Interference from heterophilic antibodies in TSH assays

We read with interest the report on heterophilic antibodies as a case of false positive interference in thyroid-stimulating hormone (TSH) assays. ¹ As stated by the authors the terms ‘heterophile’, human anti-mouse and human anti-animal (HAA) can be used imprecisely. HAA antibodies are of high affinity and directed against animal epitopes, whereas heterophilic antibodies are weak polyspecific antibodies formed early in the immune response before affinity maturation. ² To cause false positive assay results antibodies must be capable of cross-linking the TSH capture and detection antibody in a ‘sandwich’ assay. Rheumatoid factor is a commonly encountered subtype of heterophilic antibody with non-specific affinity for the Fc fragment of human and some animal IgGs. ³ As rheumatoid factor is an IgM this may be responsible for the second case reported by Ross et al. ¹

The situation is complicated by the presence of autoimmune anti-TSH antibodies (sometimes called ‘macro’ TSH). This has been reported by ourselves ⁴ and others as a cause of false positive results. The mechanism of autoimmune interference differs from that of heterophile antibodies as it does not invoke cross-linking. The most widely accepted mechanism is that the autoimmune antibody–hormone complex is immunoreactive but biologically inactive. We proposed ⁵ that the extent to which different assays react to the autoimmune anti-TSH complex is determined by the relative affinity of the anti-TSH-autoimmune immunoglobulin and the anti-TSH capture antibody; i.e. a high-affinity capture antibody could ‘strip’ TSH from the autoimmune complex. This could explain the method-dependent results of Ross et al. ¹

In our hands autoimmune hormone complexes typically show an increase in immunoreactive TSH on dilution, ⁵ as does the first case reported by Ross et al. ¹ We suspect that as the autoimmune complex dissociates on dilution, more TSH is available to the capture antibody. In contrast, heterophile interferences tend to show a reduction in apparent TSH on dilution (D J Halsall, unpublished results). However for theoretical reasons, this test is unlikely to be infallible.

Gel filtration chromatography is considered the ‘gold standard’ to investigate antibody interference, but again this may not distinguish a cross-linking mechanism from an autoimmune complex as the difference in molecular mass between a cross-linking immunoglobulin and an autoimmune TSH-immunoglobulin complex is not great.

In our experience recovery studies using exogenously added TSH discriminate well between the two mechanisms. ⁵ Cases of autoimmune TSH complex interference show a vast capacity to bind exogenous TSH whereas heterophilic antibody cases do not. This test works only if the anti-TSH antibody is in excess; however, in practice we have never seen this behaviour in samples where the TSH immunoreactivity decreases on dilution.

Once assay interference has been established differentiating between these two mechanisms is academic. However the mechanism of the interference does have bearing on how reliably it can be detected. Autoimmune anti-TSH antibodies will not be removed by heterophilic blocking tubes as no anti-animal epitope is present. Paradoxically lesser degrees of interference caused by weakly bound immune complexes can be more difficult to detect than extreme cases as results may appear more clinically plausible.

DECLARATIONS

Competing interests: The authors have no competing interest.

Funding: VKC is funded by the Wellcome Trust (080237). In addition, this work was also supported by the NIHR Cambridge Biomedical Research Centre.

Ethical approval: Not required.

Guarantor: DJH.

Contributorship: DJH wrote the letter which was edited by VKC. EE developed laboratory methods to detect weakly bound TSH complexes.

Acknowledgements: We would like to thank Peter Barker for providing thyroid hormone assays.