Effects of combination treatment with tezepelumab and allergen immunotherapy on nasal responses to allergen: A randomized controlled trial

Abstract

Background: Thymic stromal lymphopoietin (TSLP) has been shown to play a central role in the initiation and persistence of allergic responses.

Objective: We evaluated whether tezepelumab, a human monoclonal anti-TSLP antibody, improved the efficacy of subcutaneous allergen immunotherapy (SCIT) and promoted the development of tolerance in patients with allergic rhinitis.

Methods: We conducted a double-blind parallel design trial in patients with cat allergy. A total of 121 patients were randomized to receive either intravenous tezepelumab plus subcutaneous cat SCIT, cat SCIT alone, tezepelumab alone, or placebo for 52 weeks, followed by 52 weeks of observation. Nasal allergen challenge (NAC), skin testing, and blood and nasal samples were obtained throughout the study.

Results: At week 52, the NAC-induced total nasal symptom scores (TNSS) (calculated as area under the curve [AUC_0-1h] and as peak score [Peak_0-1h] during the first hour after NAC) were significantly reduced in patients receiving tezepelumab/SCIT compared to SCIT alone. At week 104, one year after stopping treatment, the primary end point TNSS AUC_0-1h was not significantly different in the tezepelumab/SCIT group compared to SCIT alone, while TNSS Peak_0-1h was significantly lower in those receiving combination treatment versus SCIT. Transcriptomic analysis of nasal epithelial samples demonstrated that treatment with the combination of SCIT/tezepelumab, but neither monotherapy, caused persistent downregulation of a gene network related to type 2 inflammation that was associated with improvement in NAC responses.

Conclusions: Inhibition of TSLP augments the efficacy of SCIT during therapy and may promote tolerance after a 1-year course of treatment. (ClinicalTrials.gov NCT02237196).

Keywords: Allergy; cytokine; epithelium; inflammation; late phase; lymphocyte; mast cell; rhinitis; thymic stromal lymphopoietin; tolerance; tryptase.

FIG 1.

Longitudinal changes in TNSS AUC_0–1h and TNSS Peak_0–1h. All 4 treatment groups at all time points shown. Patients received 1 year of tezepelumab along with 1 year of cat allergen immunotherapy with follow-up for 1 year. NACs conducted during and after therapy with corresponding TNSSs are shown. (A) Primary outcome: TNSS AUC_0–1h did not differ between the tezepelumab (Teze)/SCIT group and the placebo/SCIT group at week 104. (B) Secondary outcome: Peak_0–1h showed a significant difference between Teze/SCIT and placebo/SCIT at week 104. *P < .05 and **P < .01 for cross sectional comparisons of SCIT/tezepelumab and placebo/SCIT. Mean TNSS scores and error bars are offset in order to eliminate overlap.

FIG 2.

Longitudinal antibody data. Fold change from baseline in serum cat dander–specific IgE (A), total IgE (B), cat dander–specific IgG₄ (C), and cat dander–specific IgG₄/IgE ratios (D) (data displayed by treatment group). Data are shown as means with 95% confidence intervals. *P < .05 and **P < .01 for comparisons of SCIT/tezepelumab and placebo/SCIT.

FIG 3.

Differential expression of the nasal gene module. (A) Normalized log₂ gene expression levels of the 143 gene module (mod10) by group and time point analyzed from nasal brushings collected at baseline, week 52, and week 104. Box plots show median values (horizontal line), mean values (point), interquartile ranges (box), and 1.5 interquartile ranges (whiskers). *P < .05, **P < .01, ****P < .001, *****P < .0001 for comparisons of SCIT/tezepelumab and placebo/SCIT. (B) Normalized log₂ gene expression of the same module compared to the TNSS Peak_0–1h. Shown are linear regression lines for all individual values for each group.

FIG 4.

Module expression mediation effects on TNSS Peak_0–1h. (A) Schematic of the mediation analysis showing the causal mediation effect of module expression on the TNSS Peak_0–1h. A significant mediation effect was observed only in the SCIT/tezepelumab group. (B) Shown are the density plots of the average direct effect (ADE) of each treatment on TNSS Peak_0–1h and the average causal mediation effect (ACME) of module expression on TNSS Peak_0–1h due to each treatment.

FIG 5.

Nasal gene expression module. The gene–gene association networks for the module demonstrate a significant interaction network. Genes are represented as circular nodes and known gene–gene interactions from STRING as connecting edges. Nodes are colored according to their inclusion in the KEGG (www.genome.jp/kegg/) enrichment pathways shown.

FIG 6.

Causal mediation analysis of nasal transcripts. Schematic of the mediation analysis showing the causal mediation effect of TPSAB1 expression on the TNSS Peak_0–1h. A significant mediation effect was observed only in the SCIT/tezepelumab group.