Screening of Organ-Specific Autoantibodies in a Large Cohort of Patients with Autoimmune Thyroid Diseases

Abstract

Background:

Autoimmune diseases tend to cluster in the same individual or in families. Four types of autoimmune polyglandular syndromes (APS) have been described based on the combination of endocrine and/or non-endocrine autoimmune diseases. In particular, type-3 APS is defined by the association of an autoimmune thyroid disease (ATD) and other autoimmune diseases and has a multifactorial etiology. The natural history of autoimmune diseases is characterized by three stages: potential, subclinical, and clinical.

Methods:

To determine the prevalence of organ-specific autoantibodies (anti-adrenal, anti-ovary [StCA], anti-pituitary [APA], anti-parietal cells [PCA], anti-tissue transglutaminase [tTGAb], anti-mitochondrial [AMA], anti-glutamic acid decarboxylase [GADA], anti-nicotinic acetylcholine receptor) in patients with ATD and to define the stage of the disease in patients with positive autoantibodies. From January 2016 to November 2018, 1502 patients (1302 female; age 52.7 ± 14.7 [mean ± standard deviation] years, range 18–86 years) with ATD (1285/1502 [85.6%] with chronic autoimmune thyroiditis and 217/1502 [14.4%] with Graves' disease) were prospectively enrolled.

Results:

The most common organ-specific autoantibodies were PCA (6.99%) and GADA (2.83%), while the prevalence of the remaining autoantibodies was ≤1%. All autoimmune diseases, but celiac disease, were predominant at the potential stage. Sex, ATD type, smoking habit, and coexistence of other autoimmune diseases correlated with the susceptibility to develop chronic atrophic gastritis (CAG) or autoimmune diabetes mellitus.

Conclusions:

The association between ATD and CAG was the most common manifestation of type-3 APS, mainly at the potential stage, that could lead to appropriate follow-up for early detection and timely treatment of the disease.

Introduction

Autoimmune diseases tend to cluster in the same individual or in families, possibly due to a common genetic and/or environmental background. In 1980, Neufeld et al. (1) proposed the first classification of the autoimmune polyglandular syndromes (APS), which entailed four clusters of diseases. In particular, type 1 (APS-1) included at least two among chronic candidiasis, hypoparathyroidism, and Addison's disease (AD); type 2 (APS-2) included AD plus an autoimmune thyroid disease (ATD) and/or diabetes mellitus type 1 (T1DM); type 3 (APS-3) included ATD plus another autoimmune disease but not AD; and type 4 (APS-4) included the association of diseases that did not fit into type 1, 2, and 3. APS-1 is a rare and monogenic disease occurring in children, called Autoimmune Poliendocrinopathy-Candidiasis-Ectodermal Distrophy (APECED), and caused by a mutation of the AIRE gene, while the other APS have a multifactorial etiology.

ATDs are the most common autoimmune diseases (7–8% in the general population) and the most frequent diseases in patients with other autoimmune diseases (2); another clinical autoimmune comorbidity is present in about 15% of ATD patients (3), and a further 15% of those with apparently isolated ATD may result positive, if screened, for organ- and non-organ-specific autoantibodies (4); type 3 is therefore the most frequent APS.

The autoimmune diseases may be also classified based on their natural history, which comprises three distinct phases: potential, subclinical, and clinical. The potential phase is characterized by a genetic predisposition and the presence of circulating autoantibodies and/or lymphocytic infiltration of the target organs, without any organ impairment. The subclinical phase is defined by circulating autoantibodies and/or lymphocytic infiltration of the target organs, associated with evidence of subclinical organ impairment, and the clinical phase is revealed by circulating autoantibodies and/or lymphocytic infiltration of the target organs associated with the typical signs and symptoms of the disease.

Most of the previous studies evaluated the prevalence of ATD in nonthyroidal autoimmune diseases (NTAD) rather than the reverse (5 –12), and there is not a consensus for screening for other autoimmune diseases in patients with ATD (13).

Therefore, the aims of the present study were to: (a) determine the prevalence of organ-specific autoimmune diseases in a large population of adult ATD patients by screening for organ-specific autoantibodies and by reviewing the patient medical records, (b) define the stage of the disease (potential, subclinical, or clinical) in patients with positive organ-specific autoantibodies, and (c) identify clinical parameters in patients with ATD that may confer susceptibility to develop other autoimmune diseases.

