Introduction
The prevalence of Graves’ and Basedow’s disease (GBD) is 210 persons/million/year, and thyroid orbitopathy (TO), the most common extra-thyroid manifestation of GBD, with an incidence rate of 42.2 persons/million/year [1]. TO in the moderate-to-severe stage, according to the European Group on Graves’ Orbitopathy (EUGOGO) classification, occurs with a frequency of 16.1 persons/million/year (women: 26.7; men: 5.5) or in 4.9% of GBD patients [2, 3]. TO occurs at any age, but most often affects women aged 30-50 years. However, the severity of TO tends to be worse in men and in patients who are first diagnosed over the age of 50 years [4].
Ocular symptoms associated with hyperthyroidism were first described in 1840 [5], the clinical picture of the disease can be ambiguous [7, 8], and treatment – despite increasingly modern methods – is not always satisfactory [9,10], and the pathogenesis of this condition is still not fully understood [11–14].
Among the most ubiquitous factors regulating function in the mammalian body are those belonging to the insulin and insulin-like growth factor 1 (IGF-1) family, including their respective receptors, binding and related proteins, and the extensive signalling pathways that mediate their actions [15]. Unlike insulin, IGF-1 circulates in plasma as complexes bound to specific binding proteins, i.e. insulin-like growth factor-binding protein (IGFBP). The amount of free IGF-1 does not exceed 5%. Six binding proteins have been identified (from IGFBP-1 to IGFBP-6). IGFBP-1 is the main IGF-binding protein in human amniotic fluid, IGFBP-2 in cerebrospinal fluid and seminal fluid, and IGFBP-3 in blood serum. IGFBP-3 is the most important protein responsible for the ability of IGF-1 binding [16]. In hyperthyroidism, normal or elevated levels of this peptide have been demonstrated in the blood, decreasing as a result of thyreostatic therapy or after thyroidectomy [17].
Antibodies showing the ability to interrupt IGF-1 receptor (IGF-1R) signalling were first described in the 1980s. Yamashita et al. reported the action of αIR3, which is a monoclonal antibody that, by blocking IGF-1R activation, inhibited growth hormone synthesis by IGF-1 [19]. Li et al. [19] described a second IGF-1R-blocking antibody, designated 1H7. 1H7 can inhibit basal and IGF-1- and IGF-2-dependent DNA synthesis in NIH 3T3 cells (a fibroblast cell line that was isolated from a mouse NIH/Swiss embryo). These antibodies are prototypes of those that could be used to treat TO and other autoimmune diseases in which IGF-1R plays a pathogenetic role [19].
Tsui et al. [20] reported that IGF-1R and thyroid stimulating hormone receptor TSHR can form a functional complex. They demonstrated colocalisation of TSHR and IGF-1R by confocal microscopy of orbital fibroblast (OF). Blocking IGF-1R activation by 1H7 may block signalling initiated by TSH and GBD-IgG that results in extracellular signal-regulated kinases (ERK) 1 and 2 phosphorylation. Thus, the IGF-1R/TSHR complex may account for some of the signalling thus far attributed to TSHR function alone, and it may represent a mechanism by which IGF-1 can amplify the effects of TSH.
Weightman et al. [21] found that IgG from TO patients can displace IGF-1 binding from the OF surface. IgG prepared from sera from GBD patients significantly inhibited IGF-1 binding to the OF compared to serum IgG from healthy volunteers (p < 0.002).
Pritchard et al. [22] showed that circulating antibodies can activate the protein kinase B/FKBP12-rapamycin-associated protein (also called RAFT1/mTOR)/70-kDa S6 kinase (Akt/FRAP/p70s6K) pathway and through this cascade, can induce chemoattractant, interleukin 16 (IL-16), and RANTES (regulated on activation, normal T-cell expressed and secreted) expression in OF from GBD patients. In another study, Pritchard et al. showed that autoantibodies directed against IGF-1R can be detected in almost all patients with GBD, but only in a few patients without the disease [23].
Both OF surface receptors for TSHR and IGF-1R appear to stimulate hyaluronan (HA) synthesis, but TSHR activation alone is sufficient to increase hyaluronan synthase (HAS) 1 and HAS2 expression and HA production via cyclic adenosine monophosphate (cAMP) and Akt/phosphoinositide 3 kinase (PI3K) signalling [24, 25] (Fig. 1).
