Shift in Dominance from Blocking to Stimulating Type of Thyrotropin Receptor Antibodies, Resulting in Conversion from Hypothyroidism to Hyperthyroidism during Late Pregnancy

Abstract

A 20-year-old woman with a 10-month history of treatment for Graves' disease (GD), developed hypothyroidism with a high level of thyrotropin (TSH) receptor-blocking antibodies (TBAbs). She conceived at 28 years old and was clinically euthyroid in the first and second trimester, while taking L-thyroxine. However, at 28 weeks she became hyperthyroid with an unexpected rise in TSH receptor-stimulating antibody (TSAb) levels. She was diagnosed with GD, and methimazole was initiated. Her thyroid function normalized, but the neonate became hyperthyroid. We herein report the first case of a shift in dominance from TBAbs to TSAbs in late pregnancy.

Introduction

Graves' disease (GD) is an autoimmune disease involving the thyroid and, less frequently, the orbit and skin. Hyperthyroidism is the most common manifestation of GD and is caused by the stimulating effect of thyrotropin (TSH) receptor antibodies (TRAbs) (1,2). TRAbs have in at least three varieties: TSH receptor-stimulating antibody (TSAb), TSH receptor blocking antibody (TBAb), and neutral antibody. TSAbs mimics the biological activity of TSH and stimulates the cyclic AMP pathway and thyroid hormone synthesis. In some patients, TBAbs bind to the receptor but does not activate the cyclic AMP pathway and acts as a TSH receptor (TSHR) antagonist.

To our knowledge, this is the first report of a case in which hypothyroidism due to TBAb dominance over TSAbs (called TBAb-dominant hypothyroidism) shifted to hyperthyroidism due to TSAb dominance over TBAbs (called TSAb-dominant hyperthyroidism) during gestation, with coexistence of both stimulatory and blocking types of antibodies.

Case Report

A 20-year-old woman with a 10-month history of treatment with propylthiouracil (PTU) for GD was referred to our thyroid clinic for investigation of hypothyroidism with atypical features in January 2007. At the first visit to our clinic, she was taking PTU at 50 mg three times daily and complained of general malaise and somnolence. On an examination, abnormal findings comprised mild struma without vascular bruits and bradycardia (50 beats/min). Thyroid ultrasonography (US) showed normal thyroid echogenicity and regular internal texture of the thyroid. The thyroid weight on US was 28 g (reference, 10 to 25 g). On a chemiluminescent immunoassay (3,4), thyroperoxidase antibody (TPOAb) (267.8 U/mL; reference range, ＜50 U/mL) and thyroglobulin antibody (TgAb) (500 U/mL; reference range, ＜40 U/mL) were positive. Her personal and family histories were unremarkable for thyroid disease. Thyroid function data before therapy were unavailable.

According to a primary physician referral, in May 2006 she became euthyroid with elevated TSH-binding inhibitory immunoglobulin (TBII) on a first-generation assay, of 45.1% (reference range, ＜15%), after 2 months of PTU at 100 mg three times daily. The dose was thus lowered to 100 mg twice daily. However, as her GD worsened in July 2006 with triiodothyronine (FT3) 6.92 pg/mL (reference range, 2.00-4.40), free thyronine (FT4) 2.42 ng/dL (reference range, 0.80-1.90) and TSH ＜0.01 μU/mL (reference range, 0.45-4.50), the dose was increased again to 100 mg three times daily.

Assays

FT4, FT3, and TSH values were measured by an electrochemiluminescence immunoassay (cobas e 411 plus; Roche Diagnostics, Manheim, Germany) (5). TBII levels were measured by a second-generation TRAb assay until September 2017 (6) and subsequently, by an M22-TRAb third-generation assay (7).