Patients and Methods

Patients

From January 2016 to November 2018, 1502 patients (1302 female, age 52.7 ± 14.7 [mean ± standard deviation, SD] years, range 18–86 years) with ATD (1285/1502 [85.6%] with chronic autoimmune thyroiditis [CAT] and 217/1502 [14.4%] with Graves' disease [GD]) were prospectively enrolled. The inclusion criteria were as follows: age >18 years and a previous diagnosis of ATD, based on the positivity of anti-thyroperoxidase and/or anti-thyroglobulin antibodies for CAT or anti-thyrotropin receptor antibodies for GD, and the presence of a typical hypoechoic thyroid ultrasound pattern. The exclusion criteria were as follows: age <18 years and the use of drugs affecting the immune system. About 5% of the eligible patients declined to participate in the study. All ATD patients, included in the study, were evaluated for NTAD by careful review of their medical history or assessment of selected autoantibodies. Patients with a previous diagnosis of NTAD, at the time of the enrollment in the study, were screened for all organ-specific autoantibodies except for that correlated with the known NTAD, while patients without a previous diagnosis of NTAD were screened for all selected organ-specific autoantibodies. The study was approved by the local ethical committee, and written informed consent was obtained from all study participants.

Biochemical tests

We screened the patients for the following autoimmune organ-specific diseases by measuring the corresponding autoantibodies: autoimmune diabetes mellitus (anti-glutamic acid decarboxylase antibodies [GADA], anti-tyrosine phosphatase [IA2A] and anti-zinc transporter 8 antibodies [ZnT8Ab]), celiac disease (anti-tissue transglutaminase [tTGAb] and anti-endomysium antibodies [EMA]), AD (anti-adrenal antibodies [ACA]), premature ovarian failure (anti-ovarian antibodies [StCA]), autoimmune hypophysitis (anti-pituitary antibodies [APA]), autoimmune chronic gastritis (anti-gastric parietal cell antibodies [PCA]), primary biliary cholangitis (anti-mitochondrial [AMA] and AMA-M2 subtype antibodies [M2]), and myasthenia gravis (anti-nicotinic acetylcholine receptor antibodies [ARAb]).

GADA were assessed by enzyme immunoassay (ELISA) (RSR Limited, Cardiff, UK), and serum GADA levels >5 U/mL were considered positive. If GADA were positive, IA2A and ZnT8Ab were measured, with ELISA assay (RSR Limited, Cardiff, UK) with a cutoff value of 7.5 and 15.0 U/mL, respectively. Immunoglobulin A (IgA) tTGAb were measured by fluoroenzyme immunoassay (FEIA) (Thermo Fisher Scientific, Inc., Uppsala, Sweden), and serum tTGAb levels >10 U/mL were considered positive. The following antibodies were assessed by indirect immunofluorescence: EMA on sections of monkey esophagus (Euroimmun; Medizinische Labordiagnostika AG, Lübeck, Germany), ACA on monkey adrenal tissue (Euroimmun; Medizinische Labordiagnostika AG), StCA on monkey ovary tissue (Euroimmun; Medizinische Labordiagnostika AG), APA on monkey pituitary tissue (Euroimmun; Medizinische Labordiagnostika AG), and PCA on rat stomach–liver–kidney substrate (Euroimmun; Medizinische Labordiagnostika AG), which also allows detection of AMA; if AMA was positive, M2 subtype were measured with FEIA (Thermo Fisher Scientific, Inc.). ARAb were assessed by radioimmunoassay (IBL International GmbH, Hamburg, Germany); serum ARAb levels >0.4 nmol/L were considered positive. The performance of the assays (14 –19) is reported in Supplementary Table S1.

Disease staging

GADA-positive patients underwent oral glucose tolerance test (OGTT). Diabetes was considered: at clinical stage for blood glucose levels >200 mg/dL, after OGTT; subclinical for values between 140 and 200 mg/dL, after OGTT or if fasting blood glucose was between 100 and 126 mg/dL; and potential for levels <140 mg/dL, after OGTT. tTGAb-positive patients were also screened for EMA, and if positive, they had endoscopy. Celiac disease was considered at clinical stage after histological confirmation and in the presence of symptoms, otherwise it was defined at potential stage.

Basal morning cortisol, ACTH, renin, and aldosterone were assessed in ACA-positive patients, and ACTH stimulation test was performed in patients with morning cortisol <180 ng/mL. In case of increased renin levels and/or reduced aldosterone and/or cortisol values <180 ng/mL after ACTH test, the disease was considered at subclinical stage. If morning cortisol levels were <50 ng/mL, the disease was defined at clinical stage. In contrast, if renin and cortisol levels before and after stimulation with ACTH were normal, the disease was considered at potential stage. In StCA-positive patients, two consecutive measurement of FSH in early follicular phase were performed, and if the levels were >25 μU/mL and oligoamenorrhea was present for at least four months, the disease was considered at clinical stage, otherwise it was defined as potential. In APA-positive patients, basal pituitary hormones were assessed and pituitary magnetic resonance imaging was performed in case of hormonal deficiency, headache, and/or nausea; the disease was considered at clinical stage if one or more hormonal deficiencies were documented, otherwise it was defined at the potential stage. In PCA-positive patients, a complete blood count, iron status, vitamin B12, folic acid, and gastrin levels were evaluated, and those with one or more impaired parameters had endoscopy. In the presence of anemia, vitamin deficiency, and/or gastric atrophy, the disease was considered at clinical stage, while in the remaining cases at potential stage. AMA-positive patients were also screened for M2 subtype, and in case of positivity of AMA-M2, autoimmune liver blot and liver function tests (liver enzymes, serum protein electrophoresis, M immunoglobulin, total cholesterol) were performed; in cases with impairment of one or more parameters, the disease was considered at clinical stage, otherwise it was defined as potential. ARAb-positive patients underwent a neurological assessment, and if normal, the disease was defined at potential stage.

