Georges B. Nassrallah MD, FRCSC
Running Head: Teprotumumab-Related Adverse Events
1Temerty School of Medicine, University of Toronto, Toronto, Ontario, Canada
2Department of Ophthalmology and Vision Sciences, University of Toronto, Toronto, Ontario, Canada
3Department of Ophthalmology, St. Michael’s Hospital/Unity Health Toronto, Toronto, Ontario, Canada
4Department of Ophthalmology and Visual Sciences, University of Alberta, Edmonton, Alberta, Canada
5Department of Ophthalmology, Mount Sinai Hospital, Toronto, Ontario, Canada
Correspondence Address:
Oculoplastic and Reconstructive Surgery, Department of Ophthalmology & Vision Science
Mount Sinai Hospital, University of Toronto
Suite 1–001, 1–60 Murray St., Toronto, Ontario, Canada, M5T 3L9
A
Abstract
Background/Objectives:
Teprotumumab is the first FDA-approved medication for thyroid eye disease (TED), with proven efficacy in reducing proptosis and inflammation. However, emerging reports have raised concerns about systemic adverse events (AEs). This study aimed to evaluate systemic AEs associated with teprotumumab using a population-based pharmacovigilance approach.
Subjects/Methods:
This population-based observational study analyzed teprotumumab-related AEs reported in the Food and Drug Administration Adverse Event Reporting System (FAERS) database from January 2020—December 2024 using Open Vigil 2.1 software. Drug-specific AEs attributed to teprotumumab were compared to background FAERS reporting rates for other drugs. Disproportionality analyses were performed using reporting odds ratios (RORs) and Bayesian confidence propagation neural network algorithms to identify significant safety signals. Subgroup analyses by sex and age were also performed.
Results
A total of 3,787 AEs were attributed to teprotumumab over the study period. Teprotumumab was overreported for 28 systemic AEs (≥ 10 AEs with an ROR ≥ 10). Notably, many AEs pertained to ear and labyrinth disorders: permanent deafness (3.2%, n = 121/3,787; ROR = 9,827.3, 95%CI=[6,968.5, 13,859.1]), autophony (0.4%, n = 16/3,787; ROR = 2,366.6, 95%CI=[1,202.5, 4,657.5]), eustachian tube dysfunction (0.3%, n = 12/3,787; ROR = 147.8, 95%CI=[82.9, 263.6]), neurosensory deafness, auditory disorder, and tinnitus. Several other safety signals including but not limited to hyperglycaemia, muscle spasms, dysgeusia, menstrual disorders, onychoclasis, and gingival recession were also identified.
Conclusions
A
Heightened vigilance is needed to monitor early systemic AEs in patients administered teprotumumab. Proactive monitoring and intervention may mitigate the impact of these AEs, improving patient safety and outcomes. Future research should explore strategies to minimize or manage these AEs while preserving efficacy.
A
Introduction
Thyroid-associated orbitopathy, or thyroid eye disease (TED), is an autoimmune disorder affecting the orbital tissues and extraocular muscles [1, 2]. The most common sign of TED is lid retraction, and active disease typically presents with inflammation, proptosis, conjunctival redness, photophobia, diplopia, and, in severe cases, vision impairment [1, 3, 4]. Women are more likely than men to develop TED, though men tend to experience more severe symptoms at older ages [3, 5, 6]. Treatment for TED depends on disease severity, focusing on symptom management, inflammation reduction, addressing ocular complications, smoking cessation and other lifestyle modifications [2]. Pharmacologic treatments including corticosteroids, off-label tocilizumab and other immunosuppressive agents may not prevent disease progression, and all have significant side effects [7].