Using a porcine thyroid cell cAMP system, TSAb enzyme immunoassay (EIA) was measured with a TSAb kit for EIA bioassay (Yamasa Corporation, Chiba, Japan), according to the manufacture's instruction until July 2022 (5,8). This bioassay specifically stimulated TSAb activation via wild-type receptor expression on cultured porcine thyroid cells. Binding of TSAbs in patient sera to TSH receptors on porcine thyroid cells leads to adenyl cyclase activation, which increases the release of intracellular cAMP into the culture medium. After cell lysis with Triton X100, the total cAMP in porcine thyroid cells and culture medium was measured by a solid-phase EIA (total assay time, 6 hours). The estimated cut-off value of TSAbs is 120%. The intra- and interassay coefficients of variation (CVs) ranged between 3.2-5.9% and 3.7-5.6%, respectively.

Using the same porcine thyroid cell AMP system, TBAb activity was assayed by measuring the ability of patient sera to prevent bovine TSH from stimulating cAMP production in comparison to the control serum response (5,9). The estimated cut-off value of TBAbs is 34%, and the intra- and interassay CVs ranged between 1.2-7.4% and 1.3-5.3%, respectively.

The cAMP levels for TSAbs were measured by a conventional assay for TSAbs (10) until June 2014, namely by an EIA [TSAb (EIA)] (5.8) until July 2022, and subsequently by a cAMP Biosensor [TSAb (Bio)] (GloSensor cAMP, ProMega KK, Madison, USA) (9).

The conventional assay of TSAbs (10) was measured with cAMP production from porcine cells, which was further augmented more by polyethylene glycol (22.5%). The estimated cut-off value is 180%. There was a significant and close correlation (r=0.848, p＜0.001; y=0.840x+115.640) between TSAbs, as estimated by a conventional assay and those estimated by an EIA.

In the TSAb (Bio) assay, binding of cAMP to the luciferase biosensor promotes the oxidation of firefly luciferin, followed by a large increase in oxyluciferin with light output. Relative luminescence units (RLU) were measured using a luminometer, and the TSAb level was calculated for each serum sample (S) (BIOSENSOR TSAB YAMASA, Yamasa, Chiba, Japan), which is expressed as the percentage corrected by normal control (N, TSAb, 100%) and positive control (P, TSAb, 750%), as follows:

TSAb (%)=100+(S-N)/(P-N)×650

The cut-off value for TSAb (Bio) was 110% (sensitivity, 99.1%; specificity, 99.1%). Intra- and interassay CVs were both below 5%. There was a significant correlation (r=0.794, p＜0.001; y=1.120x+193.354) between the results of TSAb (EIA) and TSAb (Bio). and the TSAb (Bio)-negative thyrotoxic patients could exclude the diagnosis of GD.

Positive thyroid autoantibodies estimated by the chemiluminescent immunoassay were defined at a TPOAb concentration ＞50 U/mL and TgAb ＞40 U/mL (3,4).

The vascularity index representing the microvessel density, was estimated using color Doppler US (ARETTA 850; Fujifilm Healthcare, Tokyo, Japan) (12). Subjects laid supine on the examination table with a hyperextended neck supported with a pillow. The device creates a square of approximately 2 cm in the right lobe. US can automatically calculate the ratio of the number of color pixels due to the blood flow signal in the defined region to the total number of pixels in the same region (×100) The estimated cut-off value was 50% (sensitivity, 83.4%; specificity, 95.8%).

The diagnosis, clinical course, and treatment

Taken together, her clinical course supported the diagnosis of GD. Although the PTU dose was lowered, the thyroid function test showed hypothyroidism with TBII at 82.1%, consequently, she was referred to our clinic. Thyroid function tests confirmed the diagnosis of overt hypothyroidism: TSH level: 283.4 μU/mL, FT3: 1.50 pg/mL, and FT4: 0.55 ng/dL. The second-generation TRAb assay showed remarkably elevated TBII levels at 92.7% (reference range, ＜20%).

