Clinical Applications of TSH Receptor Antibodies in Thyroid Diseases

Transcription

1 J Korean Med Si 2002; 17: ISSN Copyright The Korean Aademy of Medial Sienes REVIEW Clinial Appliations of TSH Reeptor Antibodies in Thyroid Diseases The loning and sequening of thyroid-stimulating hormone (TSH) reeptor (TSHR), ombined with advanes in moleular tehniques, have failitated the understanding of the interation of the TSHR antibodies (TSHRAbs) with the TSHR at the moleular level and have allowed the delineation of their linial role. TSHRAbs in vivo are funtionally heterogeneous; the stimulating TSHRAbs ause hyperthyroidism and diffuse goiter in patients with Graves disease, whereas, the bloking TSHRAbs ause hypothyroidism in some patients with autoimmune hypothyroidism and are the ause of transient neonatal hypothyroidism. Measuring TSHRAbs has potential linial impliations in differential diagnosis of Graves disease, prediting the outome of Graves disease after antithyroid drug treatment, and prediting the fetal/neonatal hyperthyroidism or neonatal hypothyroidism. The existene of epitope heterogeneity in a patient, i.e., of stimulating TSHRAbs with epitopes other than on the N-terminal region of the extraellular domain, is signifiantly assoiated with favorable long-term linial response to antithyroid drug treatment. Measuring these subtypes for thyroid-stimulating antibody (TSAb) has potential linial impliations, for example, in prediting responsiveness to treatment in untreated patients with Graves disease. Key Words : Reeptors, Thyrotropin; TSH Reeptor antibody; Epitope heterogeneity; Graves Disease; Hyperthyroidism; Hypothyroidism Bo Youn Cho Department of Internal Mediine, Seoul National University College of Mediine and Hospital, Seoul, Korea Reeived : 25 January 2002 Aepted : 26 February 2002 Address for orrespondene Bo Youn Cho, M.D. Department of Internal Mediine, Seoul National University Hospital, 28 Yungon-dong, Chongno-gu, Seoul , Korea Tel : , Fax : byho@plaza.snu.a.kr INTRODUCTION Autoimmune thyroid diseases onsist of a wide spetrum of linial presentations, with Graves disease, Graves ophthalmopathy, Hashimoto s thyroiditis, and atrophi thyroiditis (primary myxedema) (1, 2). Autoimmune thyroid diseases share ommon immunologi evidenes; lymphoyti infiltration of the thyroid, irulating thyroid autoantibodies, T ell immunity, HLA assoiation and familial aggregation. Autoantibodies against the thyroid-stimulating hormone (TSH) reeptor (TSH reeptor antibodies, TSHRAbs) are diretly involved in the pathogenesis of Graves disease and autoimmune hypothyroidism. The TSHRAbs ompete with TSH for TSH reeptor (TSHR) binding and mimi or blok the TSH ation, resulting in hyperthyroidism and goiter or hypothyroidism and thyroid atrophy, respetively. The heterogeneous nature of TSHRAbs is well established. Antibodies apable of interating with multiple epitopes on the TSH reeptor moleule and showing variable TSH ativities, both agonisti and antagonisti, are generally present in a given patient. This variation in epitope speifiity is refleted by the influene of TSHRAbs on the TSHR funtion, as demonstrated by the ourrene of transplaentally transmitted hyperthyroidism or hypothyroidism in the fetuses of mothers with high levels of irulating stimulating or bloking TSHRAbs (1-4). In reent years, a number of investigators have advaned our understanding of how TSH and TSHRAbs interat with TSHR (2, 5), delineating linial impliations of TSHRAbs. The purposes of this review are, first, to summarize reent advanes in TSH reeptor and its autoantibodies, and seond, to desribe their potential linial impliations. CHARACTERISTICS OF TSH RECEPTOR Gene and protein struture The TSHR is a glyoprotein expressed on the ell membrane of thyroytes and a member of the large superfamily of G- protein-oupled seven transmembrane reeptors. The TSHR gene has been loned and sequened (6, 7) and has been loated on the long arm of hromosome 14q31 (8), spans more than 60 kb and ontains 10 exons. Exons 1-9 enode for the extraellular domain (ECD) of the reeptor, and exon 10 odes for the transmembrane and the ytoplasmi domain. Exons 2-8 ode for the leuine-rih repeats (LRRs) (7). The open leading frame onsists of 2,295 nuleotides enoding a 764-amino aid 293

2 294 B.Y. Cho N-glyan A subunit NH2 Helix Sheet Insertion Site (38-45) Ieuine rih repeats (58-277) COOH 764 Fig. 1. Shemati representation of TSH reeptor struture. Modified from ref. 2. protein. In the ECD, TSHR has 35-45% of homology with lutenizing hormone/horioni gonadotropin reeptor (LH/ CGR) and follile stimulating hormone reeptor (FSHR). The TSHR differs from LH/CGR by the presene of two unique insertions of 8 (residues 38-45) and 50 amino aids (residues ) in the ECD (2, 5) (Fig. 1). The ECD of human TSHR ontains six potential asparagines-linked glyosylation sites and at least two sites are glyosylated (Fig. 1). The primary translation produt of 85 kda TSHR is glyosylated, resulting in a high-mannose form of ~100 kda, with subsequent maturation to a 120 kda form with omplex arbohydrates (2, 4). TSH binding is inreased on mature, glyosylated form of the reeptor, and mutation of glyosylation sites redues TSH binding (9). There are 11 ysteine residues in the ECD of TSHR that ould form five pairs, with one residue remaining as the orphan. Disulfide bonding is vital for the orret folding of the TSHR and the maintenane of its tertiary onformation. At present, however, there are no diret data to indiate whih ysteines form the five pairs. Two subunits model of the TSH reeptor Insertion Site ( ) ? 