Real-World Safety Profile of Cenegermin Per FDA Adverse Event Reporting System

Introduction

Cenegermin is a recombinant human nerve growth factor (rhNGF) medication that is applied topically in the management of neurotrophic keratitis (NK). ¹ Cenegermin was approved by the U.S. Food and Drug Administration (FDA) in 2018 and is currently being sold under the brand name Oxervate. ² NK is a degenerative corneal disease that is characterized by corneal hypoesthesia or anesthesia, epithelial defects, or corneal ulcers ³ due to degradation of the trigeminal nerve. As a rhNGF, cenegermin serves to treat NK by promoting sensory nerve survival. ⁴ Though management of NK often indicates surgery, cenegermin has demonstrated efficacy as a nonsurgical therapeutic. A multicenter randomized trial found that an 8-week course of six times daily cenegermin significantly improved corneal healing compared with a vehicle control. ⁵ A separate retrospective study has demonstrated that these effects are long-lasting, persisting for 48 months following the end of treatment. ⁶ One recent study demonstrated that patients who received cenegermin have more desirable outcomes and higher levels of satisfaction when compared with other treatments for NK. ⁷ Given that cenegermin is an effective treatment for NK, it is of great importance that we understand its safety profile in order to better counsel and treat the patients that use it.

Phase II clinical trials such as REPARO found that the most common adverse effects reported after cenegermin treatment were eye pain (3.8%) and blurry vision (3.8%) followed by several localized, transient ocular complications, none of which were serious. ⁸ Further trials produced similar results with eye pain (13%) being the most common adverse events (AEs) and nonocular complications being rare if ever present. ⁵ Though clinical trials have proven effective in identifying AEs associated with cenegermin use in its early stages, more work is needed to characterize the real-world, postmarket safety profile of cenegermin. The FDA adverse event reporting system (FAERS) is a public database that collects AE data from healthcare professionals and pharmaceutical manufacturers for any approved product, allowing for postmarket safety surveillance. Past literature shows that FAERS analysis can be used to determine specific ocular AEs associated with medication use, including anti-VEGF therapy ⁹ and immune checkpoint inhibitors. ¹⁰ In this study, we use FAERS to characterize the safety profile of cenegermin.

Materials and Methods

We conducted a pharmacovigilance study on cenegermin, focusing on a comparative analysis with other ophthalmical topical agents. ¹¹ Our study utilized data from the FAERS database, spanning the period from 2019 to 2023. The FAERS database adheres to the International Safety Reporting Guidelines (ICH E2B) established by ICH, and it categorizes AEs based on the Medical Dictionary for Regulatory Activities (MedDRA), ¹² encompassing 27 MedDRA system organ classes. ¹³ Each case report within the FAERS database includes information on demographic details, geographic and temporal data, reason for drug utilization, patient outcomes, counts of serious and nonserious cases, as well as AEs.

We extracted AE reports related to both cenegermin and a comprehensive array of ophthalmological topical agents listed in the FAERS database, including but not limited to Xiidra, Rhopressa, Systane, Durezol, and others. These ophthalmological topical agents serve diverse functions, such as antiangiogenic, anti-infective, anti-inflammatory, glaucoma management, steroid-based treatments, lubricants, and irrigations, among others. ¹² A full list of the ophthalmological topical agents included in the analysis can be observed in Supplementary Table S1.

The assessment of the association between cenegermin and AEs was carried out utilizing four statistical algorithms: the reporting odds ratio (ROR), the proportional reporting ratio (PRR), the Bayesian confidence propagation neural network, and the multiitem gamma Poisson shrinker. ¹⁴ Detailed equations and criteria for these four algorithms can be found in Supplementary Table S2. Similarly, we applied these four algorithms to evaluate the association between other ophthalmological agents and AEs. For the purposes of reporting results, the ROR and PRR were primarily cited.

Results

From the FAERS database, there were 4,231 cases of AEs associated with cenegermin intake. A total of 31.74% of these events occurred in males, while 65.16% of these events occurred in females. A total of 85.06% of AEs were classified as nonserious, while 6.69% resulted in hospitalization and 2.39% resulted in death. The most common AEs reported were eye pain (46.94%), eye irritation (14.46%), ocular hyperemia (9.86%), and product dose omission issues (9.12%). Age, reporting country, and reporting year were not specified for a majority of cases (Table 1).

