2014 Volume 31 Issue 4 Pages 303-309
６例のMarin-Lenhart症候群を経験した。これら６例はGraves病と，過機能性甲状腺結節の合併と見なし得た。この症候群の結節と結節以外の部分は131Iシンチグラムをみても，或いは文献によれば，組織学的にみても，大きな差はなかった。従って，この両部分の病因の少なくとも一部は共通しているはずである。ところが現在の通説によれば，結節部分はTSHと無関係に甲状腺ホルモンを過剰産生する自律性腫瘍であり，結節以外の部分は抗TSH受容体抗体で刺激されている自己免疫疾患である，という説明になる。かつ二種類の甲状腺機能亢進症が，１個の甲状腺の中に隣り合って共存しているという，理解し難い説明になってしまう。これらふたつの学説は，片方或いは両方ともに，誤りである可能性が高い。しかし，これらの両甲状腺部分が，ともにTSH刺激に対して過敏すぎる，或いはTSH受容体が多すぎる，と仮定すれば，すべての現象が矛盾なく説明できる。過機能性甲状腺結節が，TSH刺激に対して過剰反応することは，既に証明されている。Graves病甲状腺腫の中のTSH受容体が，異常に多いことも既に知られている。その多すぎるTSH受容体を中和して，この疾患を自然治癒させる為に，抗TSH受容体が作られると推察される。抗体は抗原に結合すると離れないから，不運にもTSHに似てしまった，この抗体は抗原であるTSH受容体に結合したまま，いつまでも甲状腺を刺激し続けて，甲状腺機能亢進症を悪化させていると推察される。治療はTSH受容体を減らすことである。
Gravesʼ disease occurring in association with functioning thyroid nodules is now known as Marine-Lenhart syndrome［１］.
Six cases of this syndrome were encountered in our clinic. All of them could be regarded as being Gravesʼ disease combined with hyperfunctioning thyroid nodules. A major problem with the syndrome is that there is essentially no major difference in the 131I scintigram or in the microscopic findings between the nodule and the extranodular part, as is mentioned in the literature. Therefore, the mechanism of development of both diseases must be at least partially a single, common one. However, this mechanism is currently explained using either of two widely held theories: the first, stating that the nodule automatically produces an excess of thyroid hormone independently of TSH; and the second, that the extranodular part is stimulated by anti-TSH-receptor antibody. Still these two types of hyperthyroidism are said to coexist in a single thyroid gland. At least one of these contradictory explanations, both of which are widely accepted, must therefore contain errors and misunderstandings. A new hypothesis that can clarify these problems without discrepancy is necessary. The present author believes that both parts are hypersensitive to TSH or have too many TSH receptors. The reasons are given in this article.
Over a period of 5 years (2004～2009), 1,418 patients with Gravesʼ disease satisfying the criteria of the Japanese Thyroid Association were seen and treated in Kurihara Thyroid Clinic. Six of these patients (0.42％) had thyroid nodules that showed hot silhouettes in radioactive iodine scintigrams, indicating the presence of Marine-Lenhart syndrome［１］. Their clinical features were investigated.
The following hormone and antibody levels, with normal ranges indicated in parentheses, were determined using Eclusys kits (Roche Diagnostics GmbH, Penzberg, Germany): TRAb (＜10％), free T3 (fT3) (2.3～4.0 pg/ml), free T4 (fT4) (0.90～1.70 ng/dl). The 24-hour thyroid radioiodine uptake (TU, normal range 10～35％) was determined after a 1-week elimination of iodine-rich seaweed (a popular food in Japan), from the patientsʼ diet. The goiter volume and nodule size were estimated by comparison with goiter models specially prepared by our original method［２］ and by measurement using ultrasound images. Diagnoses of Gravesʼ disease were made following the diagnostic criteria of the Japanese Thyroid Association: thyrotoxic symptoms (tachycardia, weight loss, finger tremor, hyperhidrosis, etc.), diffuse goiter, high fT4, low TSH, positive TRAb and a high radioactive iodine uptake rate with a thyroid scintigram showing a homogenous thick silhouette. The author designated areas of the thyroid that showed dense, moderate, weak iodine uptakes, and no silhouettes in a radioactive iodine scintigram as hot, warm, cool, and cold areas, respectively.