1 Cleavage sites ? B subunit In vitro studies suggest that the TSHR is present not only as a single hain, but also in a two-subunit form (10, 11). The glyosylated ECD, the A subunit, is leaved from the membrane-spanning B subunit and these two subunits are linked by disulfide bonds (2). The amino-terminal leavage site is loated between residues 302 and 317 in proximity to a basi domain formed by residues (10, 12). The arboxyterminal leavage site seems to be loated between residues (Fig. 1). Although there is some debate whether the leavage site is two or not, reent studies suggest that the initial leavage at the amino-terminal site is followed by progressive leavage or degradation to the amino-terminal site or, alternatively, rapid degradation of the putative onneting peptide (10, 12). The physiologi signifiane of the leavage and shedding phenomenon remains unresolved. Intraellular signaling through TSH reeptor TSH binding to its reeptor leads to oupling Gs and stimulates AMP prodution through ativation of adenylyl ylase (13). At higher onentrations of TSH, the TSHR ouples Gq/11, resulting in ativation of phospholipase C, and stimulates the inositol phosphate (IP) pathway (13). The inrease in AMP leads to phosphorylation of protein kinase A and to ativation of transription fators, suh as CREB, resulting in the inrease of iodide uptake, thyroid peroxidase, and thyroglobulin synthesis (13). In addition, stimulation of Gq/11 and the phospholipase C-dependent inositol phosphate /diaylglyerol pathway at the higher doses of TSH ativates hydrogen peroxide generation and iodination. Graves immunoglobulins (IgG) an enhane both AMP and IP prodution in FRTL-5 ells as well as in ells transfeted with human TSHR DNA (14). The physiologial signifiane of this dual signaling system and the linial role of the IP pathway in Graves disease still remain to be defined. TSH and autoantibody binding to the TSH reeptor Several approahes have been used to map the sites on TSHR responsible for TSH or TSHRAbs binding, all having serious problems. These inlude the use of syntheti peptide sequenes and anti-peptide antibodies as well as site-direted mutagenesis. Although studies using syntheti peptides have provided useful information on funtionally important sites on TSHR, they failed to provide onlusive evidene for TSHRAbs interation with TSHR (2), for the following reasons; they ould identify only linear and non-glyosylated binding sites, but TSH and TSHRAbs were expeted to interat with nonlinear onformational epitopes. Site-direted mutagenesis has been used to failitate the searh for TSH-binding sites and for epitopes of TSHRAbs. Several studies have shown that TSH and TSHRAbs interat with TSHR at different sites. From himeri TSHR-LH/CGR data, important TSH ontat points are present in the midregion (residues ) and arboxyl-terminal segments (residues ) of the TSHR etodomain (15). Homologous or nonhomologous substitutions in the TSHR suggest that amino aids between , , , and are the binding sites for TSH (15, 16). Thus, the binding sites of TSH on the ECD of TSHR are multiple, nonlinear, and disontinuous. In several reports using himeri TSHR-LH/CGR, most patients with Graves disease were found to have one or more stimulating TSHAbs epitope(s) on the N-terminal portion of the ECD of TSHR, whereas the major bloking TSHRAb epitope in patients with hypothyroidism and idiopathi myxedema was loated in the C-terminus (17-19). The N-terminal third of the TSHR (amino aids 8-165) is neessary for TSHRAb stimulation. The epitope for bloking TSHRAb on the ECD of TSHR overlaps with, but is not preisely the

3 Clinial Appliations of TSH Reeptor Antibodies in Thyroid Diseases 295 N N N TSAb-1 TSAb-2 TSBAb TSH/TBII TSH/TBII C WT C Ml+2 C M2 Table 1. Classifiation of TSHRAbs based on funtional ativities and detetion methods Antibody Funtion Detetion method TBII Inhibits 125 I-TSH Radioreeptor assay: inhibition of 125 I-TSH binding to TSHR binding to soluble porine TSHR or reombinant human TSHR TSAb Stimulates AMP Bioassay: % inrease in AMP over prodution, iodine normal serum in ultured thyroytes uptake, Tg synthesis (FRTL-5, htshr expressed CHO ells) TSHR type Fig. 2. Shemati struture of the wild type TSH reeptor and two TSH and LH/CG reeptor himeras designated as M1+2 and M2, illustrating binding sites for TSH, TBII, TSBAb and major and minor binding epitopes for TSAb subtypes. The amino aid numbers indiate the restrition sites used for himeras. TSH, TSAb, and TBII responses in these ell lines are depited as graded positive or negative in the table (bottom). The TSAb-2 epitope was onsidered to be minor beause of its low level of reativity in M1+2 or M2 ell lines and is boxed in dotted lines. Modified from ref. 17 and 18. same as, the TSH-binding site and the stimulating TSHRAb epitope. Overlap between these three ligands is greatest in the C-terminal region of the ECD of TSHR (Fig. 2). TSHRAb ASSAYS TSHRAbs an be deteted by two tehniques. 1) TSHRAbs ompete with TSH for reeptor binding, and are deteted by radioreeptor assay based on the ompetitive inhibition of 125 I-TSH binding on TSHR. The TSHRAb deteted by this assay is designated TSH binding inhibitor immunoglobulin (TBII). 2) TSHRAbs stimulate the adenylyl ylase system that leads to the prodution of AMP, or bloks TSH-stimulated AMP prodution in an in vitro thyroid ell bioassay. The TSHRAb deteted by this bioassay is designated thyroid-stimulating antibody (TSAb, stimulating TSHRAb) or thyroid stimulation bloking antibody (TSBAb, bloking TSHRAb), respetively. Eah of these assays has different diagnosti effetiveness, limitations, and linial utility. The lassifiation of TSHAbs based on their effet on thyroid ell funtion and their detetion methods are listed in Table 1. TBII assay htsh-r sequene AMP response TSH TSAb-1 TSAb-2 LH/CG-R sequene TBII WT M M In the TBII assay system, TSHR preparation and 125 I-labeled TSBAb Inhibits TSH-indued Bioassay: % inhibition of AMP prodution, TSH-indued AMP prodution iodine uptake, ompared to normal serum Tg synthesis TBII, TSH binding inhibitor immunoglobulin; TSAb, thyroid-stimulating antibody; TSBAb, thyroid stimulation bloking antibody; Tg, thyroglobulin. TSH are essential omponents. Porine thyroid reeptor protein has been established as an adequate alternative to human TSH