Elevated serum thyroid stimulating immunoglobulin linked to failure of first-line intravenous methylprednisolone monotherapy in moderate-to-severe thyroid eye disease

Abstract

Purpose

To assess factors associated with failure of intravenous methylprednisolone (IVMP) monotherapy as the first-line treatment for thyroid eye disease (TED) and to identify patients who might benefit from supplementing mycophenolate mofetil (MMF) to IVMP.

Methods

Data for all patients with TED treated with IVMP according to the EUGOGO protocol in our center between 2016–2021 were retrospectively analysed.

Results

Forty-seven patients (mean age 51.32 ± 14 years, 27 females) were enrolled. The mean time from first reported symptoms to first IVMP treatment was 12.1 ± 5.59 months (range 0–120). The mean clinical activity score (CAS) before treatment and at a mean of 5 and 12.2 weeks after treatment initiation was 6.00, 2.96, and 1.81, respectively (P < 0.01). Twenty-one patients (44.68%) were recommended second-line treatment: nine due to no response or worsening of CAS, six due to partial response, four with good response but early relapse after completion of treatment, and one due to late relapse. Eighteen of those 21 patients received second-line treatment which included rituximab (n = 7), MMF (n = 6), a second course of IVMP (n = 4), and tocilizumab (n = 1). Serum thyroid-stimulating immunoglobulin (TSI) levels were higher in patients who received second-line treatment compared with patients who responded well to first-line IVMP monotherapy at presentation (2135% vs 1159%, P = 0.05) and after completion of first-line treatment (2201% vs. 986%, P = 0.043).

Discussion

TED patients requiring second-line treatment after failed IVMP monotherapy had higher baseline and post-first-line treatment serum TSI levels. Those with elevated TSI may benefit from dual therapy (IVMP and MMF) and require closer monitoring.

Introduction

Graves’ disease (GD) is an autoimmune disorder in which autoantibodies stimulate the thyroid stimulating hormone receptor (TSHR) on follicular cells in the thyroid gland causing hyperthyroidism [1]. Thyroid stimulating immunoglobulin (TSI) can also bind and stimulate TSHR, which is expressed on orbital fibroblasts and adipocytes in higher than normal levels [2–4]. This activation plays a role in the pathogenesis of thyroid eye disease (TED), causing extraocular muscle enlargement and orbital fat hypertrophy. Elevated serum TSI levels were detected in 98% of patients with untreated active TED, and those levels correlated with disease activity and severity [2].

The mainstay medical first-line treatment for moderate-to-severe TED recommended by the European Group on Graves Ophthalmopathy (EUGOGO) remains intravenous (IV) methylprednisolone (IVMP) [5]. Reported success rates for this therapy vary widely in various studies, with cited response rates of 77–88% in some but as low as 44–53% in others real-life studies [6–10]. A EUGOGO observer-masked multicentre trial found that the addition of mycophenolate sodium 0.72 g daily to IVMP for 24 weeks improved response to treatment from 49% to 63% in week 12 and from 53% to 71% in week 24 in patients with moderate-to-severe TED [9, 11]. That supplement was therefore recommended as first-line combination treatment with IVMP in their 2021 guidelines [5]. However, medical treatment options for TED have expanded in recent years, with multiple studies showing promising results with the use of other drugs, such as rituximab, tocilizumab, teprotumumab (which was the first drug specifically approved for the treatment of TED by the FDA but remains unavailable outside the USA), and combinations of oral prednisolone with cyclosporin or azathioprine [5].

The myriad of new treatment options and the moderate response rate to standard first-line IVMP therapy have provided a stronger incentive to search for clinical and laboratory markers to predict response to treatment. TSI has been shown to be positive in almost all patients with TED and to correlate with disease activity and severity, representing one such potential marker. In the current study, we examined the correlation between various clinical and laboratory features, including TSI, and response to first-line IVMP.

Methods

This is a retrospective cohort study. We analysed all patients with moderate-to-severe TED who had first-line treatment with IVMP according to the EUGOGO protocol (500 mg of IVMP weekly for 6 weeks followed by 250 mg of IVMP weekly for an additional 6 weeks) [5, 12] between 2016 and 2021 at the Sheba Medical Center. Patients who had received previous treatments with IVMP or other drugs, those who had undergone orbital surgeries (e.g., orbital decompression), and those with dysthyroid optic neuropathy who had undergone urgent orbital decompression were excluded from the study. The enrolled patients’ demographic, clinical, and laboratory data, together with their clinical activity score (CAS), quality of life questionnaire responses, and past and present treatments were retrieved from the medical center’s database.

The patients in our multidisciplinary TED clinic undergo a comprehensive evaluation at each visit, including a complete ophthalmic examination and an assessment of optic nerve functions. Evaluated are visual acuity, color vision by means of the Hardy-Rand-Rittler plates, pupillary response, and automated visual field (Humphrey 24-2). They undergo applanation tonometry, ocular motility and orthoptic examinations, an eyelid evaluation and measurements of upper eyelid margin to reflex distance, lower eyelid margin to reflex distance, lagophthalmos, grading of eyelid oedema, and exophthalmos measurements by means of an exophthalmometer. The patients are also seen by an endocrinologist who conducts a systemic evaluation, a physical examination, and records the findings of all available blood tests.

TED activity is graded according to the clinical activity score (CAS) as defined by the EUGOGO (up to 7 points for the first visit and up to 10 points for the follow-up visits). Quality of life (QoL) was evaluated by the quality of life standard EUGOGO questionnaires.

Serum TSI activity was measured with a functional cell-based TSHR bioassay (Thyretain, Quidel, CA, USA) according to the manufacturer’s instructions. Briefly, the levels of TSI activity were measured in triplicate with the Infinite M200 microplate reader (Tecan, Crailsheim, Germany). All measured values were corrected for the plate’s internal auto-luminescence by reduction of the mean value in blank wells. The results were reported as percentage of specimen-to-reference ratio (SRR%). The SRR% values were calculated according to the following formula: SRR% = Average TSI specimen relative light units (RLU)/average reference standard RLU × 100, as previously described elsewhere. The patient’s serum was considered positive for the presence of TSI activity if the resultant SRR % measured ≥140% over the reference control [13].

Response to treatment was assessed mainly according to the change in CAS, with a CAS equal to or greater than 3 generally being considered active disease. Failure to respond to the first line of therapy was defined as a CAS ≥ 3 after completion of therapy. Early or late relapses were defined as an initial improvement of the CAS ≤ 2 following treatment with reactivation of the disease and a CAS ≥ 3 before or after 12 months following completion of therapy, respectively. The main outcome measures were the correlation between failure to respond to systemic first-line IVMP and the patient’s demographic, clinical, and laboratory parameters.

The study was conducted in accordance with good clinical practice guidelines and adhered to the tenets of the Declaration of Helsinki. Ethical approval was granted by the institutional review board, which waived informed consent for this retrospective anonymized investigation. In order that the data will be available, an approval from the local institutional review board is required.

Statistical analysis

Quantitative variables were described as mean ± standard deviation. Categorical variables were described as absolute and relative frequencies. A matched paired analysis was performed to evaluate differences in CAS before treatment, during first-line treatment, before second treatment and at the end of follow-up. An ANOVA test was used to compare between the group of the patients who were recommended to receive a second-line treatment and those who responded well to the first-line IVMP and needed no additional treatment. The overall significance level was set to an alpha of 0.05. The statistical analysis was carried out with Microsoft Excel 2017 (Microsoft Corporation, Redmond, WA) and IBM SPSS software version 24.0 (SPSS, Inc., Chicago, IL, USA).

Results

Forty-seven patients (20 males and 27 females) with a mean age of 51.32 ± 14 years (range 21–77) were treated with IVMP as first-line therapy according to the EUGOGO protocol for TED. Eight patients (17.02%) were active smokers. The clinical characteristics and medical history of the study cohort are summarized in Table 1.

Table 1.

Clinical characteristics and medical history of the 47 TED patients.

Coexisting systemic diseases	Number	Percentage
Autoimmune	7	14.9
Cardiovascular	11	23.4
Diabetes mellitus	2	4.3
Neoplasm	0	0
None	27	57.4

The mean time from the first reported symptoms to first IVMP treatment was 12.1 ± 5.59 months (range 0–120). The mean CAS before treatment and at mid-treatment examination (a mean of 5.3 weeks after treatment onset) were 6.00 and 2.96, respectively (matched pairs, P < 0.01). The mean CAS at 12.2 weeks after the initiation of treatment was 1.81, and that decrease in CAS was significant compared to baseline (matched pairs, P < 0.01).

Second-line treatment was recommended for 21 patients (44.68%) for the following reasons: no response or worsening of CAS after first-line treatment (n = 9, 19.14%), partial response (n = 6, 12.75%), good response with early (<12 months) relapse of symptoms after completion of treatment (n = 4, 8.51%) or late (≥12 months) relapse (n = 1, 2.12%), and having a history of an autoimmune disease and continuing treatment with systemic steroids for that indication (n = 1, 2.12%). Eighteen of those 21 patients (85.71%) agreed to receive second-line treatment which included: rituximab (n = 7, 38.89%), MMF (n = 6, 33.33%), a second course of IVMP per EUGOGO protocol (n = 4, 22.22%), and tocilizumab (n = 1, 5.55%). There was a significant decrease in the CAS after undergoing the second-line treatment compared to the CAS before it (1.13 vs. 3.33, respectively, matched pairs, P < 0.01). There was also a significant decrease in the CAS between pre-treatment to last follow-up (6 vs. 1, respectively, matched pairs, P < 0.01).

Serum TSI levels at presentation were higher among patients who were recommended to receive second-line treatment compared to patients who responded well to first-line IV methylprednisolone and needed no additional treatment either at presentation (2135% vs 1159%, respectively, ANOVA, P = 0.05) or after completion of first-line treatment (2201% vs. 986%, respectively, ANOVA, P = 0.043). There were no differences in TSH, T3, or T4 levels between patients who were recommended to receive second-line treatment and those who were not.

Finally, the CAS before the first treatment of patients who were recommended to receive second-line treatment was higher than that of patients who responded well to first-line IVMP and needed no additional treatment at presentation (6.21 vs. 3.78, respectively, ANOVA, P = 0.032). There were no group differences in CASs during or after first-line treatment.

Discussion

With more treatment options than systemic steroids becoming available, physicians are now looking for markers to help tailor specific treatment for each patient diagnosed with TED. In the current real-life study, we found that 44.68% of patients needed additional treatment following first-line IVMP monotherapy, representing a moderate response rate but similar to that reported by other recent studies [9, 10]. That group of patients had significantly higher TSI levels compared to patients who responded well to first-line IVMP monotherapy.

Few studies have examined the association between TSI and the clinical features of TED, and none specifically addressed the patients’ response to first-line treatment with IVMP [10, 14]. Ko et al. reported the results of a 12-month follow-up of 76 patients with newly diagnosed TED and found that both the TSI and the TSHR binding inhibiting immunoglobulin (TBII) levels declined over the first year of treatment and that both correlated to various degrees with clinical scores, such as CAS and NOSPECS [14]. That study did not assess response to treatment, but those authors did support our conclusion that the monitoring of TSI levels may help evaluate prognosis and aide with clinical decision making. TSI reportedly had a stronger and more consistent association with clinical activity and severity of TED than TBII [2]. Other works also found that TSI was more specific than TBII for TED prognosis and disease severity [15]. Based upon those reports, we prefer to follow-up TSI levels in our patients rather than TBII levels.

A small number of studies investigated factors associated with failure of first-line steroid treatment for TED. Liyun et al. reported an association between baseline serum microRNA-224-5p and response to treatment with IVMP in a cohort of 9 patients with TED [16]. Those authors found that the combination of microRNA-224-5p with TRAb levels had a 91.67% positive predictive value and a 69.56% negative predictive value for steroid resistance. However, such testing may not be readily available in clinical practice. Ahn et al. reported a relatively low response rate of 44.7% to treatment with IVMP and demonstrated that extraocular muscle enlargement, younger age, and lower levels of TBII were predictive of a good response to steroids [10]. The correlation between elevated TSI levels and failure of first-line IVMP therapy that emerged in the current study serves as a useful and practical tool to predict which patients might be more difficult to treat.

The success rate of first-line IVMP monotherapy can be as low as 44–53% [9, 10]. It was 55.32% in our current study. The addition of first-line concurrent mycophenolate in the recent EUGOGO 2021 guidelines is based upon the results of two randomized trials, which found superior response rates in the combined treatment group at 12, 24, and 36 weeks [5, 9, 17]. In the EUGOGO study, which included 164 euthyroid patients with moderate-to-severe active TED, the combined treatment group demonstrated a response rate of 71% vs. only 53% in the IVMP-only group [5, 9]. The safety profile of mycophenolate was examined in several reports and found to be satisfactory [18, 19]. However, one study reported an increase in side effects in the combined treatment group vs. the IVMP-only group, which included gastrointestinal disorders (8.8% vs. 5.4%) and infections (7.1% vs. 5.4%) [18].

Despite the favorable risk-benefit ratio of the combined treatment, it is clear that a significant proportion of patients (about 53%) will do well with IVMP monotherapy [9]. Our challenge should be to identify the patients who can be expected to benefit from the combined treatment over steroids monotherapy. Our findings in the current study revealed that patients with lower TSI levels may do well with first-line IVMP monotherapy, thereby sparing them the potential side effects of MMF. Future prospective studies should verify this possibility.

This is a retrospective study with its inherent limitations. However, a strict and consistent treatment protocol at our center enabled us to present real-life reliable data of all patients who met the inclusion criteria without exception. Additionally, the question of cause and effect remains unanswered. Future studies will need to determine whether higher TSI activity reflects a different disease which is “steroid-resistant” or that it may represent a different phase of TED in a specific patient (i.e., on the patient’s specific theoretical Randall’s curve). If the former possibility is true for a given patient, then another treatment option should be considered. If the latter is true, it would be reasonable to repeat the use of steroids as a second-line treatment.

In conclusion, given the wealth of new treatment options, the pursuit of novel clinical and laboratory markers that can guide personalized treatment for patients with TED is of utmost importance. It could optimize the risk-benefit ratio for individual patients. The results of our study are one step in this direction.

Summary

What was known before

• First-line treatment of thyroid eye disease with intravenous methylprednisolone (IVMP) has a modest response rate. There are few clinical markers to predict poor response to first-line treatment.

What this study adds

• Patients requiring second-line treatment after failed IVMP monotherapy had higher baseline and post-first-line treatment serum thyroid stimulating immunoglobulin (TSI) levels. Those with elevated TSI may benefit from dual therapy (IVMP and mycophenolate) and require closer monitoring.

Author contributions

Conception/design of the research: OZ and OS. Data collection: AR. Analysis/interpretation: OZ and OS. Drafting the manuscript: OS. Revision of the manuscript: OZ, AR, AP, DLP, TCY, RS, NAL, and GJBS. Final approval of the manuscript: OZ, AR, AP, DLP, TCY, RS, NAL, GJBS, and OS.

Competing interests

The authors declare no competing interests.