| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Studies |
Divisione di Endocrinologia, University of Messina School of Medicine, 98125 Messina, Italy
Address all correspondence and requests for reprints to: Dr. Salvatore Benvenga, Divisione di Endocrinologia, Policlinico Universitario, Padiglione H, via Consolare Valeria, 98125 Messina, Italy.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We found the following. 1) Serum Tg most often peaks 1–3 h after FNAB (61 ± 45% of the baseline level; mean ± SD). 2) Only 7% of the initially TgAb-negative patients converted to positive, and only 12% of those initially positive had an increase in the levels of TgAb. 3) THAb were detected in 0 of 400 patients before FNAB, but were found in 9 of 214 (4.2%) after FNAB. This proportion is 2 orders of magnitude higher than that (149 of 369,000 or 0.04%) found in consecutive patients attending European thyroid clinics. Of the 9 cases, 6 had Hashimoto’s thyroiditis (HT), 2 had euthyroid colloid goiter, and 1 had Hurthle cell carcinoma. In the 5 of 9 cases who were TgAb negative, the post-FNAB increment in Tg was 21–99%, i.e. lower than that of the majority of patients (101–12,500%). 4) THAb were of the IgM class in all 9 (6 against T3 and 3 against T4), and were accompanied and/or followed up to 3 months after FNAB by IgG-THAb of the same specificity (2 against T3 and 1 against T4) in 3 cases. In a fourth case, IgM-T3 were followed by a long-lasting synthesis of IgG-T3 (i.e. up to 1 yr post-FNAB). All 4 cases with IgG-THAb had HT and remained TgAb positive. 5) In the 2 HT and the 3 non-HT patients with undetectable TgAb, THAb were of the IgM class only. 6) In the HT group, 2 risk factors for the development of post-FNAB THAb appeared to be pre-FNAB TgAb levels below 400 U/mL that did not increase after FNAB and Tg released from a colloid nodule.
We conclude that Tg release from the thyroid is sufficient to elicit THAb synthesis. In patients with autoimmune thyroid disease (HT), this synthesis occurs with a frequency 10-fold higher than that in patients with nonautoimmune thyroid diseases (21% vs. 2%). However, in only a fraction of patients with autoimmune disease, who need to be TgAb positive by a sensitive assay, the primary immune response (IgM) is followed by a secondary one (IgG). As, once present, this secondary response is long lasting in only a minority of our patients, we think that this could contribute to the rarity of naturally occurring THAb.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
From our review of the literature (1), we found that prevalence of THAb in thyroid autoimmune diseases is greater than that in thyroid nonautoimmune diseases. In the former group prevalence ranges from 1–40% (a 40-fold difference that reflects the 55-fold difference in the number of patients studied and the differences in methodologies). In agreement with the said European survey (2), the specificities are the following: about half of the cases are T3 antibodies (T3Ab), and the rest are equally distributed between T4Ab and T3/T4Ab. Only 50% of the THAb-positive patients are TgAb positive.
Immunologically, thyroid hormones are haptens and, as such, are unable to elicit antibody synthesis (1, 4, 5, 6). Tolerance to thyroid hormones may be overcome when thyroid hormones are coupled to carrier protein(s) and in this complexed form are presented to the immune system. The possibility that such a carrier is one or more of the three major physiological plasma transport proteins [albumin, transthyretin (TTR), and T4-binding globulin (TBG)] was considered pure speculation by Premachandra and Blumenthal (5). Their preliminary observations on the absence of antibodies to TBG-T4 complexes received support from our observation (7) that there are no circulating IgM or IgG antibodies against either TTR or TBG. The prevalent opinion is that the carrier is Tg, although conclusive evidence is lacking (1, 3, 4).
At least two reasons are cited against Tg being the carrier: 1) about 50% of THAb-positive patients are TgAb negative; and 2) circulating Tg is devoid of T3 and T4 (1). However, negativity for TgAb can well be due to poor sensitivity of passive hemoagglutination, a technique that was and remains of popular use. We sought to better substantiate the antigenic role of Tg by taking advantage of the leakage of thyroid-stored Tg caused by diagnostic fine needle biopsy of the thyroid (FNAB) (8, 9). Accordingly, we evaluated the appearance of THAb (and TgAb) in serum after FNAB in consecutive patients who were THAb negative before FNAB.
