Disproportionality analysis of drug-related interstitial lung disease in patients with head and neck squamous cell carcinoma: a signal mining study.

BACKGROUND

With the widespread use of novel therapeutic agents, an increasing number of drugs have been confirmed to be associated with interstitial lung disease (ILD). However, real-world evidence regarding ILD reporting signals in patients with head and neck squamous cell carcinoma (HNSCC) remains limited.

METHODS

We performed a signal mining study using data from the FDA Adverse Event Reporting System (FAERS) database from Q1 2004 to Q1 2025. Eligible adverse event reports related to ILD in HNSCC patients were included. Disproportionality analysis was performed using the combination of reporting odds ratio (ROR) and information component (IC). Time to onset (TTO) was analyzed using descriptive statistics and Weibull distribution modeling.

RESULTS

A total of 513 patients with ILD related to HNSCC treatment in the FAERS database were included. The median age was 68.00 years (IQR 60.00-75.00), the median weight was 56.00kg (IQR 47.80-68.00), and a total of 42 drugs were involved. Based on ROR and IC, three drugs showed definite positive ILD signals: cetuximab [ROR = 1.46 (95%CI: 1.22-1.74), IC = 0.36 (95%CI: 0.12-0.60)], nivolumab [ROR = 1.67 (95%CI: 1.35-2.06), IC = 0.61 (95%CI: 0.30-0.90)], and pembrolizumab [ROR = 2.39 (95%CI: 1.93-2.96), IC = 1.06 (95%CI: 0.74-1.35)]. In addition, etoposide showed a positive ROR [ROR = 4.51 (95%CI: 1.41-14.43)] but a negative IC [IC = 2.11 (95%CI: -0.26-2.72)]. Drug class analysis revealed that anti-EGFR monoclonal antibodies [ROR = 1.45 (95%CI: 1.21-1.72), IC = 0.35 (95%CI: 0.11-0.59)] and PD-1/PD-L1 inhibitors [ROR = 2.14 (95%CI: 1.80-2.54), IC = 0.72 (95%CI: 0.49-0.95)] were classes with positive safety signals for ILD. The median TTO was 32.00 days (IQR 14.00-73.50), α=61.70 (95%CI 52.98-71.85), β=0.83 (95%CI 0.76-0.90), and ILD was mainly characterized by early onset.

CONCLUSION

ILD reports associated with HNSCC treatment were mainly concentrated among anti-EGFR monoclonal antibodies and PD-1/PD-L1 inhibitors, with the early treatment stage representing a critical window that warrants strengthened monitoring for ILD in clinical practice. Given spontaneous reporting data, causality cannot be confirmed and further verification is required.