A presumptive diagnosis of TBAb-induced hypothyroidism was made because of overt hypothyroidism and markedly elevated TBII values. L-thyroxine (LT4) at 50 μg daily was therefore initiated. The antithyroid drug (ATD) dose was gradually titrated and discontinued because TBAb was detected at 92.9% (reference range, ＜34%). In January 2009, her TSAb level on a conventional assay was negative at 147% (reference range, ＜180%), but her TBAb level was elevated at 99.1%. In August 2009, the TSAb level on a conventional assay become positive at 330% and the TBAb level was elevated at 98.4%. Thereafter, a euthyroid state was maintained with LT4 at 100 μg per day, and she continued well until 28 years old, when she became pregnant. The results of her serial thyroid function tests and TRAbs during pregnancy and postpartum are summarized in Table.

Table.

Serial Thyroid Function Tests during the Pregnancy and Postpartum Periods.

Weeks gestation and periods postpartum	TSH	FT4	FT3	M22-TRAb	TBAb	TSAb	TSAb	Medication after testing
	(μU/mL)	(ng/dL)	(pg/mL)	(IU/L)	(%)	(EIA) (%)	(Bio) (%)
5	2.09	1.52	2.33	27.1				LT4 (100 μg/d)
13	0.30	1.51	2.79	26.1	97.2	327	10,693	LT4 (100 μg/d)
18	0.21	1.61	3.03	29.9	95.3	579	14,336	LT4 (75 μg/d)
28	0.01	3.85	7.80	34.2	102.9	521	26,629	Stopped LT4
31	0.01	2.87	7.23	32.7	98.3	541	26,209	Started MMI (15 mg/d)
32	0.01	1.96	4.51	31.6	100.7	651		MMI (alternate 15mg and 10mg/d)
34	0.01	1.16	2.44	29.4	96.3	371	21,887	MMI (10 mg/d)
35	0.01	1.14	2.81	26.9	96.0	531	21,870	MMI (10 mg/d)
2 weeks postpartum	0.01	2.21	4.67	35.2	91.5	406		MMI (10 mg/d)
3 months postpartum	91.84	0.46	1.60	61.7	97.1	402		MMI (5 mg) and started LT4 (50 μg/d)
5 months postpartum	11.73	2.58	1.28	36.2	93.7	369		MMI 5 mg every other day with LT4 (50 μg/d)
8 months postpartum	14.86	1.02	2.17	28.0	94.1	285		Stopped MMI and LT4 (50 μg/d) continued
10 months postpartum	0.12	1.90	3.11					Stopped LT4
1 year postpartum	8.40	1.43	2.48	29.0		777		None
2 yeas postpartum	104.70	0.73	1.95	35.8		881		Start LT4 (75 μg)
4 years postpartum	8.07	1.61	2.27	120.7	99.9	219	13,423	LT4 (87.5 μg)
5 yeas postpartum	1.76	2.09	2.57	37.1				LT4 (100 μg)

TSH: thyrotropin, FT4: free thyroxine, FT3: free triiodothyronine, TBAb: TSH receptorblocking antibodies (reference range <34%), TSAb (EIA): TSH receptor-stimulating antibodies, in which CAMP was measured by an enzyme immunoassay (reference range <120%), TSAb (Bio): TSH receptor-stimulating antibodies in which CAMP was measured with cyclic AMP biosensor (normal range <110%), LT4: levothyroxine given once daily, MMI: methimazole given once daily

In the first and second trimesters, she was clinically and biochemically euthyroid, while taking LT4. Thyroid function test study at 5 and 13 weeks revealed a euthyroid status, as follows: FT4, 1.52 and 1.51 ng/dL (reference range, 0.73-1.95 ng/dL); FT3, 2.33 and 2.79 pg/mL (reference range, 1.90-4.15 pg/mL), and TSH, 2.09 and 0.30 μU/mL (reference range, 0.01-2.36 μU/mL). The tests for M22-TRAb, TBAb, TSAb (EIA) and TSAb (Bio) levels at 13 weeks were all positive, with values of 26.1 IU/L (reference range, ≤2.0 IU/L), 97.2% (reference range, ＜34%), 327% (reference range, ≤120%), and 10,693% (reference range, ＜110%), respectively (Table). At 18 and 28 weeks, the TSAb (Bio) value rose unexpectedly and gradually to 14,336% and 26,629%, respectively, from 10,693% at 13 weeks. The TSAb (Bio) value mildly declined to 21,887% at 34 weeks and 21,870% at 35 weeks, and the TSAb (EIA) value was also increased to 579% at 18 weeks and peaked the value of 651% at 32 weeks.

