The authors have declared that no competing interests exist.
Conceived and designed the experiments: CNK FH AN. Performed the experiments: CNK FH. Analyzed the data: CNK. Contributed reagents/materials/analysis tools: CNK. Wrote the paper: CNK FH AN.
Thyroid dysfunction is the commonest endocrinopathy associated with HCV infection due to interferon-based treatment. This comprehensive and systematic review presents the available evidence for newly developed thyroid antibodies and dysfunctions during interferon treatment (both single and combination) in HCV patients.
This systematic review was conducted in accordance with the PRISMA guidelines. The data generated were used to analyze the risk for thyroid dysfunctions during interferon (IFN) treatment in HCV patients. There was a wide range in the incidence of newly developed thyroid dysfunctions and thyroid antibodies in HCV patients during IFN treatment (both single and combination). The wide range of incidence also denoted the possibility of factors other than IFN treatment for thyroid-related abnormalities in HCV patients. These other factors include HCV viral factors, genetic predisposition, environmental factors, and patho-physiological factors. Variations in IFN dosage, treatment duration of IFN, definition/criteria followed in each study for thyroid dysfunction and irregular thyroid function testing during treatment in different studies influence the outcome of the single studies and jeopardise the validity of a pooled risk estimate of side effects of thyroid dysfunction. Importantly, reports differ as to whether the thyroid-related side effects disappear totally after withdrawal of the IFN treatment.
The present review shows that there is a wide range in the incidence of newly developed thyroid dysfunctions and thyroid antibodies in IFN treated HCV patients. This is a comprehensive attempt to collate relevant data from 56 publications across several nations about IFN (both mono and combination therapy) related thyroid dysfunction among HCV patients. The role of each factor in causing thyroid dysfunctions in HCV patients treated with IFN should be analyzed in detail in future studies, for a better understanding of the problem and sounder clinical management of the disease.
As per the World Health Organization (WHO), nearly 3% of the global population suffers from Hepatitis C Virus (HCV) infection, prevalence of the same ranging from 0.1–5% is reported for different European countries
A higher prevalence of thyroid disorders has been reported in HCV-infected patients than in the general population
IFN has important immunomodulatory properties due to which it can induce autoimmune phenomena like autoimmune thyroiditis with hypo - or hyperthyroidism
A common drug used with IFN α in HCV treatment is Ribavirin (RIBA)
Such is the correlation between the therapy and the gland malfunction that clinicians have often reduced the dose or sometimes even discontinued IFN α treatment in patients who develop thyroid dysfunction, thus possibly compromising the therapeutic response
This background, a comprehensive and systematic review presenting the available evidence for the newly developed thyroid antibodies (Tab) and dysfunctions during interferon treatment (both single and combination) in HCV patients was conceived. We have included herein 19591 case studies/patient histories (16149 from mono-therapy and 3442 from combination therapy) from 56 publications (31 mono and 25 combination treatments) to understand the frequency of risk associated with thyroid dysfunctions during IFN treatment (single and combination) among HCV patients.
To the best of our knowledge, this systematic review has included the highest number of case studies and publications to analyze the risk of thyroid dysfunction in patients during both single and combination IFN α treatment compared to earlier studies that were based either on single or combination therapy of IFN α or dealt with limited numbers of patients and publications in earlier narrative and systematic reviews with meta-analysis
A systematic literature search was performed using PubMed, EMBASE and Google. The keywords used were ‘interferon treatment’ combined with ‘thyroid’, ‘hepatitis C’, ‘antibodies’, ‘autoimmunity’, ‘dysfunctions’, ‘pegylated’, ‘meta-analysis’, ‘pathogenesis’, ‘molecular mimicry’, ‘genetic predisposition’, ‘Levovirin’, ‘consensus Interferon’, ‘diagnosis’, ‘management’ and ‘ribavirin’ for the period between January 1990 to November 2012. Identification, screening, eligibility and inclusion of database for the study have been depicted in a flow chart (
The inclusion of publications for the present study was based on the following criteria:
Design of study: Case-control, prospective, retrospective
Availability of data on thyroid disease ie., newly developed thyroid dysfunctions (hyper and hypothyroidism),newly developed thyroid antibody (Tab’s) during IFN treatment
Treatment must include at least one of the following therapy regimes:
IFN α
IFN α+RIBA
IFN in combination with RIBA [IFNα 2b+RIBA]
Consensus Interferon-1 (IFN α Con-1)+RIBA
Pegylated IFN (PEG-IFN) α+RIBA
Pegylated IFN (PEG-IFN) α+Levovirin
Consensus Interferon (CIFN) α
German and English Language articles were screened for the study.
