Teprotumumab for Thyroid Eye Disease: Efficacy and Impact on Orbital Decompression Utilization

ABSTRACT

Objectives

We evaluated the impact of teprotumumab therapy (TT) introduction on orbital decompression (OD) utilization for thyroid eye disease (TED) and compared rates of significant exophthalmos reduction (SER) and diplopia between TT and OD.

Methods

Newly diagnosed TED patients treated pre (2015–2020) and post teprotumumab (2020–2025) FDA approval were identified from the EHR. Utilization trends, diplopia rates, and immediate (≤ 6 weeks post‐intervention) and long‐term follow‐up SER (≥ 2 mm reduction) were studied.

Results

Of 406 TED patients, 53 (36.8%) underwent OD in the pre‐TT cohort (n = 144). Post‐teprotumumab approval (n = 262), primary OD was performed in 29 (11.1%), of which 3 (10.3%) were subsequently placed on TT; primary TT was used in 68/262 (25.9%) with subsequent OD performed in 9 (13.2%). The remaining patients used other treatments (corticosteroids, orbital radiation, rituximab, tocilizumab, etc.). In the overall cohort, immediate SER was achieved in 94.4% of OD versus 83.8% of TT patients (p = 0.028), with greater median reduction in the OD group (5 vs. 4 mm; p = 0.001). At long‐term follow‐up (median 13 vs. 12 months), SER was sustained in 85.4% of OD and 62.9% of TT patients (p = 0.042). Pre‐therapy diplopia was more prevalent in TT vs. OD patients (79.4% vs. 59.7%; p = 0.019). New‐onset posttreatment diplopia was higher in the OD group (20.7% vs. 5.9%; p = 0.008).

Conclusions

Rates of OD surgery declined following FDA approval of teprotumumab. OD achieved sustained and greater exophthalmos reduction but carried a higher risk of diplopia. Contemporaneous OD indications may have evolved to use in TT‐recalcitrant TED or those in need of urgent decompression.

Level of Evidence

4.

Teprotumumab has been proven effective in thyroid eye disease and associated exophthalmos management. Since its FDA approval in 2020, teprotumumab therapy has been a popular choice for managing exophthalmos which has resulted in decreased utilization of orbital decompression surgery, from 36.8% between 2015 and 2020 to 11.1% TED patients between 2020 and 2025, in a multicenter tertiary healthcare setting. Orbital decompression provided ≥ 2 mm exophthalmos reduction immediately post‐intervention as well as on long‐term follow up, and also resulted in greater reduction compared to teprotumumab. However, new onset post‐intervention diplopia rates were higher after decompression.

1. Introduction

Thyroid eye disease (TED) is the leading cause of adult proptosis, affecting 90–155 per 100,000 individuals [1]. Driven by thyroid‐stimulating autoantibodies and overexpression of thyroid‐stimulating hormone receptors on orbital fibroblasts, TED results in extraocular muscle and orbital fat hypertrophy [2, 3, 4]. Clinically, it presents with proptosis, lid retraction, and strabismus, which may progress to keratopathy or dysthyroid optic neuropathy [5].

Disease activity is assessed with the clinical activity score (CAS), with intravenous methylprednisolone historically serving as first‐line therapy, though its effect on proptosis is limited [6, 7]. Orbital decompression (OD) has therefore been the traditional treatment for significant proptosis.

The therapeutic landscape shifted with FDA approval of teprotumumab in 2020, an IGF‐1R inhibitor shown to reduce proptosis, inflammation, and diplopia [7, 8]. Early database studies suggest declining OD rates after its introduction [9, 10], but real‐world practice patterns remain unclear.

In this study, we first studied practice patterns and examined trends in OD utilization before and after the introduction of teprotumumab within a large, multi‐state healthcare system. We also compared the immediate and long‐term magnitude of exophthalmos reduction between patients undergoing OD and those receiving teprotumumab as first‐line therapy. A secondary objective was to evaluate and compare the incidence of new‐onset diplopia following each intervention, a topic not well characterized in existing literature.