The data of the patients with the syndrome are shown in the Table1. To summarize: there were 5 women and 1 man, ranging in age from 33 to 58 years (average, 48.0 y.o.). Two patients (patients 2 and 6) had exophthalmos. All were hyperthyroid and had low TSH levels. All were positive for TRAb. In the radioactive iodine scintigram, gradation silhouettes from moderately hot to hot were observed in the extranodular parts, but the nodule silhouettes were all hot. When radioiodine was administered to 4 patients (patients 1, 4, 5 and 6), the goiter containing the nodule became smaller in 3 (patients 1, 4 and 5), while, in patient 6, the extranodular part became smaller, and the nodule became cystic 4 years later. All four became euthyroid or hypothyroid. Thiamazole was administered to 3 patients (patients 2, 3, and 5), 2 of whom (patients 2 and 3) became euthyroid. In the remaining patient (patient 5), because the hyperthyroid condition occurred repeatedly, radioiodine was administered, resulting in a hypothyroid state.
Patients.
Charkes［３］ reported 10 cases of Gravesʼ disease, each associated with a functioning thyroid nodule. The nodules of these patients showed hot silhouettes in the 131I scintigrams, and there was also strong accumulation in the extranodular parts. Charkes［３］ considered the disease to be a new entity differing from Plummer disease. He named it Marine-Lenhart syndrome, after the researchers who first described it in 1911［１］. The hyperthyroidism of the syndrome is represented by Gravesʼ disease and a functioning thyroid nodule. It is difficult both to measure accurately how great a part each of these two diseases plays, overall, in hyperthyroidism, and determine whether the nodule is hyperfunctioning or functioning normally. In all six, the silhouettes of the extranodular part of the 131I scintigram showed gradation from warm to hot, whereas those of the nodules were hot in each case, therefore, it is possible that these nodules were themselves hyperfunctioning. In case 6, the silhouettes of the nodule and the extranodular part are both hot. Among the Marine-Lenhart cases reported by 18 authors［３～20］. 6 cases (25％) reported by 5 authors［３,６,９，11，14］ showed this type of “hot in hot” silhouette. In Konnoʼs case［11］, for example, the nodule can not be distinguished from the extranodular part （Fig.1）. On the other hand, six nodules among 19 subtotal thyroidectomized goiters by 8 authors［３～５，10～12，17，19］ reporting similar cases were all histologically identified as follicular adenomas. This indicated that, histologically, there was no qualitative difference between them. In Takiʼs case［19］ for example, both tissues showed similar follicular structures （Fig.2）. The nodule and the extranodular part were pathologically classified as adenoma and hypertrophy, although these finding suggest that there may not have been any great difference in biological nature between them. The mechanism of development of these two tissues must be, at least partially, common to both. However, it is currently explained using two well-known theories: first, that the nodule automatically produces an excess of thyroid hormone independently of TSH; and second, that the extranodular part is stimulated by anti-TSH-receptor antibody. These two types of hyperthyroidism reportedly coexist in a single thyroid gland, but this suggests the interference of some kind of error of judgement or of a misunderstanding. A new hypothesis which can explain these problems without discrepancy is therefore necessary. The present author believes that, both a hyperfunctioning thyroid nodule and a Gravesʼ thyroid may be hypersensitive to TSH or have too many TSH receptors. The reasons are as follows.
A 131I thyroid scintigram (left) of a Marine-Lenhart syndrome patient, and a corresponding diagram (right). The diagram represents the 2×3cm functioning nodule that was present in the left lobe. However, in the scintigram, it cannot be distinguished from the extranodular part because both parts have equally thick silhouettes. Reproduced with permission from Konno［11］.
Histological findings in a subtotal thyroidectomized goitor of Takiʼs Marine-Lenhart syndrome patient［19］. The left upper region shows Gravesʼ disease, while the right lower region shows Plummer disease. Both regions show similar follicular structures. A fibrous capsule separates the two tissues. Reproduced with permission from Taki［19］.