The methodological quality of the literature was assessed as “moderate” or “good”. A study was deemed to be of “moderate” quality if it did not include any of the key words given for search and did not follow the inclusion criteria of publication mentioned above. A study was rated as “good” if publications were relevant to the topic, any two of the above keywords were mentioned in the publication, and it followed the inclusion criteria discussed previously. Only the “good” quality publications were selected for the study.
The study found 168 publications from PUBMED and EMBASE and 114 documents from other sources like Google during the systematic database search. Of these, 56 publications were synthesized on the basis of the inclusion criteria and PRISMA guidelines (
Si No | Country | Treatment | No.(M/F) | Newly developed thyroid antibody (Tab’s)n (%) | Newlydevelopedthyroid dysfunction n (%) | Reference |
1 | USA | IFN α | 237 | NR | 6 (2.5) |
|
2 | France | IFN α | 68 (39/29) | 4 (5.9) | 8 (12) |
|
3 | Italy | IFN α | 11241 | NR | 67 (0.6) |
|
4 | Japan | IFN α | 677 | NR | 18 (2.7) |
|
5 | Japan | IFN α | 439 | NR | 17 (3.9) |
|
6 | Japan | IFN α | 109 (77/32) | 2 (1.9) | 9 (8.2) |
|
7 | Japan | IFN α | 58 (37/21) | 19 (32.8) | 2 (3.4) |
|
8 | Italy | IFN α | 114(79/35) | 36 (31.5) | 8 (7) |
|
9 | Italy | IFN α | 120 | NR | 40 (33.3) |
|
10 | Italy | IFN α | 75 (50/25) | 26 (34.6) | 5 (6.7) |
|
11 | Italy | RIFN 2α | 78 | 31 (40) | 27 (34.6) |
|
12 | Spain | IFN α | 144 (95/49) | 7 (4.9) | 4 (2.8) |
|
13 | Italy | RIFN α | 32 (26/6) | 3 (9.3) | 11 (34.4) |
|
14 | Italy | IFN α | 114 (79/35) | 36 (31.6) | 12 (10.5) |
|
15 | Spain | IFN α | 134 | 27 (20) | 16 (12) |
|
16 | Norway | 128 | 16 (6.5) | 15 (11.7) |
|
|
-do- | 126 | 0 | 9 (11.9) |
|
||
17 | China | IFN α | 150 | NR | 28 (18.7) |
|
18 | Australia | IFN α | 246 | NR | 9 (3.7) |
|
19 | Italy | Ribavirin | 72 (25/47) | 17 (23.6) | 11 (15.3) |
|
20 | Italy | IFN α | 75 (23/52) | 17 (22.7) | 3 (4) |
|
21 | Japan | IFN α | 439(278/161) | NR | 17 (3.9) |
|
22 | Italy | IFN α | 136 (96/40) | 64 (47) | 16 (11.8) |
|
23 | Italy | IFN α | 130 | 27 (21.1) | NR |
|
24 | France | IFN α | 301 | NR | 30 (10) |
|
25 | Japan | IFN α | 439 | NR | 17 (3.9) |
|
26 | Japan | IFN α | 42 | 5 (12) | 6 (14.3) |
|
27 | Canada | IFN α | 54 | NR | 3 (5.5) |
|
28 | France | IFN α | 12 | NR | 2 (16.7) |
|
29 | Japan | IFN α | 50 | NR | 6 (12) |
|
30 | China | IFN α | 88 | NR | 7(7.9) |
|
31 | Germany | IFN α | 21 (12/9) | 5 (23.8) | 3 (14.3) |
|
Overall frequency | 16149 | 342 (20.6) |
432 (2.7) |
Abbreviations and symbols in Table: RIBA, Ribavirin; IIT, Interferon induced thyroiditis; M/F, male/female ratio; NR, not reported; Tab, thyroid antibodies;
IFNa 2b (total dose 366 MIU);
IFN alpha therapy (total dose 234 MIU);
frequency % calculated out of studies reporting Tab’s in 1656 patients and 16 studies;
frequency % calculated out of studies reporting thyroid dysfunctions in 16019 patients and 30 studies; RIFN, Recombinant IFN.