At 28 weeks, she became mildly thyrotoxic with easy fatigability, excessive sweating, heat tolerance, increased appetite, and insomnia and thyroid function tests confirmed hyperthyroidism, as follows: FT3, 7.80 pg/mL (reference range, 1.59-3.08 pg/mL), FT4: 3.85 ng/dL (reference range, 0.85-1.30 ng/dL), and TSH: 0.01 μU/mL (reference range, 0.02-2.45 μU/mL). Her vascularity index as measured by color Doppler US was elevated at 99.5% (reference, ＜50%), and Doppler flow images showing increased blood flow, were called a thyroid inferno, which was compatible with the diagnosis of GD (Figure). LT4 was discontinued. In subsequent follow-up, thyroid function tests at 31 weeks, after 3 weeks of cessation of LT4, were indicative of hyperthyroidism, as follows: FT3, 7.23 pg/mL, FT4, 2.87 ng/dL, and TSH, ＜0.01 μU/mL.

Figure.

Thyroid vascularity at 28 weeks gestation. Thyroid vascularity was estimated by color Doppler sonography. The device creates a square of approximately 2 cm in the right lobe. Sonography can automatically calculates the ratio of the number of color pixels due to blood flow signal in the defined region to the total number of pixels in the same defined region (100×) (9). The estimated cut-off value was 50% (sensitivity, 83.4%; specificity, 95.8%).

High TSAb (EIA) and TSAb (Bio) levels persisted at 541% and 26,209% at 31 weeks, respectively, as did high levels of TBAb and M22-TRAb. The above data were consistent with the diagnosis of GD, and therapy with methimazole (MMI), 15 mg daily was initiated. FT4 and FT3 levels normalized with suppressed TSH levels and 10 mg of MMI was subsequently continued throughout gestation and after delivery.

A female infant was born after 35 weeks gestation by C-section with a birthweight of 2,882 g. Umbilical cord blood tests revealed subclinical hyperthyroidism (FT4: 1.72 ng/dL, FT3: 1.33 pg/mL and TSH: ＜0.01 μU/mL) with an M22-TRAb level of 19.0 IU/L and a TSAb (EIA) level of 454%. But repeat blood testing demonstrated the evolution of neonatal passive-transfer GD (13) by day 2 after birth (FT4: 4.26 ng/dL, FT3: 6.82 pg/mL, TSH: ＜0.01 μU/mL, M22-TRAb: 17.3 IU/L and TSAb (EIA): 410%).

Thyroid function test results after delivery are shown in Table. The patient developed overt hypothyroidism at 3 months postpartum during treatment with MMI 10 mg daily. MMI was decreased to 5 mg/day and LT4 50 μg/day was initiated. Two months after hypothyroidism, her thyroid function tests spontaneously improved, as shown in Table. Based on the clinical course, postpartum thyroiditis was retrospectively suspected, although no thyrotoxic phase was demonstrated. At 8 months postpartum, MMI was stopped because of subclinical hypothyroidism with FT3 of 2.17 pg/mL, FT4 of 1.02 ng/dL and TSH of 14.86 μU/mL. At 10 months postpartum, LT4 was stopped because of subclinical hyperthyroidism. At 2 years postpartum, thyroid function tests became overtly hypothyroid with FT3 of 1.95 pg/mL, FT4 of 0.73 μg/dL, and TSH of 104.7 μU/mL. LT4 75 μg daily was initiated. On February 13, she was subclinical hypothyroid under treatment with LT4 at 87.5 μg, and her TBAb (99.9%), TSAb (EIA) (219%) and TSAb (Bio) (13,423%) levels were consistent with TBAb-dominant hypothyroidism. At 5 years postpartum, she was clinically and biochemically euthyroid while taking 100 μg of LT4 daily.