One truly TSH-independent tissue is anaplastic thyroid cancer. This tissue will almost certainly show no response to TSH, because it probably secretes hardly any thyroid hormone. A report by Schuppert et al［21］ supports this speculation. They found that the tissues of 6 resected anaplastic thyroid cancers accounted for only 7％ of the amount of TSH receptor mRNA in healthy thyroid tissue （Fig.3a）. However, a hyperfunctioning thyroid nodule produces a large amount of thyroid hormone, whereas anaplastic thyroid cancer tissue produces very little. Therefore, it is logical to conclude that a hyperfunctioning thyroid nodule must be hypersensitive to TSH stimulation or must have an excess of TSH receptors. Two reports support this idea. Burke et al［22］ found that 2 removed hyperfunctioning nodules of Plummer disease showed 2.4 and 6 times the response of the surrounding healthy tissue to in vitro TSH stimulation, as indicated by the increase measured in c-AMP production. Those authors stated that this disassociation of in vivo and in vitro “autonomy” may play a role in the genesis of an autonomous hyperfunctioning thyroid nodule. Larsen et al［23］, on the other hand, found that 2 resected functioning thyroid adenomata responded normally or excessively to in vitro stimulation of TSH, as measured by c-AMP production and iodine organization, but they stated that this phenomenon will require further study. The present author believes that hyperfunctioning thyroid nodules are hypersensitive to TSH stimulation or have too many TSH receptors. If this is so, nodules will become hyperthyroid, and later, the TSH level will fall as a result of a negative feedback mechanism. In other words, the pituitary causes a reduction of the high level of thyroid hormone by secreting less TSH. A hyperfunctioning thyroid nodule demonstrates that any goiter hypersensitive to TSH or having too many TSH receptors can be hyperthyroid. This also suggests that anti-TSH-receptor antibody is not always necessary for the development of hyperthyroidism. The novel concept that Gravesʼ disease goiter may also be hypersensitive to TSH or may have an excess of TSH receptors, should also be considered. One report by Schuppert et al［21］ and another by Mizukami［24］ support this hypothesis. Schuppert et al［21］ found an excess of TSH receptor mRNA in 12 resected goiters with Gravesʼ disease, the amount reaching, on average, 2.2 times that of a healthy thyroid （Fig.3b）. These authors considered that no negative regulation of TSH receptors by either TSH or TSAb occurs in Gravesʼ disease. Mizukami et al［24］, using immunohistochemical staining, observed many TSH receptors in the follicular cells of Gravesʼ disease goiters （Fig.4a,b）. This was contrary to those authorsʼ expectations of decreases in the number of TSH receptors due to down-regulation. The cause of Gravesʼ disease is thus suggested to be a TSH-hypersensitive goiter or the presence of too many TSH receptors. This speculation indicates that the thyroid itself may be responsible for hyperthyroidism in Gravesʼ disease. All thyroidologists know that the pituitary is healthy in Gravesʼ disease. Therefore, the speculation is a very straightforward and sensible one. The difference in the silhouette densities of 131I scintigrams between the hyperactive thyroid nodules and the extranodular tissue of Gravesʼ disease may depend on the difference in the number of TSH receptors. Felicio et al［25］ reported that there are a few TSH receptors in the orbit of a healthy human, but that the number is greater in Gravesʼ disease patients. Their report suggests that, in Gravesʼ disease patients, goiter-like hypertrophy may occur similarly in the orbits. Parma et al［26］ found somatic mutations in the TSH receptor gene in hyperfunctioning thyroid adenoma tissue－a finding that suggests that the TSH receptors may play an important role in hyperthyroidism. These findings strengthen the hypothesis that either TSH hypersensitive goiter or an excess of TSH receptors could be the cause of hyperthyroidism.
a）Results of Northern blot analysis for TSH-receptor mRNA, Pax-8, TTF-1, TPO and Tg expression in 45 representative human thyroid glands. Twenty μg of total RNA was loaded per lane. The transcript size was assessed with reference to a 9.5 kb RNA ladder (BRL, Gaithersburg, MD) in an adjacent lane, of which the 7.5 kb, 4.4 kb, 2.4 kb and 1.5 kb fragments are depicted. Slight differences in the amount of RNA per lane were normalized with ribosomal 28S RNA. To allow detection of weaker bands in anaplastic thyroid carcinomas, blots used for this Figure were exposed for a longer time than those used for Figure 3b. Fig.3a demonstrates TSH-receptor mRNA in 6 cases of anaplastic thyroid carcinoma and two healthy thyroids. In 5 of the 6 cases, the TSH-receptor mRNA was drastically reduced to 7％ on average (p=0.0033) in comparison with the TSH-receptor mRNA expression level in the two healthy thyroids, which was arbitrarily defined as 1.
