Objective. To evaluate the effect of a humanized monoclonal antibody to immunoglobulin E, omalizumab (Xolair, Novartis Pharmaceuticals, East Hanover, NJ; Genentech Inc, South San Francisco, CA), on airway inflammation in asthma, as indicated by the fractional concentration of exhaled nitric oxide (FENO), a noninvasive marker of airway inflammation. Xolair was approved recently by the US Food and Drug Administration for moderate-to-severe allergic asthma in adolescents and adults.
Study Design. As an addendum at 2 sites to a randomized, multicenter double-blind, placebo-controlled trial, FENO was assessed in children with allergic asthma over 1 year. There were 3 consecutive study periods: 1) stable dosing of inhaled beclomethasone dipropionate (BDP) when the dose was optimized (period of 16 weeks); 2) inhaled steroid-reduction phase (period of 12 weeks), during which BDP was tapered if subjects remained stable; and 3) open-label extension phase, during which subjects receiving placebo were switched to active omalizumab (period of 24 weeks). The primary outcome was area under the FENO versus time curve (AUC) for adjusted FENO, defined as the ratio of FENO at each time point compared with the value at baseline.
Results. Twenty-nine subjects participated and were randomized to omalizumab (n = 18) and placebo (n = 11) treatment groups in a 2:1 ratio dictated by the main study. There was a significant difference for age, resulting in a difference in absolute forced expiratory volume in 1 second but no difference in asthma severity based on the forced expiratory volume in 1 second percentage predicted. Baseline BDP dose was comparable between groups, as were baseline values of mean FENO (active: 38.6 ± 25.6 ppb; placebo: 52.7 ± 52.9 ppb). The degree of BDP dose reduction during the steroid-reduction and open-label phases was equivalent between the omalizumab and placebo-treated groups; subjects in the omalizumab- and placebo-treated groups had reduced their BDP dose by an average of 51% and 60%, respectively, at the end of the steroid-reduction phase and by 68% and 94%, respectively, by the end of the open-label period. In the active and placebo groups, 44% and 27% and 75% and 73% of subjects had stopped use of inhaled corticosteroids at the end of the steroid-reduction and open-label phases, respectively. There was no significant difference between the active and placebo groups during the steroid-stable phase for AUC of adjusted nitric oxide (1.31 ± 1.511 vs 1.45 ± 0.736). However, during the steroid-reduction phase, the variability of adjusted FENO in the placebo-treated group was greater than that of the omalizumab-treated group at most visits, with a significant difference between groups for AUC of adjusted nitric oxide (0.88 ± 0.69 vs 1.65 ± 1.06). FENO fell from 82.1 ± 55.6 ppm at the end of the steroid-reduction phase to 33.3 ± 21.6 ppb at the end of the open-label period in the placebo group who were placed on active omalizumab. This decrease occurred while the mean dose of BDP remained very low. Analysis of FENO over 52 weeks of omalizumab treatment in the active group demonstrated that there was a significant reduction from baseline to the end of the open-label period (41.9 ± 29.0 to 18.0 ± 21.8 ppb) despite a high degree of steroid reduction.
Conclusion. In this preliminary study based on FENO, a noninvasive marker of airway inflammation, treatment with omalizumab may inhibit airway inflammation during steroid reduction in children with allergic asthma. The degree of inhibition of FENO was similar to that seen for inhaled corticosteroids alone, suggesting an antiinflammatory action for this novel therapeutic agent in asthma. This is in keeping with recent evidence that omalizumab inhibits eosinophilic inflammation in induced sputum and endobronchial tissue.