The interaction between TSHR and IGF-1R is initiated by TSHR-stimulating monoclonal antibody (TSAb), M22, and purified TO-IgG in the absence of IGF-1R activation; some IGF-1R-blocking antibodies and IGF-1R kinase inhibitors have been effective in suppressing TSHR signalling [26].
IGF-IR is a receptor tyrosine kinase (RTK) that binds IGF-1 and IGF-2 and activates 2 important signalling pathways: MAPK and PI3K/Akt. Treatment with an IGF-1R-blocking monoclonal antibody (1H7) or an Akt inhibitor significantly reduced HA levels. IGF-1 treatment increases phospho Akt (pAkt) expression levels; 1H7 and PI3K blockers decrease PI3K and pAkt protein expression, and Akt inhibitors decrease PI3K, Akt, and pAkt expression [27].
IGF-1R is overexpressed in OF and T and B lymphocytes in TO patients and regulates HA synthesis and adipogenesis in orbit [28]. 1H7 has also been shown to inhibit autophagy and induce apoptosis by inhibiting IGF-1R signalling. This leads to the death of both OF and adipocytes, which may be the main mechanism for the efficacy of teprotumumab (a human monoclonal antibody that blocks IGF-1R) in patients with TO [29].
Krieger et al. showed that simultaneous activation of TSHR and IGF-1R causes rapid, synergistic phosphorylation/activation of extracellular signal-regulated kinases (ERK1 and ERK2) in primary cultures of TO-OF and human thyrocytes, as well as human embryonic kidney (HEK) cells with TSHR overexpression [30].
Nevertheless, how TSHR and IGF-1R interact is not clear. They may interact through the activation of overlapping signalling pathways or physically through direct heterodimerisation. Moreover, beta-arrestin 1(βARR 1) and beta-arrestin 2 (βARR 2) are key factors in the regulation of TSHR-mediated signalling: βARR 2 plays a major role in TSHR desensitisation, and βARR 1 is mainly involved in the activation of TSHR signalling [31, 32].
Synergistic interactions between TSHR and IGF-1R occur rapidly upon ERK phosphorylation, demonstrating that the interaction between TSHR and IGF-1R occurs early in the signalling cascade, proximal to the receptors [30]. βARR 1 is required for stimulation of ERK phosphorylation by TO-IgG, and HA secretion assays have shown that βARR physically provides a scaffold for TSHR and IGF-1R in the protein complex [33].
Also, the process of de novo adipogenesis in TO is enhanced, as shown by increased expression of adipocyte-specific genes leptin, adiponectin, fatty acid synthase, adipocyte fatty acid binding protein (AP2), and peroxisome proliferator activator gamma (PPAR-γ) mRNA. Both TSHR and IGF-1R are involved in adipogenesis; in fact, they share the same intracellular Akt/PI3K signalling. Stimulatory TSHR antibody increases Akt protein and cAMP levels, and enhances adipogenesis through the PI3K signalling cascade [34]. IGF-1 has been shown to mediate the proliferation and differentiation of preadipocytes into adipocytes through binding to IGF-1R and phosphorylation of Src homology domain-containing protein 2 (Shc) and insulin receptor substrate 1 (IRS) and the Akt/PI3K pathway [34–36] (Fig. 1).
The aim of the study was to evaluate the levels of IGF-1RAb, IGF-1, and IGFBP-3 in a group of GBD patients with or without TO.
Material and methods
The study was carried out in patients referred to the Department of Endocrinology and Neuroendocrine Tumours, Department of Pathophysiology and Endocrinology, Silesian Medical University in the years 2021 to 2023 due to GBD with and without TO.
Inclusion criteria for the study group were confirmation of GBD based on laboratory tests (TSH, free thyroxine [FT4], TRAb) and thyroid ultrasound. Ophthalmological examination to assess activity (clinical activity score [CAS] >3 as active TO) and severity of the TO (on the NOSPECS scale of grade 2 to 6 and no lower severity than moderate-to-severe according to the EUGOGO classification), and orbital tissue inflammation confirmed by magnetic resonance imaging (MRI).
Exclusion criteria were other orbital diseases, other autoimmune and inflammatory diseases, liver dysfunction and previous immunosuppressive treatment, and lack of informed consent to participate in the study.
Sixty-seven patients were included in the study, including 47 GBD and 20 control patients. In the GBD group, 31 patients were diagnosed with active TO. GBD and the first onset of active TO were diagnosed no later than 12 months before enrolment to the study. All patients with active TO were treated with thiamazole at the time of the study’s initiation and 7 patients were still in a hyperthyroid state, while 22 were euthyroid and 2 patients were hypothyroid.