In January 2020, the U.S. Food and Drug Administration (FDA) approved Teprotumumab (TepezzaR) as the first targeted therapy for TED. Teprotumumab is a human monoclonal antibody that inhibits the insulin-like growth factor type I receptor (IGF-1R) [7, 8]. Clinical trials have demonstrated teprotumumab’s efficacy in reducing TED signs and symptoms, such as proptosis and inflammation [9, 10]. The phase III trial for Treatment of Graves' Orbitopathy to Reduce Proptosis with Teprotumumab Infusions (OPTIC trial) demonstrated that 83% of patients treated with teprotumumab achieved a ≥ 2 mm reduction in proptosis, compared to only 10% in the placebo group at 24 weeks [10]. There have been many reports concerning the safety profile of teprotumumab and the need for retreatment [10]. Due to the ubiquity of IGF-1 signaling, systemic inhibition can lead to various adverse effects [11–19]. Commonly reported side effects include muscle spasms, nausea, alopecia, fatigue, diarrhea, hearing impairment, dry mouth and hyperglycaemia [10, 12, 20]. On April 17, 2025, and May 7, 2025, the drug received regulatory approval in Canada and the United Kingdom, respectively, expanding global access despite persistent concerns over cost and toxicity. In the present investigation, we aim to conduct an updated, real-world pharmacovigilance analysis to evaluate the safety of teprotumumab using the FDA Adverse Event Reporting System (FAERS).
Methods
Database and Drug Criteria
Real-world pharmacovigilance data were sourced from the FAERS database, which compiles post-marketing adverse events reported to the FDA by healthcare professionals, patients, pharmacists, and the pharmaceutical industry. Adverse event reports were queried for the drug “teprotumumab” and classified using the Medical Dictionary for Regulatory Activities (MedDRA). Only AEs with at least 10 reports and those recorded as the primary suspect in these reports were considered. This criterion aimed to minimize confounding factors, as drugs not listed as the primary suspect (i.e., secondary suspect, concomitant, or interacting drugs) may show a statistical association with disproportionate adverse event reporting but may have less clinical significance. For each eligible AE report, the following data were extracted when available: generic drug name, year of report, patient age, sex, country, and indication.
Statistical Analysis
AEs attributed to teprotumumab and reported to the FAERS database between January 2020 and December 2024 were analyzed using Open Vigil 2.1 (Kiel, Germany) data-mining software. We compared the drug-specific AEs attributed to teprotumumab to the background rates for other drugs in the FAERS database. Disproportionality analyses were performed using reporting odds ratios (RORs). In this context, RORs indicate the likelihood of an adverse event being reported for teprotumumab relative to all other drugs in the FAERS database. Our criteria for identifying positive adverse drug reaction signals reflected the methodology outlined by Evans, Waller & Davis (2001), but instead of using a minimum case count of n > 2, we applied a more conservative threshold of n > 10 to increase confidence in the findings [21]. To further mitigate false-positive associations, we applied Bayesian confidence propagation neural network (BCPNN) algorithms, where drug-event combinations were considered disproportionately over-reported if the lower bound of the 95% confidence interval (CI) for the information component (IC) was positive (i.e., IC025 > 0). Subgroup analyses were also conducted to assess reporting patterns by patient sex and age. Age was stratified into four categories: 0–17, 18–44, 45–74, and 75 + years.
Results
Out of 13,237,811 AEs documented in the FAERS database during the study period, 3,787 (0.03%) were attributed to teprotumumab. The mean age of patients in these reports was 56.6 ± 14.7 years old, with 1,053 (27.8%) being female and 275 (7.3%) being male. The gender in 2,459 (64.9%) reports was not specified. Cases were reported from 2020 onwards, with half being reported in 2024 alone (47.6%, n = 1,801/3,787). Most cases were also reported in the United States (99.2%, n = 3,756/3,787), with a few reported from other countries (Table 1).
Table 1
Clinical characteristics of patients with teprotumumab AEs from the FAERS database.