Statistical analysis

Study parameters are reported as mean ± SD or as median + interquartile range according to their normal or nonnormal distribution. Patient groups were compared with the Mann–Whitney test or the t-test, based on the normality distribution evaluated with the Anderson–Darling test and the homogeneity of variances evaluated with the Bartlett test, when variables were quantitative. Chi-square test or Fisher's exact test was performed, accordingly with expected frequencies, when the variables were categorical. The presence of autoimmune comorbidities was correlated with clinical and biochemical parameters by using the Spearman correlation coefficient (r). The following variables were studied by univariate analysis: sex, age, smoking habit, and family history of autoimmune diseases. Variables with a p-value <0.05 from the univariate analysis and some clinical parameters, known to increase the prevalence of APS, were entered in a multivariate analysis (log-binomial stepwise regression based on the Akaike information criterion) to identify the variables with more predictive value and to estimate the relative risk with their 95% confidence interval [CI]. The positive predictive value of the test (Ab positive) was calculated as the percentage of the true APS+ of all Ab+ cases. Statistical analysis was performed using the StatView software for Windows version 5.0.1 (SAS Institute, Cary, NC, USA) and with R version 3.6.2. A p-value of ≤0.05 was considered statistically significant.

Results

Clinical characteristics of the study population

GD patients were younger and mostly female, the number of smokers was higher, and the autoimmune gastrointestinal comorbidities were less common than in those with CAT. Family and personal history of autoimmune diseases, number, and type (but not gastrointestinal) of autoimmune comorbidities were not significantly different between the two groups. Clinical features of the patients are summarized in Table 1.

Table 1.

Clinical Features of Patients

	ATD (N = 1502)	CAT (n = 1285)	GD (n = 217)	p-value; CAT versus GD
Sex (F/M)	1302/200	1134/151	168/49	<0.0001
Age at screening (mean ± SD)	52.7 ± 14.7	53 ± 14.9	49.8 ± 14.9	0.012
Smoking habit
Current/past/no	227/1126/133	166/106/999	61/27/127	<0.0001
Family history of autoimmune diseases
Yes/no	615/870	534/743	81/127	0.43
Personal history of autoimmune diseases	245/1257	215/1070	35/187	0.28
No. of autoimmune comorbidities
None/one/two or more	1257/206/39	1070/183/32	187/23/7	0.3
Type of autoimmune comorbidities
Endocrine	35	31	4	0.6
Gastrointestinal	89	85	4	0.0059
Dermatological	76	60	16	0.09
Rheumatological	67	60	7	0.34
Hematological	5	5	/	0.36
Neurological	11	9	2	0.72

Bold indicates statistical significance.

ATD, autoimmune thyroid disease; CAT, chronic autoimmune thyroiditis; GD, Graves' disease; SD, standard deviation.

Prevalence of organ-specific autoantibodies

ACA, GADA, PCA, tTGAb, and AMA were measured in the whole cohort of patients (N = 1502), while ARAb, APA, and StCA were measured in a subgroup of patients (n = 989; StCA only in females in the fertile age: 18–49 years). Patients with a previous diagnosis of NTAD were excluded from the prevalence calculation of the corresponding autoantibodies. Ten patients were known to be affected by AD (0.66%), 21 (0.14%) by T1DM or latent autoimmune diabetes in adults (LADA), 44 (2.02%) by chronic atrophic gastritis (CAG), 24 (1.66%) by celiac disease, 4 (0.27%) by primary biliary cholangitis, and 1 (0.1%) by myasthenia gravis. PCA were the most common organ-specific autoantibodies (6.99%), followed by GADA (2.83%), while the prevalence of the remaining antibodies was ≤1% (ACA 0.6%, tTGAb 0.47%, AMA 1.13%, ARAb 0.3%, APA 1%, StCA 0.28%).