The patients were divided into the following groups:
- — group I (n = 47): GBD patients regardless of the presence of TO. In this group 21 subjects had a TRAb level below 10 IU/mL (group IA), and a level above this value was present in 26 subjects (group IB);
- — group II (n = 31): GBD patients with active TO before immunosuppressive treatment. In this group, 21 patients were in the moderate-to-severe (group IIA) and 10 in the sight-threatening (group IIB) stage of TO severity according to EUGOGO classification [9, 37];
- — group III (n = 24): GBD patients with active TO after completion of immunosuppressive treatment. This group was incomplete in size at the start of the study because 4 patients were still on immunosuppressive treatment at the time when biochemical tests were performed, one patient did not report for a follow-up examination after completing treatment, and 2 patients did not complete the immunosuppression cycle (because of side effects of treatment — liver damage with 3-digit value of liver enzyme levels) and were referred to other treatments options (rituximab and radiotherapy);
- — group IV (n = 16): GBD patients without TO.
The control group consisted of 20 healthy subjects of comparable age range without clinical features and hormonal thyroid dysfunction, and a history of autoimmune diseases, liver diseases, and immunosuppressive treatment. In previous studies, it was shown that the incidence of positive IGF-1RAb tests was 14% in TO patients and 11% in healthy subjects [38], so we decided to recruit a minimum of 20 patients per group, which would exceed the above estimate.
Ophthalmologic examination including exophthalmometry (Hertel exophthalmometer, Oculus, Germany), measurement of eyelid width, assessment of ocular motility and diplopia, slit-lamp corneal examination, fundus examination, intraocular pressure, visual acuity test, and assessment of CAS and severity of TO according to NOSPECS and EUGOGO classification [37]. To assess diplopia, a self-modification of the Bahn-Gorman scale (with the point: 0 = no diplopia, 1 = intermittent, 2 = inconstant, and 3 = constant) was used, in which the absence of diplopia was defined as the absence when looking straight ahead and downward, upward more than 30 degrees, and to the right and left more than 45 degrees. The diplopia examination was conducted with Medmont Studio 4.1.0 software using the Binocular Single Vision test (Medmont International Pty Ltd, Vermont, Australia).
A positive response to treatment was assessed as a reduction in CAS scale activity by 2 points, a reduction in Bahn-Gorman diplopia scale by ≥ 1 degree (or the acquisition of single frontal and downward vision), and a reduction in TO severity according to the EUGOGO scale to “mild” (eyelid retraction < 2 mm, mild soft tissue inflammation, exophthalmos < 3 mm above normal, with no or only intermittent diplopia).
The demographic and clinical characteristics of the study group are shown in Tables 1 and 2.
|
|
Grupa I (n = 47) |
Grupa II (n = 31) |
Group III (n = 24) |
Group IV (n = 16) |
|
Age [year] |
42.7 (min 21, max 60) |
45.2 (min 27, max 60) |
44.7 (min 27, max 60) |
38.7 (min 21, max 59) |
|
Sex [women/men] |
31/16 (64.6/35.4%) |
19/12 (59.4/40.6%) |
13/11 (54.2/45.8%) |
12/4 (75/25%) |
|
Smokers |
29 (60.4%) |
21 (65.6%) |
16 (66.7%) |
8 (50%) |
|
TRAb (mean, minimum, maximum) Reference value: < 1.75 IU/mL |
17.52 Min 2.15 Max 97.47 |
21.09 Min 2.15 Max 97.47 |
8.05 Min 1.24 Max 20.5 |
10.62 Min 2.9 Max 31 |
|
TSH (mean, minimum, maximum) Reference value: 0.27–4.2 µIU/mL |
1.49 Min 0.01 Max 5.6 |
1.2 Min 0.01 Max 4.04 |
2.02 Min 0.48 Max 5.4 |
2.05 Min 0.01 Max 5.6 |
|
FT4 (mean, minimum, maximum) Reference value: 0.93–1.7 ng/dl |
1.89 Min 0.39 Max 4.2 |
1.98 Min 0.39 Max 4.2 |
1.51 Min 1.1 Max 2.1 |
1.73 Min 0.8 Max 3.5 |
|