Discussion

In this report, we described a patient, with a 10-month history of treatment with ATDs for GD, who was referred to our clinic for an investigation of overt hypothyroidism. Her test results were positive for TBAbs and negative for TSAbs. Therefore, she was diagnosed with TBAb-induced hypothyroidism. Tamai et al. (14) reported that TSH-blocking antibodies may account for hypothyroidism in approximately one third of patients with GD who were previously treated with ATDs. Similar sporadic cases have been reported previously (15-17).

At 28 years old, the patient became pregnant. In the first and second trimesters, she was clinically and biochemically euthyroid under LT4 100 μg daily. TBAbs, TSAbs (EIA), and TSAbs (Bio) were all positive. Unexpectedly the TSAb (Bio) value gradually increased from 10,693% at 13 weeks gestation to 14,336% at 18 weeks and peaked at 26,629% at 28 weeks. The TSAb (Bio) value then declined to 21,887% at 34 weeks. A similar up and down was also observed in TSAbs (EIA).

At 31 weeks, a diagnosis of Graves' disease was made and MMI 15 mg/day was initiated. Although her thyroid function returned to a euthyroid status at 34 weeks and MMI was continued, she delivered a female infant with neonatal hyperthyroidism.

In summary, the sequential changes in TBAb and TSAb during gestation suggested that unsuspected rise in TSAbs were followed by the onset of GD during gestation, although little is known of its mechanism including effect on TBAbs (18).

To our knowledge, this is the first case report to show a shift from TBAb-dominant hypothyroidism to TSAb-dominant hyperthyroidism in late pregnancy. After delivery, the patient eventually returned to TBAb-dominant hypothyroidism and was treated with her pre-pregnancy LT4 at 100 μg/day.

Our study has some weaknesses, including data of TSAb (Bio) were retrospectively available, submitted to the presented paper, and TBII as well as TSAb which were measured using various different assays which were available at each time, presented to this manuscript because of long history of present illness.

Another possible limitation is the performance of a TBAb assay in sera with high TSAb levels which is reported to provide false-positive results for TBAbs. In our assay system of TBAbs, the TBAb activity in the sera of patients whose TSAb (EIA) value were ＞3,000% cannot be determined due to assay limitations (9).

Tagami et al (18) showed that the amount of each stimulating and blocking antibody played an important role in GD and that for the onset of blocking-type hypothyroidism, blocking activity require high amounts of antibodies compared to stimulating activity. Throughout pregnancy and the postpartum period of alteration in the thyroid function in our case, both TSAb and TBAb activities were present. Another interesting finding of our case report was that the TSAb (EIA) level is ＞500% during hyperthyroid episode and ＜500% during hypothyroidism, while the TSAb (Bio) level is ＞20,000% during hyperthyroid episodes and approximately 10,000% during hypothyroidism. In contrast, whenever tested, high activities of M22-TRAb and TBAb were invariably present, not only during hypothyroid phases, but also during the hyperthyroid episodes. We therefore hypothesized that our patients developed a hyperthyroid episode due to dominance in TSAb activity over high TBAb activities during gestation and rapidly returned to hypothyroidism because of a decrease in TSAb activity and vice versa.

TSAbs causes hyperthyroidism in patients with GD in the presence of a mixture of heterogeneous antibodies, differing in biological characteristics. TBAbs were detected in 18.5% of patients with untreated hyperthyroid GD (19). Interestingly, Diana et al. (20) reported that, of 10 patients positive for both TSAbs and TBAbs, 4 were hypothyroid, 2 were euthyroid, and 4 were hyperthyroid.

In this context, Kraiem et al. (21) reported a patient with the coexistence of both stimulatory and blocking types of antibodies who underwent, over a mere three-year period, three successive cycles of oscillation from hypothyroidism to hyperthyroidism and back to hypothyroidism. These findings suggest the possibility, as pointed by Kasagi et al. (22) that both types of TSH receptor antibodies may coexist in one patient, and the thyroid function may change depending on the alteration in balance between these two types of antibodies.