Characteristics | Value |
|---|
Total teprotumumab-related events | 3,787 |
Mean age (± SD) | 56.6 (14.7) |
Sex | N (%) |
Female | 1,053 (27.3%) |
Male | 275 (7.3%) |
Not specified | 2,459 (64.9%) |
Reporting year | |
2020 | 205 (5.4%) |
2021 | 499 (13.2%) |
2022 | 605 (16.0%) |
2023 | 677 (17.9%) |
2024 | 1,801 (47.6%) |
Reporting location | |
United States | 3,756 (99.2%) |
Other | 19 (0.5%) |
Not specified | 12 (0.3%) |
By number of reports, the most common AEs for were muscle spasms (n = 671), fatigue (n = 429), tinnitus (n = 357), hypoacusis (n = 328), alopecia (n = 279), headache (n = 268), increased blood glucose (n = 265), nausea (n = 253), diarrhoea (n = 231) and deafness (n = 192). Teprotumumab was disproportionately overreported for 52 adverse events (≥ 10 events with an ROR ≥ 10). After excluding all the MedDRA AE terms related to eye disorders or ocular symptoms of TED, thyroid dysfunction, product issues, or social circumstances, we identified 28 systemic AEs associated with TED (Table 2).
Table 2
The top AE signals related to teprotumumab by signal strength.
| | Adverse Event | N (%) | ROR (95% CI) | IC025 |
|---|
1 | Deafness permanent | 121 (3.2%) | 9,827.3 (6,968.5, 1,3859.1) | 8.0 |
2 | Autophony | 16 (0.4%) | 2366.6 (1202.5, 4657.5) | 5.3 |
3 | Eustachian tube dysfunction | 12 (0.3%) | 147.8 (82.9, 263.6) | 4.0 |
4 | Deafness neurosensory | 69 (1.8%) | 130.0 (100.3, 168.5) | 5.4 |
5 | Auditory disorder | 24 (0.6%) | 102.7 (68.3, 154.3) | 4.4 |
6 | Gingival recession | 27 (0.7%) | 100.5 (68.5, 147.5) | 4.5 |
7 | Onychoclasis | 88 (2.3%) | 90.0 (72.6, 111.6) | 5.2 |
8 | Nail growth abnormal | 15 (0.4%) | 80.5 (46.4, 139.6) | 3.6 |
9 | Ear discomfort | 113 (3.0%) | 76.4 (63.1, 92.6) | 5.1 |
10 | Deafness bilateral | 16 (0.4%) | 60.8 (37.1, 99.8) | 3.6 |
11 | Tinnitus | 357 (9.4%) | 55.3 (49.6, 61.8) | 5.0 |
12 | Nail disorder | 53 (1.4%) | 47.1 (35.8, 61.9) | 4.2 |
13 | Deafness | 192 (5.1%) | 43.1 (37.2, 49.9) | 4.6 |
14 | Hypoacusis | 328 (8.7%) | 38.9 (34.7, 43.6) | 4.6 |
15 | Amenorrhoea | 91 (2.4%) | 32.4 (26.3, 40.0) | 4.0 |
16 | Deafness unilateral | 26 (0.7%) | 30.3 (20.6, 44.6) | 3.3 |
17 | Muscle spasms | 671 (17.7%) | 28.2 (25.9, 30.7) | 4.2 |
18 | Taste disorder | 53 (1.4%) | 18.8 (14.4, 24.7) | 3.2 |
19 | Glycosylated haemoglobin increased | 73 (1.9%) | 18.2 (14.4, 23.0) | 3.3 |
20 | Blood thyroid stimulating hormone decreased | 11 (0.3%) | 18.1 (10.0, 32.7) | 2.3 |
21 | Hyperglycaemia | 85 (2.2%) | 15.9 (12.8, 19.7) | 3.2 |
22 | Discomfort | 132 (3.5%) | 13.8 (11.6, 16.4) | 3.1 |
23 | Gingival pain | 15 (0.4%) | 13.3 (8.0, 22.1) | 2.2 |
24 | Diabetes mellitus | 162 (4.3%) | 12.3 (10.5, 14.4) | 3.0 |
25 | Menstrual disorder | 15 (0.4%) | 11.5 (6.9, 19.1) | 2.1 |
26 | Type 1 diabetes mellitus | 13 (0.3%) | 11.4 (6.6, 19.6) | 2.0 |
27 | Ear pain | 30 (0.8%) | 11.0 (7.7, 15.8) | 2.4 |
28 | Hair growth abnormal | 13 (0.3%) | 10.1 (5.7, 17.9) | 1.9 |
Ear and Labyrinth Disorders
Notable safety signals for teprotumumab were permanent deafness (3.2%, n = 121/3,787; ROR = 9,827.3, 95%CI=[6,968.5, 13,859.1], IC025 = 8.0), autophony (0.4%, n = 16/3,787; ROR = 2,366.6, 95%CI=[1,202.5, 4,657.5], IC025 = 5.3), eustachian tube dysfunction (0.3%, n = 12/3,787; ROR = 147.8, 95%CI=[82.9, 263.6], IC025 = 4.0), neurosensory deafness (1.8%, n = 69/3,787; ROR = 130.0, 95%CI=[100.3, 168.5], IC025 = 5.4,) auditory disorder (0.6%, n = 24/3,787; ROR = 102.7, 95%CI=[68.3, 154.3], IC025 = 4.4), ear discomfort (3.0%, n = 113/3,787; ROR = 76.4, 95%CI=[63.1, 92.6], IC025 = 5.1), bilateral deafness (0.4%, n = 16/3,787; ROR = 60.8, 95%CI=[37.1, 99.8], IC025 = 3.6), tinnitus (9.4%, n = 357/3,787; ROR = 55.3, 95%CI=[49.6, 61.8], IC025 = 5.0), deafness (5.1%, n = 192/3,787; ROR = 43.1, 95%CI=[37.2, 49.9], IC025 = 4.6), hypoacusis (8.7%, n = 328/3,787; ROR = 38.9, 95%CI=[34.7, 43.6], IC025 = 4.6), unilateral deafness (0.7%, n = 26/3,787; ROR = 30.3, 95%CI=[20.6, 44.6], IC025 = 3.3), and ear pain (0.8%, n = 30/3,787; ROR = 11.0, 95%CI=[7.7, 15.8], IC025 = 2.4) (Fig. 1).
Gingival Events
Gingival recession (0.7%, n = 27/3,787; ROR = 100.5, 95%CI=[68.5, 147.5], IC025 = 4.5) and pain (0.4%, n = 15/3,787; ROR = 13.3, 95%CI=[8.0, 22.1], IC025 = 2.2) were disproportionately reported for teprotumumab compared to other drugs in the FAERS database (Fig. 1).
Integumentary System Disorders
Safety signals related to integumentary system disorders such as onychoclasis (2.3%, n = 88/3,787; ROR = 90.0, 95%CI=[72.6, 111.6], IC025 = 5.2), abnormal nail growth (0.4%, n = 15/3,787; ROR = 80.5, 95%CI=[46.4, 139.6], IC025 = 3.6), nail disorder (1.4%, n = 53/3,787; ROR = 47.1, 95%CI=[35.8, 61.9], IC025 = 4.2), and abnormal hair growth (0.3%, n = 13/3,787; ROR = 10.1, 95%CI=[5.7, 17.9], IC025 = 1.9) were identified for teprotumumab (Fig. 1).
Endocrine Disorders
AEs related to endocrine disorders that were overreported with teprotumumab include increased glycosylated haemoglobin (1.9%, n = 73/3,787; ROR = 18.2, 95%CI=[14.4, 23.0], IC025 = 3.3), decreased blood thyroid stimulating hormone (0.3%, n = 11/3,787; ROR = 18.1, 95%CI=[10.0, 32.7], IC025 = 2.3), hyperglycaemia (2.2%, n = 85/3,787; ROR = 15.9, 95%CI=[12.8, 19.7], IC025 = 3.2), diabetes mellitus (4.3%, n = 162/3,787; ROR = 12.3, 95%CI=[10.5, 14.4], IC025 = 3.0), and type 1 diabetes mellitus (0.3%, n = 13/3,787; ROR = 11.4, 95%CI=[6.6, 19.6], IC025 = 2.0) (Fig. 1).
Reproductive System and Breast Disorders
Reproductive disorders such as amenorrhoea (2.4%, n = 91/3,787; ROR = 32.4, 95%CI=[26.3, 40.0], IC025 = 4.0) and menstrual disorder (0.4%, n = 15/3,787; ROR = 11.5, 95%CI=[6.9, 19.1], IC025 = 2.1) were overreported for teprotumumab (Fig. 1).
Other AEs
Other overreported AEs for teprotumumab included muscle spasms (17.7%, n = 671/3,787; ROR = 28.2, 95%CI=[25.9, 30.7], IC025 = 4.2), and taste disorder (1.4%, n = 53/3,787; ROR = 18.8, 95%CI=[14.4, 24.7], IC025 = 3.2) (Fig. 1).
Subgroup Analyses
The reporting of certain AEs differed by sex (Fig. 2). Females showed higher reporting of autophony (ROR = 1,407.9), auditory disorder (ROR = 75.8), gingival recession (ROR = 99.9), abnormal nail growth (ROR = 137.9), bilateral deafness (ROR = 42.6), amenorrhea (ROR = 23.4), unilateral deafness (ROR = 41.2), taste disorder (ROR = 22.1), decreased blood TSH (ROR = 27.4), gingival pain (ROR = 9.9), menstrual disorder (ROR = 10.6), ear pain (ROR = 7.4), and abnormal hair growth (ROR = 6.8). Males demonstrated higher reporting eustachian tube dysfunction (ROR = 566.4).
Certain AEs were reported more often in particular age groups (Fig. 3). The youngest group (0–17) did not show overreporting of any AE. The 18-44-year-old age group was more likely to report amenorrhea (ROR = 33.2) compared to all other age groups. The middle-aged group (45–74) was more likely to report permanent deafness (ROR = 10,776.9), nail growth abnormalities (ROR = 266.1), auditory disorders (ROR = 181.1), gingival recession (ROR = 86.4), unilateral deafness (ROR = 60.9), decreased blood TSH (ROR = 41.3), ear pain (ROR = 16.0), and hyperglycaemia (ROR = 9.1) compared to younger and older adults. Older adults (75+) did not report any AEs more frequently than middle aged adults but did show higher reporting of neurosensory deafness (ROR = 1,290.1) compared to the younger age group (18–44). All groups except 0–17 showed disproportionate reporting of ear discomfort and tinnitus.
Discussion
This study represents the most recent pharmacovigilance analysis to date on systemic AEs associated with teprotumumab use in a real-world population. Our findings emphasize notable safety signals for teprotumumab, particularly involving auditory, gingiva, integumentary, metabolic and reproductive systems. Given that over half of reported AEs occurred in 2024 alone and prescriptions for teprotumumab continue to rise, these emerging trends highlight the need for enhanced clinical monitoring and updated prescribing practices. Moreover, the potential latency in onset of certain adverse effects, coupled with recent data suggesting a 24% retreatment or failure rate, underscores the importance of long-term surveillance and individualized patient counselling [22].
Our analysis revealed a substantial burden of auditory AEs associated with teprotumumab, with over a third of signals involving the ear and labyrinth. Leading safety signals include permanent deafness, autophony, neurosensory deafness, eustachian tube dysfunction, auditory disorder, ear discomfort, bilateral deafness, tinnitus, and hypoacusis. These toxicities are likely caused by IGF-1R inhibition in cochlear tissue, disrupting cochlear homeostasis and outer hair cell survival, critical for auditory nerve function and contributing to sensorineural hearing loss (SNHL) [11, 13, 23, 24]. Conductive hearing loss may also result from nasopharyngeal fat pad atrophy impairing eustachian tube function [10, 13]. Auditory AEs such as hypoacusis, autophony, ear pressure, tinnitus, and SNHL have been consistently reported in clinical trials, post-marketing data, and real-world studies [25–29]. While hearing-related AEs occurred in only 7–12.2% of patients in the OPTIC trial, more recent observational data suggest a broader and potentially more severe ototoxicity profile and burden [9, 10, 29]. Sears et al. reported new auditory complaints in 81.5% of patients after an average of 3.8 infusions, while Belinsky et al. documented cases ranging from mild hearing changes to irreversible SNHL and permanent deafness [13, 27].
Updated 2024 data show a marked increase in both the frequency and strength of otologic AE reports compared to earlier analyses performed through early 2023 [24, 30, 31]. Reports of permanent deafness rose from 14 to 121, with the ROR escalating from 1,552.4 to 9,827.3—making it the most disproportionate AE [24]. In contrast, autophony decreased from an ROR of 4,188.3 to 2,366.6, albeit still a strong signal [24]. Other events, such as eustachian tube dysfunction (ROR = 147.8) and neurosensory deafness (ROR = 130.0), also demonstrated strong signals, while earlier signals such as ear swelling declined [24]. These evolving trends may warrant regulatory reassessment and highlight the need for proactive audiologic monitoring, particularly for patients with pre-existing hearing vulnerabilities.
Middle-aged adults (45–74 years) disproportionately reported auditory AEs, such as permanent deafness and auditory disorder, likely reflecting TED’s typical onset and greater symptom awareness compared to older or younger age groups. Females reported more autophony and ear pain, consistent with higher TED prevalence and healthcare engagement, compared to males that are less frequently affected [3, 5]. Males, however, showed higher reporting of eustachian tube dysfunction—possibly due to anatomical differences, though this finding may also reflect reporting bias, random variation, or the still-limited sex-stratified data on teprotumumab-associated AEs. Notably, both sexes demonstrated high reporting of permanent deafness. Although ototoxicity can occur irrespective of baseline SNHL, ototoxic drugs, or noise exposure, Sears et al. found increased hearing AEs in patients with prior audiologic conditions [27]. The authors also found that although there was a higher incidence in older age groups, the difference was not significant, aligning with our age group distribution (45–74) [27].
Previous studies found that symptom onset typically occurred between the third and fourth infusions (6–9 weeks), though it can range from the first infusion up to 27 weeks post-treatment initiation [32]. While many symptoms resolved after treatment cessation, SNHL often persisted. Remission rates vary (33–100%, median ~ 50%), with tinnitus and ear fullness resolving nearly completely, but SNHL persisted in about 45.5% of cases [27, 32]. Given teprotumumab’s ~ 20-day half-life, monitoring should continue up to 100 days after discontinuation [27, 33, 34]. Early tinnitus may serve as a warning for impending hearing loss and justify drug discontinuation [31]. However, currently, no proven strategies exist to prevent teprotumumab-induced hearing loss. Routine otolaryngologic evaluation—including baseline and serial audiometry (preferably at high-frequencies), and PET testing in high-risk patients (e.g., older adults, smokers, pre-existing hearing loss or ototoxic medication use) is recommended [25, 27]. Given the wide reported AE prevalence and risk of irreversible damage, the integration of standardized auditory monitoring into treatment protocols may be considered. Physicians should educate patients about ototoxic risks, encourage early symptom reporting, and maintain a high index of suspicion [27, 28]. Prospective studies are needed to establish effective screening and management guidelines.
Our analysis supports novel findings by Wang et al. which suggest an association between teprotumumab and gingival AEs–specifically gingival recession and pain–which were not documented in the original prescribing information [31]. Gingival recession demonstrated a notably high ROR, indicating a strong pharmacovigilance signal. These findings are significant, as oral and dental side effects were absent from initial clinical trial data and official labeling. IGF-1 is critical for gingival homeostasis, promoting collagen synthesis via insulin-like growth factor binding protein-5 (IGFBP-5); thus, IGF-1 receptor (IGF1R) inhibition may impair tissue repair and contribute to gingival recession, fibrosis, or sensitivity [14]. Additionally, mucosal barrier disruption may increase susceptibility to secondary oral complications, highlighting the need for heightened clinical awareness and routine oral health monitoring during treatment.
Nail abnormalities also emerged as a notable AE of teprotumumab in our investigation. Our analysis showed strong signals for onychoclasis, abnormal nail growth, and general nail disorders. IGF-1 is essential for keratinocyte turnover and nail matrix health; its inhibition may lead to fragility, slowed growth, and dystrophic changes such as brittleness or splitting [15, 16]. These events were reported more often in middle-aged (45–74) and younger adults (18–44), possibly due to faster baseline nail growth and greater sensitivity to cosmetic or functional changes. In contrast, older adults may underreport these effects due to slower nail turnover or attribution to aging [35]. AEs like onychoclasis, abnormal nail growth, and general disorders have shown increased signal strength over time, while earlier signals such alopecia and dry skin have declined, possibly reflecting reduced incidence, reporting variability, or stricter detection criteria [31].
Metabolic-related AEs were also prominent, with hyperglycaemia, increased glycosylated haemoglobin, and new-onset diabetes mellitus disproportionately reported for teprotumumab. These effects are biologically plausible given the role of IGF-1R in glucose metabolism and insulin sensitivity [17, 18]. IGF-1R shares partial homology with the insulin receptor (IR), allowing formation of hybrid receptors in insulin-responsive tissues [36]. Teprotumumab’s inhibition of IGF-1R may disrupt this signaling, leading to insulin resistance, impaired glucose uptake, and hyperglycaemia–a recognized class effect of IGF-1R inhibitors [19]. Clinical trials initially reported mild-to-moderate hyperglycaemia in approximately 10% of patients, primarily those with pre-existing diabetes, with cases managed by adjusting diabetes medications and no need for treatment discontinuation [9, 10, 37]. However, real-world data suggest a higher incidence and severity than initially reported. In a cohort of 42 teprotumumab-treated patients, Amarikwa et al. found that 52% developed hyperglycaemia and 31% experienced an HbA1c increase of ≥ 0.5%, a clinically significant rise [38]. Notably, 41% of initially normoglycemic patients developed prediabetes, and only 36% of those who developed hyperglycaemia returned to their baseline glycemic status by the end of follow-up [38]. Risk was higher among older adults, individuals with pre-existing diabetes, and those of Asian or Hispanic ethnicity [38]. While patients with pre-existing diabetes or glucose intolerance may be at greater risk, hyperglycaemia can also develop de novo in those without prior metabolic dysfunction [38]. Furthermore, pharmacovigilance data and case reports have described more severe events, including hyperosmolar hyperglycaemic states and diabetic ketoacidosis, particularly in patients with baseline prediabetes [31, 39, 40]. Although these events were rare in our dataset and did not reach high reporting thresholds, it remains essential to recognize the potential for serious endocrine complications. In our subgroup analysis, middle-aged patients (45–74) exhibited higher reporting rates of hyperglycaemia, underscoring the value of vigilant monitoring of blood glucose levels throughout teprotumumab therapy.
Teprotumumab was also associated with various reproductive system AEs, notably amenorrhea and menstrual disorders, particularly in females aged 18–44. These effects are likely due to inhibition of IGF-1R, which is highly expressed in the endometrial epithelium and plays a key role in menstrual regulation [41]. In a retrospective study by Terrarosa et al., 75% of women aged 25–53 who completed teprotumumab treatment reported menstrual irregularities, although most cases resolved within three months [42]. Similarly, Shah et al. reported reproductive AEs (e.g., amenorrhea, vaginal dryness, and erectile dysfunction) in 12.2% of 131 patients, including amenorrhea in 11% (14 out of 131) of the total cohort [43]. These AEs typically began around 6.8 weeks after treatment initiation; 37.5% resolved within an average of 25.3 weeks, while 62.5% remained persistent at nearly one-year follow-up [43]. Our study supports these associations using real-world reports, underscoring the importance of counselling reproductive-aged women about the risk of menstrual disturbances during and after teprotumumab therapy.
Teprotumumab was also associated with musculoskeletal and dermatologic AEs in our analysis. Muscle spasms may occur, likely due to IGF-1R inhibition or electrolyte disturbances [34]. Hair texture or growth changes may also occur due to effects on follicular cycling. Common systemic AEs such as fatigue, dysgeusia, headache, diarrhea, alopecia, and nausea had low RORs (< 10), suggesting weaker or non-specific associations and possible underreporting. Because systemic administration is required to reach orbital fibroblasts, other IGF-1R-expressing tissues can be affected, resulting in off-target effects. While teprotumumab is primarily degraded via proteolysis rather than renal or hepatic clearance, the impact of organ impairment on systemic adverse events has not been fully studied.
Sex- and age-based analyses revealed important patterns in teprotumumab-associated AEs. Females were disproportionately affected by both auditory (e.g., autophony, auditory disorder, bilateral and unilateral deafness, and ear pain) and non-auditory AEs (e.g., amenorrhea, thyroidectomy, gingival recession, abnormal hair and nail growth, decreased blood TSH), likely reflecting a combination of biological susceptibility and greater healthcare engagement. Males, in contrast, had higher reports of eustachian tube dysfunction, although the mechanism is unclear. Middle-aged adults (45–74) had the highest RORs for a broad range of AEs, particularly permanent deafness, auditory disorders, metabolic disturbances (e.g., hyperglycaemia), and tissue-related effects like gingival recession and nail growth abnormalities. Surprisingly, older adults (75+) had lower reporting rates, including for auditory symptoms, despite being at higher baseline risk—likely reflecting under recognition, reduced attribution to medication, or more conservative prescribing.
Several limitations should be acknowledged. FAERS is a spontaneous reporting system subject to underreporting, reporting biases, and incomplete data, limiting causal inference or incidence estimation [44]. Differentiating AEs due to teprotumumab from those intrinsic to TED (e.g., ocular symptoms, thyroid dysfunction) is difficult, even with indication-based exclusions, potentially leading to missed or misattributed signals [6]. The absence of comprehensive clinical context, such as baseline comorbidities or concomitant medications, limits nuanced interpretation. Some ocular AEs may have been excluded as TED-related, possibly underestimating the drug’s safety profile. Common AEs like muscle spasms, nausea, and alopecia appeared frequently but had low reporting odds ratios, likely due to their nonspecific nature. In contrast, rarer, drug-specific AEs such as hearing loss showed high RORs despite fewer reports. Serious events demonstrated in literature such as diabetic ketoacidosis were infrequently reported, possibly reflecting low incidence or detection limits. Duplicate reports from multiple sources may also affect data interpretation. Overall, pharmacovigilance signal detection is hypothesis-generating, with identified signals requiring validation through prospective studies and ongoing monitoring.
Despite the limitations of spontaneous reporting systems, our study offers key strengths that build on the foundational work of Huang et al. By leveraging a more recent and demographically detailed FAERS dataset—47.6% of reports were from 2024 alone—we provide an updated view of teprotumumab’s evolving safety profile [24]. Our stratified analysis revealed a sharp rise in reports of permanent deafness (from 14 to 121) with a corresponding ROR increase (1,552.4 to 9,827.3), now the most disproportionate adverse event to date. We also identified emerging signals such as eustachian tube dysfunction and noted the decline of others like alopecia and dry skin, likely reflecting improved data granularity and growing clinical awareness.
Conclusion
In conclusion, our findings emphasize significant systemic risks associated with teprotumumab, particularly otologic and metabolic AEs, with notable sex- and age-related differences. As teprotumumab use continues to expand globally—despite its high cost, a reported 24% retreatment rate within one year of follow-up, and recent approval in other countries—enhanced screening protocols, targeted counselling for high-risk groups, and interdisciplinary monitoring are essential to safeguard patient outcomes [22]. Given global variability in access and the financial implications of therapy, pharmacovigilance studies such as this are increasingly important for resource planning and informed reimbursement decisions. Furthermore, with several other drug classes currently under investigation for thyroid eye disease—including additional IGF-1R inhibitors, IL-6 antagonists, and anti-FcRn agents—ongoing surveillance and post-marketing comparisons will be critical to guiding safe and evidence-based use across evolving treatment landscapes [45].