reeptors beause of its ready availability and stability (20). Purified bovine TSH has higher biologial ativity than human TSH and is a more suitable 125 I-labeled ligand for this assay. This test s simpliity, preision, and ost-effetiveness, along with its ommerial availability, have made it the most widely used test in linial laboratories. However, there are a number of limitations to this assay. The TBII assay does not differentiate between the TSH-binding inhibitory antibodies that stimulate (TSAbs) and those that blok the TSH ativity (TSBAbs). In addition, about 10% or more of linially hyperthyroid Graves patients test negative with the TBII assay (21), a finding that may be aused by the use of TSHR and TSH from heterologous speies (porine TSHR and bovine TSH) that may influene the detetion of autoantibodies to the human TSHR (22). A reently developed seond-generation assay for TBII uses reombinant human TSHR expressed in eukaryoti ells, grown in suspension at a high density (23). The htshr was then aptured in the test tubes oated with a monolonal antibody to native TSHR. This assay showed better diagnosti sensitivity (99%) and speifiity (99%) than the onventional TBII assay. Bioassays for TSHRAbs measurement Thyroid-stimulating antibody assay After a sensitive AMP response to TSH or TSHRAbs in ultured human thyroytes using a medium free of sodium hloride for inubation was demonstrated (24), diret measurement of AMP in the medium (instead of within the ells) simplified this assay and signifiantly enhaned its sensitivity. Later the assay was further simplified and made pratial by using rat thyroid ell line alled FRTL-5 instead of human

4 296 B.Y. Cho thyroid ells. Nevertheless, the major limitation of FRTL-5 ells has been the metiulous ulture onditions they require, inluding expensive 6-hormone (inluding TSH) medium and growth for 3-5 days in TSH-deprived medium (3). Furthermore, in our experiene (18), the FRTL-5 bioassay fails to detet stimulating ativity in roughly 10 to 20% of patients with Graves disease. The reported diagnosti sensitivity generally ranges between 75 and 90%. The stable expression of the human TSHR in CHO (CHO- TSHR) ells has ombined the advantage of deteting TSAb ativity with a homologous reeptor and the ready availability ompared to FRTL-5 ells (2). The TSHR density is 5- to 10-times higher in CHO-TSHR ells than in FRTL-5 ells or human thyroid ells. This enhanes analytial sensitivity, thus allowing detetion of some borderline positive patients (2, 3). Thus, the use of CHO-TSHR ells redues the frequeny of false negative results seen with the FRTL-5 ells (18). The reported diagnosti sensitivity ranges between 90 and 95% (18). Further advantages are that CHO-TSHR ells are easier to maintain in ulture, they do not require lengthy pre-ulture time for assay, and they are superior to FRTL-5 ells in auray, simpliity, and ost (3). Thyroid stimulation bloking antibody assay Some patients with atrophi hypothyroidism have antibodies that blok TSH binding and ation (25). These bloking antibodies (TSBAbs) ompete with TSH for reeptor binding and blok the biologial effets of TSH on thyroid ells, inluding AMP prodution, iodine uptake, and thyroid ell growth (26-28). They indue transient neonatal hypothyroidism when transferred aross the plaenta (29, 30). TSBAb assay quantitates inhibition by autoantibodies of the inrease in AMP prodution or the iodide uptake in response to a standard onentration of TSH in the FRTL-5 ell (26-28) or CHO- TSHR ell (31, 32). HETEROGENEITY OF TSHRAbs The heterogeneous nature of TSHRAbs in terms of funtion and epitopes is well reognized. A single patient has funtionally different kinds of TSHRAbs. In addition, sine the moleular loning of the TSHR, it has beome possible to disern epitopi differenes even within a population of autoantibodies with similar funtion. Funtional heterogeneity It has been well known that there are two primary types of funtional TSHRAbs: 1) TSHRAbs that stimulate the adenylyl ylase-amp asade through TSHR and ause hyperthyroidism and goiter (TSAb); and 2) TSHRAbs that blok TSH binding and ation, thereby ausing atrophi hypothyroidism (TSBAb). In addition, reently neutral TSHRAbs, whih neither stimulate the TSHR nor blok TSH ation, have been also demonstrated (33). It is well established that not all TSHRAbs that inhibit TSH binding to the TSHR are stimulatory. In fat, there has been no good orrelation between TSAb and TBII ativities in Graves disease patients in many reports (1, 2). This disrepany between TBII and TSAb ativities might be related to speies speifiity, as porine reeptor has been used for most TBII studies and rat thyroid ells for most TSAb studies. However, the orrelation between these two ativities was also weak when using reombinant human TSHR for TBII and human thyroid ells for TSAb (34). Thus, the ause of the disparity is more likely to be the TSBAbs that are deteted in the TBII assay but, in fat, mask TSAb ativity in the bioassay (32). The funtional heterogeneity of TSHRAbs was onfirmed by Kohn et al. (35), who found that the lones generated from lymphoytes from a hypothyroid woman inluded TSAbproduing lones with weak TBII ativity and a number of TBII-positive lones with TSH-bloking ativity. Furthermore, reently, we demonstrated that 18.5% of patients with hyperthyroid Graves disease had both TSBAb and TSAb ativities (32). Thus, sera from most patients with Graves disease ontain both TSAb and TSBAb/TBII ativities, and the linial effet may depend on the relative onentration and affinity of the predominating antibody. Epitope heterogeneity Heterogeneity within the stimulating TSHRAbs with respet to epitope reognition on TSHR has been demonstrated using himeras of the human TSHR and the LH/CGR (18, 19, 36). Using suh a himeri ell line, Kim et al. reently reported that patients with Graves disease have been divided into two subgroups aording to the TSHRAbs epitopes distribution (18). Of 66 patients studied, 45 patients (68%) had stimulating TSHRAbs with homogeneous epitope distribution that reognize only the N terminus (residues ) of the TSHR etodomain. Heterogeneous epitope distribution, with epitopes other than on the N-terminus of the TSHR, ourred in 21 patients (32%). Moreover, the funtional stimulating TSHRAb epitope hanged from residues to residues outside this region in half of the patients with Graves disease during treatment of hyperthyrodism (19). Patients with heterogeneous epitope for stimulating TSHRAb initially or developed during treatment are more responsive to antithyroid drug treatment (18, 19). Thus, this epitope heterogeneity in patients with Graves disease has linial impliations. CLINICAL APPLICATIONS Diagnosti values in Graves disease