2. Methodology

A real‐world retrospective cohort study was conducted within a tertiary‐level, multi‐center healthcare system, following approval from the institutional review board (IRB ID: 24–010882). The system's universal electronic health records (EHRs) were queried to identify patients diagnosed with TED during two distinct timeframes: the pre‐teprotumumab (pre‐TT) period from January 1, 2015, to December 31, 2019, and the post‐TT period from January 1, 2020, to January 1, 2025.

Patients included in the pre‐TT cohort were those newly diagnosed with TED who underwent surgical OD for exophthalmos management during that period. In the post‐TT cohort, patients newly diagnosed with TED who were managed either with OD or TT were identified. Patients in the teprotumumab group were included only if they had completed the full treatment course of eight intravenous infusions. Patients who received other biological agents for TED management, such as tocilizumab or batoclimab, were excluded.

TED patients were classified into two groups based on the first intervention received: the teprotumumab group and the OD group. The following data were extracted from individual health records and compared between groups: age, sex, date of TED diagnosis, duration of follow‐up (months), treatment details, CAS [11], and Hertel exophthalmometry measurements. Exophthalmometry values were recorded at three time points: (1) pre‐intervention (closest value within ≤ 6 weeks prior to OD or initiation of teprotumumab), (2) immediate post‐intervention (within ≤ 6 weeks following OD or after the eighth teprotumumab infusion), and (3) final follow‐up (last available measurement during the follow‐up period). In patients with bilateral exophthalmos, data from the more severely affected eye were analyzed.

Significant exophthalmos reduction (SER) was defined as a decrease of ≥ 2 mm in post‐intervention exophthalmometry compared to baseline, consistent with prior literature [12]. The presence of diplopia (pre‐ and post‐intervention), as well as the need for strabismus surgery for new onset diplopia, was also assessed in both groups.

All statistical analyses were performed using Stata BE/19.5. Data distribution was assessed through skewness and kurtosis. Continuous variables were reported as mean ± standard deviation for normally distributed data or median with interquartile range (Q₁–Q₃) when not normally distributed. Pearson's chi‐square test was used to compare categorical variables such as sex distribution, smoking status, SER rates (immediate and at follow‐up), and presence of diplopia. t‐Test was used to compare the age distribution, and the Wilcoxon rank‐sum test was used to compare other continuous variables including follow‐up duration, CAS, and exophthalmos reduction. A two‐tailed p ≤ 0.05 was considered statistically significant.

3. Results

3.1. Baseline Characteristics of the Study Cohort

A total of 406 TED patients were identified in the study period. Baseline characteristics of the pre‐TT and post‐TT cohorts are reported below.

3.1.1. Pre‐TT Cohort

Between January 1, 2015, and December 31, 2019, 144 patients were diagnosed with TED. This included 33 males and 111 females, with a mean age of 64.0 ± 14.9 years. Among these, 53 patients (36.8%) underwent OD to address proptosis and other TED‐related complications.

3.1.2. Post‐TT Cohort

Between January 1, 2020, and January 1, 2025, 262 TED patients were identified (59 males and 203 females; mean age 56.4 ± 16.4 years). Of these, 29 patients (11.1%) underwent OD as the initial intervention, and 68 patients (25.9%) received teprotumumab as first‐line intervention. Of the 68 TED patients who were managed with first‐line TT, nine (13.2%) underwent subsequent OD, and of the 29 patients who underwent first‐line OD surgery, three (10.3%) received subsequent TT for effective control of their complaints.

The remaining patients in both periods (63.2% in pre‐TT and 64.0% in the post‐TT cohort) were managed with alternative treatment options such as lubricating eye drops, selenium supplementation, other biologics (e.g., tocilizumab, batoclimab), orbital radiation, or corrective strabismus and lid surgery. Summary statistics for the pre‐ and post‐TT cohorts are presented in Table 1.

TABLE 1.

Summary statistics for the pre‐ and post‐TT cohorts.

