The effects of recombinant human thyrotropin, in normal subjects and patients with goitre

Characterization of the human thyrotropin (TSH) α and β-subunit in the late 1980s (Hayashizaki et al., 1985; Wondisford et al., 1988) paved the way for the production of recombinant human thyrotropin (rhTSH), which subsequently allowed detection of residual/recurrent disease in patients with differentiated thyroid carcinoma (DTC) without the inexpedient use of bovine TSH (bTSH) or human cadaver pituitary-derived TSH (hTSH). In the case of nontoxic multinodular goitre, a much more frequent disorder than DTC, the ideal treatment is under continuous debate (Hegedüs et al., 2003). Some clinicians favour levothyroxine (L-T4) suppressive therapy, despite little evidence to support this strategy, while others advocate thyroidectomy or 131I therapy (Bonnema et al., 2000, 2002b). Although the latter treatment is frequently used in some countries (Bonnema et al., 2000) its effectiveness can be hampered by a low thyroid radioiodine-uptake (RAIU), especially in areas with a high iodine intake. With the availability of rhTSH and its ability to stimulate thyroid function, including thyroid RAIU, 131I therapy may be an option for a larger fraction of patients with benign goitre. In recent years, several studies have reported the effect of rhTSH on thyroid function and RAIU in healthy subjects as well as in patients with nodular nontoxic goitre. In this review we discuss the effects and side-effects of rhTSH in healthy individuals with an intact thyroid gland and furthermore important aspects of the use of rhTSH in patients with benign goitre treated with 131I. Two principles are used in order to elevate circulating TSH levels in the postoperative follow-up of patients with DTC. Traditionally, a withdrawal regimen has been used by which the thyroid hormone therapy is suspended for a period of weeks; alternatively, exogenous TSH can be administered. As the only available source, bTSH was used for the first time 50 years ago (Sturgeon et al., 1953). However, several factors discouraged the use of this nonhuman glycoprotein: allergic reactions, loss of potency and development of TSH antibodies (Hays et al., 1967; Krishnamurthy, 1978; Melmed et al., 1980). Consequently, hTSH extracted from human cadavers was tried (Schneider et al., 1965), but all interest in this abated when cases of Creutzfeldt–Jakob disease, associated with pituitary GH, emerged (Will, 1991). After cloning of the human TSH-β gene in the late 1980s, highly purified rhTSH produced in Chinese hamster ovary (CHO) cells (Cole et al., 1993) made the extensive use of exogenous TSH possible. TSH is a member of the glycoprotein family and is produced in the anterior pituitary gland. Structurally, TSH consists of an α-subunit common for all glycoprotein hormones and a hormone-specific β-subunit (Pierce & Parsons, 1981). The technology for production of rhTSH involves insertion of cDNA for the human α-subunit and a partial genome clone for the human β-subunit into separate mammalian vectors and co-transfecting these into CHO cells (Cole et al., 1993). In vitro model systems were initially used to test the effects of rhTSH on thyroid function. In a human fetal thyroid cell system, Huber et al. (1991) demonstrated that rhTSH is able to activate the TSH receptor, as reflected by the accumulation of cAMP, the induction of Tg secretion and thyroid epithelial cell multiplication. Although rhTSH has an amino acid structure identical to hTSH (Cole et al., 1993), its glycosylation is different with higher sialic acid content (Thotakura et al., 1991; Szkudlinski et al., 1993). As a consequence, rhTSH exhibits lower immunological activity, lower TSH receptor binding affinity, and lower in vitro bioactivity than hTSH. The metabolic clearance rate (MCR) is significantly lower for rhTSH than for hTSH (Thotakura et al., 1991) and due to this difference in MCR it may turn out that rhTSH in vivo is equivalent to or maybe more potent than hTSH, despite the lower in vitro bioactivity (Thotakura et al., 1991). Estimated by immunoassays the specific activity of rhTSH is between 5·51 and 7·63 IU/mg based on the second International Reference Preparation (80/558) of hTSH as the standard (Rafferty & Gaines, 1999). After a single injection of rhTSH in Cynomolgus monkeys (Cole et al., 1993), a rapid clearance