5 Clinial Appliations of TSH Reeptor Antibodies in Thyroid Diseases 297 Graves disease is an organ-speifi autoimmune disease aused by stimulating TSHRAb that stimulate the thyroid funtion and growth, resulting in hyperthyroidism and diffuse goiter. Thus, theoretially, the detetion of TSHRAb in a hyperthyroid patient is diagnosti of Graves disease. However, beause 80% or more of patients with Graves disease have a diffuse goiter and homogeneous thyroid san, it is likely that most patients an be diagnosed by thyroid hormones measurements without TSHRAb testing. Thus, TSHRAb measurement has not been reommended as a routine investigation in reent guidelines (37, 38). However, assays for TSHRAbs is helpful in diagnosing patients with mild hyperthyroidism and small or no goiter and may also be useful in differentiating Graves disease from other forms of thyrotoxiosis, suh as toxi nodular goiter, painless (autoimmune) thyroiditis, or fatitious thyrotoxiosis. TSHRAbs measurements may also be helpful for pregnant women or patients with reent iodine load where radioiodine uptake or thyroid san is ontraindiated. In a patient who presents with Graves ophthalmopathy but is biohemially euthyroid, TSHRAb assay may be valuable in onfirming the diagnosis of Graves disease. TSAb measurements tend to be more sensitive (85-100%) than TBII measurements (75-96%) in untreated Graves patients (23, 39). The new more sensitive and preise assays using CHO-TSHR are positive in as muh as 98% of patients with untreated Graves disease (23). Thus, TSAb by CHO- TSHR is a better biologi marker than TBII and should be preferred for diagnosti purposes. Guiding hoie of treatment modality in Graves disease The major problem with antithyroid drug is the high relapse rate (approximately 50% or more) after 1-2 yr of treatment. Thus, it is linially important to identify patients who are likely to ahieve remission with antithyroid drug treatment. In general, however, most reported studies suggest that initial TSHRAb ativity has not had satisfatory positive and negative preditive values for remission or relapse after antithyroid drug treatment (40, 41). Nevertheless, taking into aount the initial titer of TSHRAb and other linial parameters suh as, age, gender, goiter size, the severity of hyperthyroidism, and the presene of ophthalmopathy, subgroups of patients an be identified as harboring a high or low risk of relapse. In a study by Vitti et al. (42), the ombination of a small goiter (<40 ml) and low TBII level (<30 U/L) onferred a 45% hane of remission during the 5 yr after ompletion of months ourse of antithyroid drug treatment. In ontrast, patients with a large goiter (>70 ml) and a higher TBII level had less than a 10% hane to remain in remission. As disussed above, it has been reported that patients with heterogeneous epitope for stimulating TSHRAb initially or developed during treatment are more responsive to antithyroid drug treatment (18, 19). Approximately 30% of patients initially have heterogeneous epitope for stimulating TSHRAbs whose ativity depends on sites other than the N-terminal lous. In patients with heterogeneous epitope, 94% beame euthyroid within 3 months, whereas this was true of only 70% of patients with homogeneous epitope. Furthermore, there was a derease in the duration of antithyroid drug therapy neessary to ahieve a euthyroid state in the heterogeneous epitope group. Thus, the authors (18) demonstrated for the first time that patients with a heterogeneous population of stimulating TSHRAbs involving reeptor determinants other than the N-terminus have a better linial response to antithyroid therapy. Reently we expanded this observation in 159 patients with Graves disease for 4 to 7 yr (43). In 52 patients (32.7% of 159) with heterogeneous epitope of stimulating TSHRAb, 34 patients maintained remission for more than 12 months, with 65.4% of suess rate of antithyroid drug therapy. However, 46 out of 107 patients, whose IgGs had ativities of stimulating TSHRAb with homogeneous epitope, attained remission after antithyroid drug therapy with 43.0% of rate. Thus, the suess rate of antithyroid drug therapy in the heterogeneous epitope group, was signifiantly higher than that in the homogeneous epitope group (p=0.011). While the overall average rate of suess is about 50%, the ombination of a heterogeneous TSAb epitope and low TBII titer ( 40%) onferred a 82% hane of remission, and a patient with a homogeneous TSAb epitope and high TBII titer (>40%) has a 32% hane of remission. Likewise, ombination of heterogeneity of TSAb with size of goiter and initial T3 onentration had different probability of suess ranging from 32 to 74% (Table 2). The existene of epitope heterogeneity in a patient, i.e., of stimulating TSHRAbs with epitopes other than on the N- terminal region of the extraellular domain, is, therefore, sig- Table 2. Suess rate after antithyroid drug treatment aording to ombination of heterogeneity of epitope for TSAb with other prognosti parameters Combination of parameters Suess rate Smaller goiter+heterogeneous epitope 20/28 (71.4%) Larger goiter+homogeneous epitope 19/60 (31.7%) Low T3+Heterogeneous epitope 17/23 (73.9%) High T3+Homogeneous epitope 15/46 (32.6%) Low TBII+Heterogeneous epitope 22/27 (81.5%) High TBII+Homogeneous epitope 15/47 (31.9%) Smaller goiter, none/small on linial examination; Larger goiter, medium/large on linial examination; Heterogeneous epitope, group who had IgG with residual TSAb ativity in the himeras; Homogeneous epitope, group with epitope only in the N-terminal portion of the extraellular domain; Low T3, total T3 onentration was 350 ng/dl or less than 350 ng/dl; High T3, total T3 onentration was more than 350 ng/dl; Low TBII, titer of TBII was 40% or less than 40%; High TBII, titer of TBII was more than 40%.

