OUP user menu

The natural history of interferon-α2b-induced thyroiditis and its exclusivity in a cohort of patients with chronic hepatitis C infection

H.A. Tran , G.E.M. Reeves , T.L. Jones
DOI: http://dx.doi.org/10.1093/qjmed/hcn150 117-122 First published online: 25 November 2008

Abstract

Background: Interferon-α2b (IFN-α2b) is well known to cause both hyper- and hypo-thyroidism. In the former, the commonest aetiology is thyroiditis. As there is no previous data to fully characterize the entity of IFN-related thyroiditis, the aim of this study is to document in detail its evolution in a cohort of hepatitis C patients treated with pegylated IFN-α2b and Ribavirin (RBV).

Methods: A prospective observational study was conducted in patients who developed thyroid diseases whilst receiving combination of pegylated IFN-α2b and RBV for hepatitis C. The patients were followed with monthly thyrotropin (TSH). Where TSH was undetectable, free tetra- (fT4) and tri-iodothyronine (fT3) were added. Anti-thyroperoxidase (TPO), anti-thyroglobulin (Tg) and thyroid stimulating immunoglobulin (TSI) levels were also performed at diagnosis, during and at the end of IFN therapy. All patients were assessed and followed up closely with monthly TSH, fT4 and fT3 levels until the completion, after 6 and 12 months of treatment.

Results: There were seven females and four males over a 30-month period. All patients were found to have thyroiditis. On average, the time to the development of thyroid disease was 10 weeks and duration of disease 9 weeks. All patients eventually recovered normal biochemical thyroid function although two required short-term supplementation.

Conclusions: Thyroiditis was found exclusively in our patients. Both the hyper- and hypo-thyroid phase can be short lived, extreme and transient in nature which warrants strict monthly TSH monitoring. Careful follow-up of all patients is mandatory as complete recovery is expected.

Introduction

Treatment for hepatitis C infection often results in many endocrinological disturbances of which thyroid dysfunction is most prevalent.1 Within the realm of thyroid-related diseases, thyroiditis is the commonest cause of interferon (IFN)-associated dysfunction. Others include Graves’ like thyrotoxicosis and the tri-phasic thyroiditis in which the patient oscillated between increased and absent thyroid pertechnetate uptake on nuclear scan whilst being profoundly and clinically thyrotoxic.2–4 Although the subject of much discussion, there are no previous reports that follow the clinical course of this peculiar clinical condition on a monthly basis. Previous studies measured thyroid parameters in an ad hoc basis. Kee et al.7 and Dalgard et al.8 measured thyroid function tests (TFTs) every 3 months; Moncoucy et al.6 every 2 or 3 months and Hsieh et al. every 4 weeks for 24 weeks, followed by 8 weeks for another 24 weeks.9 Minelli et al.10 evaluated thyroid functions at 1, 2, 3 and 6 months after diagnosis. Many studies were also performed in the setting of mono-IFN therapy10 and/or combination regular IFN and Ribavirin (RBV),7,9 not the pegylated form. To fully characterize, better understand and manage the condition, we prospectively studied 11 patients who developed thyrotoxicosis during the course of treatment. Their natural history and outcome were carefully recorded in the final analysis.

Methods

Patients

The Hunter Area Hepatitis C treatment centre assesses and treats all cases of hepatitis C in Northern New South Wales, Australia. It is part of the John Hunter Hospital, a major tertiary referral centre in the state. In total, 11 patients who developed thyroiditis whilst undergoing combination treatment for hepatitis C were included over a 24-month period. All were drug naive, i.e. all were undergoing therapy for the first time. All other causes of chronic hepatitis were excluded including hepatitis B and chronic alcoholic liver disease. Baseline characteristics of all studied subjects are summarized in Table 1. All patients were assessed for thyroid disease clinically and biochemically using TSH level prior to the beginning of therapy. No goiter was detected and baseline TSH levels were normal in all cases.

View this table:
Table 1

Baseline characteristics, hepatitic outcomes and auto-antibody profiles in 11 thyroiditis patients

Subject No.GenderAgeHepatitis C genotypeDuration of therapy (weeks)SVRAnti-thyroglobulinAnti-TPOTSI
BeforeDuringAfterBeforeDuringAfterBeforeDuringAfter
1M261a48Y<116<1<1<1<1<1010<10
2M51224Y<116<1<1<1<1<10<10<10
3M54324Y<1<1<1<1<1<1<10<10<10
4M49448Y<1<1<1<1<1<1<10<10<10
5F42224Y<1<1<1<12561024<10<10<10
6F34324Y<1<1<1<1256256<10<10<10
7F49148Y<1<1<1<164<1<1010<10
8F49148Y<116<164102416<10<1019
9F50224Y<1<1<1<116<1<1010<10
10F37148Y<1<1<1<1<1<1<1014<10
11F43448Y<1<1<1<1<1<1<10<10<10
  • Mean age: 43.3 (95% CI, 25.7–60.9). All are Caucasians.