phase half-life of 35 min is followed by a postdistribution clearance phase half-life of 9·8 h. In patients with DTC, receiving 0·9 mg rhTSH the mean half-life is 22 h (Anonymous, 2003). It is well known that iodide uptake across the basolateral membrane of the thyroid follicular cells is catalysed by the Na+/I− symporter (NIS). Under normal conditions there is only minimal expression of the NIS in the plasma membrane of the thyroid follicular cells (Jhiang et al., 1998). Kogai et al. (1997) demonstrated, in a rat thyroid cell line (FRTL-5 cells), that rhTSH induces a significant increase in NIS-mRNA after 3–6 h, reaching a maximum at 24 h. Furthermore, an increased thyroid 125I uptake was detected 12 h after rhTSH stimulation with a maximum after 72 h (Kogai et al., 1997), showing that optimal expression and activation of the NIS may take some time. Braverman et al. (1992) confirmed this effect of rhTSH in vivo. In addition to a significant stimulation of T4 and T3 secretion, a doubling of the 6 and 20 h thyroid 123I-uptake was observed in Rhesus monkeys following intramuscular (i.m.) injections of 2 units of rhTSH for 3 days (4·6 units of rhTSH corresponds to 1 mg rhTSH). Observing these potent physiological effects of rhTSH in in vitro systems and in animal studies, the next step was to test rhTSH in humans with an intact thyroid gland. Ramirez et al. (1997) were the first to investigate the effects of 0·1 mg rhTSH i.m. in six healthy euthyroid subjects. With this dose, serum TSH (s-TSH) increased significantly within 2 h and peaked 4 h after injection. Twenty-four hours after rhTSH injection, the s-TSH level declined. Parallel to the rise in s-TSH, serum T4 (s-T4) and serum T3 (s-T3) levels increased by 54% and 89%, respectively, within 4–8 h. Both peaked at 24 h, whereas serum Tg (s-Tg) had a slower rise, peaking at 48 h after rhTSH stimulation. The suppressed s-TSH observed seven days after injection is probably best explained by the increased levels of s-T4 and s-T3 at this time. After 3 weeks all thyroid hormone levels had normalized. These effects on thyroid function, both in healthy subjects and in patients with nodular goitre, have been confirmed in subsequent studies by us and others (Huysmans et al., 2000; Torres et al., 2001; Nielsen et al., 2004; Table 1). Although different doses of rhTSH (0·01, 0·03, 0·3 and 0·9 mg) were used, the same patterns in the various biochemical markers were observed. A clear dose–response effect seems to exist, as a greater response in serum levels of T4, T3 and Tg is achieved when giving 0·3 mg rhTSH compared to lower doses. However, a maximal stimulatory dose also seems to exist as 0·9 mg rhTSH does not stimulate thyroid function more than 0·3 mg, when investigated in the same subjects (Torres et al., 2001; Fig. 1). When using very small doses of rhTSH (0·01 mg and 0·03 mg), as has been done in patients with nodular nontoxic goitre (Huysmans et al., 2000), the increases in serum T4 and T3 levels are blunted and most patients retain thyroid hormone levels within the normal range. Thus, the minimum thyroid stimulatory dose of rhTSH is probably in the range of 0·01 mg but experience is so far very limited (Huysmans et al., 2000). Relative increases in serum T4 or serum FT4 (%), following stimulation with 0·9 mg, 0·3 mg, 0·1 mg, and 0·03 mg rhTSH. Modified from Nielsen et al. (2004); Silva et al. (2003); Torres et al. (2001); Huysmans et al. (2000). Huysmans et al. (2000) evaluated the changes in thyroid RAIU following rhTSH in 15 patients with nontoxic nodular goitre and they confirmed the findings from animal studies (Braverman et al., 1992). The administration of 0·01 mg rhTSH 24 h before 131I increased the mean 24 h thyroid RAIU significantly more (from 29% to 51%) than if the same dose was given just 2 h before 131I (from 30% to 42%). Thus, the time interval between administration of rhTSH and 131I appears to be crucial, which is of no surprise taking into consideration that the NIS is stimulated by rhTSH with some time delay (Kogai et al., 1997). Hence, a single injection of either 0·01 or 0·03 mg rhTSH given 24 h before 131I administration approximately doubles the thyroid RAIU in patients with nontoxic nodular goitre without significantly changing the biological half-life of iodine (Huysmans et al., 2000; Fig. 2). A dose of 0·03 mg rhTSH is probably slightly