3INSERM UMR S1144, Université Paris Descartes, Paris, France; 4INSERM U1139, Université Paris Descartes, Paris, France.


Introduction: Testosterone circulates under different forms in blood, mainly bound to proteins i.e. Sex Hormone Binding Globulin (SHBG) and albumin. Free testosterone (FT), the biologically active form, represents 2% of total testosterone (TT). FT measurement is mainly indicated when TT level is discordant with clinical picture but remains technically challenging. Indeed, as for all free hormones, gold standard method relies on equilibrium dialysis, unusable in routine. Direct immunoassays by competition have thus been designed, traditionally based on sensible radioactive detection signal (RIA). FT can also be calculated from TT, SHBG and albumin levels. Our work aimed to compare a new automated immunoassay to preexisting dosages and to propose adapted usual values.

Materials and methods: Analytical performances of this new FT assay were evaluated. FT was therefore determined in 164 patients (68 women, 96 men) using the new immunoassay (IS-5300, IDS-iSYS Free Testosterone), a RIA immunoassay (KIPI19000, DIAsource), and a calculation based on TT (RIA TESTO-CT2, Cisbio), SHBG, and albumin (Cobas ROCHE) concentrations. Usual values for the new dosage were established.

Results: Analytical performances of the new assay claimed by the manufacturer were confirmed and comparable with those of the RIA assay except for a higher detection limit. Correlation between immunoassays was satisfactory in men (R2=0.77) but weaker in women (R2=0.45), results with the new automated dosage being globally 30% lower. Correlation between both immunoassays and calculated FT was also satisfactory in men (respectively R2=0.68 for automated and 0.76 for RIA immunoassays) and poor in women (respectively R2=0.15 and 0.13). Calculated FT was much higher than measured FT, as the corresponding reference values proposed by the manufacturers. This discrepancy was confirmed by the analysis of external quality controls results whatever the direct immunoassay. We proposed preliminary usual values (minimal and maximum values observed in the subgroup of patients with normal testosterone and SHBG levels): 18.9–51.7 pmol/l in men <50 years old (n=23); 7.4–39.5 pmol/l in men > 50 years old (n=33); < 6.2 pmol/l in women < 50 years old (n=34) and < 4.3 pmol/l in women > 50 years old (n=23).

Conclusion: IDS-iSYS FT assay is one of the first automated assays allowing FT dosage. Its analytical performances are suitable and provide valuable results in comparison to both RIA immunoassay and calculated FT, at least in men. Clinicians should pay attention to FT usual values indicated by the laboratories, given the large differences observed, particularly between direct immunoassays and calculated FT.

Article tools

My recent searches

No recent searches.

My recently viewed abstracts