Therapy

All patients were treated with combination pegylated IFN-α2b and RBV therapy. The duration of treatment depended on the HCV genotypes; genotypes 2 and 3 were treated for 24 weeks and types 1 and 4 for 48 weeks, respectively. For the latter group, treatment was continued for the full 48 weeks irrespective of the HCV RNA status at 24 weeks. The dosage for pegylated IFN-α2b was 180 µg weekly with RBV dose ranging from 1000 to 1200 mg daily according to bodyweight.

Thyroid function assessments

All patients received routine TSH levels at the start of treatment and at monthly intervals. When the TSH level was undetectable, free tetra- (fT4) and tri-iodothyronine (fT3) levels were performed, followed by an endocrinological and clinical assessment from which further imaging investigations were determined. All had a thyroid ultrasound and thyroid pertechnetate uptake scan. Anti-thyroperoxidase (anti-TPO), anti-thyroglobulin (anti-Tg) antibodies and thyroid stimulating immunoglobulin (TSI) levels were performed at diagnosis, 4 weeks after thyroiditis and at the completion of IFN course. All patients were followed up at 4 weekly intervals until the end of therapy, in 6 months’ time as part of the routine HCV treatment review and again in 12 months’ time. Sustained virologic response (SVR) was defined as undetectable hepatitis C virus RNA at 24 weeks after the completion of treatment.5

Laboratory assay characteristics

Third generation serum TSH, serum fT4 and fT3 were determined by two-site sandwich immunoassay using an automated chemiluminescent system (Diagnostic Products Corporation, Immulite 2000, Los Angeles, CA, USA). The reference range (RR) for TSH was 0.4–4.0 mU/l, fT4 10.0–26.0 and fT3 3.5–5.5 pmol/l. The coefficients of variation (CV) were 5.0% and 5.1% at TSH concentrations of 4.0 mU/l and 10.0 mU/l, respectively. For fT4, the CV was 6.5% at 10.0 pmol/l and fT3 8.9% at 3.5 pmol/l.

Serum autoantibodies to thyroglobulin and TPO were measured by agglutination (Serodia-ATG and Serodia-AMC, Fujirebio, Inc., Tokyo, Japan). Titres of less than 1:400 were considered normal for both.

TSI was measured using cell culture and radio-immunoassay. This is an in-house bioassay using Chinese Hamster Ovary (CHO) cells in culture to detect the presence of thyroid-stimulating activity. The CHO cells are transfected with the TSH receptor genes and thus are responsive to TSI. Thyroid-stimulating activity is measured by evaluating the intracellular release of cAMP induced by the patient's serum immunoglobulin on the CHO cells. The results are reported as U/ml. TSI should be absent in the normal population. A TSI level of <10 is considered negative, 10–50 as weakly, 50–100 as moderately and >100 U/ml as strongly positive.

Definitions

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, with normal or low (<10.0 pmol/l) fT4 levels.

Thyroiditis

Thyroiditis is defined as the triad of clinical and/or thyrotoxicosis, with a reduced/negligible thyroid pertechnetate uptake scan. All uptake scans were reviewed by a specialist nuclear physician consultant. Thyroid autoantibodies may be present but are not considered essential to the diagnosis.

Statistical analysis

Data are presented as mean with 95% CI.

Results

There were 11 patients who developed the condition (seven females and four males) over the study period. Their characteristics are listed in Table 1. All thyroid ultrasound and pertechnetate uptake studies were performed within 14 days of diagnosis. In the former, the findings showed normal thyroid gland in seven patients (five females and two males). In the other four patients (two males and two females), there was multinodularity but without a size increase. All nuclear uptake results were absent or negligible. There appears to be no relationship between gender, genotype and the risk of developing thyroid disease. The duration to the development of disease, hyper- and hypo-thyroid phases are all quite variable. All cases returned to normality eventually. In some cases, the titres of all autoantibodies increase with the development of thyroiditis but then recover. Two patients (cases 9 and 10, Table 2) developed significant hypothyroidism, one quite severe and prolonged, required thyroxine supplement but both were able to be weaned off therapy 12 months after the completion of HCV therapy. The range of fT4 and fT3 levels can be quite elevated. Equally, TSH levels can also be quite high, with undetectable fT4 level, representative of severe thyroid failure. All patients who developed thyroiditis achieved SVR and this was considered curative.5 The evolutionary features of thyroiditis are listed individually in Table 2.

