In 2015 I gave a talk to a TUK forum on the poor quality control and performance of free T4 and free T3 assays then used. The situation hasn't changed in the meantime, but I got a letter accepted in a journal on this very subject: It will appear very soon but since it will be behind a paywall I'm giving it in full here (sorry for the length):
Global FT4 immunoassay standardization.
Response to: Kratzsch J et al. Global FT4 immunoassay standardization: an expert opinion review
John E. Midgley
jem.midgley@gmail.com
North Lakes Clinical, Ilkley LS29 8PT,
UK Corresponding author: John E. Midgley, North Lakes Clinical, Ilkley LS29 8PT,
To the Editor,
Even after many years of use, harmonisation of free thyroid hormone assays still appears to be a distant prospect [1]. Forty years after the invention and production of the one-step tests for free thyroxine (FT4) and free triiodothyronine (FT3), in which I had a role as an inventor [2], rationalisation and harmonising of such assays (including the two-step alternatives from different manufacturers) has still not been resolved. Why is this so?
I would like to briefly comment on three intertwined obstacles that must be overcome to achieve a successful outcome.
A major barrier to progress has been the various later attempts at exploring the basic nature of these assays. All of them misapprehend how the assays work, being either theoretically bogus [3], or practically invalid either from a general mistaken examination by disassembly of the assay components [3–7]. Illegitimate claims result both from such an approach, while in some studies the wrong conditions were used to compare so called “gold standard” methods with only a few examples of working assays [8]. There can be technical difficulties when performing “gold standard” methods such as equilibrium dialysis and especially tandem mass spectrometry/ultrafiltration of undiluted sera where, despite its apparent validity, potential problems with membrane integrity under pressure, and other extrusion artefacts can seriously affect the results for free hormone measurement [6, 8]. The technical pressure on equilibrium dialysis, on the other hand, is somewhat less demanding, readily demonstrating its validity through its robust response to dilution of the serum sample, by maintaining constancy of measurement of free hormone concentrations [9]. However, in many studies using these techniques, general conclusions were derived from only examining a small selection of available tests, thereby tarring all with the same brush. This is neither justified nor acceptable.
In a more general sense, the first problem for assay rationalisation is the general failure of assay designers to properly understand the complexity of devising and testing these assays for routine use. In the initial phase and trialling of the first such tests, to thoroughly examine the assays’ performance, the development used sera from many classes of individuals This was done not only in the frequently encountered circumstances, such as differentiation of hypo-, eu- and hyperthyroid groups, those on T4 therapy, and in the three trimesters of pregnancy, but additionally, individuals with variations in thyroxine binding protein concentrations (TBG) from zero to four times the mean concentration of the reference range, analbuminemics, familial dysalbuminemic hyperthyroxinemics (FDH), sera with high concentrations of T4 and T3-containing autoantibodies, and various states and forms of nonthyroidal illness (NTI) [10]. The tests were additionally supported by comparison with equilibrium dialysis. Correlations between TBG, transthyretin and albumin with FT4 or FT3 were also examined in healthy persons. By such comprehensive testing protocols, appropriate findings in the rare groups consolidated the assays’ accuracy and precision in the more common cases. Rarely, if at all, has any other test since been validated using a similar rigorous examination before submission for licence to market the method. Rather methods were merely copied and justified in commonly encountered patient groups without applying a deeper understanding of the care needed to produce a comprehensively acceptable performance in more challenging situations. The exhaustive protocol of testing used in the first assays uncovered hitherto unknown problems, which were largely corrected in later developments [10]. Such duty of care was no longer maintained by recent relaxed protocols. Crucially, awareness of the especial difficulties of producing a valid FT3 test appears to be lacking. This requires an order of magnitude of greater understanding of the tests even than for FT4. The apparent success of such imperfect assays in general cases does not validate their performance in rare sera. Modification costs may contribute to an unwillingness to refine existing assays, because the cost of improvement may outweigh the perceived additional financial return
.