Si No | Country | Treatment | No.(M/F) | Newly developed thyroid antibody (Tab’s)n (%) | Newlydevelopedthyroid dysfunction n (%) | Reference |
1 | Italy | IFN-α 2b+RIBA | 36 | 10 (27.8) | 10 (27.7) |
|
-do- | CIFN α+RIBA | 15 | 5 (33.3) | 15 (100) |
|
|
2 | USA | IFN-α 2b+RIBA | 225 | NR | 15 (6.7) |
|
3 | Germany | PEG-IFN α+RIBA | 59 | NR | 11 (18.6) |
|
4 | Pakistan | IFN α-2b+RIBA | 107 | NR | 20 (18.7) |
|
5 | Brazil | IFN α-2b+RIBA | 65 | 4 (6.15) | 3 (4.6) |
|
6 | China | IFN α+RIBA | 161 | NR | 14 (8.69) |
|
-do- | Peg IFN α +RIBA | 343 | NR | 47 (13.70) |
|
|
7 | UK | IFN α+RIBA | 260 (172/88) | NR | 58 (22.3) |
|
8 | Brazil | IFN α+RIBA | 107 | 1(0.93) | 5 (4.6) |
|
9 | Australia | IFN α 2b (IFN α)+RIBA | 272 | NR | 18 (6.7) |
|
10. | Pakistan | IFN α+RIBA | 100 (77/23) | NR | 18 (18) |
|
11. | Germany | Peg IFN α 2b +RIBA | 61 |
7 (11.5) | 6 (9.8) |
|
12. | Taiwan | IFN α 2b | 391 | 0 | 49 (84.8) |
|
-do- | Peg IFN α 2b +RIBA | 70 | 0 | 9 (12.8) |
|
|
13 | Taiwan | IFN α | 95 | 11 (11.6) | 14 (14.7) |
|
14 | Germany | IFN α+RIBA | 40 (19/21) | 2 (5) | 3 (7.5) |
|
-do- | Peg IFN α +RIBA | 62 (29/33) | 7 (11.3) | 6 (9.6) |
|
|
15 | Poland | Peg IFN α +RIBA | 30** | NR | 2 (6.7) |
|
16 | Germany | Peg IFN α +RIBA or Levovirin | 21 | NR | 0*** |
|
17 | Australia | Peg IFN α +RIBA | 18 (6/12) | 1 (5.56) | 0 |
|
18 | Australia | Peg IFN α +RIBA | 11 (4/7) | 0 |
|
|
19 | Greece | Peg IFN α +RIBA | 61 | NR | 13 (21.3) |
|
20 | Australia | Peg IFN α+RIBA | 511 | NR | 45 (8.8) |
|
21 | Turkey | Peg IFN α +RIBA | 119 (21/98) | 5 (25) | 20 (16.8) |
|
22 | Korea | Peg IFN α +RIBA | 1 |
|
1 (100) |
|
23 | Greece | Peg IFN α+RIBA | 109 (56/53) | 5 (7) | 26 (23.8) |
|
24 | Poland | IFN α+RIBA | 89 (57/32) | 7 (7.6) | 12 (13.5) |
|
25 | Australia | Peg IFN α 2b +RIBA | 3 |
NR | 1 (33.4) |
|
Overall frequency | 3442 | 65 (5)## | 441 (12.8) |
Abbreviations and symbols in Table: RIBA, Ribavirin; IIT, Interferon induced thyroiditis M/F, male/female ratio; NR, not reported; Tab: thyroid antibodies;
Children and adolescent (2–17 yrs); **Children between 8–19 years; ***Although remaining within the reference interval. TSH was reported as increasing during therapy in this study;
although 6 cases showed thyrotropin outcome profile variation during treatment, all recovered at the end of the study;
Case report; ##Out of 1292 patients and 13 studies.
