To the Editor: Akimaya et al1 conducted a meta-analysis of interleukin-12/-23 (IL-12/23) and IL-23 inhibitors on respiratory tract infections (RTIs), concluding that use of IL-12/23 or IL-23 antagonists for autoimmune diseases increased upper respiratory tract infection (URTI) risk but not viral URTI or lower RTI. In our meta-analyses of pivotal psoriasis trials, we reported odds ratios of RTI associated with biologics that target tumor necrosis factor, IL-17, and IL-23 as 1.08 (95% confidence interval [CI] 0.84-1.38), 1.56 (95% CI 1.04-2.33), and 1.24 (95% CI 0.98-1.56), respectively.2, 3, 4
We note that the estimates of difference in RTI are the same between the 2 studies. The risk difference (RD) as presented by Akimaya et al1 was 0.02 (95% CI 0.01-0.03; P = .007) whereas ours was 0.02 (95% CI −0.01 to 0.04; P = .18). We estimate 1 additional RTI for every 63 patients treated with IL-23, ie, the number needed to harm, over the average trial period (or 25 per year). The main differences between the 2 studies are 1) the inclusion of trials of IL-12/23 inhibitors versus IL-23 inhibitors only, 2) the inclusion of only trials of patients with psoriasis versus patients with other diseases, and 3) the definition of RTI outcomes.
Akiyama et al1 combined biologics with different mechanisms of action (briakinumab and ustekinumab also block IL-12, which may result in different infection risk) with biologics which are more targeted, only blocking IL-23. Furthermore, including studies of IL-23 inhibitors in nonpsoriasis populations introduces estimates that may be influenced by the coadministration of other immune suppressants or different underlying risk profiles. Therefore, to yield a clinically informative estimate, we believe researchers should only combine biologics with similar mechanism of action (ie, not combining IL-12/23 data with IL-23 data) within a single disease indication (ie, not combining psoriasis with Crohn's disease, ankylosing spondylitis, etc). In addition, nonapproved therapies should be excluded from the primary analysis because their results may be driven by doses that will not be approved.
Finally, the motivating factor behind our 3 meta-analyses is that we recognized that the analysis of safety data on RTI were flawed as >10 adverse event terms are used to identify this outcome in a safety dataset. Analyzing adverse events individually can miss safety signals, making it necessary to pool terms to investigate RTI. However, as we emphasized, evaluating the risk of RTI in clinical trials is difficult because the diagnosis is made clinically without objective testing, and therefore the etiology of these symptoms, be they viral, bacterial, fungal, or allergic, is unknown, and therefore we recommend specific testing for severe acute respiratory syndrome coronavirus 2 in ongoing and future trials (Table I ).2 In addition, meta-analyses of specific AEs depend on reporting consistency, which is difficult because of threshold-dependent reporting criteria for publicly available summaries.
Table I.
Recommendations for analysis of respiratory tract infections in clinical trials of psoriasis and related diseases
| Recommendation | Rationale |
|---|---|
| Pool drugs of the same mechanism of action when conducting meta-analyses | Treatments with different biologic targets may have different rates of RTIs |
| Pool studies within the same disease indication when conducting meta-analyses | Different diseases will have different risk trajectories for RTI and also may vary in use of combination immune modifying treatment vs monotherapy in clinical trials |
| Use data of approved drugs | The general patient population is not subject to risk profiles of treatments that are not approved, especially as it relates to drug dose |
| Combine all RTI terms | RTI terms are generally nonspecific and not confirmed using objective testing. Pooling terms increases power for signal detection |
| Report all adverse events on publicly available domains consistently (for sponsors/investigators) | Threshold-dependent reporting criteria results in variation and missing events that can result in invalid results |
| Make SARS-CoV-2 testing available to trial subjects (for sponsors/investigators) | COVID-19 is a highly variable illness with nonspecific symptoms. Systematic testing for SARS-CoV-2 in trial subjects presenting with RTI symptoms will allow for robust evaluation of drug safety |
COVID-19, Coronavirus disease 2019; RTI, respiratory tract infection; SARS-CoV-2, severe acute respiratory syndrome coronavirus 2.
Therefore, we believe the impact of biologics targeting IL-12/23 or IL-23 on the risk of viral URTIs is uncertain and cannot be ruled out based on currently available data. These findings emphasize the importance of pooling data in a manner that is biologically and clinically relevant as well as understanding the limitations of trial data.
Footnotes
Funding source: Supported in part by National Institute of Arthritis and Musculoskeletal and Skin Diseases grant P30AR069589-03 (to George Cotsarelis, MD).
Conflicts of interest: Dr Gelfand served as a consultant for Bristol-Myers Squibb, Boehringer Ingelheim, GlaxoSmithKline, Janssen Biologics, Novartis Corp, Regeneron, UCB (Data Safety and Monitoring Board), Sanofi, and Pfizer Inc, receiving honoraria; in addition, he receives research grants (to the Trustees of the University of Pennsylvania) from Abbvie, Janssen, Novartis Corp, Sanofi, Celgene, Ortho Dermatologics, and Pfizer Inc, and he has received payment for CME work related to psoriasis that was supported indirectly by Eli Lilly and Company and Ortho Dermatologics. In addition, Dr Gelfand is a copatent holder of resiquimod for treatment of cutaneous T cell lymphoma, and he is a deputy editor for the Journal of Investigative Dermatology, receiving honoraria from the Society for Investigative Dermatology.
Dr Syed is supported in part by a grant from Pfizer. Dr Shin has no conflicts of interest to declare.
IRB approval status: Not applicable.
References
- 1.Akiyama S., Yamada A., Micic D., Sakuraba A. The risk of respiratory tract infections and interstitial lung disease with IL-12/23 and IL-23 antagonists in patients with autoimmune diseases: a systematic review and meta-analysis. J Am Acad Dermatol. 2021;84:676–690. doi: 10.1016/j.jaad.2020.08.026. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Wan M.T., Shin D.B., Winthrop K.L., Gelfand J.M. The risk of respiratory tract infections and symptoms in psoriasis patients treated with interleukin 17 pathway–inhibiting biologics: a meta-estimate of pivotal trials relevant to decision-making during the COVID-19 pandemic. J Am Acad Dermatol. 2020;83:677–679. doi: 10.1016/j.jaad.2020.05.035. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Syed M.N., Shin D.B., Wan M.T., Winthrop K.L., Gelfand J.M. The risk of respiratory tract infections in psoriasis patients treated with interleukin 23 pathway–inhibiting biologics: a meta-estimate of pivotal trials relevant to decision-making during the COVID-19 pandemic. J Am Acad Dermatol. 2020;83:1523–1526. doi: 10.1016/j.jaad.2020.06.1014. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Syed M.N., Shah M., Shin D.B., Wan M.T., Winthrop K.L., Gelfand J.M. Effect of anti-tumor necrosis factor therapy on the risk of respiratory tract infections and related symptoms in psoriasis patients - a meta-estimate of pivotal phase 3 trials relevant to decision-making during the COVID-19 pandemic. J Am Acad Dermatol. 2020 doi: 10.1016/j.jaad.2020.08.095. [DOI] [PMC free article] [PubMed] [Google Scholar]