Serum Chemokine (C-X-C Motif ) Ligand 10 Levels Are Elevated in Patients with Graves' Disease in Long-Term Remission

Dear Editor:

Chemokine (C-X-C motif ) ligand 10 (CXCL10) is one of the CXC chemokines and is associated with T-helper 1-mediated immune responses (1). Recently, many studies have reported increased serum CXCL10 levels in the initial phase of Graves' disease (GD) (2). In addition, it has been reported that the increased serum CXCL10 levels are decreased by methimazole (MMI) treatment in some studies (3 –5). On the other hand, there are only a few investigations that have reported on serum CXCL10 levels in remission or relapse of GD after termination of antithyroid drug (ATD) treatment (3). This investigation (3), however, included insufficiently treated patients. If serum CXCL10 values in remission are lower than those at onset, they could be useful for distinguishing between complete and incomplete remission of GD when following-up the patients with a history of thyrotoxicosis due to GD. Therefore, we performed a study of serum CXCL10 levels in patients with GD who were in apparent long-term remission. Patients with clinical evidence of destructive thyroiditis were not included. The study was approved by the ethics committee of our hospital.

We enrolled patients who had been treated with MMI for at least 2 years before they met our criteria for discontinuing MMI. Our criteria required those patients taking less than 5 mg of MMI daily, whose test for serum antithyrotropin receptor antibody (TRAb) is negative, and who are euthyroid with a serum thyrotropin of >0.5 μIU/mL. We divided these patients into two groups. The remission group consisted of euthyroid patients who were not taking ATDs and had been euthyroid for at least 2 years while not on ATDs. The relapse group consisted of patients who developed thyrotoxicosis after MMI had been discontinued. We also studied a third group that consisted of thyrotoxic patients with GD who were in the early onset stage of their disease before ATD treatment had been started as this is a group in which elevated serum CXCL10 levels have been noted in many studies. We measured CXCL10 and other laboratory data at the same time (Table 1). We excluded pregnant patients, those taking corticosteroid therapy, or those with autoimmune diseases.

Serum CXCL10 levels were determined by quantitative sandwich immunoassay using a commercially available kit (R&D System, Minneapolis, MN). Values are presented as means ± standard deviation for normally distributed variables. Mean group values were compared by the Kruskal–Wallis test. Proportions were compared by the χ ² test.

The mean values of CXCL10 were 146 ± 100 pg/mL in the remission group, 101 ± 70 pg/mL in the relapse group, and 133 ± 55 pg/mL in the early onset thyrotoxic group. There were no significant differences in serum CXCL10 levels among the three groups. To minimize a possible influence of age, we eliminated patients younger than 30 years or older than 60 years. Serum CXCL10 levels were similar in the three groups: 116 ± 80 pg/mL in the remission group (n = 12), 128 ± 62 pg/mL in the relapse group (n = 8), and 143 ± 55 pg/mL in the early onset thyrotoxic group (n = 13). The mean age of each group was 46 ± 8, 46 ± 12, and 42 ± 10 years. In the remission group, there was no significant difference in serum CXCL10 levels between patients who had positive TRAb (n = 4) and negative TRAb (n = 27) (data not shown). The mean serum CXCL10 value in healthy volunteers (n = 14) was 62.5 ± 13.5 pg/mL. This is in agreement with previous reports and is significantly less than the remission group with GD. Therefore, our data did not support the hypothesis that serum CXCL10 levels would not be elevated in patients with a history of GD who were in apparent remission.

Domberg et al. also reported that there was no difference in serum CXCL10 levels between 25 patients with GD who were thought to be in remission and 25 patients who had a relapse of thyrotoxicosis after ATD treatment was stopped (6). However, this report was not clear both about the treatment duration and the definition of apparent remission. This might not have required consideration of serum TRAb value. Their high relapse rate of 50% among all patients suggests that the duration or some aspect of ATD treatment was less efficacious than for the patients in our remission group. Further, the study of Domberg et al. (6) cannot be compared with our (present) study because in their study the serum CXCL10 analysis was done during ATD treatment. Moreover, though 38 of 50 cases were euthyroid when studied, 12 were hyperthyroid. Importantly, serum CXCL10 levels are reported to decrease during MMI treatment (3 –5).

Our results appear to contrast with those of Antonelli et al., who reported that serum CXCL10 levels were low in 18 patients with GD who were in remission and similar to that of euthyroid patients with GD while they were taking MMI (3). Details of their 18 cases concerning age or duration of remission were unclear, and their treatment duration was shorter (2–41 months) than in our series. These factors might lead to different results compared with ours. They also suggested that the serum CXCL10 levels could possibly be a prognostic maker after MMI treatment. Our results contradict this suggestion because serum CXCL10 levels in GD patients in remission were as high as those in relapse. Romagnani et al. reported that serum CXCL10 levels decrease gradually with long-term MMI treatment of 1–4 years (4).

In conclusion, we did not find a relationship between serum CXCL10 levels and the state of disease activity related to thyrotoxicosis in patients with GD. Therefore, we hypothesize that immune mechanism(s) other than those directly relating to CXCL10 production may prevent relapse of thyrotoxicosis in patients with GD who are in apparent remission.