Staging of the disease in patients with autoantibody positivity

After staging, all autoimmune organ diseases were defined at the potential stage in the majority of cases but celiac disease. Thirty of 42 GADA-positive patients underwent OGTT, and 2 of 30 were newly diagnosed with diabetes mellitus. In 78 of 102 PCA-positive patients, a complete blood count, iron status, vitamin B12, and folic acid were evaluated; 11 of 78 (14.1%) patients had one or more impaired parameter, in particular 4 of 11 showed vitamin B12 deficiency, associated with anemia and macrocytosis in 2 cases, and 7 of 11 anemia with iron deficiency. Thirty-four of 78 (43.6%) patients had hypergastrinemia, and 21 of 34 underwent esophagogastroduodenoscopy (EGD) with corpus and fundus biopsies: gastric atrophy was documented in 19 of 21 (90.5%) patients and in 12 of 21 (57.1%) cases intestinal metaplasia was also present. Four of 19 (21%) patients with gastric atrophy, confirmed at histology, showed also anemia and/or vitamin deficiency. Ten of 17 AMA-positive patients had M2 positivity, and 6 of 10 patients had autoimmune liver blot and liver function test evaluation: in three cases, a diagnosis of primary biliary cholangitis was made. In the tTGAb-positive group, EMA were measured and found to be positive in six of seven cases. Five of six tTGAb-positive and EMA-positive patients had EGD with duodenal biopsy, and celiac disease was confirmed in all cases, and they were also symptomatic. Patients with positive autoantibodies (APA, StCA, and ARAb) and normal hormonal function at screening were followed for at least 2 years, and only in one case, 30 months later, myasthenia gravis was diagnosed after new onset of diplopia.

A summary of the prevalence of the newly diagnosed and total number (including the pre-existing cases) of organ-specific diseases, based on the disease stage, is shown in Tables 2 and 3, respectively.

Table 2.

Prevalence of the Newly Diagnosed Organ-Specific Diseases Based on the Disease Stage

Disease	Stage
Disease	Subclinical/clinical	Potential	Patients with positive Ab not staged
Addison's disease	0	9/1502 (0.6%)	0/9
Autoimmune diabetes mellitus	2/1490 (0.13%)	28/1490 (1.88%)	12/42
Gastric chronic atrophy	26/1502 (1.7%)	76/1502 (5.06%)	0/102
Celiac disease	5/1501 (0.33%)	1/1501 (0.07%)	1/7
Primary biliary cholangitis	3/1498 (0.2%)	10/1498 (0.67%)	4/17
Myasthenia gravis^a	1/989 (0.1%)	2/989 (0.2%)	0/3
Hypophysitis^a	0	10/989 (1.01%)	0/10
Premature ovarian failure^a,b	0	1/352 (0.28%)^b	0/1

Myasthenia gravis, hypophysitis, and premature ovarian failure were evaluated in subgroup of patients (n = 989).

Premature ovarian failure only in females in the fertile age (n = 352).

Ab, antibodies.

Table 3.

Prevalence of the Organ-Specific Diseases (Pre-Existing Plus Newly Diagnosed Cases)

Disease	Prevalence (pre-existing plus newly diagnosed)
Disease	Subclinical/clinical	Total (including potential)
Addison's disease	10/1502 (0.66%)	19/1502 (1.26%)
Autoimmune diabetes mellitus	23/1490 (1.54%)	51/1490 (3.4%)
Gastric chronic atrophy	70/1502 (4.66%)	146/1502 (9.7%)
Celiac disease	29/1501 (1.93%)	30/1501 (1.99%)
Primary biliary cholangitis	7/1498 (0.47%)	17/1498 (1.13%)
Myasthenia gravis	2/989 (0.2%)	4/989 (0.4%)^a
Hypophysitis	0	10/989 (1.01%)^a
Premature ovarian failure	0	1/352 (0.28%)^b

Myasthenia gravis, hypophysitis, and premature ovarian failure were evaluated in subgroup of patients (n = 989).

Premature ovarian failure only in females in the fertile age (n = 352).

Correlation of patient clinical features with autoantibody positivity

We evaluated the correlation of several clinical parameters (sex, age at screening, type of ATD, smoking habit, positive family history for autoimmune diseases, number, and type of autoimmune comorbidities) with the susceptibility to develop the most common autoimmune comorbidities (diabetes mellitus or CAG). Patients with a previous diagnosis of autoimmune diabetes mellitus or atrophic gastritis were excluded from the analysis.

On univariate analysis, Graves' disease and AD correlated with GADA positivity (p = 0.003 and 0.002, respectively), and the correlation was confirmed on multivariate analysis (relative risk [RR] = 2.87 [CI 1.41–5.82] for GD and RR = 10.41 [CI 2.02–53.65] for AD).

Male sex, smoking habit, and the number of autoimmune comorbidities were correlated, on univariate analysis, with PCA positivity (p = 0.003, 0.028, and 0.028, respectively). On multivariate analysis, the correlation with PCA positivity was confirmed for male sex (RR = 1.96 [CI 1.22–3.15]), smoking habit (RR = 1.53 [CI 1.00–2.32]), and personal history positive for more than one autoimmune comorbidity (RR = 3.14 [CI 1.23–8.01]).

Discussion

We determined the prevalence of organ-specific autoantibodies in a large cohort of adult patients, affected by CAT or GD, followed-up, and prospectively enrolled at the same institution. A similar aim was pursued by three previous studies (3,20,21), but in one study (20), both adults and children were included, the number of adult patients was smaller, and it did not include GD patients, while in the other two studies (3,21), the prevalence of coexisting autoimmune diseases was established only by patient recall and cross-checking with records and medications. Conversely, in our study, the presence of other autoimmune comorbidities was determined both by reviewing the patient medical records and by screening the patients for organ-specific diseases by measuring the corresponding autoantibodies.

