Evolving Ocular Safety Signals of EGFR Inhibitors: A FAERS Disproportionality Analysis of Amivantamab, Mobocertinib, and Classic Agents

This study utilizes a FAERS disproportionality analysis to confirm a strong, class-wide signal of ocular toxicity ranging from benign eyelash abnormalities to sight-threatening complications across both classic and newer EGFR inhibitors, underscoring the critical need for proactive ophthalmologic monitoring in patients receiving these therapies.

The Gist

Imagine your body is a bustling city, and the EGFR is a super-efficient construction manager. This manager's job is to keep the city's buildings (your cells) growing, repairing themselves, and staying organized. When cancer shows up, it's like a rogue construction crew taking over, building chaotic, dangerous structures.

EGFR Inhibitors are the "stop-work" orders sent by the oncologists. They tell the construction manager, "Stop! We need to halt this specific building project." This works wonders for stopping the cancer, but there's a catch: the construction manager is also responsible for maintaining other parts of the city, like the eyes.

This paper is like a massive detective investigation into what happens to the "eye district" of the city when we send out these stop-work orders.

The Detective Work: The "FAERS" Database

The researchers didn't just look at a few patients; they acted like digital detectives scanning a giant, public library of millions of medical reports called FAERS (the FDA Adverse Event Reporting System). Think of this library as a giant "Complaint Box" where doctors and patients write in if something goes wrong after taking a medicine.

They looked at 12 different types of "stop-work" orders (drugs) used over 24 years (2001–2025) to see if they were causing specific problems in the eye. They used a special mathematical filter (called a "disproportionality analysis") to separate real patterns from random noise. If a specific eye problem showed up way more often with a specific drug than with any other drug in the world, they flagged it as a Signal.

What They Found: The "Eye Trouble" Report

The investigation revealed that while these drugs are great at fighting cancer, they leave a very distinct "fingerprint" of trouble on the eyes. They found 20 strong signals of trouble, which they grouped into three main categories:

1. The "Eyelash Explosion" (Trichomegaly)

• The Metaphor: Imagine the construction manager is also in charge of the city's decorative vines (eyelashes). When you stop the manager, the vines go wild. They grow longer, darker, and curlier than ever before.
• The Reality: Drugs like Panitumumab and Erlotinib caused eyelashes to grow excessively long and curly. In fact, for Panitumumab, the reports of this were 465 times higher than for any other drug. It's so common it's almost a signature side effect. Sometimes, these long lashes even curl inward and scratch the eye (trichiasis).

2. The "Dry and Irritated City" (Ocular Surface Disease)

• The Metaphor: The construction manager also runs the city's water supply and street cleaning crews. When the manager stops, the streets get dusty, the water supply gets low, and the streets get inflamed.
• The Reality: Many drugs caused dry eyes, conjunctivitis (pink eye), and blepharitis (inflamed eyelids). This wasn't just one drug; it was a "class-wide" effect, meaning almost all these drugs caused similar irritation to the surface of the eye.
• New Discovery: The study found that newer drugs, like Amivantamab, had a surprisingly strong signal for Keratitis (inflammation of the clear window of the eye), which doctors might not have expected yet.

3. The "Structural Collapse" (Vision-Threatening Events)

• The Metaphor: In rare cases, the lack of maintenance leads to a building actually cracking or collapsing.
• The Reality: While rare, some patients on Erlotinib experienced serious issues like corneal perforation (a hole in the eye's surface) or optic neuropathy (damage to the nerve connecting the eye to the brain). These are the "emergency alarms" that need immediate attention.

The "New Kids on the Block"

The study also looked at the newest drugs, like Amivantamab and Mobocertinib.

• Amivantamab was found to be a "Keratitis Specialist"—it had a very high signal for causing inflammation of the cornea.
• Mobocertinib was uniquely linked to Dry Eye.
• Osimertinib (a popular third-generation drug) seemed to have a "safer" profile for the eyes compared to the older drugs, though it still had some minor signals.

The Big Takeaway: Don't Ignore the "Check Engine" Light

The authors conclude that these drugs are like powerful tools that fix a major problem (cancer) but have a side effect of "messing up the eyes."

The Analogy: Think of taking these drugs like driving a high-performance race car. The car goes incredibly fast (cancer treatment), but the tires (eyes) wear out faster and might blow out if you don't check them.

The Recommendation:
Because these signals are so strong and consistent, the authors suggest that every patient starting these drugs should get a "baseline eye check-up" before they even take the first pill.

• Why? To know what your eyes look like before the drugs start.
• What then? If you start feeling gritty, dry, or see your eyelashes getting weirdly long, you shouldn't just ignore it. You need to tell your doctor immediately so an eye specialist can step in.

Summary

This paper is a wake-up call for doctors and patients: EGFR inhibitors are great for cancer, but they are also "eye irritants." By spotting these patterns early, doctors can protect patients' vision, ensuring they can enjoy the benefits of cancer treatment without losing their sight to a preventable side effect. It's about balancing the fight against cancer with the care of the eyes.

Technical Summary

1. Problem Statement

Epidermal Growth Factor Receptor (EGFR) inhibitors are cornerstone therapies for malignancies, particularly non-small cell lung cancer (NSCLC). While their dermatologic side effects (e.g., acneiform rash) are well-documented, their ocular adverse events (AEs) remain under-characterized.