Here we show that FNAB causes 1) the release into the circulation of iodinated, heterologous (and therefore potentially immunogenic) Tg molecules; and 2) the appearance of IgM-THAb in 9 of 214 cases (4.2%) and of IgG-THAb in 4 of 214 cases (T3Ab in 3 and T4Ab in 1), all TgAb positive by sensitive immunoradiometric assays, resulting in a prevalence 48-fold higher than expected (1.9% vs. 0.04%). Only 1 of these 4 IgG-THAb (1 in 214 = 0.5%) appeared to be persistent. The appearance of serum THAb after FNAB in patients who were THAb negative before FNAB rules out any antigenic role of albumin, TTR, or TBG, which are of hepatic, nonthyroid, production.
We conclude that an important factor contributing to the rarity of THAb of the IgG class is the rarity with which an already rare primary immune response (IgM) switches to a long lasting secondary one (IgG). For the latter to occur, an important requisite should be a prolonged vs. occasional or acute leakage of thyroid-stored Tg in an individual prone to synthesize Ab against Tg.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
One of the first patients enrolled gave her consent to have additional amounts of blood drawn before and 3 h after FNAB to perform this study. In these two serum samples heterogeneity of Tg was studied (in the laboratory of Dr. Arthur B. Schneider, Chicago, IL) by equilibrium gradient (isopycnic) centrifugation (10). That laboratory had previously demonstrated that Tg density determined with this technique is directly and linearly proportional to iodine content (10).
Protocol
Patients were enrolled after giving informed consent, and the study protocol was approved by the ethical committee of our university. None of the enrolled patients received thyroid palpation in the week before FNAB, had been thyroidectomized, or were receiving L-T4 therapy. These exclusion criteria were to avoid previous occasions of Tg release into the bloodstream or possible suppression of THAb synthesis by TSH-suppressive therapy. As THAb had been reported in patients with antecedent 131I therapy or external irradiation of the neck (1), such patients were also excluded. Patients had to be THAb negative in the baseline (pre-FNAB) sample. Patients were sampled within 30 min before FNAB and at the following times after FNAB: 60–180 min, 3 and 15 days, and 1, 3, 6, and 12 months. The following parameters were measured in duplicate: Tg, TgAb, and THAb. Other minor parameters that we measured in only a small number of patients were TPOAb and organ- and nonorgan-specific antibodies (see below).
Assays
Tg (immunoradiometric kit from CIS, Gif-sur-Yvette, France) was assayed from 0–15 days, because of its early release and short half-life (8, 9). The normal range is 0–50 ng/mL. Tg was assayed only in sera found to be TgAb negative, to avoid interference by these Ab. TgAb (immunoradiometric kit from CIS) were measured at all time points except 1–3 h and 3 days. Like all commercial kits, our kit measures TgAb of the IgG class. The normal range is to 0–50 U/mL.
Four types of THAb were measured: IgM-T3, IgG-T3, IgM-T4, and IgG-T4. THAb of the IgM class were assayed through the third month after FNAB; THAb of the IgG class were assayed at the same time points as TgAb. THAb of either class were measured by the radioimmunoprecipitation technique described in detail previously (11), using antihuman IgM or antihuman IgG serum (Behringwerke, Mahrburg, Germany) and [125I]T3 or [125I]T4 (Johnson and Johnson, Milan, Italy). In brief, 500 µL serum were incubated with 0.5 µCi [125I]T3 or [125I]T4 for 60 min at 23 C. Twenty microliters of this mixture were then incubated with 150 µL antihuman IgM or antihuman IgG, both prediluted 1:10 with saline containing BSA at a final concentration of 0.5%. After 24-h incubation at 4 C, tubes were centrifuged at 2000 x g for 20 min, and the supernatant was aspirated. Sera were considered THAb positive when precipitated radioactivity (percentage bound over the total) was more than 2 SD from the normal mean. These cut-off points were 3.9% (IgM-T3), 3.4% (IgM-T4), 3.6% (IgG-T3), and 3.9% (IgG-T4). When levels were above normal (positive) or borderline, THAb were reassayed twice.