The prevalence of ACA (22,23) and GADA (24 –28) was consistent with that reported in the literature, while the prevalence of PCA, tTGAb, and APA was lower in our cohort. In previous studies, PCA positivity ranged between 12% and 40% (29 –32) and APA positivity was about 11% (33), while in our study population, the prevalence of PCA and APA was about 7% and 1%, respectively. These discrepant results may be due to the assays used. In our study, PCA positivity was confirmed by a different method (FEIA; data not shown). The prevalence of tTGAb was reported to be 3–4% (34,35), while in our cohort, it was about 0.5%. We speculate that the use of gluten-free products in the general population or undiagnosed IgA deficiency may have affected the tTGAb true prevalence. No data are available about the prevalence of AMA, ARAb, and StCA in large cohort of ATD patients; in our study, the prevalence of these autoantibodies was negligible for ARAb and StCA (both about 0.3%) and slightly higher for AMA (about 1%) but similar to that reported in the female general population (36).

We further determined the stage of the disease in patients with positive organ-specific antibodies and their predictive value.

In our study, AD, T1DM/LADA, chronic autoimmune atrophic gastritis, myasthenia gravis, hypophysitis, and primary biliary cholangitis were mainly at the potential stage as reported in the literature in asymptomatic patients (24,25,27 –30,32,33,37 –41), while celiac disease was clinical in nearly all cases.

In previous studies, about 60% (29,32,42) of ATD patients with PCA positivity had atrophic gastritis. The presence of hypergastrinemia or anemia increased the predictive value for atrophic gastritis up to 80–100% (43). Accordingly, in our study, the predictive value of PCA and hypergastrinemia in combination was about 90%.

The risk of developing T1DM is associated with the number of positive antibodies and a family history (44 –46). Accordingly, in our study, the predictive value for GADA was negligible (∼7%), and in about half of the remaining GADA-positive patients, with a normal glucose tolerance, other anti-pancreas antibodies were negative, while in one patient with a new diagnosis of diabetes, anti-IA2 antibodies were positive.

The predictive value of combined positivity for tTGAb and EMA (100%) was extremely high in agreement with the literature (35).

Previous studies showed a predictive value of ACA ranging from 0% to 90%, while in our study, none of the ACA-positive patients had adrenal insufficiency. These conflicting results may be due to several factors (sex, age, co-existing autoimmune disease, duration of follow-up, antibody levels, and function of the adrenal gland at the baseline) based on which our study patients can be considered at low-to-moderate risk for the development of AD (40).

In our study, none of the APA-positive patients had symptoms of hypophysitis and/or pituitary hormonal deficiencies; however, subclinical hypopituitarism cannot be ruled out since we did not perform dynamic tests. Only few studies documented a relationship between the presence of APA and the onset of anterior pituitary dysfunction (mainly growth hormone [GH] deficiency) in patients with ATD, with a positive predictive value ranging from 18% to 35% (33,37). Whether APA positivity, detected by immunofluorescence, should be considered a diagnostic marker of lymphocytic hypophysitis and a predictor of pituitary failure still remains a controversial issue.

The positive predictive value of AMA in our population is similar to a previous study, in which 469 AMA-positive patients were followed-up for 28 years: 50% of them developed symptoms after 5 years and 95% after 20 years (47).

The positive predictive value of ARAb is undefined in asymptomatic subjects. In a previous study, 50 patients with Graves' ophthalmopathy underwent measurement of ARAb, but none of the four positive patients (8%) developed signs of myasthenia gravis during the median follow-up period of 4.5 years (38), while in our study, one of three ARAb-positive patients developed the disease 30 months after the screening. Given the very low prevalence of ARAb, studies with a larger number of cases are needed to define their predictive value as well as for StCA.

Finally, we tried to identify clinical parameters in patients with ATD that could confer susceptibility to developing other autoimmune diseases. On multivariate analysis, Graves' disease and AD correlated with GADA positivity. This result could be expected since ATD, autoimmune diabetes mellitus, and AD are components of APS-2; however, the prevalence of autoimmune diabetes mellitus in GD is controversial in the literature. Some studies suggested that T1DM was more prevalent in CAT than in GD (48), while in a recent study, the pattern of the associated autoimmune disorders in GD was not significantly different from that observed in CAT (21). In our cohort, the result could have been affected by the higher frequency of males in patients with GD, and autoimmune diabetes mellitus is known to be more prevalent in males (49).

Male sex, smoking habit, and other autoimmune comorbidities were significantly correlated with PCA positivity. PCA were reported to be more prevalent in females (50), conversely to our study, a possible explanation for this discrepancy could be that males are also more commonly smokers than females.