View this table:
Table 2

Characteristics and evolutionary characteristics of 11 patients who developed thyroiditis and their thyroid status at the completion of IFN therapy

Subject No.Time to onset of thyroiditis (weeks)Duration of Toxic phase (weeks)Duration of hypothyroid (weeks)Hyperthyroid phaseHypothyroid phaseSonographic findingsThyroid outcome
TSHPeak fT4Peak fT3TSHFT4FT3
12035072.721.99.29.93.3MNN
23174025.15.934.68.23.5NHYPO
32864021.06.627.24.33.1MNN
42155033.87.814.68.63.4MNN
51462048.711.418.26.83.5NN
61693038.9711.212.33.5NN
71780020.55.16.517.85.2NN
82088032.98.46.818.74.8MNN
91244023.55.8115.83.72.1NHYPO
1011516027.14.760.24.82.4NHYPO
11864025.58.234.88.33.3NN
Mean18.26.85.10.033.96.918.28.23.4
95% CI2.5–33.90.3–13.30–13.1N/A23.9–45.55.0–11.67.7–51.55.9–12.72.8–4.1
  • All three hypothyroid cases eventually recovered and did not require thyroxine supplement at 12 month follow-up. All cases had no/negligible uptake on Technetium scintigraphy. HYPO: hypothyroid; MN: multinodularity; N: normal.

Discussion

Although there are many previous reports addressing thyroid dysfunction,6–10,12 this is the first study to closely detail, on a monthly basis, the evolution of IFN-induced thyroiditis, to the authors’ best knowledge. Thyroiditis is the most common thyroid-related disease in relation to IFN-based therapy. In general, there are four forms of thyroid dysfunction in this setting: non-thyroidal illness, thyroiditis (bi-phasic), hypothyroidism and Graves’ like thyrotoxicosis.4 The thyroiditis form, similar to other non-IFN mediated forms, demonstrates a bi-phasic nature, shifting from hyper- to hypo-thyroidism with subsequent full recovery, albeit at times prolonged. It is the single and exclusive form of thyroid disease observed in our cohort of 201 patients. Although the number of subject is modest, it offers a unique opportunity to comprehensively study this peculiar type of thyroid dysfunction. Graves’ like thyrotoxicosis, although well described, was surprisingly and completely absent in our cohort. The reason for this discrepant observation is unknown, especially when compared with previous series described in Australia, Europe and North America.4,8,11,12 It is possible that this is due to an ascertainment bias where pre-existing Graves’ disease is detected due to the increased thyroid surveillance under IFN therapy. The contrasting findings on nuclear scintigraphy should rule out any potential misdiagnosis between the two conditions however. Similarly, tri-phasic thyroiditis was not seen although admittedly this pattern is rare, with a single report.13

The pathogenesis of thyroiditis in this particular setting is currently not fully understood and has been the subject of many recent reviews.1,14 The opportunity to study the condition may also be made more difficult due to the condition's low frequency of <1%.15 Concisely, there must be synergy between genetic predisposition, the presence of HCV virus and IFN therapy. The latter will result in an induction of IFN-α and -β production as part of the innate immune response.16 IFN also causes the activation of natural killer cells, maturation of dendritic cells, proliferation of dendritic cells, proliferation of memory T cells and prevention of T-cell apoptosis.17 These will induce a rise the thyroid auto-antibodies titre, which will in turn cause destructive changes in the thyroid gland in some of the cases. The presence of hepatitis C viral particle within the thyroid cells may additionally contribute further damage to the thyroid gland.18 The addition of IFN-α then inflames an already vulnerable thyroid gland. Furthermore, IFN therapy has also been shown to have a direct toxic effect on thyroid cells, resulting in thyrocyte apoptosis, rupture of follicles and release of thyroid hormones.16 These pathophysiologic events manifest themselves in the form of the bi-phasic thyroid response that is so classical of this type of thyroiditis, Figure 1.

Figure 1.

The evolving biochemical profile of thyroid hormones in IFN-induced thyroiditis. Each time point represents the mean value of each thyroid parameters with 95% CI bar. Testing at point [A] will falsely reassure with normal thyroid tests. Testing at point [B] will detect hyperthyroidism. Point [C] indicates hypothyroidism, which may be deemed permanent and required unnecessary life-long thyroxine therapy. fT4 levels (solid line); fT3 levels (dashed line), TSH concentrations (dotted line). The right Y-axis is the secondary axis for TSH.