For any given parameter, all samples from the same individual were
measured in the same assay. Sera had been stored at -20 C before
assay. An aliquot of the pre-FNAB sample was used to measure the four
THAbs to determine whether the patient could be enrolled. The
intraassay coefficients of variation for Tg, TgAb,
IgM-T3, IgM-T4,
IgG-T3, and IgG-T4 were
3.3% (but 6.5% for values 
5 ng/mL), 5.7%, 2.4%, 2.6%, 2.9%, and
2.3%. The corresponding interassay coefficients of variation were
5.2% (but 7.4% for values 5 ng/mL), 7.0%, 3.1%, 3.3%, 3.5%, and
3.5%.
TPOAb were measured at the same time points as TgAb only in those patients who were TPOAb positive (>10 U/mL) before FNAB. TPOAb were assayed with a RIA kit from Sorin (Saluggia, Italy).
To ascertain whether the patients who tested THAb positive could have a general propensity toward humoral autoimmunity, the following autoantibodies were also determined in all of these patients as well as in a number of randomly selected controls (THAb negative) for a total of 50 patients: antinuclear, antimitochondrial, antismooth muscle, antiparietal cell, antiinsulin, and antiislet cell. These autoantibodies were measured at the same time points as TgAb.
Patients
We enrolled 400 consecutive patients admitted to our Day Hospital for FNAB of a scintigraphically cold nodule. However, only 213 (174 women and 39 men) presented regularly at all of the scheduled times (see Protocol above). Of the remaining 187 patients, only 1 developed THAb at 2 of the examinations (see Results). As we have included this Hashimoto’s thyroiditis (HT) woman in our analysis, the total becomes 214. Some of the initially enrolled patients were dropped from the study if they took corticosteroids, even as nasal drops or unguents. Drop-out of these steroid-treated patients occurred at the third or sixth month, and until then patients were THAb negative. Of these 214 patients, 178 (83.2%) had euthyroid nodular goiter, 7 (3.3%) had histologically proven cancer, and 29 (13.6%) had HT. FNAB was repeated in 9 patients because the first puncture resulted in an inadequate sample for cytological analysis.
Because of our interest in THAb of the IgM class, we were careful to avoid interference in the natural course of the primary immune response. Thus, repetition of FNAB and subtotal or total thyroidectomy were performed after a minimum of 6 weeks from the first FNAB. Fifty-three patients with either colloid, but large, goiters or cytologically suspicious goiters were operated on 8–52 weeks after FNAB, some of them after a 6-month trial of TSH-suppressive therapy commenced after the third month of sampling. This therapy was also given to 100 of the remaining 160 patients, again after the third month of sampling. As shown in Results, IgM-THAb were always detected at 15 or 30 days only, whereas IgG-THAb started to become detectable at 1 or 3 months and were always preceded by IgM. Thus, it is extremely unlikely that TSH-suppressive therapy commenced after the third month could have depressed THAb synthesis. All thyroidectomized patients continued to be sampled through 1 yr after FNAB, because of the known release of Tg during excision.
Statistics
Data are given as the mean ± SD. Differences
between means were evaluated by two-tailed Student’s t
test, and differences between proportions were assessed by
2 or Fisher’s exact test as appropriate.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
After FNAB, there was an increase in the density of Tg, as shown
by its shift to the right (Fig. 1
). There
was also an increase in the heterogeneity of Tg, as shown by the
increased width of the peak. These observations are consistent with
FNAB having caused the release of iodinated and heterogeneous molecules
of Tg.