Omalizumab is a recombinant humanized monoclonal anti-immunoglobulin E (anti-IgE) antibody that recognizes IgE at the same site as the high-affinity receptor for IgE (FcϵRI). The monoclonal antibody complexes with free IgE, thereby blocking the binding of IgE to cell membrane FcϵRI and inhibiting cell activation and mediator release.1,2. Phase III studies of omalizumab in patients with moderate-to-severe allergic asthma administered as a subcutaneous injection at intervals of 2 or 4 weeks have demonstrated that omalizumab reduces the incidence and frequency of exacerbations and has a steroid-sparing effect, as indicated by reduced use of inhaled corticosteroid (ICS).3–5 Omalizumab dramatically reduces free serum IgE immediately after the first injection6 and attenuates both early- and late-phase asthmatic reactions to inhaled allergens after a course of therapy.7,8 The latter studies have shown additionally an association between reductions in circulating and sputum eosinophilia and nonspecific bronchial hyperresponsiveness. Similarly, studies in seasonal allergic rhinitis9,10 have demonstrated that omalizumab prevents the seasonal increase in nasal and ocular symptoms and decreases the use of rescue medication. Collectively, these studies support the concept that anti-IgE therapy is likely to be efficacious in the management of allergic airway disease, probably as a consequence of its antiinflammatory effects. The US Food and Drug Administration approved omalizumab for moderate and severe adult allergic asthma in July 2003.
Nitric oxide (NO) is a highly reactive central mediator in biological systems, including the vascular endothelium, the immune system, and the nonadrenergic, noncholinergic inhibitory nervous system.11 NO is produced from l-arginine by the action of NO synthase (NOS), which exists as 3 isoforms: 2 constitutively expressed (types I and III NOS) and 1 inducible (type II). Although the constitutive isoforms of NOS synthesize endogenous NO at picomolar concentrations and mediate physiologic functions (eg, smooth muscle relaxation), type II NOS is expressed in pathologic states in response to proinflammatory cytokines and synthesizes NO at nanomolar concentrations. These large concentrations of NO lead to the generation of toxic metabolites such as peroxynitrite, which can subsequently lead to widespread oxidative damage.12,13
Immunohistochemical studies have demonstrated that not only is there increased expression of type II NOS in asthmatic airway epithelial cells but also that ICSs decrease the expression of this enzyme.12,14 Correspondingly, several studies have demonstrated that the fractional concentration of exhaled NO (FENO) is increased in patients with asthma15 and that these levels fall after treatment with ICSs and oral corticosteroids.15,16 Other studies have shown correlations between levels of FENO and other variables of asthma, in particular bronchial hyperresponsiveness,17 airway eosinophilia,18 and corticosteroid responsiveness,19 thereby suggesting that measurement of FENO may provide a surrogate, easy-to-use, and noninvasive means for monitoring inflammation in asthmatic airways.20
In this study, we evaluated the effects of omalizumab on FENO in children with moderate-to-severe allergic asthma before and during ICS withdrawal. We hypothesized that, in children treated with omalizumab, FENO would remain stable during reduction in beclomethasone dipropionate (BDP) dose but would rise in children receiving placebo. Preliminary data from this study have appeared in an abstract21 that also appeared in a book chapter.22
This preliminary study, conducted as an addendum to a multicenter study of omalizumab,6 was performed at 2 US centers (National Jewish Medical and Research Center, Denver, CO [study center 1] and Creighton University, Omaha, NE [study center 2]). After randomization, 29 children entered a 16-week, double-blind, steroid-stable phase of treatment with omalizumab or placebo while maintained on stable doses of BDP (Table 1). Thereafter, patients entered a 12-week, double-blind, steroid-reduction phase, during which patients’ BDP dose was reduced by 25% every 2 weeks (provided that their asthma did not worsen). In a 24-week open-label extension, all patients previously receiving placebo were switched to omalizumab treatment. Study treatments were administered by subcutaneous injection at intervals of 2 or 4 weeks. The dose of omalizumab was based on each patient’s serum total IgE level and body weight at baseline to provide a dose of at least 0.016 mg/kg per IU/mL of IgE per 4-week period.
The study enrolled children aged 6 to 12 years with allergic asthma who were well controlled with ICSs at doses equivalent to BDP >168 to 420 μg/day. Baseline forced expiratory volume in 1 second (FEV1) was ≥60% predicted, with a short-acting β2-agonist reversibility of ≥12% documented within the previous 12 months. All subjects had positive skin-test reactivity to at least 1 perennial allergen and ongoing exposure to that antigen during the study. The addendum and main study were approved by protocol review boards at both study centers. The subjects and their legal guardians signed separate assent and consent forms, respectively, to the addendum.