The limitations of our study could be represented by the selection bias of the patients and the duration of the thyroid disease. However, primary care physicians and other specialists, including the diabetologists, usually refer patients to our institute, which is a referral center, and do not manage thyroid diseases on their own. Therefore, we may assume that the selection bias is minimal in our study. About two-third of our patients had a duration of the thyroid disease, defined as the time elapsed from the diagnosis to the screening, lower than 10 years; however, the median age of this group (50.75 years) was similar to that of patients with a duration of disease greater than 10 years (long-group), and by that age, most of the autoimmune diseases are supposed to have developed based on their peak age of incidence. However, in the long-group, there is a risk of underestimating the prevalence of some autoantibodies known to potentially decline or disappear over time; in particular, we cannot rule out to have missed some cases of hypophysitis and atrophic gastritis because of antibodies against APA and PCA converting to negative. However, it should be pointed out that APA disappear only in patients with hypopituitarism (51) and that the fall of PCA levels is supposed to happen in an advanced stage of the disease (31) often when symptoms and signs related to anemia appear and warrant further investigation (e.g., gastroscopy that ultimately leads to the diagnosis of the disease despite of PCA positivity). Another limitation could be represented by the stage of diseases defined only in a subset of patients; we informed the patients and fully discussed the potential risks associated with the positivity of the detected antibody/ies, but not all of them were willing to undergo further testing and/or some results were not available yet at the time of writing the article.

The strengths of our study are the considerable number of patients and the screening of the autoimmune organ-specific diseases performed by both medical record review and measurement of the corresponding autoantibodies, including some (AMA, ARAb, and StCA) that were not previously determined.

In conclusion, PCA are the most prevalent organ-specific autoantibodies in ATD patients and in combination with hypergastrinemia are very accurate in predicting atrophic gastritis, therefore a routine screening is highly suggested especially in smokers and in patients with other autoimmune diseases. GADA confer a risk of developing autoimmune diabetes mellitus, in particular if other anti-pancreas antibodies are positive, rather than predict pancreatic damage, therefore routine screening is not recommended. Same consideration applies to StCA. APA, measured with indirect immunofluorescence, are not able to predict pituitary damage in asymptomatic ATD patients; thus, unless signs or symptoms of hypophysitis are present, routine screening is not recommended. Screening for ACA may be worthwhile in female and young ATD patients with other autoimmune diseases, in particular diabetes mellitus, and measurement of tTGAb is suggested in patients with chronic anemia, diarrhea, or abdominal pain. Since myasthenia gravis and primary biliary cholangitis are rare diseases and the prevalence of ARAb and AMA in ATD patients was very low, routine screening is not recommended, unless an increase of alkaline phosphatase is observed, in this case, it may be useful to measure AMA, which have a high predictive value for primary biliary cholangitis.

Footnotes

Authors' Contributions

T.P. was responsible for study conception and design and carried out clinical data collection and statistical analysis of the results. G.D. carried out clinical data collection and statistical analysis of the results. B.P., S.C., A.T., A.P., and A.S. were in charge of material preparation and performed biochemical analysis. A.C. carried out statistical analysis of the results. R.F. carried out clinical data collection and statistical analysis of the results. P.F. was responsible for study conception and design. C.S. and M.G.C. supervised the whole project. All co-authors have reviewed and approved the article before submission.

Ethical Approval

All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Declaration of Helsinki and its later amendments or comparable ethical standards.

Disclaimer

The article has been submitted solely to this journal and is not published, in press, or submitted elsewhere.

Author Disclosure Statement

No competing financial interests exist.

Funding Information

The study was supported by the Grant Number RF-2011-02350673 by the Italian Ministry of Health.

Supplementary Material

Supplementary Table S1

References

Neufeld

, Maclaren

, Blizzard

. 1980. Autoimmune polyglandular syndromes. Pediatr Ann, 9:154–162.

Betterle

, Zanchetta

. 2003. Update on autoimmune polyendocrine syndromes (APS). Acta Biomed, 74:9–33.

Boelaert

, Newby

, Simmonds

, Holder

, et al. 2010 Prevalence and relative risk of other autoimmune diseases in subjects with autoimmune thyroid disease. Am J Med 123:183 e181–e189.

Betterle

, Garelli

, Coco

, et al. 2014. A rare combination of type 3 autoimmune polyendocrine syndrome (APS-3) or multiple autoimmune syndrome (MAS-3). Auto Immun Highlights, 5:27–31.

Punzi

, Betterle

. 2004. Chronic autoimmune thyroiditis and rheumatic manifestations. Joint Bone Spine, 71:275–283.

Acay

, Ulu

, Ahsen

, et al. 2014. Assessment of thyroid disorders and autoimmunity in patients with rheumatic diseases. Endocr Metab Immune Disord Drug Targets, 14:182–186.

Kalkan Ç, Soykan

. 2016. Polyautoimmunity in autoimmune gastritis. Eur J Intern Med, 31:79–83.

Song

, Yao

, Wang

, et al. 2019. Thyroid disorders in patients with myasthenia gravis: a systematic review and meta-analysis. Autoimmun Rev, 18:102368.