|
Data for the 156 TgAb-negative patients are summarized and
contrasted with those from the pertinent literature in Table 1. Baseline Tg levels above 300 ng/mL
(range, 302-4409) and above 1000 (range, 1015–4409) were measured in
16 and 4 patients, respectively. A significant increase occurred in
three fourths of the patients (115 of 156). In 11 (10%) and 3 (3%)
patients, all with colloid goiter, this increase was more than 500%
and more than 4000%, respectively. Of these 115 patients, 78 (68%)
had the Tg peak at 1–3 h, and 27 (23%) had the Tg peak at 3 days. On
the average, Tg increased by 66% at 1–3 h and by 35% at 3 days and
returned to baseline by the 15th day. In the patients who developed
post-FNAB THAb (see below), post-FNAB Tg increment was modest (+21 to
99%), possibly because not all Tg molecules released were recognized
by the TgAb of our kit. Although no patients with a post-FNAB Tg
increment between 101–12.500% developed THAb, THAb did not occur
either in 35 patients whose post-FNAB Tg increment was between
20–100%. As previously reported (8), we found no significant
correlation between elevation of serum Tg and type of nodule (solid,
cystic, or mixed), size of nodule, volume or color of the aspirate
fluid, or diagnosis (data not shown).
|
Of the 156 patients TgAb negative before FNAB, 145 (93%) remained so, and 11 (7%) converted to positive 1 yr later. Of the 57 patients TgAb positive before FNAB, 7 (12%) had TgAb levels at 12 months post-FNAB that were higher than baseline. Four of these 7 showed a clear time-dependent pattern, e.g. in patient A.S. TgAb levels at 0, 1, 3, 6, and 12 months were 109, 199, 290, 555, and 734 U/mL, respectively.
Of the nine patients who underwent two FNABs, all five who were TgAb negative before FNAB remained so at all samplings. In the remaining four, all with HT, there were final increases only in two (from 305 to 707 and from 1152 to 1804 U/mL).
THAb
THAb were detected in 0 of 400 patients before FNAB, but were
found in 9 of 214 (4.2%; P < 0.001) after FNAB
(Tables 2 and 3). THAb prevalence in non-HT patients
(cases 1–3) was 3 of 185 or 2%, which is lower than the prevalence (6
of 29 or 21%; P < 0.001) in patients with HT (cases
4–9). None of these 9 THAb-positive cases belonged to the subgroups of
patients punctured twice or thyroidectomized (see Materials and
Methods).
|
|
).
IgM-T3 appeared on the 15th (n = 4) or 30th
day (n = 2); IgM-T4 were always detected on
the 30th day. In no case were either IgM-T3 or
IgM-T4 detected on the 90th day after FNAB,
consistent with the early and transient nature of the primary immune
response. Similarly, THAb of the IgG class were always detected not
before the 30th day post-FNAB, consistent with the nature of the
secondary immune response.
In addition to two cases of IgM-T3 (no. 8 and 9)
and one of IgM-T4 (no. 6), there was another case
(no. 4) with THAb of the IgG class. In this subject,
IgG-T3 were detected up to the last sampling
(Table 3).
Relation of THAb with TgAb
IgM-THAb in patients who were TgAb positive before FNAB were
detected with a frequency not significantly higher than that in
patients who were TgAb negative (5 of 58 or 8.6% vs. 5 of
156 or 3.2%; P > 0.10). However, the corresponding
proportions for IgG-THAb were statistically different (4 of 58 or 6.9%
vs. 0 of 156; P < 0.01). All 3 cases of
isolated IgM-T3 (no. 1–3) remained both TgAb and
TPOAb negative as well (Table 2). In contrast, the 2 cases with
isolated IgM-T4 (no. 5 and 7) were TgAb negative
and TPOAb positive. All 4 cases (no. 4, 6, 8, and 9) with IgG-THAb were
TgAb positive HT patients. Three of these 4 were TPOAb positive.