more effective than 0·01 mg in this context. Interestingly, the effect seems to be inversely correlated to the baseline thyroid RAIU, meaning that patients with the lowest thyroid RAIU have most benefit from rhTSH prestimulation. Torres et al. (2001) found similar results in six healthy euthyroid subjects, in whom the thyroid RAIU, at 6 h as well as at 24 h, approximately doubled after 0·9 mg rhTSH. Despite using a much higher rhTSH dose, the mean change in the 24 h thyroid RAIU was lower than that obtained by Huysmans et al. (2000) studying patients with nontoxic nodular goitre. This discrepancy is probably due to differences in iodine intake, but an influence of the difference in thyroid morphology cannot be excluded. The fact that Torres et al. (2001) found a wide interindividual variation in the thyroid RAIU response to rhTSH – to some extent also found by Huysmans et al. (2000) – indicates the involvement of yet unidentified factors. In case of multinodular goitre, the thyroid RAIU undoubtedly is dependent on the general iodine load, extent of nodular autonomy and the s-TSH level, the latter being of importance for the paranodular tissue. These factors also contribute to the inhomogeneous scintigrams typical for multinodular goitre. Nieuwlaat et al. (2001) investigated planar 123I thyroid scintigrams in patients with multinodular goitre before and 24 h after administration of either 0·01 or 0·03 mg rhTSH. rhTSH resulted in a considerably more homogeneous distribution of the isotope in some patients compared to the baseline scintigrams. In some cases, cold areas were transformed into relatively warm areas and vice versa. Effects of 0·01 and 0·03 mg rhTSH on 24 h thyroid RAIU, in patients with nodular goitre. Values in boxes indicate the factor by which the 131I activity is reduced while still attaining the same thyroid dose, due to the rhTSH-induced increase in 24 h thyroid RAIU. Modified from Nieuwlaat et al. (2003). To what extent is rhTSH able to increase the thyroid RAIU in iodine loaded individuals? In healthy subjects, in whom the thyroid uptake was markedly suppressed by 7 days of iodide ingestion, Lawrence et al. (2001) measured the change in 16 h thyroid RAIU following 0·9 mg rhTSH administered 8 or 32 h before 123I. The thyroid RAIU increased by a mean of 62% and 97%, respectively, confirming the importance of the time interval between rhTSH stimulation and the effect obtained. However, rhTSH did far from normalize the thyroid RAIU (only a mean increase from 3% to a mean of 6% compared to the baseline mean of 19% before iodide ingestion). Thus, besides being able to stimulate the thyroid function, rhTSH clearly augments the thyroid RAIU in normal subjects as well as in patients with nodular goitre, and even very low doses of rhTSH seem effective for this purpose, apparently without any particular concomitant rise in thyroid hormone levels. The time interval between rhTSH and 131I administration is a critical factor and should probably be no less than 24 h. Until recently it remained an open question whether this marked increase in thyroid RAIU amplifies the effect of 131I therapy in patients with benign symptomatic goitre. In essence, there are three kinds of therapy available when dealing with nodular goitre: L-T4 suppressive therapy, surgery and 131I therapy. However, there is no consensus regarding the most favourable treatment (Bonnema et al., 2000, 2002b). In some countries, including ours, 131I has been used for decades in the treatment of symptomatic nontoxic nodular goitre, resulting in a mean thyroid volume reduction ranging from 40% to 60% within 1–2 years after treatment (Hegedüs et al., 1988; Nygaard et al., 1993; Le Moli et al., 1999) as described comprehensibly by Hegedüs et al. (2003). However, the efficacy of 131I therapy is hampered by the inhomogeneous 131I uptake seen in multinodular goitre. Additionally, in some areas of the world, the thyroid RAIU is low due to a high dietary iodine intake. Thus, if 131I therapy is to be used, a relatively high amount of radioactivity must be administered, often hindering outpatient treatment. At present, four studies have investigated whether rhTSH stimulation influences the effect of 131I therapy in patients with benign nodular goitre (Duick & Baskin, 2003; Graf et al., 2003; Nieuwlaat et al., 2003; Silva et al., 2003; Table 2). In a