Among the 34 eye disorders that were found to be associated with cenegermin use, the three with the highest odds ratios were periorbital pain [ROR: 25.34, confidence interval (CI): 20.70–31.01, PRR: 16.45], eye lid pain (ROR:22.96, CI: 19.39–27.19, PRR: 15.50), and eye pain (ROR: 20.02, CI: 18.77–21.35, PRR: 16.08). Patients taking cenegermin were also at higher risk for hospitalization (ROR:16.39, CI: 12.84–20.94, PRR: 12.17) and surgery (ROR: 9.57, CI: 6.14–14.93, PRR: 8.01). Patients taking cenegermin were less likely to develop vitreous floaters than patients taking other eye drop medications (ROR:0.47, CI: 0.26–0.87, PRR:0.488). Patients taking cenegermin were also more prone to specific infections, including COVID-19, urinary tract infections, and influenza; gastrointestinal disorders, including nausea, vomiting, and diarrhea; nervous system disorders including headache, migraine and somnolence; respiratory and thoracic disorders including oropharyngeal pain and nasopharyngitis; musculoskeletal disorders such as facial pain and hip fracture; and skin disorders such as Herpes Zoster. Patients taking cenegermin were less likely to develop death (ROR: 0.74, CI: 0.59–0.93, PRR: 0.74) and hypersensitivity (ROR:0.40, CI: 0.25–0.64, PRR: 0.41) when compared with patients taking other eye drop medications (Table 2).

Patients taking cenegermin were significantly more likely to develop eye disorders (ROR: 2.21, CI: 2.12–2.31, PRR: 1.46) and undergo surgical or medical procedures (ROR: 3.20, CI: 2.66–3.67, PRR: 30.94) than patients taking other eye drop medications. Patients taking cenegermin were not more likely to develop disorders of any other MedDRA system organ class. ¹³ Patients taking cenegermin were less likely to develop blood and lymphatic system disorders, cardiac disorders, ear and labyrinth disorders, endocrine disorders, gastrointestinal disorders, general disorders and administration site conditions, hepatobiliary disorders, immune system disorders, injury, neoplasms, psychiatric disorders, reproductive system and breast disorders, skin and subcutaneous disorders, and vascular disorders (Table 3).

AEs in the FAERS database are organized into serious and nonserious cases. Compared with nonserious cases, serious cases had higher rates of hospitalization (5.86% vs. 0.00%), death (4.94% vs. 0.00%), eye operation (2.60% vs. 0.00%), and pneumonia (1.89% vs. 0.00%). Nonserious cases had higher rates of eye pain (21.44% vs. 7.61%), product dose omission issues (3.87% vs. 3.12%), eye irritation (6.55% vs. 2.73%), ocular hyperemia (4.40% vs. 2.15%), blurred vision (2.91% vs. 1.69%), and wrong product usage (2.26% vs. 1.76%) (Table 4).

Discussion

In our analysis of the FAERS database, we found that patients taking cenegermin are more likely to develop eye disorders and undergo surgical or medical procedures than patients taking other eye drop medications. At the system organ class (SOC) level, they are not more likely to report AEs of any other organ system. Patient’s taking cenegermin were less likely to experience AEs in 14 SOCs when compared with patients taking other eye drop medications, suggesting the effects of cenegermin are more localized than that of other medications applied topically to the eye. Cenegermin’s specific ocular toxicity is expected, given that it only minimally enters circulation, ¹⁵ and past studies have found few to no cases of nonocular AEs. ¹⁶ Our findings that patients taking cenegermin more frequently undergo surgical operation may be explained by the fact that treatment nonresponsive NK is often corrected with surgery such as amniotic membrane grafts or tarsorrhaphy. ¹⁷ Similarly, our findings that patients taking cenegermin are hospitalized at higher rates are also likely due to the underlying NK, rather than the effects of cenegermin itself. Other outcomes, including death, were not seen in any clinical trial and are almost certainly due to associated comorbidities rather than cenegermin use.^5,8