No THAb of either class were detected among the 11 patients with TgAb
levels above 400 U/mL (range, 410–52,600). Nine of these 11 patients
were TPOAb positive (range, 327–16,000 U/mL), 1 was borderline (13
U/mL), and the other was negative. In addition to THAb-positive
patients 4, 6, and 8, there were 11 patients TgAb positive at levels
below 400 U/mL (range, 103–364). In the HT patients, the proportion of
positivity for THAb was greater in the group with TgAb levels below 400
U/mL than in the group with pre-FNAB TgAb levels above 400 U/mL (6 of
18 or 33% vs. 0 of 11; P = 0.039). With
regard to TPOAb, 7 of 10 were clearly positive (range, 20–1,250 U/mL),
2 of 10 were marginally positive (10 and 12 U/mL), and 1 was negative.
Although in these 10 THAb-negative patients TgAb levels always
surpassed the pre-FNAB levels at least once between 1–12 months
post-FNAB, such behavior was never seen in THAb-positive patients 4, 6,
and 8 (Table 3).
In addition to THAb-positive patients 5 and 7, there was one other TgAb negative but TPOAb positive patient. In this patient, TgAb became positive after FNAB and fluctuated between 52–146 U/mL, whereas TPOAb remained positive between 90 U/mL (baseline) and 164 U/mL. By contrast, in both cases 5 and 7, TgAb remained negative, and TPOAb fluctuated but never surpassed the pre-FNAB levels.
Relation of THAb to the cytology of the punctured nodule
Of the 29 HT patients, the punctured nodule was defined as colloid or colloid cystic by FNAB in 16. THAb appeared in 6 of these 16, but in none of the remaining 13 patients in whom the punctured nodule had other cytological aspects (adenoma, hyperplasia, or lymphocytic infiltration). The difference between the 2 proportions (6 of 16 vs. 0 of 13) was statistically significant (P = 0.0169). Of the 4 HT patients subjected to 2 FNABs, none developed THAb, and only 1 had a colloid nodule.
Taken together, the data for cytology and TgAb (see preceding section) indicate that perhaps THAb do occur only in subsets of HT patients, particularly those 1) with pre-FNAB TgAb levels below 400 U/mL that do not increase further thereafter, and 2) with Tg released from a colloid nodule.
Relation of THAb to nonthyroid autoantibodies
All of the 50 studied patients (i.e. the 9 with and the 41 without post-FNAB THAb) tested negative for circulating antinuclear, antimitochondrial, antismooth muscle, antiparietal cell, antiinsulin, and antiislet cell antibodies at all time points. Thus, patients 1–9 did not develop post-FNAB THAb because of general propensity toward humoral autoimmunity.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
50–300K). Not only molecular
mass but also half-life and, most importantly, immunoreactivity toward
a TgAb were different. The researchers concluded that the "different
antigenic reactivity must be caused by changes in structure of the
molecules leading to qualitative or quantitative differences in the
antigenic determinants" (13). The range of increase in serum Tg after FNAB is extremely wide. In 26% of the patients we found no significant increase in Tg, a prevalence matching the 25% found by Catania et al. (9) in a much smaller study group. In the 5 patients who developed THAb and who were TgAb negative, so that Tg could be measured, the extent of Tg increase was modest (21–99%), possibly because not all species of Tg molecules released were recognized by the TgAb of our kit. This explanation is consistent with the findings of altered immunoreactivity of Tg released after thyroidectomy (13). Overall, there were 40 patients in whom post-FNAB Tg increased by 20–100%, but only 5 of 40 (12%) developed IgM-THAb. The 3 cases with IgM-T3 were non-HT patients, whereas the 2 cases with IgM-T4 were HT patients. In our study we found that the ratio between the prevalence of IgM-T3 and that of IgM-T4 is 1.5:1, which contrasts with the 3:1 ratio between the prevalence of IgG-T3 and that of IgG-T4. This means that in terms of primary immune response, THAb to either iodothyronine tend to occur with almost equal frequency, but often a secondary response to T4 is aborted or is transient. Indeed, only 1 (no. 6) of 4 IgM-T4-positive patients converted to IgG-T4 positive. However, the IgG-T4 of this patient disappeared at the samplings beyond the 1st month post-FNAB, whereas in case 4 (both IgM-T3 and IgM-T4 positive), the secondary response was only against T3 and was detected up to the 12th month. In contrast, 3 (no. 4, 8, and 9) of 5 IgM-T3-positive patients became IgG-T3 positive, and these Ab were detected at least through the third month post-FNAB. However, regardless of iodothyronine specificity, the switch to IgG-THAb positivity was always associated with positivity for TgAb (as measured by a sensitive assay). Of the 2 IgM-T4-positive patients who were TgAb positive (no. 4 and 6), only 1 (50%) developed IgG-T4. In contrast, of the 2 IgM-T3 patients who were TgAb positive (no. 4 and 8), both (100%) developed IgG-T3. The greater likelihood with which a primary immune response to T3 switches to a secondary response and, once it has occurred, its tendency to be prolonged can explain at least in part the greater prevalence of IgG-T3 vs. IgG-T4 reported previously (see introduction). On the other hand, the rate of TgAb negativity within IgG-THAb-positive subjects is, in our patients, lower than that reported previously (2 of 6 or 1 of 3 vs. 1 of 2) because we used a sensitive immunoradioassay, not passive hemoagglutination, to measure TgAb.