Thyroid status: Hyperthyroid/euthyroid/hypothyroid |
11/33/3 |
7/22/2 |
3/19/2 |
4/11/1 |
|
Parameters |
Group II (n = 31) |
Group III (n = 24) |
|
Exophthalmometry Mean in millimetres |
RE 20.8 LE 20.9 |
RE 18.5 LE 19.1 |
|
Diplopia in the primary eye position (Bahn-Gorman scale) |
Absent: 2 (6.5%) Intermittent: 1 (3.1%) Inconstant: 6 (19.4%) Constant: 22 (70.9%) |
Absent: 6 (25%) Intermittent: 4 (16.7%) Intermittent: 8 (33.3%) Constant: 6 (25%) |
|
CAS (points) |
3.84 ± 1.15 |
1.71 ± 0.85 |
|
EUGOGO classification |
Mild: 0 Moderate-to-severe: 21 Sight threatening: 10 |
Mild: 9 Moderate-to-severe: 15 Sight threatening: 0 |
|
NOSPECS classification |
Grade 2: 1 Grade 3: 0 Grade 4: 18 Grade 5: 2 Grade 6: 10 |
Grade 2: 6 Grade 3: 3 Grade 4: 15 Grade 5: 0 Grade 6: 0 |
Patients with active TO in moderate-to-severe and sight-threatening stages received immunosuppressive treatment in the form of intravenous infusions (i.v.) of methylprednisolone (MP) along with mycophenolate sodium according to the standard of EUGOGO guidelines [9, 37, 39]. Depending on the response, radiotherapy to the orbit at a total dose of 20 Gy was used as second-line treatment in some patients.
IGF-1 and IGFBP-3 levels were determined with the use of chemiluminescence immunoassay (CLIA) methods (reagent kits Cat. No 130205007M and 130298005M, respectively) on a Maglumi 2000 plus analyser (Snibe, China). Both methods are non-competitive, sandwich immunoassays based on monoclonal antibodies. Assays were performed in one analytic run with 2.15% and 1.91% intra-assay imprecisions for IGF-1 and IGFBP-3, respectively. According to the manufacturer of the kit, the reference range for IGF-1 (as 2.5th–97.5th percentiles range for healthy adults over 20 years of age) was 60–350 ng/mL. For IGFBP-3 references the ranges were age-dependent and varied between 2.72 and 7.64 ug/mL for 19–80 years of age.
Autoantibodies against the receptor for IGF-1 (IGF-1RAb) were measured by our own constructed enzyme-linked immunosorbent assay (ELISA) method. ELISA plates (Maxisorp F96, Nunk, Denmark) were coated with IGF-1R recombinant protein (BioTechne, Cat. No 391-GR-050) at 0.5 mg/mL in PBS (4oC, at night), washed with phosphate-buffered saline with Tween® detergent (PBST) (PBS + 0.05% Tween 20), and blocked with SeramunBlock (Seramun GmbH, Germany). For calibration, the IGF-1RAb goat affinity pure polyclonal antibody (BioTechne, Cat No AF-305-NA) was used at a range from 0 to 25 ng/mL.
For the “free” autoantibodies assay, the patient’s serums were 100-fold diluted in 1% bovine serum albumin (Jackson Immunoresearch, USA) in PBST. For the “total” autoantibodies assay, the serums were first diluted in 0.2 M glycine-hydrochloric acid (HCl) buffer (pH 2.5), and after 15 minutes, neutralised with 0.2 M sodium diethylbarbiturate solution with bovine serum album (BSA) and Tween20 (final pH = 7.4, 1:100 dilution of sera, 1% BSA). Diluted samples and standards were pipetted on antigen-coated plates (100 uL/well) and incubated for 60 minutes at 37oC. After the first incubation, plates were washed in PBST, and horseradish peroxidase (HRP) conjugate of protein G conjugate (Merck, Cat No. 18–161, 100 ng/ml in 1% gelatine/PBST) was added (second incubation: 60 minutes, 37oC). TMB substrate (TMB slow, Seramun GmbH, Germany) was added after subsequent washing, and after 20 minutes the reaction was stopped with 0.5 M HCl. Absorbances at 450 nm (reference wavelength 630 nm) were measured in an ELx800 ELISA-plate reader (BioTek, USA). Calculation of results was done using the KCJunior computer program (BioTek, USA) from the 4-parameter calibration curve. Results were expressed as ng antibodies/mL undiluted serum. An assay of anti-IGF-1R antibodies was performed as a single run with 3.08% and 3.82% intra-assay imprecision (for “free” and “total” IGF1-Rab, respectively).
