To continue the saga of measuring low testosterone 
concentrations, especially the effect of SHBG. As Roy F. 
points out there are one or two reports on the effect of 
SHBG on the measurement of testosterone but I think that 
this only becomes a problem at relatively high 
concentrations of the binding protein i.e. at levels above 
150 nmol/L. These are very rare perhaps in pregnancy or 
Dianette treated patients. I have had a look at the effect 
of SHBG on several automated testosterone assays by 
altering the SHBG concentrations but keeping the 
testosterone concentration constant at e.g. ~3 nmol/L and 
at ~10 nmol/L but I concur with the published data that the 
SHBG has to be really quite high for any significant effect 
to be seen.
I think we should keep in mind that we are using methods 
which claim to have a 100-fold (~0.5-50nmol/L)assay 
range.The same obtains for Oestradiol (~50-15000 pmol/L)-a 
300-fold range!!! I have been trying to think of a 
'Core'analyte method with such a wide calibration range, 
 at such low concentrations, with such avid 
protein-binding.
Michael J. Diver 
On Mon, 21 Aug 2000 17:55:07 +0100 "Fisher,  Roy - RCHT" 
<[log in to unmask]> wrote:

> I would whole heartedly agree with Jonathan and Mike Diver regarding the
> sub-optimal performance of testosterone assays on female samples. As some
> assays clearly perform better than others it would be useful to try and
> identify the characteristics of those assays. However, the method inserts
> rarely give information on the antibody affinity constants, the nature of
> the testosterone-SHBG release agent or cross reactions with the major
> androgen metabolite conjugates. The effect of SHBG may be a major factor in
> partially determining the testosterone result.  SHBG levels can inversely
> influence the testosterone concentration (Clin Chem 1987; 33: 300-2)and a
> study comparing 6 testosterone kits, found that the between kit variablity
> was greatest for female samples, with the lowest SHBG samples having the
> highest variances ( Fertil Steril 1998; 69:286-92). 
> 
> Perhaps Jonathan you could consider assessing this further through the
> Female Testosterone NEQAS, by producing pools with similar testosterone
> levels by extraction or GCMS methods, with varying SHBG levels from low
> (hirsute female) to high(female on OCP).
> 
> With regards to the main androgen conjugates causing interference, we looked
> at this about 7 years ago with 3 direct testosterone RIA methods( DPC,
> Medgenix and Farmos). Spiking charcoal stripped serum with levels beyond
> that expected in hirsutism, we found no interference. Has anyone looked at
> the current automated methods?
> 
> Can we afford not to address this problem? Inundating the three or four
> laboratories running extraction assays with all our suspiciously raised
> results would be one solution, which would be expensive and possibly not
> welcomed. Designing robust precise and specific assays for female samples
> must be the the way forward and ideally the diagnostics industry's
> aim,hopefully with our assistance.
> 
> Roy Fisher
> Royal Cornwall Hospital
>   
> 
> -----Original Message-----
> From: [log in to unmask] [mailto:[log in to unmask]]
> Sent: Monday, August 21, 2000 10:56
> To: [log in to unmask]
> Subject: testosterone assays
> 
> 
> Hello list members  
> 
> Having just returned from holiday, I have read with interest the 
> messages starting with Helen's about testosterone assays and 
> their internal quality control.   
> 
> The issue of the availability of internal quality control materials 
> within the normal female reference range has been a source of 
> irritation and confusion for many years, and in my capacity as 
> organiser of the UK NEQAS for this analyte I have often spoken to 
> instrument manufacturers about this.  Some have suggested 
> diluting with 'zero standard' to achieve a target value between 1.5 & 
> 2.0 nmol/L.  
> 
> The issue of whether a laboratory should perform an assay without 
> a suitable control material within the normal range is a professional 
> and moral one.  A purist (like me) would say 'NO'.  The assay 
> should be optimised to cover the required analytical range for the 
> analyte and controls placed strategically throughout that range and 
> ideally at the main clinical decision point(s).  I believe that 
> manufacturers of instruments and reagents (and IQC materials) 
> should face up to this and provide suitable materials.  
> 
> The problem is that manufacturers see the testosterone assay as a 
> continuum from 'low' (say 1.0 nmol/L in a fermale) to 'high' (say 50 
> nmol/L in a male) and provide a number of controls throughout this 
> range, often without regard to the main (male & female) decision 
> points.  This is scientifically and clinically unsound!  
> 
> It has long been a hobby horse of mine that testosterone assays 
> should be separately optimised and controlled for female and male 
> applications (this is why we separated the UK NEQAS schemes 
> nearly a decade ago).  
> 
> Assay systems behave differently: (specificity, recovery, 
> relationship to GCMS reference values etc) with the different 
> matrices.  Separation of assay systems with improved specificity 
> and accuracy for the female matrix is urgently required to address 
> problems of 'true' reference limits, interference with metabolites, 
> spurious high values (with need for repeat extraction assays) etc.  
> Analysis of testosterone in the male matrix seems much less 
> problematical and is performed adequately by most systems.   
> 
> Perhaps it is time for laboratories to question their suppliers as to 
> what they are doing to improve optimisation for the female matrix 
> and/or separate the assays.  The tools exist now (eg SysRef 
> materials) with which accurate comparisons with GCMS target 
> values can be made and optimisation improved.  My guess is that 
> no manufacture will be able to create an assay that performs well 
> for both male and female SysRef materials which will further 
> enforce the need for separate assay systems.
> 
> I already anticipate the response - "it is 'too difficult' to implement 
> and 'market' separate female and male testosterone assay 
> systems on the one platform" and "customers would have the 
> additional work load of having to sort out their samples".
> 
> I believe that the clinical science of testosterone measurement 
> should be paramount and drive the necessary technology not the 
> other way round. What do other list members and manufacturers 
> think?
> 
> JGM
>     
>   
> Jonathan Middle PhD
> Organiser UK NEQAS for Steroid Hormones
> Chairman UK NEQAS Executive
> Deputy Director UK NEQAS (Birmingham)
> UK NEQAS PO Box 3909 Birmingham B15 3NY
> tel 0121 414 7300, fax 0121 414 1179
> [log in to unmask]  http://www.ukneqas.org.uk

----------------------
[log in to unmask]



%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%