nonrandomized study, Nieuwlaat et al. (2003) investigated 22 patients with nodular goitre pretreated with either 0·01 or 0·03 mg rhTSH 24 h prior to 131I therapy. The amount of 131I activity (MBq) was reduced according to the rhTSH-induced increase in thyroid RAIU, (Fig. 2). Only a very modest increase in the thyroid hormone levels was observed following 131I therapy. Thyroid size was monitored by magnetic resonance imaging (MRI). On average, the goitre size as well as the smallest tracheal cross-sectional area was unaltered 1 week after 131I therapy in the 0·01 mg group. However, in those patients who were stimulated with 0·03 mg rhTSH the mean goitre volume 1 week after 131I was in fact increased significantly by 5%. In studies not using rhTSH, on average no acute goitre swelling following 131I therapy has been observed (Nygaard et al., 1994; Bonnema et al., 1999). After 1 year the mean thyroid volume reduction was 35% in the 0·01 mg rhTSH group and 41% in the 0·03 mg rhTSH group with no significant difference between the two groups (Nieuwlaat et al., 2003). Although this study did not include a control group, the results are comparable with those found in previous studies not using rhTSH (Hegedüs et al., 2003). Duick & Baskin (2003) investigated the effect of either 0·3 mg or 0·9 mg rhTSH in 16 patients with low thyroid RAIU and nodular goitre, nine of whom had suppressed s-TSH levels consistent with subclinical hyperthyroidism. The authors found a more than fourfold increase in thyroid RAIU 72 h after administration of 0·3 mg rhTSH. Between 3 and 7 months after 131I therapy, the goitre was reduced by 30–40%, irrespective of the rhTSH-dose. A total of 69% of the patients reported remission of the compressive symptoms. However, the lack of a control group, and thyroid size determination by palpation alone, constitute drawbacks of this Graf et al. in a – and yet – investigated the effect of two injections h of 0·1 mg rhTSH in patients with nodular goitre and a low baseline thyroid RAIU. The 24 h thyroid RAIU markedly from to an outpatient dose of 131I. months after 131I therapy, mean goitre reduction evaluated by was Silva et al. (2003) investigated patients with a very nodular goitre were and seven were to 131I therapy or to 131I therapy by mg rhTSH given 24 h before 131I to increase the thyroid The 131I activity was without taking the thyroid RAIU into hindering a 131I dose In the group receiving rhTSH, the thyroid RAIU increased from to This resulted in a mean goitre volume reduction of at 12 This was significantly higher than the obtained in the control group (Fig. in thyroid volume in patients with nodular goitre, 6 and 12 months following either with mg rhTSH prior to 131I therapy or 131I therapy Values in in the two indicate the thyroid volume compared to Modified from Silva et al. (2003). these treatment studies (Duick & Baskin, 2003; Graf et al., 2003; Nieuwlaat et al., 2003; Silva et al., some can be with rhTSH seems to a reduction of the 131I activity while still a mean goitre reduction of approximately 40% within the first 12 an may 131I therapy more for patients and may an increased of patients to be treated on an in low the same dose of the goitre size reduction seems to be from approximately to 60% within a if rhTSH is This may to more results of 131I therapy in patients with very in whom thyroidectomy is the treatment option (Bonnema et al., 2000, Hegedüs et al., 2003). In et al. described that of healthy individuals stimulated with bTSH thyroid swelling and These were followed by of similar but more bTSH in patients with goitre et al., rhTSH is well in patients with treated with doses of 0·9 mg rhTSH. In studies only a of patients had as and et al., 1994; et al., et al., 1999). no effects have been reported when similar or lower doses of rhTSH in subjects with an intact benign thyroid gland (Huysmans et al., 2000; Lawrence et al., 2001; Torres et al., there have been recent of more as swelling and from resulting from injections of 0·9 mg rhTSH in patients with et al., et al., 2000; et al., 2001; et al., et al., 2003). The in addition to the on the effects of us et al., to the acute effects of 0·9 mg rhTSH on thyroid size and function in nine healthy euthyroid in a of rhTSH resulted in a significant increase in mean thyroid size of after 24 h, peaking at 48 h to 35% (Fig. On thyroid size had to