In the statistical analysis, we used the software from TIBCO Software Inc. (2017) and Statistica (data analysis software system), version 13. The normality of the data distribution was checked with the Kolmogorov-Smirnov test, and the homogeneity of variance was checked with Levene’s test. To compare continuous variables in the study groups, Student’s t-test was used for data with a normal distribution, and the Mann-Whitney U-test was used for data with a non-normal distribution. Results are presented as median, upper (Q75), and lower (Q25) quartiles. The distribution of dichotomous variables was compared using Fisher’s exact test. P < 0.05 was taken as statistically significant.
All patients gave their informed written consent to participate in the study, which followed the tenets of the Declaration of Helsinki. The study protocol was approved by the Bioethical Board of the Medical University of Silesia (PCN/CBN/0052/KB1/12/I/22).
Results
The concentrations of IGF-1RAb, IGF-1, and IGFBP-3 in the study groups are shown in Tables 3–5 and Figures 2–4.
|
|
Group I (n = 47) |
Group II (n = 31) |
Group III (n = 24) |
Group IV (n = 16) |
Control (n = 20) |
p ANOVA |
|
IGF-1RAb total [ng/mL] |
||||||
|
Median |
190.48 |
151,11 |
141.05 |
453.48 |
124.29 |
0.007 |
|
Q25 |
105.36 |
94.25 |
89.25 |
204.10 |
108.14 |
|
|
Q75 |
588.61 |
227.90 |
228.54 |
869.94 |
232.48 |
|
|
IGF-1 [ng/mL] |
||||||
|
Median |
132.0 |
130.00 |
119.50 |
139.50 |
169.50 |
0.023 |
|
Q25 |
115.0 |
115.00 |
107.00 |
115.50 |
142.00 |
|
|
Q75 |
157.0 |
177.00 |
151.00 |
153.50 |
238.00 |
|
|
IGFBP-3 [µg/mL] |
||||||
|
Median |
3.58 |
3.32 |
3.31 |
3.73 |
4.29 |
0.009 |
|
Q25 |
2.68 |
2.62 |
2.71 |
3.17 |
3.40 |
|
|
Q75 |
4.03 |
3.96 |
3.62 |
4.10 |
4.67 |
|
|
|
Group I (n = 47) |
Control (n = 20) |
p |
|
IGF-1RAb total [ng/mL]* |
|||
|
Median |
190.48 |
124.29 |
0.175 |
|
Q25 |
105.36 |
108.14 |
|
|
Q75 |
588.61 |
232.48 |
|
|
IGF-1 [ng/mL]* |
|||
|
Median |
132.0 |
169.50 |
0.018 |
|
Q25 |
115.0 |
142.0 |
|
|
Q75 |
157.0 |
238.0 |
|
|
IGFBP-3 [µg/mL]** |
|||
|
Median |
3.58 |
4.29 |
0.016 |
|
Q25 |
2.68 |
3.40 |
|
|
Q75 |
4.03 |
4.67 |
|
|
|
IIA group (n = 21) |
IIB group (n = 10_ |
p |
|
IGF-1RAb total [ng/mL]* |
|||
|
Median |
189.35 |
107.4 |
0.014 |
|
Q25 |
105.36 |
85.37 |
|
|
Q75 |
292.14 |
164.57 |
|
|
IGF-1 [ng/mL]** |
|||
|
Median |
123.0 |
177.50 |
0.009 |
|
Q25 |
113.0 |
130.0 |
|
|
Q75 |
135.0 |
216.0 |
|
|
IGFBP-3 [ug/mL]** |
|||
|
Median |
3.30 |
3.52 |
0.203 |
|
Q25 |
2.56 |
3.04 |
|
|
Q75 |
3.88 |
4.19 |
|
Because there are no established reference values for IGF-1RAb, in our study, we took positive values above Q75 obtained in the control group for consideration.
Including our cut-off value (Q75 — 232.48 ng/mL), positive serum IGF-1RAb was found in 25% of patients in the control group (5 out of 20 patients) and 38.3% (18 out of 47 patients) of patients with GBD. The median (Me), lower (Q25), and upper (Q75) quartiles of IGF-1RAb concentration in the control group were Me: 124.29, Q25: 108.14, and Q75: 232.48 ng/mL, and we found no difference (p = 0.175) compared to GBD patients regardless of the presence of TO (group I) (Tab. 4).
