Thyroglobulin as a biomarker of iodine status

Abstract

New Zealand (NZ) soils are low in iodine and consequently NZ has a history of iodine deficiency. Mild iodine deficiency has a number of adverse consequences including impaired cognition in children and an increased risk of thyroid cancer in adults. Mild iodine deficiency is difficult to detect using biochemical tests such as thyroid-stimulating hormone (TSH), triiodothyronine (T3) and thyroxine (T4) because the concentrations of TSH, T3 and T4 fall within the normal reference ranges. Thyroglobulin (Tg), a glycoprotein exclusively produced by the thyroid gland, has been proposed as a more sensitive biomarker of iodine status. A median Tg cut-off ˂13 μg/L has been shown to indicate iodine sufficiency in children. It is not clear, however, whether Tg can be used to assess iodine status in adults. The aim of this thesis was to determine if Tg can be used to assess iodine status in adults, both in groups and in individuals.

The objective of the randomised, double-blind, placebo-controlled intervention trial was to determine the efficacy of Tg to assess iodine status in a group of adults. Mildly iodine deficient adults were randomly assigned to consume either 150 μg potassium iodate (KIO3) or placebo daily for 24 weeks (wk). At baseline and 24 wk, participants were asked to provide five spot urine samples for the determination of urinary iodine concentration (UIC), give a blood sample for the measurement of TSH and free thyroxine (FT4), and complete an iodine-specific food frequency questionnaire (FFQ). Thyroglobulin antibody (TgAb) and thyroid peroxidase antibody (TPOAb) were measured at baseline. Tg was determined at baseline (0 wk), 8, 16, and 24 wk. The main outcome of the study was a difference in Tg concentration between the iodine-supplemented and placebo groups at 24 wk. At baseline, the overall median UIC of all participants (n=112) was 65 μg/L, confirming mild iodine deficiency in this group (i.e. median UIC <100 μg/L). The overall median Tg at baseline was 16.6 μg/L and the concentrations of TSH and FT4 were within the normal reference range. At 24 wk, the UIC of the placebo and iodine-supplemented groups were significantly different (median UIC: 79 and 168 μg/L, respectively; P<0.001), with the iodine-supplemented group categorised as iodine sufficient. Tg concentration in the iodine-supplemented group decreased by 12, 20 and 27% compared to the placebo group at 8 (P=0.045), 16 (P<0.001) and 24 wk (P<0.001), respectively. However, there was no significant change in the concentration of TSH and FT4 at 24 wk for either group. This study demonstrated that iodine supplementation of 150 μg iodine/day in mildly iodine deficient adults for 24 wk improved iodine status and was associated with a concomitant decrease in Tg concentration. Adequate iodine status in this group of adults was associated with a median Tg concentration of 13 µg/L, in agreement with findings in children.

The screening phase of the intervention trial provided an opportunity to evaluate the efficacy of Tg to assess iodine status in individual adults. In this study, the following data from each of 178 adults aged 18-40 years was utilised: mean UIC of five spot urine samples; determination of Tg, TSH, FT4, TgAb and TPOAb from a blood sample; and the iodine content of the diet measured via an iodine-specific FFQ. The main aim of this study was to evaluate the sensitivity and specificity of Tg using the mean UIC of five spot urine samples as the reference standard. When a UIC cut-off <100 μg/L and a Tg cut-off ≥13 μg/L were used to indicate iodine deficiency, the sensitivity and specificity of Tg was 68% and 40%, respectively, well below the acceptable range for sensitivity and specificity of ≥80%. Furthermore, all receiver operating characteristic (ROC) curves for Tg were situated close to the line of chance and the area under the curve neared 0.50. Using UIC as the reference standard, this study indicated that Tg was unable to accurately assess iodine status in individuals.

In conclusion, although Tg appears to be a poor biomarker of individual iodine status, the work presented in this thesis clearly demonstrated that Tg can be used as a biomarker of iodine status in groups of mildly iodine deficient adults. Furthermore, the median Tg cut-off ˂13 μg/L used for children can also be used to indicate iodine sufficiency in a group of adults. These findings demonstrate that Tg is responsive to an improvement in iodine status, and should be included as biomarker in studies monitoring the impact of iodine supplementation and fortification programmes.