Floreani

, Mangini

, Reig

, et al. 2017. Thyroid dysfunction in primary biliary cholangitis: a comparative study at two European Centers. Am J Gastroenterol, 112:114–119.

10.

Vallianou

, Stratigou

, Koutroumpi

, et al. 2019. Autoimmune thyroiditis in patients with type 1 diabetes mellitus: a long-term follow-up study. Diabetes Metab Syndr, 13:608–611.

11.

Khoury

, Kadah

, Mari

, et al. 2020. Thyroid dysfunction is prevalent in autoimmune hepatitis: a case control study. Isr Med Assoc J, 22:100–103.

12.

Vachiramon

, Harnchoowong

, Onprasert

, et al. 2017. Prevalence of thyroid abnormalities in Thai patients with Vitiligo. Biomed Res Int, 2017:7502935.

13.

Weetman

. 2005. Non-thyroid autoantibodies in autoimmune thyroid disease. Best Pract Res Clin Endocrinol Metab, 19:17–32.

14.

Törn

, Mueller

, Schlosser

, et al.; Participating Laboratories 2008. Diabetes Antibody Standardization Program: evaluation of assays for autoantibodies to glutamic acid decarboxylase and islet antigen-2. Diabetologia 51:846–852.

15.

Husby

, Koletzko

, Korponay-Szabó

, et al. 2020. European Society Paediatric Gastroenterology, Hepatology and Nutrition Guidelines for Diagnosing Coeliac Disease. J Pediatr Gastroenterol Nutr, 70:141–156.

16.

Andreetta

, Rinaldi

, Bartoccioni

, et al. 2017. Diagnostics of myasthenic syndromes: detection of anti-AChR and anti-MuSK antibodies. Neurol Sci, 38:253–257.

17.

Conrad

, Schossler

, Falk

, et al. Autoantigens, autoantibodies, autoimmunity. In: Autoantibodies in Organ Specific Autoimmune Diseases. A Diagnostic Reference. Second edition. PABST Science Publ, Lengerich, Germany, 2017, Vol 8.

18.

De Bellis

, Pane

, Bellastella

, et al.; Italian Autoimmune Hypophysitis Network Study 2011. Detection of antipituitary and antihypothalamus antibodies to investigate the role of pituitary or hypothalamic autoimmunity in patients with selective idiopathic hypopituitarism. Clin Endocrinol (Oxf), 75:361–366.

19.

Bagnasco

, Saverino

, Pupo

, et al. 2018. Estimate of the prevalence of anti-gastric parietal cell autoantibodies in healthy individuals is method dependent. Am J Clin Pathol, 150:285–292.

20.

Ruggeri

, Trimarchi

, Giuffrida

, et al. 2017. Autoimmune comorbidities in Hashimoto's thyroiditis: different patterns of association in adulthood and childhood/adolescence. Eur J Endocrinol, 176:133–141.

21.

Ferrari

, Fallahi

, Ruffilli

, et al. 2019. The association of other autoimmune diseases in patients with Graves' disease (with or without ophthalmopathy): review of the literature and report of a large series. Autoimmun Rev, 18:287–292.

22.

Laureti

, De Bellis

, Muccitelli

, et al. 1998. Levels of adrenocortical autoantibodies correlate with the degree of adrenal dysfunction in subjects with preclinical Addison's disease. J Clin Endocrinol Metab, 83:3507–3511.

23.

Falorni

, Laureti

, Nikoshkov

, et al. 1997. 21-hydroxylase autoantibodies in adult patients with endocrine autoimmune diseases are highly specific for Addison's disease. Belgian Diabetes Registry. Clin Exp Immunol, 107:341–346.

24.

Sallorenzo

, Silva

, Kasamatsu

, et al. 2017. Prevalence of pancreatic autoantibodies in non-diabetic patients with autoimmune thyroid disease and its relation to insulin secretion and glucose tolerance. Arch Endocrinol Metab, 61:361–366.

25.

Moriguchi

, Noso

, Kawabata

, et al. 2011. Clinical and genetic characteristics of patients with autoimmune thyroid disease with anti-islet autoimmunity. Metabolism, 60:761–766.

26.

Jonsdottir

, Jönsson

, Lantz

. 2019. Prevalence of diabetes and presence of autoantibodies against zinc transporter 8 and glutamic decarboxylase at diagnosis and at follow up of Graves' disease. Endocrine, 64:48–54.

27.

Aksoy

, Yurekli

, Yildiz

, et al. 2006. Prevalence of glutamic acid decarboxylase antibody positivity and its association with insulin secretion and sensitivity in autoimmune thyroid disease: a pilot study. Exp Clin Endocrinol Diabetes, 114:412–416.

28.

Kawasaki

, Abiru

, Yano

, et al. 1995. Autoantibodies to glutamic acid decarboxylase in patients with autoimmune thyroid disease: relation to competitive insulin autoantibodies. J Autoimmun, 8:633–643.

29.