The frequency of newly developed Tab during IFN mono-treatment was in the range between 1.9–47% in 16 studies whereas the newly developed thyroid dysfunction ranged from 0.6–34.6% in 30 studies (
The frequency of newly developed Tab during IFN treatment in combination therapy was in the range between 0–33.3% in 13 studies whereas the newly developed thyroid dysfunction ranged from 0–100% in 25 studies (
The 31 publications (mono-therapy) included studies from Japan (8 studies), USA (1 studies), France (3 studies), Italy (12 studies), Spain (2 studies), Norway (1 study), China (1 study), Australia (1 study), Germany (1 study), Pakistan (1 study) and Canada (1 study). 13 studies from Italy showed frequency of the newly developed Tab and thyroid dysfunctions in the range of 9.3–47% and 4–34.6%, respectively. Eight studies from Japan showed the frequency in the range of 1.9–32.8% for newly developed thyroid antibody and 2.7–14.3% for thyroid dysfunctions. Studies from other countries (with 1–3 studies) also demonstrated similar wide variations in the frequency of newly developed thyroid antibody and thyroid dysfunctions (
The 25 publications (combination therapy) included studies from USA (1 study), Italy (1 study), Brazil (2 studies),UK (1 study), China (1 study), Australia (5 studies), Taiwan (2 study), Germany (4 studies), Pakistan (2 studies), Poland (2 studies), Greece (2 studies), Korea (1 study) and Turkey (1 study). 5 studies from Australia showed frequency of the newly developed Tab and thyroid dysfunctions in the range of 0–5% and 0–33.4%, respectively. Four studies from Germany showed the frequency in the range of 5–11.5% for newly developed thyroid antibody and 7.5–18.6% for thyroid dysfunctions. Studies from other countries (with 1–3 studies) also demonstrated similar wide variations in the frequency of newly developed thyroid antibody and thyroid dysfunctions (
Intriguingly, neither the IFN α dosage nor the virological treatment response was found to be related to the incidence of thyroid dysfunction as per one report
In contrast, the long-acting pegylated IFNα (PIFN) treatment had a lower incidence of thyroid-related side effects compared to non-pegylated IFNα
Patients treated with IFN α+RIBA have a relative risk of 4.3 for developing thyroid dysfunction
Our study further brings to front the following findings observed in an earlier study:(i) the addition of RIBA to IFN α therapy for Chronic Hepatitis C (CHC) was associated with a higher risk of hypothyroidism, (ii) Patients without thyroid autoantibodies after treatment with IFN α alone were protected from the development of thyroid autoimmunity and/or dysfunction in a second course of antiviral treatment with IFN α+RIBA, (iii) the development of hypothyroidism in patients with thyroid autoantibodies undergoing treatment with IFN α+RIBA was significantly associated with the long-term remission of CHC
As discussed earlier, a wide range in the prevalence of thyroid-related side effects was observed in the same study locations - for instance, in Italy and Japan (
Pre-existing thyroid autoimmunity can emerge as an important risk factor for developing thyroid dysfunction during IFN therapy. The presence of thyroid peroxidase antibodies (TPO-Ab) before treatment was identified as a risk factor for the incidence of thyroid disease in 60% of HCV patients receiving IFN α
Women were found to be more susceptible than men to develop IFN-related thyroid disease in some studies
A genetic predisposition to thyroid autoimmune disease is probably necessary for the development of thyroid disease in patients treated with IFN
HCV infection in a patient can lead to development of thyroid autoimmune disease
However, a population-based study excluded a specific role of HCV infection in determining the development of thyroid disease
Epidemiological and clinical evidence suggest that iodine supplementation in an iodine-deficient population may precipitate the onset of thyroid autoimmunity