Exhaled NO was measured by using a standardized single-breath method, which conformed to American Thoracic Society recommendations for FENO measurement.23 In brief, the seated patient inspired humidified medical compressed air from a reservoir connected to a mouthpiece fitted with a 2-way valve. After inspiration to total lung capacity, the patient exhaled immediately at a constant flow rate of 42 mL/s. The exhaled air was sampled via a side port close to the mouth and analyzed online for NO by using an Ionics-Sievers (Boulder, CO) 280 NOA analyzer, sensitive to 1 ppb and calibrated daily with a standard NO gas of 25 ppm (Scott Specialty Gases, Plumsteadville, MA). Identical NO analyzers were used at both study centers, and NO was measured at several time points over the course of the study as shown in Table 1. NO data from the 2 study centers were checked at the beginning and end of the study by using a NO standard calibration gas and found to be similar for the 2 study centers: 4.9 and 4.7 ppm in study centers 1 and 2, respectively, at the start of the study and 4.8 and 4.6 ppm in study centers 1 and 2, respectively, at the end of the study.
Exhaled NO data were stored securely on personal computers in Microsoft Excel (Microsoft Corporation, Seattle, WA). The study sponsor provided supplementary data (including demographic data and drug dosage) after the study was unblinded.
Definition of Outcome Variables
Adjusted FENO was defined as the ratio of FENO at time point tL compared with baseline (visit 3). The percent FENO change at time tL compared with baseline was defined as: (1)
Summary statistics (n, mean, and standard deviation [SD]) were presented for both active and placebo treatment groups at each visit.
Primary Outcome Variable
The primary outcome variable was the change in area under curve (AUC) of adjusted FENO during the steroid-reduction phase. Between-group differences were analyzed by using an independent 2-sample t test.
The secondary outcome variable was the change in AUC of adjusted FENO during the steroid-stable phase. Between-group differences were analyzed by using an independent 2-sample t test.
Algorithm for Calculation of AUC of Adjusted FENO
Subjects with missing FENO at baseline were excluded from calculation. Missing FENO at a later time point tL was assigned a value (imputation) by using the last-observation-carry-forward method.
The ratio of FENO at tL (after imputation) versus FENO at baseline was called adjusted FENO. AUC for each subject then was calculated by the trapezoidal rule using the adjusted FENO values as follows: (2) where yti = adjusted FENO at time point ti with i = 1, 2, 3… L (final time point), with time in units of weeks.
Finally, the AUC was standardized by time period (weeks), ie, it was divided by 12 in the primary analysis and 16 in the secondary analysis. All analyses were performed using SAS 8.2 (Cary, NC).
A total of 29 subjects participated and were randomized to omalizumab (n = 18) and placebo (n = 11) treatment groups in a 2:1 ratio, as dictated by the main study. The only significant difference between the groups was age, resulting in a difference in absolute FEV1. However, the 2 groups were comparable in terms of mean FEV1 percent predicted, indicating that asthma severity was similar for both patient groups (Table 2). Baseline BDP dose was also comparable between groups (Table 2), as were baseline (randomization) values of mean FENO (active: 38.6 ± 25.6 ppb; placebo: 52.7 ± 52.9 ppb [P = .44]).
ICS Dose Reduction
The degree and pattern of BDP dose reduction during the steroid-reduction and open-label phases were equivalent between the omalizumab- and placebo-treated groups (Table 3). Subjects in the omalizumab- and placebo-treated groups had reduced their BDP dose by an average of 51% and 60%, respectively, at the end of the steroid-reduction phase and by 68% and 94%, respectively, by the end of the open-label period. In the active and placebo groups, 44% and 27% of the subjects, respectively, had stopped use of ICS at the end of the steroid-reduction phase. Corresponding values at the end of the open-label period were 75% and 73%, respectively.