The prevalence of thyroiditis is low in our cohort with 11 cases over a 2-year period and this is consistent with our previous report.19 The time from exposure to the development of the disease, except for one case is long, up to 4 months, suggesting the need for prolonged IFN exposure to stimulate and modulate the immune system. Because of this unpredictability, it is important that thyroid surveillance is performed very early and frequently in the course of treatment to diagnose and manage the condition. The degrees of biochemical hyper- and hypo-thyroidism are quite extreme and can be significantly incongruent with the clinical state. Symptomatology is also unreliable because of the side effects of IFN, which often override or mask thyroid-related symptoms. The urge to treat the biochemical values can be quite pressing and may need to be resisted as the natural history is relatively benign. Similarly, the hypothyroid phase can be prolonged and profound and thus should be closely followed up. The need to implement thyroxine replacement therapy should be judiciously considered in light of the clinical state as the potential to recover is high. Even if thyroxine is required, patients should be given a trial off therapy at the completion of the HCV treatment and observed for any potential recovery. This is particularly important as patients can be erroneously classified depending on the timing of the testing and the frequency. Patients who are found in the hypothyroid phase can be easily categorized as hypothyroidism, especially in the presence of positive thyroid auto-antibodies. These patients are then subjected to life-long thyroid supplement without realizing the great potential of recovery and normalization of thyroid function.

This report also suggests that patients undergoing combination treatment for hepatitis C should have baseline and monthly TSH thereafter in order to fully assess thyroid status. The potential for misdiagnosis, depending on the timing of thyroid testing, is illustrated in Figure 1. The National Academy of Biochemistry is yet to recommend thyroid testing in this clinical scenario.20 Three monthly TSH as suggested by Mandac et al.21 may completely miss the diagnosis as thyroid functions have completely normalized in many cases. Of note, both the British Society of Gastroenterology and the American Gastroenterological Association recognize the potential thyroid effect of IFN and recommend thyroid screening.22,23 However, only the former specifies that thyroid function test is recommended at each treatment visit, rather than monthly. The National Institute of Health consensus statement on hepatitis C management surprisingly makes no mention of the thyroid issue.24 Alternately, testing in another phase can detect up hypothyroidism. In the presence of significantly slow recovery, this may be diagnosed as primary hypothyroidism leading to permanent therapy. It is also a concern that many reports in the literature may have indeed inadvertently misclassified these patients without ever recognizing the true nature of the diagnosis.25 It is thus important that all cases of abnormal thyroid function in this setting undergo thyroid nuclear uptake scan to carefully define the diagnosis. Ultrasound studies may contribute and detect incidental nodularity in four of our cases but are not essential to the diagnosis. It may be debatable that due to the benign nature of the condition, thyroid surveillance is not important given the fact that recovery to normality is expected. Furthermore, the detection of thyroiditis does not necessarily warrant cessation of therapy. Many patients are willing to compromise thyroid function and risk hypothyroidism by completing their treatment regimen in order to achieve SVR and the potential cure. It is however important that the condition is recognized so that appropriate thyroid treatment is delivered. Furthermore, re-exposure can result in recurrence, albeit in isolated cases.26 The rare possibility of IFN-mediated Graves’ disease should also be fully excluded as management strategy is completely different. As IFN toxicity can easily mimic thyroid disease symptoms, it is equally critical to detect the two polarised phases of the thyroiditis so that temporary and alleviating measures can be implemented. It is important to identify thyroid involvement so that erroneous and premature termination of IFN therapy does not occur. Patients are often psychologically labile and finding an organic explanation for these symptoms can be additionally very reassuring. The hyperthyroid phase can be managed symptomatically with β-blockade. Glucocorticoid should be reserved as last line therapy, especially in the setting of chronic hepatitis. As the hypothyroid phase is temporary, there may be a strong argument to use tertroxin (fT3 supplement) which can be rapidly withdrawn at the completion of IFN with the expectation of complete thyroid recovery.

Conclusion

This study details clearly the natural history of IFN-induced thyroiditis. It also highlights the need for regular monthly thyroid testing to fully document and diagnose this prevalent and exclusive thyroid dysfunction in our cohort of hepatitis C patients. Short-term and rapid onset treatment should be considered to reduce the burden of psychological and physical symptoms contributed by thyroid diseases to patients undergoing IFN treatment.

Acknowledgement

We would like to sincere thank Geoffrey M. Kellerman for his review and constructive comments of this article.

Conflict of interest: None declared.

References

View Abstract