Several studies have put forth contradictory results regarding the reversibility of the effect of IFN therapy on thyroid function after withdrawal of the treatment. As per one study, IFN alpha-related thyroid autoimmunity was not a completely reversible phenomenon because some patients developed chronic thyroiditis
Autoimmune thyroiditis may not be reversible after IFN therapy
These contrasting results may be due to either the variable length of follow-up after IFN withdrawal or differences in the criteria used to define the recovery from thyroid disease
The uncertainty in the clinical management of patients developing IFN-induced thyroid disease may also be due to the variable expressions and different long-term outcomes of this side effect
Perhaps the true prevalence of thyrotoxicosis or hypothyroidism is much higher than that reported in literature
Hence, the systematic screening of thyroid gland function and TPO-Ab titers in all patients with HCV - before, during and after IFN alpha therapy - should be recommended. Also, patients should be informed of the associated risk of thyroid dysfunction
Considering the significant association between HCV infection and autoimmune thyroid diseases (AITD), the detection of TPO-Ab and TG-Ab in all HCV patients, independent of IFN therapy, is suggested
IFN therapy has shown to have direct toxic effect on thyroid cells, resulting in thyrocyte apoptosis, rupture of follicles and release of thyroid hormones
The current standard approach of European Association for the Study of Liver (EASL) and well accepted standard of care for chronic hepatitis C is treatment with a combination of pegylated INF alpha plus ribavirin
The wide variation among the frequency of side effects was observed in both single and combination therapy studies. The different dosage and treatment schedule and measurements of thyroid parameters at different time intervals viz., 3 months
References | Definition given in method section |
Thyroid dysfunction (TD) was defined as having hypo- orhyper-thyroidism, (clinically and/or biochemically based). Thyrotoxicosis was defined as having TSH <0.1 mU/L, either fT4 level >26.0 and/or FT3 level >5.5 pmol/l, respectively. Hypothyroidism was defined as having TSH level >4.0 mIU/L, withnormal or low (<10.0 pmol/L) fT4 levels. | |
|
Patients developing TD were classified as either hyperthyroid orhypothyroid on the basis of their first serum TSH abnormality.Patients with a serum TSH <0.27 mU/L were classified as hyperthyroid.Patients with hyperthyroidism identified with a serum TSHsuppressed to <0.01 mU/l were subject to a diagnostic thyroid isotope scan to identify those with Graves’ disease. All patients with hyperthyroidism not developing Graves’ disease were classified as having a transient thyroiditis [associated with transient, overt hyperthyroidism (free T4>22.0 pmol/L and/or free triiodothyronine (T3) >6.8 pmol/L) or a transient subclinical hyperthyroidism (free T4 and free T3 in the normal range)].Patients with hypothyroidism (serum TSH >4.2 mU/L) were categorized according to whether hypothyroidism was transient (acute or subclinical) or permanent requiring long-term levothyroxine replacement therapy following consultation with a specialist endocrinologist at completion of IFN/RBV as described later. |
|
TSH (ultrasensitive third-generation method with a reference normal range of 0.35–5.50 mcIU/L), and FT4 (reference normal range of 0.58–1.40 ng/dL) were assayed using commercially available kits by immunometric assays.TPO-Ab was detected by solid phase 2-site sequential chemiluminescent immunometric assay (normal:<40.0 IU/mL).Patients were classified as positive (TPO-Ab >40.0 IU/mL) or negative (TPO-Ab <40.0 IU/mL) for thyroid autoimmunity |
|
TSH elevated |
|
Thyroid dysfunction (TSH <0.1 or >5 mU/L) |
|
Along with testing auto-antibodies, thyroid function was evaluated by measuring the serum levels of free triiodothyronine (FT3; normal values: 1.8–4.6 ng/L), free thyroxine (FT4; normal values: 0.9–1.7 ng/dL) and thyroid-stimulating hormone (TSH; normal values:0.3–4.2 mU/L).Determined the anti-thyroglobulin antibody (anti-thyroid peroxidase antibodies (TPO; normal values: <35 IU/mL) in the samples. |