Positive serum IGF-1RAb was found in 22.5% of GBD patients with active TO (group II) (7 out of 31 patients), and Me was 151.11, Q25: 94.25, and Q75: 227.90 ng/mL, with no differences to the control group but with a significantly lower level when compared to group IV (GBD patients without active TO) (Fig. 2).
In the group of patients with active TO in sight-threatening stage (group IIB), were also significantly lower values of IGF-1RAb compared to the group of patients with GBD without the presence of TO (group IV): Me: 107.4, Q25: 85.37, and Q75: 164.57 vs. Me: 453.48, Q25: 204.10, and Q75: 869.94 ng/mL, p = 0.004, Mann-Whitney U-test).
The was also a difference in IGF-1RAb concentration between the group in moderate-to-severe (group IIA) and sight-threatening stage (group IIB) of TO according to EUGOGO classification before starting immunosuppressive treatment (p = 0.014) (Tab. 5).
After the end of immunosuppressive treatment (group III), we showed no significant difference in IGF-1RAb concentrations compared to the control group and the group before the start of immunosuppression (group II). There was a significant difference in IGF-1RAb concentrations compared to group IV (GBD patients without TO symptoms) (p = 0.007) (Tab. 3, Fig. 2).
IGF-1 levels in GBD patients, regardless of the presence of TO (group I) were significantly lower compared to the control group (p = 0.018) (Tab. 4). There was no difference in IGF-1 levels between the control group and GBD patients with active TO before starting immunosuppressive treatment (group II) and group IV (GBD patients without active TO). We showed a reduction in IGF-1 levels after the end of immunosuppressive treatment, significantly compared to values in the control group (p = 0.023) (Tab. 3, Fig. 3).
There was a significant difference in IGF-1 concentration between the group with moderate-to- severe (group IIA) and sight-threatening stage (group IIB) of TO according to EUGOGO classification before starting immunosuppressive treatment (p = 0.009, Student’s t-test) (Tab. 5).
We obtained similar results by examining IGFBP-3 concentrations. We found significantly lower IGFBP-3 concentrations in GBD patients regardless of the presence of TO (group I) compared to the control group (p = 0.016, Student’s t-test) (Tab. 4), and a non-significant reduction in its concentration after treatment with the difference compared to the control group (p = 0.009) (Fig. 4). There was no difference in IGFBP-3 concentrations between group IIA (moderate-to-severe TO) and group IIB (sight-threatening TO) patients (p = 0.203, Student’s t-test, Tab. 5).
Table 6 shows IGF-1RAb, IGF-1, and IGFBP-3 concentrations according to TRAb levels. All patients with GBD (Group I) were divided into 2 subgroups: group IA with TRAB levels below 10 IU/mL and group IB with levels above 10 IU/mL. There were significantly lower IGF-1RAb concentrations in the IB group (p = 0.012, Mann-Whitney U test) with no significant difference in IGF-1 and IGFBP-3 concentrations between the groups.
|
|
Group IA (n = 21) |
Group IB (n = 26) |
p |
|
IGF-1RAb total [ng/mL]* |
|||
|
Median |
439.33 |
146.07 |
0.012 |
|
Q25 |
180.33 |
102.02 |
|
|
Q75 |
692.37 |
227.90 |
|
|
IGF-1 [ng/mL]** |
|||
|
Median |
128.0 |
134.50 |
0.326 |
|
Q25 |
115.0 |
117.0 |
|
|
Q75 |
149.0 |
178.0 |
|
|
IGFBP-3 [ug/mL]** |
|||
|
Median |
3.72 |
3.35 |
0.333 |
|
Q25 |
3.04 |
2.50 |
|
|
Q75 |
3.97 |
4.03 |
|
The beneficial effect of the first-line treatment was sustained in > 85% of patients with reduction or maintenance of the exophthalmos in 20 of 24 patients (83.3%), diplopia reduction in 17 of 24 patients (70.1%), and CAS reduction in 19 of 24 patients (79.2%) (Tab. 2).
It is important to mention that none of the 10 patients with sight-threatening stage of TO required treatment with orbital decompression. After treatment according to the EUGOGO recommended standard for DON and continuation of treatment according to the first-line combination treatment standard (i.v. MP with mycophenolate sodium), a reduction in the mean CAS score from 4.44 to 2.0 points was achieved, and the severity of TO in 4 patients decreased to mild and in 6 patients to moderate-to-severe, according to the EUGOGO classification.
