A retrospective study by Kozielewicz et al. reported a probability of occurrence of TD in patients with CHB receiving Peg-IFNα treatment of 7.1% (7/99) [9]. Our results were similar, with approximately 8.8% (105/1197) of patients developing TD during treatment. The mechanism by which interferon causes TD remains unclear, although it is certain that the immune system plays an essential role. Interferon binding to the receptor activates the JAK-STAT pathway, resulting in the transcription of a large number of interferon-stimulated genes, thereby inducing autoimmune thyroiditis [10]. Prummel et al. pooled the results of multiple studies and found that TPOAb positivity before treatment was associated with a 3.9-fold increased risk of TD [11]. Koh et al. analyzed the clinical data of interferon therapy in the treatment of tumors and immune diseases and uncovered that the incidence of TD was higher in patients with thyroid antibody positivity before treatment (46.1%, 41/89) [12]. In our study, patients with baseline antibody-positive CHB were found to have a higher TD incidence (65.5%, 9/29) after interferon treatment, and the time to occurrence of TD was found to be significantly shorter in patients with antibody positivity at baseline compared with patients with antibody negativity (the median time was 18 and 24 weeks, respectively). Moreover, this study also suggested that as Peg-IFNα treatment continued, the positive conversion rate of antibodies increased significantly, from 2.4% (29/1187) at baseline to 9.6% (114/1187). Specifically, the overall positive conversion rate of thyroid autoantibody and the individual positive conversion rates of TGAb, TPOAb, and TRAb in women were all significantly higher than those in men. Meanwhile, the TD rate of women was also higher, although the negative conversion rate of TGAb in women was significantly lower than that in men. A similar study reported that when interferon was used to treat diseases other than CHB, women had a 4.4-times greater risk of TD than that of men [11]. Therefore, the administration of Peg-IFNα should be monitored more closely in patients with thyroid antibody positivity, especially among women. However, not all TD were associated with thyroid antibodies. We found that 30.5% (32/105) of patients only presented with TD and antibody negativity. It has been speculated that the pathogenesis of this type of TD is different from that described above, and some studies have suggested that genetic factors play a role. Dalgard et al. reported that among patients with chronic hepatitis C treated with interferon, the incidence of TD was significantly higher in Asian populations compared with other ethnic groups [13]. Some studies have also found that HLA-A2 is related to Peg-IFNα-induced autoimmune thyroid diseases [14]. In addition, the direct toxic effect of interferon on thyroid cells can also lead to thyroid diseases [15, 16].
Regarding the types of TD caused by interferon, this study indicated that hyperthyroidism accounted for 53.3%, followed by subclinical hypothyroidism (34.3%). Okanoue et al. studied 987 patients with chronic hepatitis C receiving short-acting interferon and found that 12 of 18 patients who developed TD had hyperthyroidism and 6 had hypothyroidism [17]. In another study that included 422 patients with chronic hepatitis C, B, and D infections, the main type of TD caused by interferon was hypothyroidism, accounting for 83.3% (25/30) of cases, while only 5 patients had hyperthyroidism [18]. We considered that the reasons for the different proportions of TD types among studies may be related to different types of diseases and interferons used. Further analysis of the correlation between the clinical types of TD and curative effect revealed that the proportion of those with > 1log reduction in HBsAg levels from baseline after 24 weeks of treatment and HBsAg seroclearance after 48 weeks of treatment were both significantly higher in patients with CHB with hyperthyroidism/subclinical hyperthyroidism than in patients with hypothyroidism/subclinical hypothyroidism. Furthermore, logistic regression analysis suggested that the odds of HBsAg levels decreasing by more than 1log from baseline after 24 weeks and HBsAg seroclearance after 48 weeks was 3.9- and 9.9-fold, respectively, in patients with hyperthyroidism/subclinical hyperthyroidism compared with patients with hypothyroidism/subclinical hypothyroidism. Reports on the correlation between TD and clinical efficacy of interferon treatment are limited, save for one study by Luo et al [19]. The authors observed positive conversion of antibodies or TD in 141 of 342 patients with CHB treated with Peg-IFNα, and there was a more pronounced decrease in HBsAg levels in these patients (p = 0.019). However, the study did not categorize TD into specific clinical types. In our study, most TD patients had a good prognosis, and the thyroid function in 78.7% of the patients returned to normal after discontinuing Peg-IFNα for 6 months. Approximately 50% of patients had negative conversion of thyroid antibodies, and antithyroid drugs or thyroid hormone replacement therapy was only required in 25% of patients with clinical TD. Studies on interferon therapy in patients with chronic HCV infection have also revealed a high normalization rate of TD after discontinuation of interferon therapy, with a normalization rate of 87.5% in 24 patients approximately 1.5 years after discontinuation [20]. Based on the pooled analysis of multiple studies, Mandac et al. believe that interferon therapy can be continued in the context of symptomatic treatment and close monitoring [21][1]. However, in patients with TD with clinical symptoms, discontinuation of interferon therapy and consultation with an endocrinologist should be considered. According to the relevant guidelines in the United States, β-blockers and other symptomatic treatments should be administered to patients with clinical hyperthyroidism during the course of interferon therapy, and antithyroid drugs and radiation therapy should only be used in patients with Graves' disease [22].
Among patients with TD, patients with CHB with clinical hyperthyroidism are more likely to achieve HBsAg seroclearance, suggesting that Peg-IFNα has an antiviral effect and also plays a role in regulating immune function. However, the up-regulated immune response may not be precise. Immune clearance is necessary for the treatment of CHB, although it will inevitably result in immune damage to the thyroid. Although the HBsAg seroclearance rate is higher in patients with hyperthyroidism/subclinical hyperthyroidism, it does not mean that TD is an effective predictor of Peg-IFNα treatment efficacy. Considering that the occurrence of TD during the course of interferon treatment is related to the curative effect and that most patients can be improved after discontinuation of the drug, whether to discontinue the drug or continue to pursue higher treatment goals after TD can be a diagnostic dilemma for clinicians. Currently, there are no well-defined laboratory-related indicators that can be used as a reference standard for discontinuation or continued use of interferon. Whether to discontinue the drug or not is largely based on the clinician’s experience and severity of the patient’s TD and treatment willingness. In antiviral therapy for CHB, virological and serological response should be pursued, and the safety of treatment should also be considered. For patients who are about to receive Peg-IFNα treatment, thyroid function and related antibody detection should be performed before treatment to assess the TD risk, and regular review is needed during the course of treatment to detect TD early. For asymptomatic patients with TD, a reduced dose of Peg-IFNα should be considered if the HBsAg has reached relatively low levels to warrant the continued pursuit of HBsAg seroclearance. For patients with unsatisfactory decline in HBsAg levels and severe TD symptoms, discontinuation of treatment should be considered.
In conclusion, this study suggests that a certain proportion of positive conversion of thyroid antibodies (7.3%) and TD (8.8%) will occur during the course of Peg-IFNα treatment for CHB and is more common in women. Overall, the clinical symptoms of TD were mild; treatment was only needed in 25% of patients, and the normalization rate of TD was up to 78.7%. Compared with patients with hypothyroidism/subclinical hypothyroidism, patients with hyperthyroidism/hypothyroidism have higher odds of reduced HBsAg levels and HBsAg seroclearance. Therefore, we believe that TD is not an absolute contraindication for Peg-IFNα therapy. In the course of treatment, patients should be monitored closely to balance efficacy and safety in the pursuit of functional cure.