The second failure has been by the regulators who determine the criteria and give permission for the tests to be marketable. The necessary basic validations appropriate for a free analyte assay are far more searching than required for the total hormone tests. Not only should precision and specificity be examined, but also the special requirements for assay validity. These include (1) robustness of the assay results to progressive serum dilution, (2) acceptable performance in rare categories, and (3) extraction of analyte by the test’s antibody sufficiently small as to negligibly affect FT4 or FT3 measurement in all categories. Regulatory requirements are instead rooted in the requirements for validation of total hormone assays, which though necessary, are here fundamentally inadequate as a test of validity.
The third culpable entity is the lenient reaction and tolerant attitude of the medical discipline to this unsatisfactory state of affairs. It is surprising that in the intercourse between producer and customer there has been no active demand by the latter for improving accuracy, precision and robustness in the various assays offered. It is usual that if offerings vary so much in a product, the customer will demand to have a more consistent performance in all. In the current environment, some financial constraints are likely to unduly influence the selection of assay groups over performance. The problem is accentuated by the present system of multi-assay offerings in a dedicated system. If within a given platform, methods for FT4 and FT3 are inadequate, it becomes very difficult to measure these separately by a different system. Such variation in test results causes unnecessary revalidation work if tests are subject to a change in product.
In conclusion, for significant progress in rationalisation of tests to occur, regulators themselves needed to strengthen the requirements for valid working assays. This must include robustness to progressive serum dilution and independence from transport protein differences. Only stricter licensing rules will force producers to deliver on the promise of rationalisation and harmonisation of contemporary FT4 and FT3 assays, which is well within their capability and the public interest.
•
1) Kratzsch J, Baumann NA, Ceriotti F, Lu ZX, Schott M, van Herwaarden AE, et al. Global FT4 immunoassay standardization: an expert opinion review. Clin Chem Lab Med 2020. doi.org/10.1515/cclm-2020-1696.
•
2.Midgley JEM, Wilkins TA. New methods of free thyroid hormone assay. In: Bizollon Ch A, editor Physiological peptides and new trends in radioimmunoassay. Elsevier/North Holland; 1981. p. 215–34.
•
3.Ekins RP. Measurement of free hormones in blood. Endocr Rev 1990;11:5–46. doi.org/10.1210/edrv-11-1-5.
•
4.Nelson JC, Nayak S, Wilcox RB. Variable underestimates of serum free thyroxine (T4) immunoassays of free T4 concentrations in simple solutions. J Clin Endocrinol Metab1994;79:1373–5. doi.org/10.1210/jc.79.5.1373.
•
5.Midgley JEM, Christofides ND. Legitimate and illegitimate tests of free-analyte assay function. Clin Chem 2009;55:439–41. doi.org/10.1373/clinchem.20....
•
6.Midgley JEM. “All that glisters is not gold”: ultrafiltration and free thyroxine measurement; with apologies to W Shakespeare. Clin Biochem 2010;44:151–3. doi.org/10.1016/j.clinbioch....
•
7.Midgley JEM. Spurious conclusions on analog free thyroxine assay performance. Clin Chem 2007;53:1714. doi.org/10.1373/clinchem.20....
•
8.Midgley JEM, Christofides ND. Inaccuracies in free thyroid hormone measurement by ultrafiltration and tandem mass spectrometry. Clin Chem 2009;55:2228–9. doi.org/10.1373/clinchem.20....
•
9.Thienpont LM, Faix JD, Beastall G. Standardisation of FT4 and harmonization of TSH measurements – a request for input from endocrinologists, and other physicians. Endocrine2015;62:855–6. doi.org/10.1507/endocrj.ej1....
•10.Christofides ND, Sheehan CP. Enhanced chemiluminescence labeled-antibody immunoassay (Amerlite-MAB) for free thyroxine: development and technical validation. Clin Chem 1995;41:17–23. doi.org/10.1093/clinchem/41....