Of note, Lu et al. (23) described a case of transient Graves' hyperthyroidism during the late second trimester in a patient under treatment for Hashimoto's hypothyroidism. The most important common finding between our report and the previous report by Lu et al. is that spontaneous transformation from hypothyroidism to GD occurred during late pregnancy. Unfortunately, Lu et al. did not report TBAb data.

These cases reported by our group and Lu et al. were completely different from cases of gestational GD, as they occurred in the first three months (24).

The mechanism underlying, the unexpected rise in TSAbs, has been suggested to be responsible for the sudden onset of hyperthyroidism due to a change in the balance between TSAb and TBAb activity which is thought to be connected to neonatal thyrotoxicosis, although further studies will be needed to clarify the mutual effect between TSAb and TBAb.

Payrat et al. (25) showed that TBII levels during pregnancy unexpectedly increased in 6 GD patients (14%) among 42 pregnant women. In the remaining 29 patients (69%), TBII levels decreased and in 7 patients (17%), they remained stable. Among the six mothers whose TBII values increased during pregnancy, the offspring of two had neonatal hyperthyroidism. Similarly, Tamaki et al. (26) also reported that TBII levels increased in 3 (30%) out of 10 active GD patients. These reports suggest that the preexisting TSAbs may have been activated in middle and late pregnancy, thereby resulting in worsening GD and neonatal thyrotoxicosis.

Takasu and Matsushita (27) found that out of 34 TBAb-positive patients on LT4 for hypothyroidism over 10 years, 2 (6%) developed TSAb-positive Graves' hyperthyroidism, suggesting that even patients with chronic hypothyroidism are at a risk of developing GD due to the appearance of TSAb.

GD tends to ameliorate or remit during pregnancy, allowing for the discontinuation or a marked decrease in the dosage of ATD during the second and third trimesters (28), flaring after delivery. The suppression of the maternal immune response to prevent rejection of the fetus (29), may explain why the levels of TBII, TSAb, TBAb (30), and MCTA (31,32) showed a decline during pregnancy. Of note, Anderson et al. (24) reported that the incidence rate of hyperthyroidism during pregnancy was high in the first 3 months of pregnancy [incidence rate ratio vs. the remaining study period: 1.50 (95% CI 1.90-2.06)], very low in the last 3 months of pregnancy [0.26 (0.15-0.44)] and reached the highest level 7-9 months postpartum [3.80 (2.88-5.029)].

Nevertheless, Payrat et al. (25) observed six unexpected cases of increasing TBII values during pregnancy. Similarly, Tamaki et al. (26) reported an increase in TBII during pregnancy in 3 out of 10 GD patients. Relatedly, Lu et al. (23) reported a case of transient Graves' hyperthyroidism during the late second trimester in a patient on LT4 replacement for Hashimoto's hypothyroidism.

In the present report, we also observed for the first time an unexpected rise in TSAb levels during middle and late pregnancy in a patient with coexistence of TBAbs and TSAbs, which seemed to result in conversion from hypothyroidism to hyperthyroidism during late pregnancy. Interestingly, in vitro (18) and in vivo studies (33) reported that thyroid-blocking activity has an inhibitory effect on TSAb activity. Accordingly, a decline in the thyroid-blocking activity may have been invoked in this increase in TSAb level. To clinically solve this puzzle, further studies will be needed to establish the exact method of directly evaluating TBAb titers or intensity without the influence of TSAb activities.

At present why TBAb-dominant hypothyroidism changed to TSAb-dominant hypothyroidism during gestation in our case is unclear.

Conclusion and Summary

To our knowledge, this is the first case report of a shift from TBAb-dominant hypothyroidism to TSAb-dominant hyperthyroidism occurring in late pregnancy.

Of potential clinical importance, we have to take special care of change in thyroid function during gestation and postpartum periods, if both types of TSH receptor antibodies may coexist in one pregnant woman.

The study was approved by the Ethics Committee of the Kamijo Thyroid Clinic and Kamijo Thyroid Research Institute. Written informed consent was obtained from the patient.

The author states that he has no Conflict of Interest (COI).