|
Hyper- and Hypo-thyroidism:Thyrotoxicosis was defined as having TSH <0.1 mU/L, either FT4 level >26.0 and/or FT3 level >5.5 pmol/L, respectively. Hypothyroidism was defined as having TSH level >4.0 mIU/L, withnormal or low (<10.0 pmol/l) fT4 levels. |
|
TD was assessed by the serum levels of free-thyroxine (FT4) and TSH. Thyroid autoimmunity was defined by elevated antithyrogobulin (TgAb) and antithyroperoxidase antibodies (TPOAb) (normal levels <60 IU/L). |
Patients with serum positivity for at least one thyroid autoantibody were defined as Abs +ive (TgAb>100U/ml and/or TPOAb >10U/ml).Overt hypothyroidism was defined by serum TSH values above normal range, serum FT4 below the normal range and serum FT3 in or below the normal range. The normal values were 3.8–7.7 pmol/L for FT3,9.0–23.1 pmol/L for FT4 and 0.3–3.5 mu/L for TSH. | |
|
Thyroid dysfunction was diagnosed when the TSH level was either >4.5 (hypothyroidism) or <0.2 MIU L−1 (hyperthyroidism). The diagnosis of symptomatic thyroid disease was based on the clinical judgment of the treating physician. |
|
The criteria of diagnosing hyperthyroidism were, apart from the typical clinical symptoms, a decrease in TSH level <0.4 mIU/ml (normal 0.4–4.0 mIU/ml) and an increase of FT4 (normal range:0.8–1.9 ng/dl) and/or FT3 (normal range: 1.8–4.2 pg/ml).Subclinical hyperthyroidism was diagnosed in case of a decrease in TSH level and normal concentration of free thyroid hormones (TH).Hypothyroidism was diagnosed when the increased serum concentration of TSH and a decreased level of FT4 concentration were revealed. In a case of increased level of TSH within the limits of 5–10 mIU/ml and normal FT4 serum level,a latent hypothyroidism was diagnosed. Autoimmune thyroiditis (ATI) was diagnosed if an increased TPOAb level (normal range: 0–35 IU/ml) and/or TgAb (normal: 0–40 IU/ml) were found. Increased concentrations of TPOAb and/or TgAb (level >100 IU/ml) were set as a criterion of ATI diagnosis. |
|
Thyroid dysfunction was defined as TSH level of either more than 4.0 (hypothyroidism) or less than 0.3 (hyperthyroidism) mU/L, irrespective of FT3/FT4 levels |
To conclude, the present review shows that there is a wide range in the incidence of newly developed thyroid dysfunctions and thyroid antibodies in IFN-treated HCV patients. IFN α therapy alone or in combination with other drugs has different effects on the incidence of thyroid dysfunctions. Several factors that pre-dispose an HCV patient to acquire thyroid related abnormalities during IFN treatment have been discussed. These may include gender and ethnicity, HCV viral factors, genetic predisposition, and environmental and patho-physiological factors among others. Variations in IFN dosage, treatment duration of IFN, definition/criteria followed in each study for thyroid dysfunction and irregular thyroid function testing during treatment in different studies influence the outcome of each study and render the pooled risk analysis of side effects of thyroid dysfunction difficult. Another aspect highlighted by this systematic review is the variability that occurs among reports discussing the reversibility of thyroid dysfunction after IFN withdrawal. This is a comprehensive attempt to collate relevant data from 56 publications across several nations about IFN (both mono and combination therapy) related thyroid dysfunction among HCV patients. The role of each factor in causing thyroid dysfunctions in HCV patients treated with IFN should be analyzed in detail in future studies, for a better understanding of the problem and sounder clinical management of the disease.
We wish to thank Dana Wendeler, Documentation Officer of the German Accident Insurance Institution for the Health and Welfare Services (BGW), Hamburg, Germany, for her support with the management of the literature.