Subjects and treatment
Treatments with 131-iodine in Skåne County (inhabitants 1 360 000 December 31 2019) are centralized to the department of Oncology in Lund where all patients with hyperthyroidism are referred if the patient´s ordinary clinician decides that radioiodine is the best choice of treatment.
This prospective observational study included GD patients, without GO, admitted to the Department of Oncology for treatment with radioiodine from August 2016 until August 2018 (n = 204) and follow up one year later for development of GO
TRAb, anti-TPO and anti-TG were measured before and 3 months after 131- iodine treatment and the fold changes were calculated.
Patients without GO at radioiodine treatment but with with risk factors for the development of GO were prescribed (by the treating clinician), prophylaxis with prednisolone. These patients (n = 45) received prednisolone 30 mg per day for 1 month and after that the dose was slowly decreased during the next 2 months and stopped after 3 months.
The patients were classified as having GD based on clinical signs, the presence of TRAb, and/or a diffuse uptake on thyroid technetium scintigraphy; all patients performed this investigation. One year later all patients were screened for the development of GO (or not) by an endocrinologist or an ophthalmologist at their home clinic in Skåne. All patients received a questionnaire and DNA swab for self-collection of DNA from the buccal mucosa which 130 out of 204 patients accepted and sent in an envelope to our research laboratory. DNA was amplified and analyzed for the presence of polymorphisms in known risk genes for GD and GO (CTLA4, HLA-DRB, CYR 61)
In total 204 patients not previously treated with radioiodine were included. The following parameters were registered; age, sex, born in Sweden, tobacco smoking, GD and GO duration, and treatment with corticosteroids (Table 1). As a clinical routine 120 Gray was used but in some patients up to 300 Gray was administered if the aim was to decrease the risk of relapse. Methimazole or propylthiouracil with or without L-thyroxine was used in some patients prior to radioiodine and was stopped 2 weeks before administering of 131-iodine. The ALARA (as low as reasonably achievable) principle was used when defining the activity needed to achieve the described doses and the activity (MBq) was estimated by use of the following formula: Dose (D) x Mass (m) / 0.043 x uptake day zero (U0) x effective half- life (Teff). U0 and Teff were calculated from the iodine uptake at 24 h and 7 days. The thyroid mass was calculated from 99mTc-pertechneate scintigraphy.
Table 1
Changes in TRAb after treatment with radioiodine and relation to clinical parameteters.
Fold change TRAb
|
˂1.1
|
≥ 1.1
|
P value*
|
Patients (%)
|
57 (31)
|
125 (69)
|
|
Females (%)
|
45 (79)
|
97 (78)
|
0.84
|
Age males and females, years
|
56 ± 17
|
54 ± 16
|
0.67
|
Age females, years
|
54 ± 17
|
52 ± 16
|
0.49
|
Smokers (%)
|
8 (14)
|
31 (25)
|
0.10
|
Born outside Sweden (%)
|
12 (21)
|
35 (28)
|
0.32
|
Treatment with Corticosteroids (%)
|
41 (73)
|
79 (63)
|
0.23
|
Treatment with ˃120 Gy
|
10 (18)
|
21 (16)
|
0.99
|
Duration of GD, months
|
12 (5–25)
|
8 (3–33)
|
0.48
|
Primary treatment with ATD (%)
|
37 (65)
|
70 (56)
|
0.33
|
Duration of GO, months
|
6 ± 6
|
5 ± 3
|
0.52
|
Treatment with Corticosteroids in GO patients
|
2 (22)
|
4 (17)
|
0.99
|
GO Smokers
|
0
|
2
|
0.99
|
The proportion of absorbed doses used were distributed as follows: 120 Gy (78%), > 120–300 Gy (22%)
Gene polymorphisms analysis and DNA extraction
In total 5 single nucleotide polymorphisms (SNPs) were genotyped in 127 patients (Table 2); Two SNPs, rs1378228 and rs12656618, in CYR61 and two in CTLA4, rs3087243 and rs231775, were chosen for analysis based on previous studies on GD and GO (13) Also, a tag SNP, rs6457617, HLA Class II, DR Beta 1 (HLA-DR-DQQ was analyzed as previously described (16).
Table 2
Characteristics of genotyped patients
|
TRAb (IU/L) ˂ median (15 IU/L)
|
TRAb (IU/L) ≥ median (15 IU/L)
|
N
|
57
|
60
|
Age
|
55 ± 18
|
54 ± 15
|
Sex
Male
Female
|
13 (23)
44 (77)
|
15 (25)
45 (75)
|
Ethnicity
Born in Sweden
Born in Europe
Born outside Europe
|
44 (77)
7 (12)
6 (11)
|
47 (78)
7 (12)
6 (10)
|
Smokers
Non-Smoker
Missing
|
12 (21)
41(72)
4 (7)
|
10 (17)
39 (65)
11 (18)
|
*TRAb IU/L (median 15 IU/L)
|
6 (3–9)
|
31 (24–39)
|
Missing Information on 10 patient |
Values expressed are means (± SD) and presented as n (%) unless otherwise stated |
DNA was collected for genotyping by using buccal swabs, then extracted using QiAamp UCP DNA Micro Kit (Qiagen, Sweden) and amplified using Repli-g Screening Kit (Qiagen, Sweden). SNPs were genotyped by TaqMan Allelic Discrimination Assay using the Quantstudio 7 Flex system (Applied Biosystems by Life Technologies, Sweden).
The minor allele frequency (MAF) for all SNPs was > 0.05. One SNP (rs12756618 in CYR61) failed the Hardy-Weinberg equilibrium and was excluded from the analysis.
The standard statistical analysis approach was used to find association of TRAb < median / TRAb > median and GO / no GO association. A linear regression model was used with smoking and gender as covariates. The data are presented as odds ratios (ORs) with 95% confidence intervals (CIs). The p-values are based on additive models for the genetic variants. All genetic analyses were performed using PLINK version 1.0 (http://pngu.mgh.harvard.edu/~purcell/plink/index.shtml).
Assays for antibodies
TRAb was measured with a competitive Electro Chem Luminiscens Immunoassay (ECLI) according to the manufactures instructions (Roche). The limit of detection was 0.3 IU/L, CV 5 % at 16 IU/L. Cut off for a positive value of TRAb was > 1 kIU/L.
Anti-TPO titer was measured with a competitive sandwich ELISA (Roche) according to the manufactures instructions (detection limit 5 kIU/L, CV 11% at 34 kIU/L. The cutoff for a positive value of anti-TPO was > 34 kIU/L.
Anti-TG was measured with a competitive sandwich ELISA (Roche) according to the manufactures instructions (detection limit 10 kIU/L, CV 10% at 73 kIU/L). The cutoff for a positive value of anti-TG was > 115 kIU/L.
Samples were analyzed in routine clinical laboratory at the Department of Clinical Chemistry in Malmö and Lund.
Statistics
The fold changes of thyroid antibodies were calculated. A change of 1.1 or more was judged as increase and if lower than 1.1 the change was judged as unchanged or decreased. The t-test (continuous variables), chi-square test (categorical variables), and bi-nomial test were used to assess the statistical significance of differences between the groups. Linear regression analysis was used to study the correlations between the parameters fold change of TRAb, anti, TPO, anti-TG. All statistical analyses were carried out using the SPSS 22.0 statistical software (SPSS, Chicago, IL, USA) or Graph Pad prism 8.0 The significance levels were;*p = < 0.05; **p = < 0.01; ***p = < 0.001, ****p = < 0.0001.