Discussion
The purpose of this study was to evaluate the presence and levels of IGF-1RAb, IGF-1, and IGFBP-3 in control and GBD patients with or without TO, and their impact on the course of TO. In our opinion, this may help in the selection of patients with active TO, who prognose to an unfavourable course of the disease while they are the group that probably prognose the best response to the use of the novel TO therapy in the form of teprotumumab.
IGF-1, IGF-1R, and IGFBP play multiple roles in mammalian tissue development and maintenance [40]. Although incompletely characterised, many aspects of the IGF-1 pathway appear to be abnormal in individuals with autoimmune disease [41].
Hansson et al. [42] showed an increase in IGF-1 concentrations in orbital fat and oculomotor fragments taken during orbital decompression surgery of patients with TO. Krassas et al. [43] found no difference in IGF-1, IGF-2, and IGFBP concentrations in the serum of patients with active TO, who were euthyroid compared to healthy subjects, which is consistent with the results of our study.
The results of studies on the effect of thyreometabolic status on serum IGF-1 and IGFBP concentrations are inconclusive [44–48].
Ramos-Dias et al. [44] showed a significant reduction in serum IGF-1 levels in hyperthyroid patients. Similar results were reported in animal studies by Frystyk et al. [45]. Other researchers found no difference in IGF-1 levels in GBD patients compared to controls [46–48].
Prummel et al. studied the effect of long-term treatment with high doses of prednisone on IGF-1 levels in a group of 10 patients with active TO [49]. The authors found a significant increase in IGF-1 levels after the inclusion of steroid therapy and a return to baseline values after the end of treatment. Other researchers have shown that long-term use of low doses of prednisone (< 0.3 mg/kg/day) did not cause statistically significant changes in serum IGF-1 and IGFBP levels [50].
In our study, we showed significantly higher IGF-1 levels in a group of patients with active sight-threatening TO compared to patients with active moderate-to-severe TO. This high level of IGF-1 coexisted with e significantly reduced IGF-1RAb levels in patients with active sight-threatening TO. IGF-1 in serum inhibits antibody binding to IGF-1R, as both IGF-1 and anti-receptor antibodies recognise the same binding site.
Paik et al. showed that IGF-1 enhances the effects of TSH and TSI on TSHR signalling. The researchers showed that OF from patients with TO showed significantly higher levels of IGF-R compared to controls. IGF-1 application increased the expression of surface TSHR in OF from TO patients, which was not observed in fibroblasts from control subjects [51].
The role of IGF-1RAb in TO patients was first reported by Weightman et al. in 1993 [21]. In 2013, two conflicting studies were published on the prevalence and function of IGF-1RAb in patients with TO. Minich et al. demonstrated the presence of IGF-1RAb in a subgroup of TO patients, defining its biological function to be antagonistic, which is consistent with our results [38]. Varewijk et al. found that TO patients with high TSAb titres were more likely to exhibit IGF-1RAb with stimulatory activity [52].
Marino et al. [55] and Lanzolla et al. [54] examined the presence and serum levels of IGF-1RAb in patients with GBD and TO using a commercial ELISA test. Marino et al. adopted a cutoff value below the 97th percentile for positive assays; the researchers showed positive IGF-1R-Abs more frequently in GBD (25%) than in healthy subjects (0%, p = 0.006). In our study, however, we showed a higher prevalence of IGF-1Rab, especially in the control group (25% of control patients, and 38.3% of GBD). Also, in contrast to our study, the researchers showed no significant difference in the presence of IGF-1RAb in patients with TO compared to patients without TO. Patients with TO showed an inverse correlation between IGF-1RAb levels and TO activity score (CAS), which is consistent with our results. The authors concluded that IGF-1RAb antibodies are antagonistic to the receptor, which we confirmed in our study.
We obtained particularly interesting results comparing the IGF-1RAb and IGF-1 concentrations between the group of patients with sight-threatening versus moderate-to-severe stage of TO and GBD patients without active TO. IGF-1RAb levels were lowest in the sight-threatening group, increased in the group with moderate-to-severe active TO, and were highest in the group of GBD patients without active TO. In contrast, we observed an inverse relationship of IGF-1 concentration, which was highest in the sight-threatening patients and significantly higher than in the moderate-to-severe group. There was no significance of its values between the moderate-to-severe group and the active TO group.