We think it is reasonable to infer that only persons previously immunized against Tg (and thus TgAb positive) can respond to the acute leakage of Tg (the extent of which is not possible to quantitate because of the interference by TgAb) by mounting a secondary immune response against the Tg hormonogenic domains. This response, however, is transient, with the possible exception of case 4, implying that a robust persistent secondary response requires a repeated continuous leakage of Tg. This reasoning would also explain why none of the 9 patients punctured twice and none of the 53 thyroidectomized patients developed THAb. Acute release of Tg rarely results in a long lasting secondary response (IgG-THAb).
In conclusion, we have shown that the acute release of Tg after FNAB is necessary and sufficient to cause the appearance of circulating THAb. Indeed, the prevalence of THAb of the IgG class among consecutive thyroid patients subjected to FNAB is about 50-fold higher (1.9% vs. 0.04%) than that in consecutive European patients attending thyroid clinics (2). Patients with autoimmune thyroid disease have an overall prevalence of post-FNAB appearance of THAb that is 10 fold-higher than that in patients with nonautoimmune thyroid diseases (21% vs. 2%). At particular risk appear to be HT patients with persistently low levels of circulating TgAb and with Tg released from benign colloid nodules. As to the first risk factor, one possible explanation is that high serum levels of TgAb could buffer the Tg released and, therefore, sequester antigen otherwise available to elicit THAb synthesis. With regard to the second risk factor, the association of THAb with Tg release from colloid nodules, but not from suspicious or malignant nodules, goes along with the significantly greater frequency of THAb detection in benign goiters vs. thyroid cancer (3–11% vs. 0–1%) (1). This would imply that benign, but not malignant, nodules are more likely to store (and release) Tg molecules potentially immunogenic with regard to THAb. TgAb-positive patients are more likely to have a primary immune response (IgM) followed by a secondary response (IgG). When this secondary response occurs, it is not persistent in most cases. We believe that autoimmunization against thyroid hormones requires the release into the circulation of heterogeneous and iodinated molecules of Tg. In vivo this release could be caused by thyroid inflammation or by intranodular degenerative processes. Although we have observed FNAB-elicited THAb (but of the IgM class only) in patients with no underlying autoimmune disease, patients with autoimmune thyroiditis are more likely to develop a sustained production of IgG-THAb. However, this occurs in only a small fraction of such patients. The rarity of a long lasting synthesis of IgG against thyroid hormones observed in our study would explain why such THAb are encountered so rarely in the clinical practice.
|
Acknowledgments |
|---|
This paper is dedicated to Prof. C. M. Barbera, whose premature death has struck us deeply. He contributed actively to our early studies on THAb.
|
Footnotes |
|---|
Received March 12, 1997.
Revised May 15, 1997.
Revised August 28, 1997.
Accepted September 9, 1997.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  S. Benvenga and F. Trimarchi Triggering of thyroid hormone autoantibodies J R Soc Med, January 1, 2003; 96(1): 50 - 50. [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
