Studies on the clinical utility of the measurement of serum thyroglobulin in man
Thyroglobulin (Tg) was extracted from human thyroid, purified by gel column chromatography, identified by its thyroxine content, and quantitated by accurate protein determination. The material was used for standards and a radioiodinated tracer in a radioimmmunoassay for serum thryoglobulin, and in an ELISA assay for thyroglobulin autoantibodies. The radioimmunoassay utilised a polyclonal rabbit anti-human-Tg antibody, and separation of bound and free fractions was achieved by a precipitating antiserum and centrifugation. Assay sensitivity was 1ug/1, and the inter and intra-assay CV 16-23% and 6-16% respectively. Thyroglobulin autoantibodies were measured by both radioimmunoassay and ELISA methods. Thyroglobulin autoantibodies at high titre caused interference in the Tg assay, and both falsely low and elevated results occurred depending on the assay conditions. Studies on the interference problem are described. Serum, Tg in newly presenting patients with thyrotoxic Graves' disease was usually elevated, but not in all individuals. No relationship with serum thyroid hormone levels was found. Serum Tg in patients with hypothyroidism was generally within the normal range. TSH dependent Tg release was demonstrated following acute withdrawal of triiodothyronine in a normal subject. The measurement of serum Tg as a marker for relapse following treatment of thyrotoxicosis with antithyroid drugs was assessed. Although a small group with a high probability of relapse could be identified in whom Tg levels exceeded 60ug/1, the test was generally insufficiently accurate to be clinically useful. Similarly serum Tg measurement following radioiodine treatment of thyrotoxicosis proved of no value in predicting the onset of subsequent hypothyroidism, although a characteristic biphasic release was observed. Serum Tg measurement did prove a discriminatory marker for recurrence of differentiated thyroid carcinoma in patients treated by thyroid lobectomy and suppressive thyroxine, despite the presence of potentially functioning residual normal thyroid tissue. The accuracy of the test was compared in patients treated in this way, and a group treated by more radical thyroid ablation.