This inverse correlation may indicate a protective role for IGF-1RAb in TO, thus showing similarity to the beneficial effects of an IGF-1R-blocking monoclonal antibody (teprotumumab) [55, 56]. In contrast, Lanzolla et al. showed that IGF-1RAb levels did not correlate with the presence of TO or its severity. The authors noted that IGF-1RAb levels negatively correlate with TSAb titres in patients with TO, which we also demonstrated in our study [54]. These authors similarly suggest a protective role for IGF-1RAb concerning the development of TO, which is consistent with the beneficial effects of teprotumumab.
In another study, Marcus-Samuels et al. [57] showed that downregulation of IGF-1R results in a 6.3–65% reduction in IGF-1-stimulated pAkt, but no effect on TO-Igs (TO patients’ immunoglobulins) stimulation of pAkt, suggesting that TO-Igs contain factors that stimulate pAkt formation, but this factor does not directly activate IGF-1R.
The interaction between IGF-1R and TSHR can be activated by autoimmunity against TSHR, while autoimmunity against IGF-1R does not play a major role [12]. The blocking IGF-1RAb has been shown to have an inhibitory effect on IGF-1R in OF cultures and a beneficial effect in TO patients [55], which may be consistent with our observation of an inverse correlation between the severity of TO and IGF-1RAb levels. Eckstein et al. showed that TRAb correlates with TO, including its inflammatory activity as measured by the CAS scale [58]; conversey, Marino et al. showed no correlation between TRAb and CAS, which is somehow consistent with the observation of a direct correlation between TRAb and IGF-1RAb, and an inverse correlation between CAS and IGF-1RAb [53].
Place et al. [59] found that despite targeting completely different IGF-1 receptor domains, linsitinib (IGF-1 receptor kinase inhibitor) and 1H7 (IGF-1 receptor blocking antibody) have similar pharmacological effects and are effective in inhibiting M22-stimulated HA secretion (monoclonal TSH receptor antibody, ECmed) and lose efficacy as M22 levels increase. These results support IGF-1R antagonism as a possible therapeutic approach for TO because it has efficacy against TSAb stimulation.
IGF-1R, which is closely related to TSHR, appears to act as its “gatekeeper” in the activation of orbital fibroblasts. An increase in IGF-1R and TSHR levels is observed during TO, and preliminary in vitro studies have shown inhibition of TSHR stimulation by IGF-1R blockade [23]. In vitro studies have shown that IGF-1R blockade leads to decreased extracellular matrix and cytokine production in activated OF [60]. Progressive research into understanding the synergistic role of TSHR and IGF-1R in the development of TED has led to the development of teprotumumab, a fully humanised monoclonal antibody directed against IGF-1R [55 ,56].
Conclusions
Autoantibodies against IFG-1R are present in the serum in most of the population, but using the upper quartile of the value obtained in the control group as a cut-off point in our study, they were positive in 25% of control patients and 38.3% of GBD patients, regardless of the presence of TO. It seems that high IGF-1RAb levels may have a protective effect against the onset or severe course of TO, and patients with low IGF-1RAb levels are at risk of severe TO.
Our results suggest that anti-receptor antibodies to IGF-1 are inhibitory antibodies, but, as in the case of anti-receptor antibodies to TSH, the presence of other fractions, i.e. neutral or even stimulatory, cannot be excluded. Measurement of IGF-1RAb in connection with the assessment of IGF-1 levels appears to be a good prognostic marker of the clinical course of TO.
Data availability statement
Data available from the authors.
Ethics statement
All patients expressed their formal written consent to participate in the study, which followed the tenets of the Declaration of Helsinki. The study protocol was approved by the Bioethical Board of the Medical University of Silesia (PCN/CBN/0052/KB1/12/I/22).
Author contributions
M.N.: concept of the paper, collection of bibliography and its analysis, preparation of the text of the article and approval of the final version for publication; T.W.: development of methodology and biochemical analyses; W.N.: collecting and analysis of bibliography, preparation of the text and graphics of the article; B.M.: preparation of the text of the article and approval of the final version for publication; B.K.-K.: preparation of the text of the article and acceptance of the final version for publication; L.S.: preparation of the article text and acceptance of the final version for publication; M.L.-O.: preparation of the article text and acceptance of the final version for publication; D.K.: analysis of bibliography, preparation of the text of the article, acceptance of the final version for publication; J.K.: analysis of the obtained research results, conducting statistical analyses and publication graphics, acceptance of the final version for publication
Funding
This work was supported by the Medical University of Silesia grant PCN-1/112/N/2/0, BNW-1-/069/N/3/0.
Conflict of interest
The authors declare no conflict of interest.