Our findings regarding the postmarket surveillance of cenegermin confirm and build upon our existing understanding of the drug’s safety profile from clinical trials. As per our FAERS analysis, the most common AEs associated with cenegermin use involved pain localized to the eye, which is consistent with the findings of previous clinical trials.^5,8,18 Though past studies have only shown an association between cenegermin use and general eye pain, our analysis shows a significant association with both periorbital and eyelid pain in addition to general eye pain. Such a finding serves to further characterize the geography of the pain caused by cenegermin use, which may prove useful to clinicians in counseling their patients. Our analysis also shows significant associations between cenegermin use and ocular AEs that were previously only reported in the REPARO trial at a frequency of two patients or less per event; these include visual impairment, ocular hyperemia, ocular inflammation, and corneal deposits. ⁸ These were among 34 ocular AEs found to be associated with cenegermin use, suggesting that administration of cenegermin has a broad range of localized, nonintended effects. It is worth mentioning that, due to the nature of this analysis, it is possible that confounding variables may influence the generated ROR values. For example, pain may be a manifestation of NK itself.

Beyond the eye, several nonlocalized AEs were also found to be more associated with cenegermin use than with other eye drop medications. These include nervous system disorders, gastrointestinal disorders, musculoskeletal and integumentary disorders, respiratory and thoracic disorders, and several infections. Apart from three cases of nervous system disorders and one case of nonspecific infection in the REPARO trial, these AEs have not been previously reported in any clinical study of cenegermin administration. ⁸ A better understanding of the full safety profile of cenegermin is of great significance in clinical decision making, especially considering cenegermin has no listed contraindications. ² It is likely that these observed systemic effects are temporally associated rather than caused by topical cenegermin. For example, cenegermin’s association with COVID-19 is likely due to the fact that the drug only received FDA approval in 2018, so a majority of its postmarket surveillance has been during the COVID-19 pandemic. Similarly, other identified systemic AEs are likely to be temporal associations rather than caused by cenegermin. Alternatively, they may be associated with the underlying NK rather than cenegermin administration.

In the REPARO clinical trial, the researchers found that no instances of death were treatment related. ⁸ In our analysis, we found that patients taking cenegermin were more likely to have AEs involving death than patients taking other eye drop medications. The reason for this can only be assessed through careful review of the cases that are not available through the FAERS database. Additionally, the majority of the total AEs occurred in females (61.56% vs. 31.74%). The main reason for this disparity is unclear and may warrant future investigation. One possibility is due to the sex differences in cenegermin prescriptions. The main indication for cenegermin is NK, which is most commonly caused by herpes keratitis, resulting from Herpes simplex virus 1 (HSV-1). HSV-1 has a higher prevalence in females than males, which would likely explain this discrepancy. While no previous investigation has shown a difference in cenegermin-related AEs by sex,^5,8 FAERS analysis has been utilized in past studies to identify real sex differences in AE profiles.

Though analysis of the FAERS database has been previously validated as a method of assessing the safety profile of medications, ¹⁹ limitations exist. For one, there is inconsistency in the reporting of age, year, and reporting country for AEs, with this information being missing for the majority of patients. If available, this information may have proven useful in identifying patients at risk for developing adverse side effects. Additionally, it is not possible to control for comorbidities, age, or other potentially confounding factors when querying the FAERS database. For this reason, an analysis of the FAERS database can only be used to draw associations between treatment and AEs—causation cannot be determined. In addition, it is possible that high ROR and PRR values may reflect overreporting or data artifacts that could not be corrected for during analysis. Finally, since FAERS is dependent on self-reporting, we are not able to account for mild AEs, which are less likely to be reported. Despite these limitations, FAERS data remains a valuable and commonly utilized tool in characterizing drug safety profiles.

Conclusions

Overall, topical cenegermin use for treatment of NK is safe and results in local AEs. Observed systemic associations are more likely to be temporally related rather than caused by cenegermin given the limited systemic absorption of the topical drug. It is important to recognize that statistical analysis alone should not overstate the clinical relevance of cenegermin-related AEs, especially when dealing with mild or infrequent reactions. Continued postmarket surveillance is essential in ensuring the safety of the cenegermin and identifying common adverse effects. Given the limitations of the FAERS database, future prospective studies are warranted to validate the results presented here.