Calsequestrin and collagen XIII antibodies are the most specific to TAO, whereas antibodies against G2s, and to a lesser extent Fp, are also markers of ophthalmopathy, but less reliable. These results are unique in that it is the first time the significance of a panel of three candidate eye muscle antibodies and a connective tissue antibody have been evaluated in the same patients with ophthalmopathy.

SummaryWe have identified several eye muscle antigens and studied the significance of the corresponding serum autoantibodies in patients with Graves' disease. Of these antigens, only calsequestrin is expressed more in eye muscle than other skeletal muscles, which could explain at least partly the specific involvement of eye muscle in patients with Graves' disease. Earlier, we found a modest relationship between anti-calsequestrin antibodies and ophthalmopathy, but in that study we used calsequestrin prepared from rabbit heart muscle and measured antibodies by immunoblotting. We have reinvestigated the prevalences of anti-calsequestrin antibodies in larger groups of well-characterized patients with thyroid autoimmunity with and without ophthalmopathy and control patients and healthy subjects, using standard enzyme-linked immunosorbent assay incorporating highly purified rabbit skeletal muscle calsequestrin, which has a 97% homology with human calsequestrin, as antigen. Anticalsequestrin antibodies were detected in 78% of patients with active congestive ophthalmopathy, in 92% of those with active inflammation and eye muscle involvement, but in only 22% of patients with chronic, 'burnt out' disease. Tests were also positive in 5% of patients with Graves' hyperthyroidism without evident ophthalmopathy (two patients) and one patient with 'watery eyes' but no other clear signs of congestive ophthalmopathy and IgA nephropathy and no known thyroid disease, but in no patient with Hashimoto's thyroiditis, toxic nodular goitre, non-toxic multi-nodular goitre or diabetes, or age-and sex-matched healthy subjects. In serial studies of all 11 patients with Graves' hyperthyroidism who had active ophthalmopathy at the time of the first clinic visit, or developed eye signs during the first 6 months, and positive anti-calsequestrin antibodies in at least one sample, anti-calsequestrin antibodies correlated with the onset of ocular myopathy in six patients. Antibodies targeting calsequestrin appear to be specific markers for ophthalmopathy and sensitive indicators of the ocular myopathy subtype of ophthalmopathy in patients with thyroid autoimmunity. However, these results must be considered preliminary until a large prospective study of patients with newly diagnosed Graves' hyperthyroidism, in which serum levels of calsequestrin antibodies are correlated with clinical changes and orbital eye muscle and connective tissue/fat volumes, has been carried out.

These findings support the notion that autoimmune attack against calsequestrin and collagen XIII in the levator palpebrae superioris (LPS) muscle may play a role in the pathogenesis of upper eyelid retraction and that lid retraction may be the dominant feature of ophthalmopathy in patients with Hashimoto's thyroiditis and non-autoimmune thyroid disease. Because calsequestrin is an intracellular protein, the corresponding autoantibodies probably do not initiate LPS muscle inflammation but may contribute to its damage. The mix of antibodies against calsequestrin and collagen XIII may shed light on the diverse presentations found in thyroid-associated ophthalmopathy.