6 298 B.Y. Cho nifiantly assoiated with favorable long-term linial response to antithyroid drug treatment. We suggest, therefore, that the measurement of differene in epitopes for TSHRAbs in patients with untreated Graves hyperthyroidism may be useful to predit the long-term outome after antithyroid drug treatment. Monitoring the response to therapy in Graves disease In general, most studies demonstrate that the titers of TSHRAb (TBII and/or TSAb) are dereased during treatment and their falling titers indiate a good response (44). One question merits onsideration: ould the determination of TSHRAb during antithyroid drug treatment ontribute to the adaptation of the treatment plan to eah ase or be benefit for the patient to monitor the level of TSHRAb during treatment? Previously Cho et al. (41) reported that in a 174 patients with an overall remission rate of 52%, the proportion of remitters was muh higher among patients treated for 24 months who had beome TBII-negative with normal basal TSH after 6 (94% remission rate) or 12 (75% remission rate) months of treatment than in patients whose TSHRAb had remained detetable for 18 (63% remission rate) or 24 months (52% remission rate). Thus, the rate of fall of TSHRAb during antithyroid drug treatment is preditive of subsequent outome. The duration of antithyroid drug treatment an be adapted to the TSHRAb status. Cho et al. (41) previously reported that the remission rate of patients, whose treatment was disontinued on the normalization of their TSHRAbs and TSH levels, was not signifiantly different from that of patients, who were treated for 24 months irrespetive of their TSH levels and TSHRAbs ativities (52% vs 63%, p<0.05). The mean treatment of duration of the former group was 10 months, 14 months shorter than the latter group. These findings suggest that the sequential monitoring of TSHRAb during the treatment of Graves disease gives valuable information onerning the optimum time for withdrawal of mediation. Beause the TSHRAb stimulates the thyroid funtion and results in hyperthyroidism in Graves disease, theoretially, the disappearane of TSHRAbs at the end of the treatment reflets remission. However, in a reent report of meta-analysis that inluded 18 studies omparing the relapse rate in TSHRAbs-positive and -negative patients, only 10 of 18 studies showed a statistially signifiant differene (44). The authors onluded that the absene of TSHRAb was signifiantly protetive against relapse, but that it was indiative of either relapse or remission after antithyroid drug treatment in individual patients. As reently disussed by Davies et al. (37), variations in the effetiveness of the assays, along with the known heterogeneity of TSHRAbs and the fat that bloking antibodies are deteted differently by TBI or TSAb assays, may have an impat on the final analysis. Furthermore, as disussed above, our data suggest that both TSAb and TBAbs are heterogeneous (18, 19, 31, 32) and that a subtype of TSAb deteted by himeri TSHR is assoiated with inreased responsiveness to antithyroid drug therapy (18, 43). Guidelines for the management of Graves disease in pregnany Although maternal TSHRAbs values normally deline during the third trimester, an elevated TSAb level at 28 to 30 weeks of gestation is assoiated with an inreased frequeny of Graves hyperthyroidism in the neonate (45). The reommendations of European Thyroid Assoiation for measuring TSHRAbs in pregnany an be summarized as follows (46): 1. Euthyroid women without mediation and previously treated with antithyroid drugs alone: no TSHRAbs testing is required. 2. Women previously treated with radioiodine or surgery for Graves disease: measure TSHRAbs early in pregnany. If the level is high, the fetus should be monitored losely for signs of hyperthyroidism and antithyroid drugs treatment for mother should be onsidered. Antibody titers should be repeated in the last trimester. 3. Women urrently treated with antithyroid drugs: adjustment of treatment is needed to ahieve a high-normal level of free T4 in the mother. TSHRAbs should be heked in the last trimester. TSHRAb and autoimmune hypothyroidism Indeed, the pathogeneti role of bloking TSHRAb in autoimmune hypothyroidism is suggested not only by their prevalene but also by their apaity to indue neonatal transient hypothyroidism. The reported prevalene of these antibodies in atrophi autoimmune thyroiditis varies greatly, raging from 14% (47) to 59% (28). In our study (28), the prevalenes of TBII in goitrous and atrophi thyroiditis were 6.3% and 48% and those of TSBAb were 10.5% and 59%, respetively. It has been well doumented that transient neonatal hypothyroidism an be indued by transplaental transfer of maternal TSBAb sine the first report of Matsuura et al. in 1980 (29). TBII and TSBAb were deteted in 5% and 4%, respetively, among the mothers of hypothyroid newborns. On the whole, antibody-related transient neonatal hypothyroidism aounts for 1% of all auses of ongenital hypothyroidism. Thus, transient transplaental neonatal hypothyroidism is as preditable as feto-neonatal hyperthyroidism by the assay of TBII and, if positive, by an assay of TSBAb during the last trimester (39). Epidemiologi data suggest that this sreening is indiated in those with atrophi thyroiditis (primary myxedema).

