Distinct endotypes of steroid-resistant asthma characterized by IL-17A(high) and IFN-γ(high) immunophenotypes: Potential benefits of calcitriol

Abstract

Background: A small population of patients with severe asthma does not respond to glucocorticoids (steroid resistant [SR]). They have high morbidity, highlighting an urgent need for strategies to enhance glucocorticoid responsiveness.

Objective: We investigated the immunologic differences between steroid-sensitive (SS) and SR asthmatic patients and the effect on immunophenotype of oral calcitriol treatment because it has been previously shown to beneficially modulate the clinical response to glucocorticoids in patients with SR asthma.

Methods: CD8-depleted PBMCs were isolated from 12 patients with SS and 23 patients with SR asthma and cultured for 7 days with anti-CD3 and IL-2 with or without dexamethasone. Cytokine production was assessed in supernatants by using the Cytometric Bead Array. Patients with SR asthma were subsequently randomized to oral calcitriol or placebo therapy, and identical studies were repeated.

Results: Patients with SR asthma produced significantly increased IL-17A and IFN-γ levels compared with those in patients with SS asthma, although it was evident that cells from individual patients might overproduce one or the other of these cytokines. Production of IL-17A was inversely and production of IL-13 was positively associated with the clinical response to prednisolone. Oral calcitriol, compared with placebo, therapy of the patients with SR asthma significantly improved dexamethasone-induced IL-10 production in vitro while suppressing dexamethasone-induced IL-17A production. This effect mirrored the previously demonstrated improvement in clinical response to oral glucocorticoids in calcitriol-treated patients with SR asthma.

Conclusions: IL-17A(high) and IFN-γ(high) immunophenotypes exist in patients with SR asthma. These data identify immunologic pathways that likely underpin the beneficial clinical effects of calcitriol in patients with SR asthma by directing the SR cytokine profile toward a more SS immune phenotype, suggesting strategies for identifying vitamin D responder immunophenotypes.

Keywords: Asthma; IL-17A; glucocorticoids; steroid resistant; steroid sensitive; vitamin D.

Fig E1

Clinical trial schematic. Patients with severe asthma were recruited and given prednisolone at screening visit 1 and returned 2 weeks later at screening visit 2 when patients with SS asthma were excluded from the trial and patients with SR asthma were retained. After a 4-week washout period, the patients with SR asthma were randomly assigned to receive calcitriol or placebo for 1 month (treatment visit 1). After 2 weeks (treatment visit 2), the patients were started on additional oral prednisolone and then returned 2 weeks later for the final visit (treatment visit 3). Lung function was assessed at all clinical trial visits except treatment visit 1. CD8-depleted PBMCs were assessed from peripheral blood collected at screening visit 1 and treatment visit 3.

Fig E2

Peripheral leukocyte counts. A, Peripheral blood total and differential leukocyte counts in the patients with SS asthma (solid symbols) and patients with SR asthma (open symbols) before and after 2 weeks of prednisolone. B and C, Total B cells (CD19⁺), T cells (CD3⁺), and CD4⁺ and CD8⁺ T cells were assessed by using ex vivo flow cytometric analysis of peripheral blood from screening visit 1 of both patients with SS and those with SR asthma (Fig E2, B), as well as in the same patients with SR asthma separated based on their assignment to calcitriol or placebo treatment (Fig E2, C). Data were assessed by using the Wilcoxon matched-pairs statistical test (Fig E2, A) or the Mann-Whitney U statistical test (Fig E2, B and C). *P ≤ .05, **P < .01, and ***P < .001.

Fig E3

A and B, Cytokine production does not associate with lung function. Cytokine production from all patients with severe asthma separated according to their absolute lung function at baseline (FEV₁) is shown. Data were assessed by using Kruskal-Wallis 1-way ANOVA with the Dunn multiple comparison post hoc statistical test.

Fig E4

Effects of calcitriol on lung function and IL-13 and IL-17A levels. Cytokine production was separated based on individual changes in absolute lung function (ΔFEV₁) in response to 2 weeks of prednisolone before (baseline; black bars) and after (after 4 weeks of calcitriol and 2 weeks of prednisolone; white bars) treatment. Numbers of patients with SR asthma are indicated in parentheses above.

Fig 1

Differential cytokine production by both patients with SS and those with SR asthma in response to dexamethasone in vitro. Cytokine production by CD8-depleted PBMCs (y-axis) in the presence of dexamethasone (10^−x mol/L) in vitro (x-axis) stimulated with CD3 plus IL-2 (A) and additional IL-4 (B) is shown. Open squares, Patients with SR asthma; solid circles, patients with SS asthma. Data are means and 95% CIs and assessed by using 2-way ANOVA with Dunnett multiple comparisons tests of cytokine production in the presence of different concentrations of dexamethasone compared with no dexamethasone. *P ≤ .05, **P < .01, and ***P < .001.

Fig 2

IL-17A is negatively and IL-13 is positively associated with clinical response to 2 weeks of prednisolone. Cytokine production (A) and ratio of IL-13/IL-17A and IL-10/IL-17A cytokine production (B) separated based on change in absolute lung function in response to 2 weeks of prednisolone (ΔFEV₁) are shown. Data are means and 95% CIs and assessed by using Kruskal-Wallis 1-way ANOVA with the Dunn multiple comparison post hoc statistical test. *P ≤ .05 and **P < .01.

Fig 3

IL-17A and IFN-γ production predict glucocorticoid responsiveness. A, Comparison of IL-17A and IFN-γ production in patients with SS asthma (solid circles) and patients with SR asthma (open squares). B, Combined production of IL-17A and IFN-γ correlated with change in absolute lung function (ΔFEV₁) in response to 2 weeks of prednisolone, as assessed by using the Spearman rank correlation statistical test. C, ROC curves for prediction of steroid resistance based on production of IFN-γ, IL-17A, and their combination.

Fig 4

Serum 25(OH)D concentration correlates with IL-13 expression in vitro. Cytokine production in cell-culture supernatants was assessed by using the Cytometric Bead Array and correlated with baseline vitamin D status (defined as the serum 25[OH]D concentration). A, Cells stimulated with CD3 plus IL-2. B, Cells stimulated with addition of IL-4 in culture. Open circles, Patients with SR asthma; solid squares, patients with SS asthma. Data were assessed by using Spearman rank correlation.

Fig 5

Calcitriol treatment decreases expression of IL-17A in cultures from patients with SR asthma. Cytokine production by CD8-depleted PBMCs from patients with SR asthma in the presence of dexamethasone (10^−x mol/L) in vitro after 4 weeks of calcitriol or placebo therapy and 2 weeks of oral prednisolone therapy and cells stimulated with CD3 plus IL-2 (A) and additional IL-4 (B) is shown. Open triangles, Calcitriol; solid triangles, placebo. Data are means and 95% CIs. Data were assessed by using 2-way ANOVA with Dunnett multiple comparisons tests. *P ≤ .05, **P < .01, and ***P < .001.