Utiyama

SRR

, De Bem

, Skare

, et al. 2018. Anti-parietal cell antibodies in patients with autoimmune thyroid diseases. J Endocrinol Invest, 41:523–529.

30.

Checchi

, Montanaro

, Ciuoli

, et al. 2010. Prevalence of parietal cell antibodies in a large cohort of patients with autoimmune thyroiditis. Thyroid, 20:1385–1389.

31.

Tozzoli

, Kodermaz

, Perosa

, et al. 2010. Autoantibodies to parietal cells as predictors of atrophic body gastritis: a five-year prospective study in patients with autoimmune thyroid diseases. Autoimmunity Rev, 10:80–83.

32.

Castoro

, Le Moli

, Arpi

, et al. 2016. Association of autoimmune thyroid diseases, chronic atrophic gastritis and gastric carcinoid: experience from a single institution. J Endocrinol Invest, 39:779–784.

33.

Manetti

, Lupi

, Morselli

, et al. 2007. Prevalence and functional significance of antipituitary antibodies in patients with autoimmune and non-autoimmune thyroid diseases. J Clin Endocrinol Metab, 92:2176–2181.

34.

Spadaccino

, Basso

, Chiarelli

, et al. 2008. Celiac disease in North Italian patients with autoimmune thyroid diseases. Autoimmunity, 41:116–121.

35.

Volta

, Ravaglia

, Granito

, et al. 2001. Coeliac disease in patients with autoimmune thyroiditis. Digestion, 64:61–65.

36.

Shibata

, Onozuka

, Morizane

, et al. 2004. Prevalence of antimitochondrial antibody in Japanese corporate workers in Kanagawa prefecture. J Gastroenterol, 39:255–259.

37.

Bellastella

, Rotondi

, Pane

, et al. 2010. Predictive role of the immunostaining pattern of immunofluorescence and the titers of antipituitary antibodies at presentation for the occurrence of autoimmune hypopituitarism in patients with autoimmune polyendocrine syndromes over a five-year follow-up. J Clin Endocrinol Metab, 95:3750–3757.

38.

Jacobson

. 1995. Acetylcholine receptor antibodies in patients with Graves' ophthalmopathy. J Neuroophthalmol, 15:166–170.

39.

Zandanell

, Strasser

, Feldman

, et al. 2020. Low rate of new-onset primary biliary cholangitis in a cohort of anti-mitochondrial antibody-positive subjects over six years of follow-up. J Intern Med, 287:395–404.

40.

Betterle

, Coco

, Zanchetta

. 2005. Adrenal cortex autoantibodies in subjects with normal adrenal function. Best Pract Res Clin Endocrinol Metab, 19:85–99.

41.

Coco

, Dal Pra

, Presotto

, et al. 2006. Estimated risk for developing autoimmune Addison's disease in patients with adrenal cortex autoantibodies. J Clin Endocrinol Metab, 91:1637–1645.

42.

Checchi

, Montanaro

, Pasqui

, et al. 2008. L-thyroxine requirement in patients with autoimmune hypothyroidism and parietal cell antibodies. J Clin Endocrinol Metab, 93:465–469.

43.

Centanni

, Marignani

, Gargano

, et al. 1999. Atrophic body gastritis in patients with autoimmune thyroid disease: an underdiagnosed association. Arch Intern Med, 159:1726–1730.

44.

Betterle

, Spadaccino

, Presotto

, et al. 2002. The number of markers of pancreatic autoimmunity is proportional to the risk for type 1 diabetes mellitus in Italian and English patients with organ-specific autoimmune diseases. Ann N Y Acad Sci, 958:276–280.

45.

Bingley

, Christie

, Bonifacio

, et al. 1994. Combined analysis of autoantibodies improves prediction of IDDM in islet cell antibody-positive relatives. Diabetes, 43:1304–1310.

46.

Bosi

, Becker

, Bonifacio

, et al. 1991. Progression to type I diabetes in autoimmune endocrine patients with islet cell antibodies. Diabetes, 40:977–984.

47.

Prince

, Chetwynd

, Craig

, et al. 2004. Asymptomatic primary biliary cirrhosis: clinical features, prognosis, and symptom progression in a large population based cohort. Gut, 53:865–870.

48.

Wiebolt

, Achterbergh

, den Boer

, et al. 2011. Clustering of additional autoimmunity behaves differently in Hashimoto's patients compared with Graves' patients. Eur J Endocrinol, 164:789–794.

49.

Diaz-Valencia

, Bougnères

, Valleron

. 2015. Global epidemiology of type 1 diabetes in young adults and adults: a systematic review. BMC Public Health, 15:255.

50.

Massironi

, Zilli

, Elvevi

, et al. 2019. The changing face of chronic autoimmune atrophic gastritis: an updated comprehensive perspective. Autimmun Rev, 18:215–222.

51.

Kajita

, Yasuda

, Yamakita

, et al. 1991. Anti-pituitary antibodies in patients with hypopituitarism and their families: longitudinal observation. Endocrinol Jpn, 138:121–129.