7 Clinial Appliations of TSH Reeptor Antibodies in Thyroid Diseases 299 Table 3. Current linial indiations of TSHRAbs assays Clinial presentation Antibody Indiations Hyperthyroidism TSAb/TBII Differential diagnosis of Graves disease Prediting the outome of Graves disease after antithyroid drug treatment Prediting the fetal and neonatal hyperthyroidism Epitope of Prediting the outome of Graves TSAb disease before treatment Hypothyroidism TSBAb/TBII Prediting neonatal hypothyroidsm Evaluation of flutuating thyroid funtion during treatment of Graves disease Euthyroid Graves TSAb/TBII Diagnosis of sublinial hyperthyroidism disease TBII, TSH binding inhibitor immunoglobulin; TSAb, thyroid-stimulating antibody; TSBAb, thyroid stimulation bloking antibody. Monitoring flutuating thyroid funtion Spontaneous evolution of hyperthyroid Graves disease to hypothyroidism has been well haraterized (48). Although thyroid destrution aused by onomitant autoimmune thyroiditis has been proposed as the major pathogeneti mehanisms of this phenomenon, the presene of bloking TSHRAbs is onsidered to be another possible ause (31, 49). TSBAbs may result from a onversion of the bioativity of TSAb or may oexist with TSAb at the time of hypothyroidism. In addition, several patients with flutuating thyroid funtion from hypothyroidism to hyperthyroidism by hanging the ativity of TSBAb to TSAb have been reported (50). Thus, the measurement of bioativities of TSHRAbs during the ourse of Graves disease is valuable in the ase, espeially, of flutuating thyroid funtion. CONCLUSION The loning and sequening of TSHR, ombined with advanes in moleular tehniques, have failitated the understanding of the interation of the TSHRAbs with the TSHR at the moleular level and have allowed the delineation of their linial role. Reently developed new tehniques onfirm that TSHRAbs in vivo are funtionally or epitopially heterogeneous, and indiate that the thyroid funtion may depend on the balane of the funtional ativities of TSHRAbs. In addition, the existene of epitope heterogeneity in a patient is signifiantly assoiated with apparently long-term linial response to antithyroid drug treatment. Measuring these subtypes for TSAb has potential linial impliations, for example, in prediting responsiveness to treatment in Graves disease. Table 3 summarizes the author s view on the urrent linial indiations for the use of TSHRAbs measurements. REFERENCES 1. Weetman AP, MGregor AM. Autoimmune thyroid disease: further developments in our understanding. Endor Rev 1994; 15: Rapoport B, Chazenbalk GD, Jaume2 JC, MLahlan SM. The thyrotropin (TSH) reeptor: interation with TSH and autoantibodies. Endor Rev 1998; 19: Gupta MK. Thyrotropin-reeptor antibodies in thyroid diseases: advanes in detetion tehniques and linial appliations. Clin Chim Ata 2000; 293: Kopp P. The TSH reeptor and its role in thyroid disease. Cell Mol Life Si 2001; 58: Kohn LD, Shimura H, Shimura Y, Hidaka A, Giuliani C, Napolitano G, Ohmori M, Laglia G, Saji M. The thyrotropin reeptor. Vitam Horm 1995; 50: Nagayama Y, Kaufman KD, Seto P, Rapoport B. Moleular loning, sequene and funtional expression of the DNA for the human thyrotropin reeptor. Biohem Biophys Res Commun 1989; 165: Parmentier M, Libert F, Maenhaut C, Lefort Gerard C, Perret J, Van Sande J, Dumont JE, Vassart G. Moleular loning of the thyrotropin reeptor. Siene 1989; 246: Rousseau-Merk MF, Misrahi M, Loosfelt H, Atger M, Milgrom E. Assignment of the human thyroid stimulating hormone reeptor (TSHR) gene to hromosome 14q31. Genomis 1990; 8: Russo D, Chazenbalk GD, Nagayama Y, Wadsworth HL, Seto P, Rapoport B. A new strutural model for the thyrotropin (TSH) reeptor as determined by ovalent rosslinking of TSH to the reombinant reeptor in intat ells: evidene for a single polypeptide hain. Mol Endorinol 1991; 5: Tanaka K, Chazenbalk GD, MLahlan SM, Rapoport B. Thyrotropin reeptor leavage at site 1 does not involve a speifi amino aid motif but instead depends on the presene of the unique, 50 amino aid insertion. J Biol Chem 1998; 273: Furmaniak J, Hashim FA, Bukland PR, Petersen VB, Beever K, Howells RD, Smith BR. Photoaffinity labelling of the TSH reeptor on FRTL5 ells. FEBS Lett 1987; 215: Graves PN, Pritsker A, Davies TF. Post-translational proessing of the natural human TSH reeptor: demonstration of more than two leavage