• The Gap: Prior pharmacovigilance studies have been limited in scope, often focusing on specific subsets of agents (e.g., only Tyrosine Kinase Inhibitors) or a narrow range of ocular events. There is a lack of comprehensive, class-wide analyses that include newer biologic agents (e.g., amivantamab, mobocertinib) and compare them against classic agents.
• The Need: As cancer therapeutics evolve with newer generations of EGFR inhibitors and combination therapies, an updated, quantitative safety profile is necessary to guide clinical monitoring and preserve patient vision.

2. Methodology

The study employed a disproportionality analysis using the FDA Adverse Event Reporting System (FAERS) database.

• Data Source: Publicly available FAERS data (via openFDA) spanning from January 1, 2001, to August 31, 2025.
• Exposure Definition: 12 systemic EGFR-targeted agents were evaluated:
  • TKIs: Erlotinib, gefitinib, afatinib, osimertinib, dacomitinib, mobocertinib.
  • Monoclonal Antibodies/Biologics: Cetuximab, panitumumab, necitumumab, amivantamab, depatuxizumab mafodotin.
  • Multi-kinase inhibitor: Vandetanib (included for secondary EGFR activity).
• Outcome Definition: A pre-specified list of ocular MedDRA Preferred Terms (PTs), including trichomegaly, dry eye, conjunctivitis, keratitis, blepharitis, corneal perforation, and optic neuropathy.
• Statistical Analysis:
  • Constructed 2x2 contingency tables for every drug-event pair.
  • Calculated the Proportional Reporting Ratio (PRR) to measure the strength of association.
  • Significance Criteria: A signal was defined only if it met all three of the following:
    1. Minimum of ≥3 co-reported cases.
    2. PRR ≥ 2.0.
    3. False Discovery Rate (FDR) adjusted p-value (q-value) < 0.05 (using Benjamini-Hochberg correction to mitigate Type I errors from multiple comparisons).
• Software: R (version 2025.05.1+513).

3. Key Contributions

• First Comprehensive Class-Wide Analysis: This is the first study to apply FDR-controlled disproportionality analysis across the entire spectrum of EGFR inhibitors, including the newest bispecific antibodies and TKIs.
• Identification of Novel Signals: The study highlights specific ocular risks for newer agents that were previously under-recognized, such as keratitis associated with amivantamab and dry eye associated with mobocertinib.
• Quantitative Risk Stratification: It provides a comparative risk profile, distinguishing agents with "poly-toxicity" (e.g., erlotinib) from those with more specific or milder ocular profiles (e.g., osimertinib).
• Methodological Rigor: The application of FDR correction distinguishes this study from many prior pharmacovigilance papers that rely solely on raw PRR thresholds, thereby reducing false-positive signals.

4. Key Results

Out of 6,976,462 drug-event combinations analyzed, 20 pairs met the strict signal detection criteria. The findings were categorized into three clinical domains:

A. Eyelash Abnormalities (Trichomegaly)

• Finding: Extremely high disproportionality observed, particularly for monoclonal antibodies.
• Top Signals:
  • Panitumumab: PRR = 465.3 (95% CI: 247.7–874.3).
  • Erlotinib: PRR = 160.0.
  • Cetuximab: PRR = 155.2.
• Context: Of 89 total trichomegaly reports in the database, 33.7% were attributed to EGFR inhibitors.

B. Ocular Surface Disease

• Finding: A consistent class-wide effect involving conjunctivitis, keratitis, and blepharitis.
• Key Agents:
  • Amivantamab: Showed a highly specific and intense signal for keratitis (PRR = 15.2, 95% CI: 5.7–40.5).
  • Erlotinib: Significant signals for keratitis (PRR = 7.60), blepharitis (PRR = 7.44), and meibomian gland dysfunction.
  • Mobocertinib: Uniquely associated with dry eye disease (PRR = 4.56).
  • Panitumumab & Cetuximab: Strong signals for conjunctivitis and blepharitis.

C. Vision-Threatening and Intraocular Events

• Finding: Rare but severe events were detected, primarily linked to erlotinib.
• Specific Signals:
  • Corneal Perforation: Erlotinib (n=7, PRR = 13.9).
  • Iridocyclitis: Erlotinib (n=11, PRR = 4.3).
  • Optic Neuropathy: Erlotinib (n=6, PRR = 2.9).
• Notable Non-Signals: Osimertinib showed a keratitis signal (PRR=2.3) but did not meet the FDR significance threshold, suggesting a potentially safer ocular profile compared to erlotinib.

5. Significance and Clinical Implications

• Biological Plausibility: The findings align with the known role of EGFR in ocular homeostasis (corneal epithelial repair, meibomian gland function, and hair follicle cycling). Inhibition leads to impaired barrier function and dysregulated growth.
• Clinical Recommendations:
  • Proactive Monitoring: The study advocates for baseline ophthalmologic assessments for all patients initiating EGFR inhibitor therapy.
  • Early Intervention: Clinicians should maintain a low threshold for ophthalmology referral for symptoms like persistent keratitis, trichomegaly causing lash-cornea touch, or visual changes.
  • Agent-Specific Vigilance: While the risk is class-wide, clinicians should be particularly vigilant for severe corneal events with erlotinib and keratitis with amivantamab.
• Future Directions: The authors call for prospective studies to determine true incidence rates and for oncology-ophthalmology collaboration to standardize management guidelines.

Limitations Noted: As a retrospective analysis of spontaneous reporting, the study cannot establish causality or absolute incidence. It is subject to under-reporting, the "Weber effect" (increased reporting for new drugs), and potential confounding from concomitant chemotherapy. However, the use of FDR correction strengthens the validity of the detected signals.