Drug-associated infections and infestations in older adults with tumor necrosis factor-alpha inhibitors: a real-world retrospective and pharmacovigilance study

Xin Yu, Jiahong Zhong, Xuemei Zhuang, Zhuomiao Lin, Hongbo Fu, Yaofeng ZhangFrontiers in Pharmacology 2025 · DOI

Abstract

The research highlighted that the difference in the risk of infection in the elderly who used TNF-α inhibitors between various TNF-α inhibitors, adverse events and therapeutic indications, respectively. The use of TNF-α inhibitors increased the infection risk in older adults. Etanercept exhibited the lowest infection risk, whereas certolizumab pegol manifested the highest risk in infections and infestations. Doctors need to pay close attention to the appearance of mycobacterial infectious disorders in older adults treating with TNF-α inhibitors, which displayed the strongest signal.

Introduction

Immune-mediated inflammatory diseases (IMIDs), including rheumatoid arthritis, Crohn’s disease, and psoriasis, are frequently managed with tumor necrosis factor-alpha (TNF-α) inhibitors. While these biologics are highly effective, they are known to suppress the immune system, thereby increasing the risk of "infections and infestations." This risk is particularly acute for older adults (aged 65 and over), who are subject to immunosenescence—the natural age-related decline of the immune system—and often present with multiple comorbidities.

Despite their vulnerability, older adults are frequently underrepresented in clinical trials. Consequently, there is a significant lack of head-to-head comparative data regarding the safety profiles of different TNF-α inhibitors in this demographic. This study utilizes real-world data from the FDA Adverse Event Reporting System (FAERS) to address this gap and provide clinicians with evidence-based prescribing guidance.

Methods and Data Selection

Researchers conducted a retrospective pharmacovigilance study analyzing FAERS data from the first quarter of 2010 through the fourth quarter of 2023. The study utilized disproportionality analysis to detect safety signals, employing both the Reporting Odds Ratio (ROR) and the Information Component (IC). The IC, calculated via the Bayesian Confidence Propagation Neural Network (BCPNN) algorithm, was used to minimize false-positive biases.

The initial dataset comprised over 18.4 million reports. After rigorous cleaning and de-duplication, the final analysis focused on 44,848 adverse event (AE) reports specifically involving infections and infestations in older adults using five primary TNF-α inhibitors: adalimumab, infliximab, etanercept, golimumab, and certolizumab pegol.

Flow diagram of the study selection process from the FAERS database.

Key Results and Comparative Safety

The analysis revealed distinct safety profiles among the five inhibitors. Etanercept emerged as the safest option, demonstrating the lowest infection risk. In contrast, certolizumab pegol and golimumab were associated with significantly higher risks for serious infections.

Comparative Risk Metrics

  • Etanercept: Showed "protective signals" (negative IC values) for severe outcomes such as septic shock (IC025 = -3.23).
  • Certolizumab Pegol: Manifested the highest overall risk, with an odds ratio (OR) of 1.85 compared to etanercept.
  • Golimumab: Exhibited the strongest signal for tuberculosis (IC025 = 2.44) and showed elevated risk specifically in patients treated for ankylosing spondylitis (ROR = 3.07).
Heatmap of IC025 values showing the association between specific TNF-α inhibitors and various infection types.

The Tuberculosis Signal

A critical finding across all five drugs was a massive signal for mycobacterial infectious disorders. TNF is vital for maintaining granulomas, the structures the immune system uses to sequester tuberculosis bacteria. By blocking TNF, these drugs can lead to the reactivation of latent TB.

Discussion and Clinical Implications

The observed safety differences may be rooted in the molecular structure of the drugs. Etanercept is a fusion protein that binds soluble TNF reversibly with a high "off-rate," whereas the other four are monoclonal antibodies that bind both soluble and transmembrane TNF more irreversibly.

Demographic analysis indicated that females and patients with lower body weight (<80kg) were at higher risk, possibly due to higher disease prevalence in women or higher drug concentrations resulting from standard dosing in lighter individuals. Furthermore, the majority of reports involved the 65–74 age group, with many infections occurring after more than a year of therapy, highlighting the necessity for long-term clinical monitoring.

Limitations

While providing valuable real-world insights, the study is limited by the nature of the FAERS database, which is subject to under-reporting and notoriety bias. The database also lacks granular details such as exact dosages and patient comorbidities. As disproportionality analysis identifies associations rather than direct causation, the authors emphasize the need for prospective, long-term clinical trials specifically targeting the elderly population to confirm these findings.

Summary updated

Abstract

The research highlighted that the difference in the risk of infection in the elderly who used TNF-α inhibitors between various TNF-α inhibitors, adverse events and therapeutic indications, respectively. The use of TNF-α inhibitors increased the infection risk in older adults. Etanercept exhibited the lowest infection risk, whereas certolizumab pegol manifested the highest risk in infections and infestations. Doctors need to pay close attention to the appearance of mycobacterial infectious disorders in older adults treating with TNF-α inhibitors, which displayed the strongest signal.
BibTeX
@inproceedings{yu2025drug,
  author = {Xin Yu and Jiahong Zhong and Xuemei Zhuang and Zhuomiao Lin and Hongbo Fu and Yaofeng Zhang},
  title = {{Drug-associated infections and infestations in older adults with tumor necrosis factor-alpha inhibitors: a real-world retrospective and pharmacovigilance study}},
  booktitle = {Frontiers in Pharmacology},
  year = {2025},
  doi = {10.3389/fphar.2025.1533902},
}
Figure