sites. Endorinology 1999; 84: Dumont JE, Lamy F, Roger P, Maenhaut C. Physiologial and pathologial regulation of thyroid ell proliferation and differentiation by thyrotropin and other fators. Physiol Rev 1992; 72: Hidaka A, Okajima F, Ban T, Kosugi S, Kondo Y, Kohn LD. Reeptor ross-talk an optimize assays for autoantibodies to the thyrotropin reeptor: effet of phenylisopropyladenosine on adenosine 3,5 - monophosphate and inositol phosphate levels in rat FRTL-5 thyroid ells. J Clin Endorinol Metab 1993; 77: Nagayama Y, Russo D, Chazenbalk GD, Wadsworth HL, Rapoport B. Extraellular domain himeras of the TSH and LH/CG reeptors reveal the mid-region (amino aids ) to play a vital role in high affinity TSH binding. Biohem Biophys Res Commun 1990; 173: Kosugi S, Ban T, Akamizu T, Kohn LD. Site-direted mutagenesis

8 300 B.Y. Cho of a portion of the extraellular domain of the rat thyrotropin reeptor important in autoimmune thyroid disease and nonhomologous with gonadotropin reeptors. Relationship of funtional and immunogeni domains. J Biol Chem 1991; 266: Tahara K, Ban T, Minegishi T, Kohn LD. Immunoglobulins from Graves disease patients interat with different sites on TSH reeptor/lh-cg reeptor himeras than either TSH or immunoglobulins from idiopathi myxedema patients. Biohem Biophys Res Commun 1991; 179: Kim WB, Cho BY, Park HY, Lee HK, Kohn LD, Tahara K, Koh C-S. Epitopes for thyroid stimulating antibodies in Graves sera: a possible link of heterogeneity to differenes in response to antithyroid drug treatment. J Clin Endorinol Metab 1996; 81: Kim WB, Chung HK, Lee HK, Kohn LD, Tahara K, Cho BY. Changes in epitopes for thyroid-stimulating antibodies in Graves disease sera during treatment of hyperthyroidism: therapeuti impliations. J Clin Endorinol Metab 1997; 82: Shewring GA, Rees Smith B. An improved radioreeptor assay for TSH reeptor antibodies. Clin Endorinol (Oxf) 1982; 17: Iliki A, Gamstedt A, Karlsson FA. Hyperthyroid Graves disease without detetable thyrotropin reeptor antibodies. J Clin Endorinol Metab 1992; 74: Vitti P, Elisei R, Tonahera M, Chiovato L, Manusi F, Rago T, Mammoli C, Ludgate M, Vassart G, Pinhera A. Detetion of thyroid-stimulating antibody using Chinese hamster ovary ells transfeted with loned human thyrotropin reeptor. J Clin Endorinol Metab 1993; 76: Costagliola S, Morgenthaler NG, Hoermann R, Badenhoop K, Struk J, Freitag D, Poertl S, Weglohner W, Hollidt JM, Quadbek B, Dumont JE, Shumm-Draeger P-M, Bergmann A, Mann K, Vassart G, Usadel K-H. Seond generation assay for thyrotropin reeptor antibodies has superior diagnosti sensitivity for Graves disease. J Clin Endorinol Metab 1999; 84: Kasagi K, Konishi J, Iida Y, Ikekubo K, Mori T, Kuma K, Torizuka K. A new in vitro assay for human thyroid stimulator using ultured thyroid ells: effet of sodium hloride on adenosine 3,5 -monophosphate inrease. J Clin Endorinol Metab 1982; 54: Endo K, Kasagi K, Konishi J, Ikekubo K, Okuno T, Takeda Y, Mori T, Torizuka K. Detetion and properties of TSH binding inhibitor immunoglobulins in patients with Graves disease and Hashimoto s thyroiditis. J Clin Endorinol Metab 1978; 46: Konishi J, Iida Y, Endo K, Misaki T, Nohara Y, Matsuura N, Mori T, Torizuka K. Inhibition of thyrotropin-indued adenosine 3,5 - monophosphate inrease by immunoglobulins from patients with primary myxedema. J Clin Endorinol Metab 1983; 57: Cho BY, Shong YK, Lee HK, Koh C-S, Min HK. Inhibition of thyrotropin stimulated adenylate ylase ativation and growth of rat thyroid ells, FTL-5, by immunoglobulin G from patients with primary myxedema: omparison with ativities of thyrotropin-binding inhibitor immunoglobulins. Ata Endorinol (Copenh) 1989; 120: Cho BY, Kim WB, Chung JH, Yi K-H, Shong YK, Lee HK, Koh C-S. High prevalene and little hange in TSH reeptor bloking antibody titres with thyroxine and antithyroid drug therapy in patients with non-goitrous autoimmune thyroiditis. Clin Endorinol (Oxf) 1995; 43: Matsuura N, Yamada Y, Nohara Y, Konishi J, Endo K, Kojima H, Wataya K. Familial neonatal transient hypothyroidism due to maternal TSH-binding inhibitor immunoglobulins. N Engl J Med 1980; 303: Cho BY, Shong YK, Lee HK, Koh C-S, Min HK, Lee M. Transient neonatal hypothyroidism due to transplaental transfer of maternal imm-unoglobulins that inhibit TSH binding, TSH-indued AMP inrease and ell growth. Endorinol