baseline Thus, the maximum thyroid between 1 and a very and thyroid between 24 and h after administration of rhTSH. The thyroid gland increased from 22 to and was and on Furthermore, rhTSH various of to thyroid thyroid in most of the subjects. The within the first 24 h following injection and within 72 h. The most of this acute effect on thyroid whether bTSH or rhTSH is administered, is an response to an and findings in study et al., support are to be but a rapid response to this et al., the goitre of in patients with nodular goitre 1 week after 131I therapy reported in the study of Nieuwlaat et al. (2003) may have been even more had the been within the first days after therapy. studies to this in order to out the of due to goitre changes in thyroid volume in healthy individuals after administration of 0·9 mg rhTSH and to with from of and Nielsen et al., It is whether rhTSH and 131I therapy in an or even It appears that a greater fraction of patients with rhTSH prior to 131I therapy (Duick & Baskin, 2003; Silva et al., 2003). In the study by Silva et al. patients had a significantly higher of within 12 months compared to those not receiving rhTSH However, L-T4 therapy is without effects this should not clinicians from using 131I therapy, especially in the and of is of The – thyroidectomy – is associated with of thyroid hormone and effects were relatively more common in the group et al., 2003). this was due either to a higher dose of in the thyroid or to a thyroid to rhTSH, or a of these factors. In more patients pretreated with rhTSH from the 131I therapy. this was to an thyroid after 131I therapy is as goitre size was not investigated at this time. However, on are (Nieuwlaat et al., 2003; Silva et al., and by studies are before rhTSH can be in the of 131I therapy. made by Silva et al. (2003) was that T4 within the first week after 131I therapy approximately increased by a factor in the rhTSH group compared with the increase in thyroid hormone secretion resulting in is to be when rhTSH is used in patients with goitre and should be into in the of rhTSH It that with doses of rhTSH of 0·1 mg or this is of little (Nieuwlaat et al., 2003). it to be whether 131I therapy in patients with nontoxic goitre increases the of (Hegedüs et al., 2003). Nieuwlaat et al. have recently that administration of a reduced 131I activity, at the same thyroid dose using either 0·01 or 0·03 mg rhTSH in patients with nodular goitre, resulted in a significantly lower especially in the and the compared with patients receiving 131I therapy. These findings are when treatment of It is most that rhTSH may be in the of not only but also of benign nodular goitre treated with 131I. However, several to be some of which have been It is well known that the of changes markedly 131I therapy without the use of rhTSH et al., et al., 2000). It was recently that the thyroid RAIU 131I therapy, in patients with nodular goitre, was lower than the RAIU, but less when rhTSH was (Nieuwlaat et al., the effective 131I half-life was by rhTSH stimulation. However, the study by Huysmans et al. (2000) has that rhTSH, may to the 131I as be from a this may the thyroid but studies are to these important In most studies, rhTSH was given 24 h before 131I therapy. However, according to both in vitro (Kogai et al., and in vivo (Duick & Baskin, studies, it may well be that the time interval should be in order to a maximum stimulation of the thyroid RAIU. the optimal rhTSH dose prior to 131I therapy to be The effect on thyroid RAIU must be as and goitre the latter to tracheal It is that with very small doses of rhTSH seem (Nieuwlaat et al., 2003). However, the of patients treated according to this has been very small an acute goitre rhTSH and 131I therapy, the to be as well as factors. of these is a for out in that is whether rhTSH in but doses is to just As by studies in nodular & 1997), it can be whether a TSH a general increase in the may of the thyroid to Thus, the goitre reduction obtained by rhTSH 131I therapy may be by factors than the well described increase in thyroid RAIU. The of rhTSH use in benign thyroid is just in its Although use of rhTSH in the of 131I therapy in nodular benign goitre the results of studies in a way that the in this review are In of the high of rhTSH, are especially On a studies may well to in the of thyroid than and benign nodular goitre & 2003).