	Pre‐TT period	Post‐TT period
N	144	262
Mean age (±SD) in years	64 (±14.9)	56.4 (±16.4)
Sex
M	33 (22.9%)	59 (22.5%)
F	111 (77.1%)	203 (77.5%)
Choice of therapy (OD vs. TT)	OD: 53 (36.8%)	OD: 29 (11.1%) TT: 68 (25.9%)

Abbreviations: OD, orbital decompression; SD, standard deviation; TT, teprotumumab therapy.

From the overall cohort of 406 patients, 82 patients (29 from post‐TT and 53 from pre‐TT cohort) were categorized into the OD group and 68 into the teprotumumab group.

3.2. Teprotumumab Group

Of the 68 patients in the teprotumumab group, 37 patients had both pre‐ and post‐intervention exophthalmometry readings. There were 13 males and 24 females, with a mean age of 58.8 ± 14.6 years. Smoking status was positive (currently smoking or history of smoking) in seven patients. CAS scores were available for 31 patients, with a median pre‐TT score of 3 (1–5). Of these, 19 patients had active disease (CAS ≥ 3), and 12 had inactive TED.

SER was achieved in 31 of 37 patients (83.8%) immediately posttreatment. The median exophthalmos reduction (n = 31) was 4 mm (2–5), with a maximum of 7 mm. Follow‐up exophthalmometry data were available for 27 patients, at a median of 12 months (6–20) after treatment completion. At follow‐up, 17 of 27 patients (62.9%) maintained SER, with a median reduction (n = 17) of 3.5 mm (2–4.5) and a maximum of 6.5 mm. Among the 10 patients who did not meet the SER threshold at follow‐up, 3 demonstrated worsening of proptosis, with a maximum increase of 1 mm.

Baseline diplopia was documented in 54 of 68 patients (79.4%). Four patients developed a new‐onset diplopia following teprotumumab therapy, all of whom were managed non‐surgically with prism correction.

3.3. OD Group

Among the 82 patients in the OD group, 54 had complete pre‐ and post‐intervention exophthalmometry readings. This group included 12 males and 42 females, with a mean age of 53.4 ± 14.2 years. Smoking status was positive (currently smoking or history of smoking) in 22 patients. CAS were available for 30 patients, who had a median pre‐OD score of 1 (1–3); 10 patients had active disease (CAS ≥ 3), and 20 had inactive disease.

SER was achieved in 51 of 54 patients (94.4%) immediately post‐OD. The median exophthalmos reduction (n = 51) was 5 mm (3–6), with a maximum reduction of 10 mm. Follow‐up exophthalmometry was available for 41 of these patients, at a median interval of 14 months (5–24.5) from the surgery date. SER was sustained in 35 of 41 patients (85.4%), with a median reduction (n = 35) of 4 mm (3–5) and a maximum of 9 mm. Of the 6 patients without SER at follow‐up, 4 showed worsening of ≥ 2 mm compared to preoperative measurements.

Pre‐intervention diplopia was noted in 49 of 82 patients (59.7%). Seventeen patients developed new‐onset diplopia post‐operatively; 7 required strabismus surgery and 10 were managed with prisms.

3.4. Comparison of Teprotumumab vs. OD

Compared to the OD group, a significantly smaller proportion of patients in the teprotumumab group achieved SER on immediate posttreatment exophthalmometry (p = 0.028). Among patients who achieved SER, the teprotumumab group also demonstrated significantly less median exophthalmos reduction (p = 0.001). A similar trend was observed at follow‐up, with fewer patients in the teprotumumab group maintaining SER (p = 0.008). However, among those who did achieve SER, even though the magnitude of exophthalmos reduction at follow‐up trended toward being lesser in the teprotumumab group, it missed the criterion for statistical significance (p = 0.058). Follow‐up durations were comparable (p = 0.60).

Pre‐treatment CAS was significantly higher in the teprotumumab group (p = 0.029). Also, fewer teprotumumab patients were active smokers or had a history of smoking (p = 0.051). Detailed comparisons between the two groups are presented in Table 2.

TABLE 2.

Summarized comparison of teprotumumab and orbital decompression groups.

	Teprotumumab group	OD group	p
N	37	54
Mean (±SD) age in years	58.8 ± 14.6	53.4 ± 14.2	0.046*
Sex			0.089
M	13	12
F	24	42
Smoking status			0.051
Active/history	7	22
N	30	32
Median pre‐intervention (Q1‐Q3) CAS	3 (1–5) (available for 31)	1 (1–3) (available for 30)	0.029
Patients achieving SER immediate post‐intervention (n 1)	31 (83.8%)	51 (94.4%)	0.028
Median exophthalmos reduction in n 1	4 (2–5)	5 (3–6)	0.001
Follow‐up exophthalmometry available for	27	41
Median follow‐up duration (Q1‐Q3)	12 (6–20)	14 (5–24.5)	0.60
Patients achieving SER on follow‐up (n 2)	17 (62.9%)	35 (85.4%)	0.008
Median exophthalmos reduction in n 2	3.5 (2–4.5)	4 (3–5)	0.058

Note: Bold text denotes statistically significant p‐values p < 0.05.

Abbreviations: CAS, clinical activity score; OD, orbital decompression; Q₁‐Q₃, Quartile 1–Quartile 3; SD, standard deviation; SER, significant exophthalmos reduction.

Baseline diplopia was more prevalent in the teprotumumab group (p = 0.019), while the incidence of new‐onset diplopia post‐intervention was significantly higher in the OD group (p = 0.008).

Proportion of patients without exophthalmometry reports was higher in the teprotumumab group than OD group (45% vs. 35%). On comparing available patient characteristics between the original cohort and those in the subgroup with complete exophthalmometry reports, no differences in age (teprotumumab group: p = 0.453, OD group: p = 0.465) or sex distribution (p = 1 for both groups) were found.

4. Discussion

This study is the first to evaluate real‐world trends in OD surgery as a first‐line intervention for TED following the approval of teprotumumab. Across our multi‐state, multihospital health system, we observed a marked decline in OD utilization as an initial intervention—from 37% (2015–2020) to 11% (2020–2025) among newly diagnosed TED patients (Figure 1). OD yielded a higher proportion of patients with SER (> 2 mm on Hertel's exophthalmometry), both shortly after intervention (< 6 weeks) and at a comparable median follow‐up (13 vs. 12 months). While OD demonstrated greater proptosis reduction at both timepoints, the long‐term difference narrowly missed statistical significance (p = 0.058). Notably, posttreatment new‐onset diplopia was significantly more frequent in the OD group, with nearly 40% requiring corrective strabismus surgery, compared to conservative management with corrective prisms in all affected teprotumumab patients.

FIGURE 1.

Diagrammatic representation of changing treatment patterns and declining utilization of orbital decompression from pre‐ to post‐teprotumumab approval era.

The clinical efficacy of teprotumumab in TED has been extensively demonstrated, particularly in reducing proptosis, inflammation, diplopia, and orbital discomfort [7, 9, 10, 12, 13, 14, 15]. As an IGF‐1R antagonist, its mechanism disrupts IGF‐1R‐mediated costimulation in orbital fibroblasts, exerting a robust anti‐inflammatory effect [16]. While early recommendations favored teprotumumab for patients with active TED (CAS ≥ 3) [12], subsequent studies have shown benefits even in chronic, inactive stages (CAS < 3) [17, 18]. This led to FDA approval for teprotumumab across all TED stages, promoting its widespread adoption [19]. These circumstances have favored utilization of TT to an increasing extent for TED patients. Consequently, the role of OD has diminished, which was primarily focused on immediate decompression in the active phase, or addressed stable proptosis in the chronic phase of TED. In contrast to OD's targeted anatomical effect and the risk of new onset diplopia [20], teprotumumab provides comprehensive therapeutic benefits while avoiding surgical risks. These differences likely underlie the evolving preference for teprotumumab as initial therapy. Our findings mirror data reported by Topilow et al. [9] and Gorelik et al. [10], who observed a post‐2020 decline in OD using CPT codes and national databases till 2023. However, our study adds novel insights by focusing specifically on first‐line intervention in newly diagnosed TED cases within a longer period, further reinforcing this treatment shift. The reduction in OD utilization from 37% to 11% within our cohort supports the increasing utilization of teprotumumab in patients who may have otherwise undergone OD surgery.

It is noteworthy that in both study periods, approximately two‐thirds of the cohort received alternative therapies, as detailed in the results. Selenium supplementation is typically the first‐line treatment for patients with mild TED, while intravenous glucocorticoids are recommended for those with moderate to severe disease [21]. A significant subset of patients respond favorably to these primary medical therapies, effectively avoiding the need for costly biologics or surgical decompression. Another subset, those with diplopia or lid retraction but without troublesome proptosis, are often managed with strabismus or eyelid repair procedures. Only the remaining portion of TED patients, primarily those with significant proptosis, proceed to choose between teprotumumab or OD. The use of orbital radiation and other biologics such as rituximab, tocilizumab, or batoclimab remains limited and is uncommon as a primary treatment modality for TED.

We also compared the efficacy of OD and teprotumumab in terms of proptosis reduction. Short‐term analysis (≤ 6 weeks post‐OD or at the 8th teprotumumab infusion) showed higher SER rates in the OD group (94%) with a median reduction of 5 mm, versus 84% and 4 mm in the teprotumumab group. While both groups included patients with active and inactive TED, the teprotumumab cohort had a significantly higher proportion of active disease (CAS ≥ 3: 61% vs. 33% in the OD group). Since teprotumumab is most effective in the inflammatory phase [7, 22], this difference may have confounded comparisons. In contrast, OD has a consistent mechanical effect on orbital volume, independent of disease activity. At longer follow‐up (median 12–13 months), 85% of OD patients maintained SER (mean reduction 4 mm) compared to 63% in the teprotumumab group (mean 3.5 mm). Interestingly, the OD group included a higher proportion of smokers, a known risk factor for TED severity and progression [23]. Despite this, they maintained greater SER, highlighting the robust effect of OD in mitigating proptosis. Nonetheless, the modest long‐term difference in proptosis reduction (4 mm vs. 3.5 mm) closely missed statistical significance (p = 0.058), suggesting a possible convergence in efficacy between modalities over longer follow up periods. Ting et al. [24] had similar findings where they observed superior proptosis reduction with OD (average 5 mm) versus TT (average 3 mm), though their measurements were limited to single post‐intervention timepoints (< 30 days of completing the last teprotumumab infusion and 3–6 months after OD).

In patients who failed to achieve SER, regression patterns differed markedly. In the teprotumumab group, only 30% of non‐responders had worsening proptosis, and none exceeded a 1 mm increase. Conversely, 67% of non‐responders in the OD group experienced worsening, all exceeding 2 mm. This divergence may reflect teprotumumab's ongoing disease‐modifying, anti‐inflammatory effect, which could attenuate progression even in partial responders. Surgical interventions like OD, lacking any influence on the underlying inflammatory process, do not provide this protective effect, making regression potentially more severe in surgical non‐responders.

Teprotumumab has also shown particular benefit in patients with lower fat‐to‐muscle ratios (FMR < 1.80) [24], indicating preferential action on extraocular muscle hypertrophy. Patients with vertical strabismus are expected to benefit most in terms of diplopia improvement, with patients with small angle deviations achieving even complete resolution [15]. On the other hand, OD is associated with the risk of precipitating diplopia, especially medial wall decompression [20]. This supports the preferable utilization of TT over OD in TED patients with baseline diplopia. Our data supports this distinction. Baseline diplopia was more prevalent in the teprotumumab group (74% vs. 60%, p = 0.019), likely reflecting a trend toward selecting TT for diplopic patients. Post‐intervention, new‐onset diplopia occurred more frequently in the OD group (20.7% vs. 5.9%, p = 0.008). Among those affected, nearly 40% in the OD group required strabismus surgery, whereas all affected teprotumumab patients were managed non‐surgically with prisms. Since prism correction is only viable for deviations without cyclotorsion and under 40 prism diopters [6], this suggests milder diplopia severity in the teprotumumab group. The underlying anti‐inflammatory action of teprotumumab likely limits muscle fibrosis and misalignment, reducing the severity of strabismus even in suboptimal responders.

Collectively, these findings reflect that teprotumumab might be the better first‐line intervention for many TED patients, particularly those with active inflammation, baseline diplopia, or a preference to avoid surgery. However, the reported adverse effects of TT (most commonly muscle cramps, hyperglycemia, and hearing loss) [17], high cost of biologic therapy and associated insurance approval issues, and lack of easy and widespread accessibility are important limitations that need to be acknowledged before wholeheartedly supporting TT as the overall preferred first–line intervention in TED. Despite the risks of inducing diplopia and its lack of disease‐modifying potential, OD may still be necessary in patients unresponsive to teprotumumab, in those presenting with DON, in an adjunctive role in those with post‐TT residual proptosis, and in those regions where teprotumumab is not available. As a result, it is essential for both otolaryngologists and ophthalmologists to maintain proficiency in this procedure. However, as clinical indications and frequency of OD continue to decline, there is a growing need to reinforce its inclusion in residency curricula to mitigate potential gaps in surgical training.

5. Limitations

Several limitations should be acknowledged. First, while the study was conducted within a single healthcare care system with both tertiary and primary practices, it may not reflect practices or demographics in community settings. As a retrospective real‐world analysis, it is subject to inherent biases, including non‐randomization and confounding from variables such as CAS, smoking status, disease severity, and patient characteristics. Follow‐up durations varied, and some pre‐ and post‐treatment data were incomplete. Teprotumumab group had slightly higher proportion of missing exophthalmometry measurements which could be due to data capture differences in surgical vs. non‐surgical interventions, although factors like follow‐up attrition and differences in treatment location need further exploration. Additionally, exophthalmometry assessments were conducted by different clinicians across three centers, introducing potential inter‐observer variability. Future prospective, randomized controlled trials are necessary to more definitively compare the efficacy and safety profiles of OD and teprotumumab and to validate these findings across broader populations.

6. Conclusions

The introduction of teprotumumab has led to a significant decline in the utilization of OD surgery as first‐line intervention for TED, as demonstrated in this real‐world, multistate, multihospital study. While OD remains more effective for achieving substantial exophthalmos reduction both in the immediate post‐intervention period and at long‐term follow‐up, teprotumumab offers additional advantages, including resolution of active orbital inflammation, improvement in diplopia, and avoidance of surgical risks that likely contribute to its increasing preference as initial intervention. However, the possible adverse effects and high cost of biologic therapy need careful consideration before acceptance of TT as the intervention of choice in TED.

The higher incidence and severity of new‐onset diplopia following OD, with a considerable proportion requiring strabismus surgery, may further dissuade its use as a first‐line option. In the current treatment landscape, indications for OD may increasingly be limited to cases of teprotumumab‐resistant proptosis or urgent decompression in sight‐threatening TED, such as dysthyroid optic neuropathy.

Given the limitations of this retrospective, non‐randomized study, further prospective, randomized controlled trials are needed to better define the evolving role of OD in the era of targeted biologic therapy.

Funding

The authors have nothing to report.

Conflicts of Interest

The authors declare no conflicts of interest.

Kumar N., Lança Gomes P., Jategaonkar G., Marino M. J., Miglani A., and Lal D., “Teprotumumab for Thyroid Eye Disease: Efficacy and Impact on Orbital Decompression Utilization,” Laryngoscope Investigative Otolaryngology 11, no. 1 (2026): e70351, 10.1002/lio2.70351.