Jpn 1988; 35: Chung H-K, Kim WB, Park DJ, Kohn LD, Tahara K, Cho BY. Two Graves disease patients who spontaneously developed hypothyroidism after antithyroid drug treatment: harateristis of epitopes for thyrotropin reeptor antibodies. Thyroid 1999; 9: Kim WB, Chung HK, Park YJ, Park DJ, Tahara K, Kohn LD, Cho BY. The prevalene and linial signifiane of bloking thyrotropin reeptor antibodies in untreated hyperthyroid Graves disease. Thyroid 2000; 10: Tonahera M, Costagliola S, Cetani F, Duobu J, Stordeur O, Vassart G, Ludgate M. Patient with monolonal gammopathy, thyrotoxiosis, pretibial myxedema and thyroid-assoiated ophthalmopathy; demonstration of diret binding of autoantibodies to the thyrotropin reeptor. Eur J Endorinol 1996; 134: Filetti S, Foti D, Costante G, Rapoport B. Reombinant human TSH reeptor in a radioreeptor assay for the measurement of TSH reeptor autoantibodies. J Clin Endorinol Metab 1991; 72: Kohn LD, Suzuki K, Hoffman WH, Tombaini D, Maroi C, Shimojo N, Watanabe Y, Amino N, Cho BY, Kohno Y, Hirai A, Tahara K. Charaterization of monolonal thyroid-stimulating and thyrotropin binding-inhibiting autoantibodies from a Hashimoto s patient whose hildren had intrauterine and neonatal thyroid disease. J Clin Endorinol Metab 1997; 82: Grasso YZ, Kim MR, Faiman C, Kohn LD, Tahara K, Gupta MK. Epitope heterogeneity of thyrotropin reeptor-bloking antibodies in Graves patients as deteted with wild-type versus himeri thyrotropin reeptors. Thyroid 1999; 9: Davies TF, Roti E, Braverman LE, Degroot LJ. Therapeuti ontroversy: thyroid-stimulating antibodies. J Clin Endorinol Metab 1998; 83: Saravanan P, Dayan CM. Thyroid autoantibodies. Endorinol Metab Clin North Am 2001; 30: Orgiazzi J. Anti-TSH reeptor antibodies in linial pratie. Endorinol Metab Clin North Am 2000; 29: Shleusener H, Shwander J, Fisher C, Holle R, Holl G, Badenhoop K, Hensen J, Finke R, Bogner U, Mayr WR, Shernthaner G, Shatz H, Pikardt CR, Kotulla P. Prospetive multienter study on the predition of relapse after antithyroid drug treatment in patients with Graves disease. Ata Endorinol (Copenh) 1989; 120: Cho BY, Shong MH, Yi K-H, Lee HK, Koh C-S, Min HK. Evaluation of serum basal thyrotropin levels and thyrotropin reeptor antibody ativities as prognosti markers for disontinuation of antithyroid drug treatment in patients with Graves disease. Clin Endorinol (Oxf) 1992; 36:

9 Clinial Appliations of TSH Reeptor Antibodies in Thyroid Diseases Vitti P, Rago T, Chiovato L, Pallini S, Santini F. Flore E, Rohi R, Martino E, Pinhera A. Clinial features of patients with Graves disease undergoing remission after antithyroid drug treatment. Thyroid 1997; 7: Kim TY, Park YJ, Kim WB, Park DJ, Kohn LD, Cho BY. Predition of outome in Graves disease after antithyroid drug treatment aording to heterogeneity in epitopes of thyroid-stimulating antibodies. J Clin Endorinol Metab (submitted). 44. Feldt-Rasmussen U, Shleusenter H, Carayon P. Meta-analysis evaluation of the impat of thyrotropin reeptor antibodies on long term remission after medial therapy of Graves disease. J Clin Endorinol Metab 1994; 78: Zakarija M, MKenzie JM. Pregnany assoiated hanges in thyroid stimulating antibody of Graves disease and the relationship to neonatal hyperthyroidism. J Clin Endorinol Metab 1983; 57: Laurberg P, Nygaard B, Glinoer D, Grussendorf M, Orgiazzi J. Guideline for TSH-reeptor antibody measurements in pregnany: results of an evidene-based symposium organized by the European Thyroid Assoiation. Eur J Endorinol 1998; 139: Tamaki H, Amino N, Kimura M, Hidaka Y, Takeoka K, Miyai K. Low prevalene of thyrotropin reeptor antibody in primary hypothyroidism in Japan. J Clin Endorinol Metab 1990; 71: Wood LC, Ingbar SH. Hypothyroidism as late sequela in patients with Graves disease treated with antithyroid drugs. J Clin Invest 1979; 64: Tamai H, Kasagi K, Takaihi Y, Takamatsu J, Komaki G, Matsubayashi S, Konishi J, Kuma K, Kumagai LF, Nagataki S. Development of spontaneous hypothyroidism in patients with Graves disease treated with antithyroid drugs: linial, immunologial and histologial findings in 26 patients. J Clin Endorinol Metab 1989; 69: Cho BY, Shong YK, Lee HK, Koh C-S, Min HK. Graves hyperthyroidism following primary hypothyroidism: sequential hanges in various ativities of thyrotropin reeptor antibodies. Ata Endorinol (Copenh) 1989; 120: