Teprotumumab-Related Adverse Events in Thyroid Eye Disease: A Multi-Center Study

Abstract

Purpose:

To assess the duration, incidence, reversibility, and severity of adverse events (AEs) in patients with thyroid eye disease (TED) treated with teprotumumab.

Design:

Multi-center retrospective observational cohort study.

Participants:

Patients with TED of all stages and activity levels treated with at least 4 infusions of teprotumumab.

Methods:

Patients were treated with teprotumumab between February 2020 and October 2022 at 6 tertiary centers. AE metrics were solicited and recorded at each visit. AEs were grouped according to the United States FDA Adverse Event Reporting System.

Main Outcome Measures:

Primary outcomes measure: AE incidence and onset. Secondary outcome measures: AE severity, reversibility, duration, proptosis response, clinical activity score (CAS) reduction, and Gorman diplopia score (GDS) improvement.

Results:

The study evaluated 131 patients. Proptosis improved by 2mm or more in 77% (101/131) of patients with 3.0±2.1mm average proptosis improvement and 3.2 points average CAS reduction. GDS improved by at least one point for 50% (36/72) of patients with baseline diplopia. AEs occurred in 81.7% (107/131) of patients. Patients had a median of 4 AEs. Most patients’ AEs were mild (74.0%, 97/131), 28.2% (37/131) moderate, and 8.4% (11/131) severe. Mean interval AE onset was 7.9 weeks after the first infusion. Resolved AEs had a mean duration of 17.6 weeks. Forty-six percent (60/131) of patients had at least 1 persistent AE at last follow-up. Patients had a mean follow-up of 70.2±38.5 weeks after the first infusion. The most common type of AEs was musculoskeletal (58.0%, 76/131), followed by gastrointestinal (38.2%, 50/131), skin (38.2%, 50/131), ear and labyrinth (30.5%, 40/131), nervous system (20.6%, 27/131), metabolic (15.3%, 20/131), and reproductive system (12.2%, 16/131). Sixteen patients (12.2%) discontinued therapy due to AEs, including hearing loss (n=4), inflammatory bowel disease flare (n=2), hyperglycemia (n=1), muscle spasms (n=1), and multiple AEs (n=8).

Conclusions:

AEs are commonly reported while receiving teprotumumab treatment. Most are mild and reversible; however, serious AEs can occur and may warrant treatment cessation. Treating physicians should inform patients about the AE risk, properly screen patients prior to treatment, monitor patients closely throughout therapy, and understand how to manage AEs should they develop.

Insulin-like growth factor 1 (IGF-1) and its receptor, IGF-1R, play a key role in regulating tissue formation and remodeling in many different tissues throughout the body.^1–5 This interaction triggers a complex signaling cascade within the cell, ultimately leading to increased cell proliferation, differentiation, and survival. The overactivation of the IGF-1 signaling pathway has been implicated in the development and progression of various diseases, including thyroid eye disease (TED).³ TED is a potentially sight-threatening and disfiguring autoimmune condition that affects up to 50% of patients with Graves’ disease (GD) and 2% of patients with Hashimoto’s thyroiditis.^6–8 The insulin-like growth factor I receptor (IGF-1R) and the thyrotropin receptor (TSHR), which co-localize and are overexpressed by orbital fibroblasts in TED patients, are thought to be the auto-antigens involved in the pathophysiology of the disease.^9,10

Inhibition of IGF-1R has long been a target of pharmacologic interest, with phase 1 clinical trials initially investigating its inhibition on sarcomas, non-small cell lung cancer, and diabetic macular edema.^11,12 In 2020, teprotumumab became the first human anti-IGF-1R monoclonal antibody to be approved by the U.S. Food and Drug Administration (FDA) for the treatment of TED based on 2 randomized, placebo-controlled trials (RCTs).^6,9,13 These studies demonstrated a significant improvement in proptosis (≥ 2mm), clinical activity score (CAS, ≥ 2 points), and Gorman diplopia score (GDS, ≥ 1 level of improvement in patients with baseline diplopia) following treatment with teprotumumab when compared with placebo.^6,9,13

The IGF-1R is involved in multiple physiologic processes, IGF-1 has neurotrophic effects on cochlear hair cells and supports synapses.^3,14 It is also key in the development of the cochleovestibular ganglion (CVG) as evidenced by sensorineural hearing loss seen in patients with congenital syndromes causing low IGF-1 levels.^15–19 In the gastrointestinal (GI) tract, IGF-1 appears to play a crucial role in promoting the growth of the GI mucosa, preventing apoptosis, supporting the barrier function of the GI mucosa, and reducing inflammation.²⁰ In glucose metabolism, IGF-1 mediates insulin signaling, promotes the uptake of glucose in peripheral tissues, and inhibits gluconeogenesis via growth hormone inhibition.²¹ IGF-1 also plays an anabolic and catabolic role in skeletal muscle, and the suppression of IGF-1R has been linked to muscle atrophy which can then lead to muscle twitching and spasms.²² As IGF-1R is ubiquitous throughout the body, off-target effects of teprotumumab have been an ongoing concern.

The FDA has reported muscle spasms, hearing loss, hyperglycemia, nausea, alopecia, diarrhea, dry skin, dysgeusia, headache, and fatigue as systemic AEs associated with teprotumumab.^3,13 The pooled-analysis of AEs from the phase 2 and 3 clinical trials demonstrated that 80% of patients experienced AEs, though only 56% were determined to be teprotumumab-related adverse events (TAEs).¹³ The majority of AEs were mild or moderate and most resolved after discontinuation of teprotumumab.¹³ Serious AEs (SAEs) were noted in 7% (7/84) of the treatment group, with 4% (3/84) attributed to teprotumumab (inflammatory bowel disease (IBD), infusion reaction, and Hashimoto’s encephalopathy).

The RCTs, however, had limited inclusion criteria. The trials excluded patients less than 18 and greater than 75 and 80 years of age in the phase 2 and 3 trials, respectively. Those who had inactive disease (CAS < 4), long-standing disease (>9 months), compressive optic neuropathy in the preceding 6 months, uncontrolled diabetes, a history of orbital decompression, orbital radiation, a cumulative dose of 1g or higher of methylprednisolone, and any treatment with rituximab or tocilizumab were also excluded.^6,9 Patients were required to have good thyroid and glycemic control. Additional exclusion criteria were a history of a bleeding disorder, history of a malignant condition in the past year, drug or alcohol use disorder in the past 2 years, or an ophthalmic condition that could complicate study results.^6,9 Given the narrow inclusion criteria, the clinical trial data may not be generalizable to the entire population of patients with TED who may be at a higher risk of developing AEs. Lastly, the RCTs provided limited data regarding the assessment, duration, and management of AEs.

This study provides a comprehensive assessment of the real-world incidence of TAEs by assessing TAE duration, severity, and reversibility and providing recommendations for proper patient selection, screening, and monitoring throughout therapy.

Methods

This was a multicenter, retrospective observational cohort study of 131 consecutive patients with TED treated with teprotumumab between February 2020 and October 2022 at six tertiary centers. Senta Clinic treated and followed 15 patients, Bascom Palmer Eye Institute 9, Scheie Eye Institute 14, the University of Minnesota Department of Ophthalmology and Visual Neurosciences 25, Cedars-Sinai Medical Center 41, and Byers Eye Clinic 27. Intravenous teprotumumab infusions, dosed at 10 mg/kg for the initial infusion and 20 mg/kg for subsequent infusions, were given every 3 weeks for a planned total of 8 infusions. Patients were seen at baseline, throughout therapy, and at regular intervals after therapy. A detailed exam, including an evaluation of all current and past AEs, was performed at the discretion of the treating provider. Patients were excluded from the analysis if they completed fewer than 4 infusions or if a detailed AE evaluation was not performed. Each institution varied in its approach to collecting AEs: 15 patients reported AEs via a questionnaire, 68 patients were prompted about AEs, and 48 patients volunteered information. Providers were responsible for assessing if AEs were related to teprotumumab treatment; only TAEs were analyzed in this study.

Charts were queried for demographic data, including age, gender, ethnicity, and past medical history, including concurrent medical issues and medications, smoking history, prior treatment for GD, thyroid status, and prior TED treatment. Thyroid status was classified according to the free T4 (FT4) and total T3 (TT3) into three categories: normal FT4 and TT3, elevated FT4 or TT3, or low FT4. Three measures were used to evaluate treatment outcomes: CAS, proptosis (Hertel exophthalmometer), and GDS.

Primary outcome measures included the incidence and time of onset of each AE. Secondary outcome measures included AE severity, reversibility, and duration. Reduction in CAS, proptosis response, and GDS improvement were also recorded. AEs were defined as any unfavorable or unintended sign (including an abnormal laboratory finding), symptom, or disease temporally associated with the use of teprotumumab. TAEs that occurred after the first infusion and no later than 90 days after the last infusion were evaluated. AEs were grouped according to the United States FDA Adverse Event Reporting System.^23–25 These categories and several of the symptoms that fall within each can be seen in Table 1.

Table 1:

Adverse events that occurred in more than 5% of patients

		Number of Patients (%)	Mean Onset after 1st infusion (weeks)	Mean Duration of Resolved AEs (weeks)	Percent Resolved
Musculoskeletal	Muscle Cramps	76 (58)	7	23.7	55.30%
Gastrointestinal	Diarrhea	36 (27.4)			80.50%
	Nausea	18 (13.7)			55.50%
	All GI Disorders	50 (38.2)	6.1	16.2	74.00%
Ear And Labyrinth	Hearing Impairment	18 (13.7)			50.00%
	Tinnitus	16 (12.2)			68.80%
	Autophony	9 (6.9)			55.50%
	Ear Fullness	17 (13)			41.20%
	All Ear Disorders	40 (30.5)	9.1	23.5	52.50%
Skin	Alopecia	35 (26.7)			28.60%
	Onychoclasis	10 (7.6)			20.00%
	Dry Skin	11 (18.4)			54.50%
	Skin Rash	10 (7.6)			100.00%
	All Skin Disorders	50 (38.2)	10.7	23.9	44.00%
General	Fatigue	36 (30.5)	38	14.5	55.60%
Reproductive System and Breast	Menstrual Irregularity	16 12.2)	6.8	25.3	37.50%
Nervous System	Headache	15 (11.5)			73.30%
	Dysguesia	11 (8.4)			36.30%
	All Nervous System Disorders	27 (20.6)	8.1	15.3	66.70%
Metabolism And Nutrition	Hyperglycemia	14 (10.6)	1.8	23.6	71.40%

AE severity was graded using the Common Terminology Criteria for Adverse Events (CTCAE).²⁶ Grade 1 or mild AEs were monitored and did not require intervention. Grade 2 or moderate AEs required minimal, local, or noninvasive intervention. Grade 3 or severe AEs were either disabling or required hospitalization. Hearing impairment was defined as any patient self-reporting partial or complete hearing loss, muffled hearing, or decreased word recognition. The presence of hearing impairment was assessed subjectively and graded using the CTCAE. Hearing impairment was considered severe if a patient required hearing aids and/or the hearing changes were disabling or limited activities of daily living.

Onset and resolution of each AE were recorded as the time in weeks after the first infusion. The AE outcome was recorded as a) resolved, b) persistent or worsening, or c) unknown. Unknown resolution status was defined as outcomes that were either undocumented or when patients were lost to follow-up. Persistent symptoms were defined as those that had not resolved before the end of patient follow-up. Management alterations (no change, concomitant treatment, dose interruption, and drug discontinuation) were also documented. Treatment discontinuation was reported only if patients experienced intolerable AEs.

The study adhered to the tenets of the Declaration of Helsinki. It was performed in accordance with the Health Insurance Portability and Accountability Act and was approved by each site’s institutional review board.

Statistics

Statistical analysis and calculation of descriptive statistics were done using SPSS and MATLAB. Descriptive statistics were used to describe demographic data, time intervals, and the number of AEs; the most appropriate summary statistic was chosen given the variable type and distribution of the data. Categorical variables were summarized with percentages. If patients had multiple symptoms within one category, the average onset of those AEs were used in category-specific calculations.

Results

Patient Summary

A total of 131 TED patients with a mean age of 56.3 years (range 15–93 years) were included. Seventy-seven percent (101/131) of patients were female and 22.9% (30/131) were male. Caucasians accounted for 39.7% (52/131) of the study population, followed by Asian and Hispanic patients at 9.9% (13/131) each, and African Americans at 6.1% (8/131). Twenty-five percent (33/131) of patients reported a history of smoking (9 current). At baseline, 52.7% (69/131) of patients were euthyroid, 31.3% (41/131) were hyperthyroid, 6.9% (9/131) were hypothyroid, and 9.1% (12/131) did not report thyroid status. Most patients (91.6%, 120/131) had prior therapy for GD (radioactive iodine, anti-thyroid medication, thyroidectomy) and more than half (52.7%, 71/131) had a history of recalcitrant TED and were previously treated with steroids (36.6%, 48/131), orbital radiation (10.7%, 14/131), orbital decompression (18.3%, 24/131), or biologic therapy (4.6%, 6/131). The most prevalent past medical conditions were diabetes mellitus type 2 (20.6%, 27/131), hypertension (28.2%, 37/131), and malignancy (13%, 17/131). There were no significant differences in the demographic and clinical characteristics of patients with or without AEs except for hyperthyroidism, which was more prevalent in patients with AEs (Supplemental Table 2).

Treatment outcomes

Over a 33-month period, 131 study patients were treated with teprotumumab. Patients received an average of 7.4 infusions (range 4–8). A total of 101 (77.1%) patients completed the entire course of eight infusions. The mean baseline CAS was 4.1±1.5 (range 1–7) and reduction was 3.2±1.6 and 77% (101/131) of patients had a proptosis response of 2 mm or more in the treatment eye, with a mean reduction of 3.0±2.1 mm. GDS was available for 91 patients, with 79% (72/91) experiencing diplopia at baseline. Of the 72 patients with baseline diplopia, 50% (36/72) experienced an improvement of 1 point or more in GDS. CAS reduction, proptosis, and GDS were all measured at an average follow-up of 38.2±13.6 weeks after the first infusion. AEs were followed for a mean of 70.2±38.5 weeks after the first infusion.

Adverse Events Overview

Of the 131 patients, 81.7% (107/131) experienced at least one TAE. The cohort experienced a combined total of 463 AEs, with each patient experiencing a median number of 4 (1–19) AEs. Eighteen percent (24/131) of patients did not report any AEs, 9.9% (13/131) of patients had only one AE, and 71.8% (94/131) of patients had multiple AEs. The mean interval of AE onset was 7.9 weeks after the first infusion, with a mean AE duration of 17.6±17 weeks for the AEs that had resolution during the study period. Most patients experienced mild AEs (74.0%, 97/131), while a smaller proportion experienced moderate (28.2%, 37/131) or severe symptoms (8.4%, 11/131). Thirty-five percent (160/463) of all reported AEs did not resolve in this cohort. Forty-six percent (60/131) of patients (48 women and 12 men) had at least 1 persistent AE at average time of last follow-up after the first infusion (60.0±39.5 weeks). Twelve percent (16/131) discontinued treatment due to AEs and 10.7% (14/131) of patients were lost to follow-up.

AEs reported by more than 5% of patients are listed in Table 1; those experienced by less than 5% are listed in Supplemental Table 3. AE incidence was notably greater in this study compared to the pooled clinical trials (Table 4).¹³ Muscle spasms, fatigue, diarrhea, alopecia, nausea, hearing loss, and hyperglycemia were the most prevalent AEs in this cohort, but the probability of remaining AE-free after starting teprotumumab varies slightly between different classes of symptoms. Patients with diarrhea and nausea were the most likely to fully recover, and their symptoms were mostly mild. While hearing loss and hyperglycemia happened more infrequently, those patients were more likely to have severe symptoms (Figure 1). Mild musculoskeletal, gastrointestinal, and skin and subcutaneous AEs affected the most patients (Supplemental Figure 2).

Table 4:

A comparison of the incidence of AEs in the RCTs with AEs in this study.

	Our Data		Clinical Trial Data: Teprotumumab-Relateda,b
Adverse Event	Affected Patients	Percent	Affected Patients	Percent
Any AE	107	82%	47	56%
Muscle Spasm	76	58%	16	19%
Fatigue	36	27%	3	4%
Diarrhea	36	27%	7	8%
Alopecia	35	27%	8	10%
Nausea	18	14%	8	10%
None	24	18%	37	44%
Hearing Change	40	31%	4	5%
Headache	15	11%	5	6%
Menstrual	14	11%	4	10%c
Dysgeusia	11	8%	4	5%
Hyperglycemia	14	11%	7	8%
Skin Rash	10	8%	3	4%
Dry Skin	11	8%	4	5%
Onychoclasis	10	8%	4	5%
AEs resulting in discontinuation of treatmentd	16	12.2%	3	4%

^aProbably or possibly related to teprotumumab as assessed by investigator.

^bSource: Kahaly et. Al. Teprotumumab for patients with active thyroid eye disease: a pooled data analysis, subgroup analyses, and off-treatment follow-up results from two randomised, double-masked, placebo-controlled, multicentre trials. Lancet Diabetes Endocrinol. 2021.

^cOnly reported in Phase II clinical trial, 4/42 patients with reproductive AEs

^dHearing impairment, IBD, hyperglycemia, and severe muscle cramps and nausea led to discontinuation in our study cohort. IBD, infusion reaction, and Hashimoto’s encephalopathy led to discontinuation in the clinical trials.

Figure 1:

Resolution status of the most common adverse events were stratified by severity. The incidence of each adverse event is reported as a number under each adverse event on the x-axis. The solid bar indicates the proportion of adverse events in each category that were resolved, and the slashed bar indicates adverse events that were persistent. The black, gray, and white shades indicate adverse events that were graded as mild, moderate, and severe according to the Common Terminology Criteria for Adverse Events.

Musculoskeletal (MSK) and Connective Tissue Disorders

Muscle spasms affected 58.0% (76/131) of patients (mean age 55.7 years) during treatment. Forty-six percent (60/131) of patients experienced mild muscle spasms, 11.5% (15/131) had moderate spasms, and 0.8% (1/131) had a severe manifestation. Seventeen patients received treatment, including vitamin/mineral supplementation (n=4), magnesium (n=3), increased fluids (n=1), massage (n=3), muscle relaxants (n=3) and non-steroidal anti-inflammatory drugs (NSAIDs) (n=3). Symptom onset was 7.0 weeks after the first infusion; spasms resolved in 55% (42/76) of patients. In patients with resolution, the mean duration of symptoms was 23.7 weeks. Forty percent (30/76) of patients had persistent spasms at average follow-up of 55.5±41.6 weeks after the first infusion. Resolution status was unknown for 5.3% (4/76) of patients. Six patients discontinued in part due to musculoskeletal AEs.

Gastrointestinal (GI) Disorders

GI disorders affected 38.2% (50/131) of patients (mean age 52.6 years). Diarrhea (27.5%, 36/131) and nausea (13.7%, 18/131) were the two most common GI AEs. Less common symptoms included constipation (3.8%, 5/131), abdominal cramps (0.8%, 1/131), vomiting (1.5%, 2/131), and bloody stools (2.3%, 3/131). Thirty-five percent (45/131) had mild symptoms and 2.3% (3/131) moderate. Two percent (2/31) had severe symptoms, one with nausea and vomiting and another with an IBD flare. Three additional patients received anti-diarrheal medications and 1 zinc sulfate monohydrate cream for painful bowel movements. The mean symptom onset was 6.1 weeks and 74% (37/50) of these patients reported symptom resolution. Resolved AEs had a mean duration of 16.2 weeks. Twenty-two percent (11/50) had persistent symptoms at average follow-up of 44.6±42.6 weeks. Resolution status was unknown for 4.0% (2/50) of patients. Four patients discontinued teprotumumab, at least in part, due to teprotumumab-related GI AEs, 2 of whom experienced IBD flares. Of note, 1 additional patient with a history of IBD did not experience GI symptoms during treatment.

Skin and Subcutaneous Tissue Disorders

Thirty-nine percent (50/131) of patients (mean age 55.9 years) experienced at least one or more skin disorders. Twenty-seven percent (35/131) had alopecia, 8.4% (11/131) dry skin, 7.6% (10/131) onycholysis, and 7.6% (10/131) skin rash. Thirty-five percent (46/131) of patients experienced mild skin AEs and 3.1% (4/131) moderate AEs. Four patients were treated with topical creams (n=2), hydrocortisone + diphenhydramine cream (n=1), and diphenhydramine (n=1). The mean onset was 10.7 weeks after the first infusion and 43% (22/51) of these patients had complete resolution. Resolved AEs had a mean duration of 23.9 weeks. Fifty-two percent (26/50) had persistent findings at average follow-up of 70.5±42.1 weeks. Resolution status was unknown for 4.0% (2/50) of patients.

Ear and Labyrinth Disorders

Thirty-one percent (40/131) of patients (mean age 56.2 years) experienced at least one otologic AE. Ten of these patients had a prior history of subjective hearing impairment; of the 12 patients with baseline audiograms, 7 demonstrated baseline hearing impairment. Nineteen had a recent history of ototoxic drug use, including NSAIDS (n=11), serotonin and norepinephrine reuptake inhibitors (n=2), vancomycin (n=1), diuretics (n=5) and aspirin (n=1). During treatment, 13.7% (18/131) of patients reported new or worsening hearing impairment, of which 8 were confirmed by audiogram. Thirteen percent (17/131) also experienced ear fullness, 12.2% (16/131) tinnitus, 6.9% (9/131) autophony, and 3.8% (5/131) otalgia. Twenty-three percent of (30/131) of symptoms were mild, 0.8% (1/131) moderate, and 6.9% (9/131) severe. Hearing aids were recommended for 8 patients with severe hearing impairment at the discretion of the audiologist. Two patients were treated with a topical vitamin supplement ear drop for ear plugging and 1 with systemic steroids for hearing impairment. The mean symptom onset was 9.1 weeks and 53% (21/40) of patients had complete resolution of otologic symptoms. Resolved AEs had a mean duration of 23.5 weeks. Forty percent (16/40) had persistent otologic symptoms at average time of last follow-up (60.8±41.3 weeks). Resolution status was unknown for 3 (2.3%). When specifically evaluating subjective hearing impairment, 50% (9/18) of patients had persistent symptoms at average follow-up of 81.1±38.4 weeks, including 4 of the 6 patients that discontinued treatment at least in part due to otologic AEs. Two patients interrupted treatment due to hearing impairment but later completed the full course. Of note, 2 additional patients with baseline hearing impairment confirmed via audiogram did not experience hearing changes with teprotumumab.

Nervous System Disorders

Neurologic disorders occurred in 20.6% (27/131) of patients (mean age 53.5 years). Eleven percent (15/131) of patients experienced headaches and 8.4% (11/131) dysgeusia. Most nervous system AEs were mild (19.1%, 25/131) and 0.8% (1/131) moderate. One percent (1/131) had a severe syncopal event. Most neurologic AEs (85.2%, 23/27) did not require management, and one required concomitant therapy with acetaminophen for a mild headache. The mean onset of nervous system AEs was 8.1 weeks and resolved in 66.7% (18/27) of patients. Resolved AEs had a mean duration of 15.3 weeks. Neurological AEs were persistent in 33.3% (9/27) of patients at average follow-up of 68.6±48.9 weeks.

Metabolism and Nutrition

Fifteen percent (20/131) of patients (mean age of 56.3 years) had metabolism or nutrition-related AEs. Hyperglycemic events were the most common (10.7%, 14/131). Ten patients had a history of type II diabetes mellitus. Decreased appetite affected 2.3% (3/131) and weight loss 2.3% (3/131) of patients. Eight percent (11/131) had mild AEs, 6.1% (8/131) moderate, and 0.8% (1/131) had a severe hyponatremic event. The mean onset of metabolic AEs was 3.2 weeks and 65% (13/20) of patients had AE resolution. Resolved AEs had a mean duration of 22.7 weeks. Thirty-five percent (7/20) had persistent AEs at average follow-up of 74.9±40.0 weeks. Five patients discontinued in part due to metabolic AEs.

Reproductive System Disorders

Reproductive AEs occurred in 12.2% (16/131) of patients (15 females and 1 male) with a mean age of 39.6 years. Eleven percent (14/131) of patients reported amenorrhea, 0.8% (1/131) vaginal dryness, and 0.8% (1/131) erectile dysfunction. All cases of menstrual irregularity were mild (10.0%, 13/131) or moderate (0.8%, 1/131). Most patients did not require management; only 1 with amenorrhea was treated with concomitant oral contraceptives and the 1 case of erectile dysfunction was treated with tadalafil. The mean onset of reproductive AEs was 6.8 weeks after first infusion and resolved in 37.5% (6/16) of patients. Resolved AEs lasted on average for 25.3 weeks. Reproductive AEs remained persistent in 62.5% (10/16) at average follow-up of 55.6±50.2 weeks.

Adverse Event-Related Teprotumumab Discontinuation

Sixteen (12.2%) patients (13 women and 3 men), mean age of 69 years (45–93), prematurely discontinued teprotumumab infusions due to AEs. This subgroup completed a mean number of 5.2 infusions (range 4–7) and an average baseline CAS of 3.6 with an average reduction of 2.8. Of these patients, 4 had isolated ear and labyrinth AEs, 2 isolated GI AEs, 1 an isolated metabolic AE, 1 an isolated musculoskeletal AE, and 8 a combination of AEs. Seven patients experienced severe AEs, 7 experienced moderate AEs, and 2 mild AEs.

Six patients, all female, discontinued teprotumumab due to otologic AEs (including hearing impairment, autophony, ear plugging, and tinnitus); 4 (ages 64, 65, 78, and 82) solely due to these symptoms and 2 (ages 69 and 84) due to hearing impairment and several other AEs. Three of these 6 patients had a history of ototoxic medication usage (albuterol, amlodipine, atenolol, atorvastatin, and telmisartan) and 1 had a history of smoking.²⁷ Five of these 6 patients had baseline audiometry testing: 2 had severe baseline hearing impairment (1 had hearing loss secondary to trauma and Meniere disease), 2 had mild baseline hearing impairment in a single ear, and 1 had normal hearing. Five patients demonstrated sensorineural hearing loss on post-treatment audiometric testing; 2 of these patients also developed patulous eustachian tube dysfunction. Five stopped teprotumumab after failure to recover hearing despite a pause in treatment, and the sixth primarily due to her concurrent AEs. Three patients began using hearing aids, one refused hearing aids, one was monitored closely by audiology without treatment. Hearing impairment persisted at last follow-up for 4 of these 6 patients; 1 patient self-reported that her hearing returned to baseline after discontinuation and 1 demonstrated return to baseline hearing status per an audiogram 33 weeks after her last infusion.

Two female patients, ages 45 and 76, discontinued due to new onset IBD flares. Neither patient had a history of smoking, recent hormone therapy, infectious gastroenteritis, or a history of GI disorders. The 45-year-old patient had a family history of IBD in one first-degree and one second-degree relative. She was hospitalized for bloody diarrhea and fecal urgency after her fifth infusion. She was diagnosed with ulcerative colitis and treated with infliximab, prednisone, methotrexate, mesalamine, pantoprazole, and budesonide. She discontinued teprotumumab after the 7^th infusion due to worsening symptoms.^28,29 The 76-year-old patient presented with a moderate IBD flare after her first teprotumumab infusion. She was co-managed with a GI specialist and treated with loperamide and budesonide. She discontinued treatment after the 7^th infusion due to her AEs and lack of a robust response; her GI symptoms resolved shortly after.

Four patients, 1 with a history of pre-diabetes and 3 with Type II diabetes, discontinued treatment due to metabolic symptoms; 1 solely due to hyperglycemia and 3 due to hyperglycemia as well as other symptoms. The first patient was a 55-year-old male with a history of pre-diabetes. He developed elevated blood sugars after his first infusion, treated with metformin. He discontinued teprotumumab due to hyperglycemia and poor response to therapy. All four patients’ worsening hyperglycemia resolved after discontinuation.

Four patients discontinued due to muscle spasms. A 67-year-old female discontinued teprotumumab after the fifth infusion due to mild but intolerable muscle spasms. Three others discontinued due to muscle spasms and other symptoms. All muscle spasm symptoms resolved after discontinuation.

Eight patients discontinued teprotumumab treatment due to multiple causes. An 84-year-old female developed hearing loss and severe hyponatremia. She also developed depression-related cognitive dysfunction and hyperglycemia leading to treatment cessation after the fourth infusion. A 69-year-old female with mild baseline hearing loss developed hearing decline during treatment; she also developed an ear infection, dry eye, and poorly controlled hyperglycemia that led to discontinuation after the fourth infusion. A 76-year-old female with a history of Type II diabetes, hypertension, and hyperlipidemia developed hyperglycemia, hair loss, fatigue, hypertension, and pruritis that led to discontinuation after 6 infusions. A 56-year-old male with Type II diabetes mellitus developed poorly controlled hyperglycemia, urinary insufficiency, muscle spasms, fatigue, and nail changes. He discontinued after the fourth infusion and symptoms resolved shortly thereafter. A 77-year-old female presented to the ED after the fifth infusion led to severe vomiting, nausea, anxiety, and worsening of already severe muscle spasms. She was treated with calcium citrate 950 mg BID for the spasms but discontinued teprotumumab after the sixth infusion. All symptoms resolved after teprotumumab was discontinued. A 54-year-old male developed moderate muscle spasms, diarrhea, and nausea treated with meclizine. Though CAS decreased by 2, he had no objective improvement in proptosis and there was no change in diplopia. His AEs, coupled with a poor response to treatment, led to discontinuation of teprotumumab after the fifth infusion. A 70-year-old female with baseline anxiety developed severe panic attacks that required treatment with alprazolam after two infusions. Though she also experienced mild muscle spasms, her psychiatric symptoms were the main cause for discontinuation after the fourth infusion. Her anxiety improved after stopping treatment. Finally, a 93-year-old female found her mild muscle spasms and fatigue intolerable and discontinued after the fourth infusion.

Discussion

Targeted therapies, including teprotumumab, have changed the treatment landscape for many ocular and orbital inflammatory diseases. This trend is expected to increase as several investigational drugs are being studied for the treatment of TED.⁷ Therefore, it is important to familiarize ourselves with the AE profile of targeted therapies to safely treat patients. Two pivotal RCTs reported TAEs in 47 of 84 patients treated with teprotumumab; however, these studies reflect data from a carefully selected population that may not be representative of the patients receiving teprotumumab in clinical practice.^6,9,13 Recent case reports and case series have highlighted SAEs with teprotumumab use; however, reports thus far include small numbers with limited follow-up.^28–32 Large studies reflecting data after the FDA’s approval of teprotumumab are lacking. This study is the first field analysis of AEs in the largest cohort of teprotumumab-treated patients to date.

This study population included patients treated with teprotumumab for TED without exclusion due to disease activity, chronicity, severity, past medical history, prior therapies, glycemic status, or thyroid function status. Additionally, the follow-up timepoints, patient compliance, and AE reporting were representative of typical practice and differed significantly from the carefully designed RCT setting. Yet, the treatment outcomes demonstrated similar reductions in proptosis and the CAS when compared to the clinical trials. This is consistent with post-approval studies demonstrating that teprotumumab is effective for the treatment of both active and chronic moderate-to-severe disease.^33,34

In this study, over 80% of patients treated with teprotumumab for TED experienced at least one TAE, and the majority of patients experienced 4 or more TAEs. While most AEs were mild, 28.2% were moderate and 8.4% were severe. Almost half of these patients were reported to have at least one persistent AE at average follow-up of 70.2±38.5 weeks after the first infusion. Of the 463 AEs collectively experienced by this cohort, muscle spasms, alopecia, fatigue, and hearing loss made up most of the 160 persistent AEs. Twelve percent of patients discontinued teprotumumab treatment due to AEs, mainly related to hearing loss, IBD and hyperglycemia.

When comparing this cohort’s AE profile to the clinical trials, we found more TAEs (81.7% vs. 56%), more teprotumumab-related SAE’s (7.6% vs. 4%), and more individual AEs in each AE category. These results also demonstrate more persistent AEs at 70-week follow-up (45.8% vs. 39%) compared to a 72-week follow-up study of the pooled AE’s from the clinical trials.¹³ Notably, this study did not include a placebo arm, however, the clinical trials demonstrated that 70% of patients in the placebo arms experienced AEs and 1% experienced a SAE.¹³ The increased frequency of AEs in this study can be explained by the differences in the patient populations being studied and differences in AE data gathering and reporting. Yet, this cohort represents 6 tertiary centers across the US, therefore, these results may be more generalizable to providers treating patients with TED.

This study found similar rates of IBD flares and hyperglycemia in teprotumumab-treated patients compared to the clinical trials, however, the clinical trials found that both AEs were associated with a history of pre-existing IBD and diabetes, respectively. The FDA warning label cautions against exacerbation of preexisting IBD; however, 2 patients in this cohort experienced new-onset IBD symptoms during treatment with teprotumumab and both discontinued therapy.³⁵ Neither patient had a history of IBD, demonstrating that patients without preexisting IBD are also at risk. Of note, the single patient in this cohort with a noted history of IBD did not develop an IBD flare throughout treatment. Similarly, at least one of the patients that developed hyperglycemia and then diabetes during teprotumumab therapy did not have a history of diabetes. Eleven percent of patients in this study experienced hyperglycemia and 3% had persistent hyperglycemia at last follow-up. However, the incidence of hyperglycemia was likely underestimated in this study, as hyperglycemia was primarily assessed via a review of HbA1c and blood glucose levels which were not uniformly assessed in all patients. Only 40 patients received a pretreatment HbA1c. One observational study evaluated hyperglycemic changes in a cohort of 40 patients treated with teprotumumab and found that 52% of patients experienced clinically significant hyperglycemia; this included patients with normoglycemia, prediabetes, and diabetes.³¹ This underscores the importance of baseline glycemic status testing and monitoring throughout therapy in all patients.

One-third of patients in this cohort experienced otologic symptoms or hearing impairment, higher than the 10% seen in the RCTs (Table 4). Half of the patients with hearing impairment (partial or complete hearing loss, muffled hearing, or decreased word recognition) had persistent subjective hearing loss at last follow-up, and a third of patients with hearing impairment discontinued treatment at least in part due to hearing loss. The difference in otologic findings may be, in part, explained by demographic differences like age. The mean age of patients with hearing impairment was 65.3 years in this study versus 51.5 years in the treatment arm of the RCTs. Other notable risk factors such as pre-existing hearing loss and prior ototoxic medication use were not reported in the RCTs. Additionally, the variation in AE assessment likely influenced reporting. A prospective observational study of 27 patients found that 81.5% of patients treated with teprotumumab complained of at least one otologic symptom, when questioned, and 40.7% experienced hearing impairment.³² The study found that patients with a history of baseline hearing loss are at higher risk for hearing loss with teprotumumab and recommends baseline audiometric testing in all patients before starting teprotumumab treatment with repeat testing during and after treatment.³² Risk of severe, and possibly permanent, hearing impairment was added to the FDA warning label in July 2023.³⁵

This study has several limitations. First, the method of assessment, follow-up timepoints, data collection, and outside medical record evaluation varied at each institution. While this may have led to variability in AE reporting, this study is reflective of the diversity in disease presentation, monitoring, and treatment across the country. Second, the AEs may be over or under-reported. The clinical trials reported the most common AEs which, if prompted by the treating physician or known by the patient, could have resulted in over-reporting of such AEs. There is also no placebo arm to this study and the determination of a TAE was often subjective and at the discretion of the treating provider, potentially leading to TAE over-reporting. AE under-reporting could have occurred due to several reasons. Patients who received less than four infusions of teprotumumab were excluded from this study; they may have discontinued therapy early due to AEs. AEs may have been missed due to the retrospective nature of the study and the method of review of systems data collection by ophthalmologists. If a patient reported an AE to an outside ER, physician, or hospital and did not report it to the treating ophthalmologist then the AE may not have been included in this study. Furthermore, not all patients underwent the same screening tests. For example, audiograms were not ordered for each patient, therefore, some may not have noticed or reported hearing loss symptoms. Third, the timing of AEs was not precise and often an estimate as dates depended on patient memory, leading to recall bias. Finally, there was no consensus on the screening, monitoring, and management of TAEs; therefore, the rates and duration of AEs may differ between groups.

The authors recommend providers utilize a standardized questionnaire (Supplemental Figure 3) to better understand and report TAEs. Further investigation into TAEs is warranted to improve patient safety. While our study predominantly observed AEs experienced by over 5% of participants, less common AEs, such as psychiatric symptoms, require investigation. Additionally, the potential additive effect of teprotumumab overtime or in retreated patients is still unknown. Finally, studies are needed to risk stratify based on subgroups, such as age, gender, comorbidities, and treatment duration.

Prescribing physicians should work closely with patients’ PCP or endocrinologist during teprotumumab therapy to monitor thyroid function tests every 4–6 weeks and adjust thyroid-related medications to maintain euthyroidism, as dysthyroidism is a known risk factor for TED progression.^36,37 Furthermore, patients with hyperthyroidism had an increased likelihood of developing AEs in this study. Importantly, all patients should be screened for hearing loss, IBD, and glycemic status prior to therapy. If risk factors are present, patients should be co-managed with the appropriate specialists. To mitigate AE risk, all patients should undergo the following baseline screening tests: thyroid function tests, HbA1C, and audiometric testing. Additional testing may be warranted based on each patient’s unique past medical history. The authors recommend pregnancy testing (when applicable) and blood sugar testing prior to each infusion, repeat HbA1C every 12 weeks during treatment, and repeat audiometry testing should otologic symptoms develop or for at risk patients. Audiometry should also be repeated after treatment cessation. Any patient who develops bloody stools or intractable diarrhea should be referred to gastroenterology for co-management of GI AEs. Treatment delay or cessation may be necessary for work-up or concern of a potential SAE. Finally, teprotumumab should be strictly avoided in pregnant women and should not be used in growing children. Patients of child-bearing potential should be instructed to use appropriate forms of contraception prior to teprotumumab therapy, during treatment, and for 6 months following therapy, and they should be advised of the risk of prolonged amenorrhea with teprotumumab therapy. Studies confirm that IGF-1 has a role in growth hormone signaling and can alter fetal development resulting in intra-uterine growth restriction and other congenital disabilities.^38–40

In conclusion, our results confirm that teprotumumab is effective for the treatment of moderate-to-severe TED; however, AEs are common and, rarely, can be serious. To safely manage patients, prescribing physicians should properly educate patients regarding the risks of therapy, carefully screen and select patients for therapy, monitor patients closely, and understand how to manage AEs in collaboration with a multi-disciplinary team. Should significant AEs occur, providers should consider holding teprotumumab to allow for proper management. SAEs may warrant discontinuation of therapy. Additional studies are needed to better understand which patients are at risk for serious TAEs, the reversibility of TAEs, and the risk benefit ratio for patients with mild or chronic forms of TED.

A multicenter study of 131 teprotumumab-treated patients found over 80% of experienced teprotumumab-related adverse events, with 12% discontinuing treatment mainly due to hearing loss, inflammatory bowel disease, and hyperglycemia. Understanding teprotumumab’s safety profile is critical.

Acronyms:

AE

Adverse event

TED

Thyroid eye disease

IBD

Inflammatory bowel disease

GD

Graves’ disease

IGF-1

Insulin-like growth factor I

IGF-1R

Insulin-like growth factor I receptor

FDA

U.S. Food and Drug Administration

RCTs

Randomized controlled trials

CAS

Clinical activity score

GDS

Gorman diplopia score

CTCAE

Common Terminology Criteria for Adverse Events

CVG

Cochleovestibular ganglion

GI

Gastrointestinal

TAE

Teprotumumab-related adverse event

SAE

Serious adverse event

FT4

Free T4

TT3

Total T3

dB HL

Decibels hearing level

NSAIDs

Non-steroidal anti-inflammatory drugs

UC

Ulcerative colitis

PCP

Primary care physician