b）Blots of TSH-receptor mRNA from Gravesʼ disease thyroids and a healthy thyroid. All 12 thyroids affected by Gravesʼ disease displayed increased TSH-receptor mRNA levels (2.2-fold) on average (p=0.01) in comparison with one healthy thyroid. From Schuppert［21］ by permission, with minor modifications.
a）A TSH receptor stained immunohistochemically using monoclonal antibody against recombinant TSH-receptors. Normal human thyroid. The red-stained reaction product indicates that TSH-receptor protein was present along the basal cell surface of the flattened follicular cells. ×200; inset×1,000.
b）Gravesʼ disease thyroid. A significantly increased amount of red-stained reaction product can be seen along the basal cell surfaces of the cuboidal and columnar follicular cells. Mildly positive staining is observed in their cytoplasms. The more intense staining in the bases of the papillary folds should be noted. In the high-power view (bottom right), the intense staining appears to be zonal rather than linear along the basal membrane. ×100; insets ×400, ×1,000. From Mizukami［24］ by permission.
The present authorʼs impression is that the anti-TSH-receptor antibody value in Gravesʼ disease patients during treatment parallels the thyroid function, but with a certain amount of lag time. Therefore, it would appear that this value is not the forerunner but a by-product of a failed natural healing process in Gravesʼ disease. The immune system of Gravesʼ disease patients may have produced the anti-TSH receptor antibody to neutralize the excess of TSH receptors. The surfaces of the TSH molecule and its receptor conform to each other, the former having a protruding shape, the latter, a complementary indented shape. Similarly, the surfaces of the TSH receptor and the anti-TSH-receptor antibody also conform to each other, and can thus perform their respective functions. Therefore, the anti-TSH-receptor antibody, which is produced by bone marrow and lymphoid tissue may possess TSH-like thyroid-stimulating properties. As a rule, once an antibody has coupled with its antigen, they never separate. The antibody may therefore stimulate the thyroid for an extended period, resulting in exacerbation of the hyperthyroidism. This speculation is compatible with the description of the mode of action of anti-TSH-receptor antibody when it was first discovered, as a “long-acting thyroid stimulator (LATS)”.
Most of the TSH-receptors of hyperthyroid Gravesʼ disease goiter may be bound to an excess of anti-TSH-receptor antibodies. Consequently, few free TSH receptors will remain. The goiter may not respond to exogenous TSH stimulation and appears to be TSH-resistant. Thus, Werner et alʼs observation in 1952［27］ that TSH failed to increase the already high thyroidal uptake of 131I in patients with hyperthyroid Gravesʼ disease is understandable. Also, the observation of Kendal et al［28］ that toxic glands with Gravesʼ disease were stimulated by both TSH and LATS, although they were less responsive than non-toxic goiters, may be accepted.
An appropriate treatment of Gravesʼ disease may involve reduction of the excess of TSH receptors.
Analysis of the mechanism of development of Marine-Lenhart syndrome, a condition which in some cases is regarded as Gravesʼ disease together with a hyperfunctioning thyroid nodule, as well as reference to the literature, leads to the speculation that both thyroids may be hypersensitive to TSH. Anti-TSH receptor antibody could be a by-product of a failed natural healing process of Gravesʼ disease. The terms “autonomous thyroid nodule” and “autonomy” may be inappropriate, at least in this context. The appropriate treatment of Gravesʼ disease may be to decrease the number of TSH receptors.
The author is grateful to Mr. C. W. P. Reynolds, associated with the Department of International Medical Communications, Tokyo Medical University, for his careful assistance with the English of this manuscript.
The author also expresses sincere thanks to Dr. Kurihara (his supervisor), who suggested and encouraged this investigation of Marine-Lenhart syndrome.
The author also thanks Dr. Atsuko Iguchi for her valuable advice.
The manuscript was presented as an abstract at the 2011 Japan Endocrine Congress, and the 2011 Japan Thyroid Congress.
The author has no commercial associations that might create a conflict of interest in connection with this article.
2014 June 4 Jun Sasaki