Change in FENO During Steroid-Stable Phase (Secondary Outcome Variable)
AUC for adjusted FENO was calculated from baseline to week 16 (end of steroid-stable phase) after imputation. There was no significant difference between the active and placebo groups for AUC for adjusted NO during this time period (1.31 ± 1.511 vs 1.45 ± 0.736 [P = .732]).
Change in FENO During Steroid-Reduction Phase (Primary Outcome Variable)
The variability of FENO in the placebo-treated group was greater than that of the omalizumab-treated group at most visits during the double-blind, steroid-reduction period (Fig 1). Overall, there was a significant difference between active and placebo groups for AUC for adjusted FENO during this phase (0.88 ± 0.69 vs 1.65 ± 1.06 [P = .031]).
Change in FENO During Open-Label Period in Placebo Group
During the 24-week, open-label extension phase, all patients in the placebo group were switched to omalizumab treatment. In the placebo group, FENO fell from 82.1 ± 55.6 ppm at the end of the steroid-reduction phase to 33.3 ± 21.6 ppb at the end of the open-label period (P = .076). This decrease occurred while the mean dose of BDP remained very low (Table 3), accompanied by a reduction in the variability of FENO that was seen among placebo-treated subjects during the steroid-reduction phase (Fig 1).
Change in FENO in Active Group During 52 Weeks of Treatment With Omalizumab
Analysis of FENO over 52 weeks of omalizumab treatment in the active group demonstrated that there was a significant reduction from baseline to the end of the open-label period (41.9 ± 29.0 to 18.0 ± 21.8 ppb [P = .032]) despite a high degree of steroid reduction (Table 3).
This is the first study to evaluate FENO via a noninvasive measure of airway inflammation during therapy with an anti-IgE antibody. The main findings were a significant elevation in FENO in the placebo group versus omalizumab during steroid reduction, whereas FENO remained stable in the active group relative to baseline. This occurred despite a profound and comparable reduction in BDP dosage in both groups.
Furthermore, FENO fell significantly during 1 year of continuous treatment with omalizumab in the active group despite continuing the profound reduction in corticosteroid therapy. In fact, omalizumab reduced FENO to a similar degree to that observed with ICSs12,16 and leukotriene-modifying agents,24 which suggests antiinflammatory activity for this agent in allergic asthma. In contrast, the observation that the concentration of FENO showed a significant increase in the placebo-treated group during the steroid-reduction phase suggests that airway inflammation worsened in some subjects as a consequence of an overall decrease in antiinflammatory therapy.
Because BDP doses did not differ between active and placebo groups at any time point, the changes in FENO were unrelated to a lower degree of steroid-reduction in the active treatment group. This finding strengthens our belief that omalizumab has antiinflammatory effects in allergic asthma. Indeed, a recent study has shown that omalizumab reduces eosinophilic inflammation in induced sputum and mucosal biopsies.25 Because BDP was reduced to a similar extent in both groups, this study cannot suggest that omalizumab is steroid sparing, although other studies have indicated that this may be so.4,5
Examining possible confounding factors, some studies have shown that the concentration of FENO increases with age in children26 and with mean predicted FEV1.27,28 This may explain the trend to higher concentrations of FENO at baseline in the placebo-treated group, who were significantly older than those randomized to omalizumab. The small magnitude of age- and FEV1-related change in FENO, however, makes it likely that the differences and dynamic changes in FENO between omalizumab- and placebo-treated groups were valid.
This study was relatively underpowered, particularly in the placebo group, due to the 2:1 recruitment ratio dictated by the main study. Furthermore, participation declined as the study proceeded, probably due to the length of the protocol. This study therefore should be considered as a preliminary examination of the effect of omalizumab on FENO, with suggestive findings that should serve as a basis for a more-definitive study.
The stability of FENO during profound steroid reduction in children receiving omalizumab suggests that this agent may prevent an increase in airway inflammation that would be expected from ICS reduction. Despite the limitation of a small number of patients, this preliminary study suggests that omalizumab may be useful as a novel long-term, antiinflammatory therapy for treatment of moderate-to-severe allergic asthma in children. Investigations of larger patient groups are required to confirm this finding.
This work was an addendum to a study supported by Genentech Inc (South San Francisco, CA) and Novartis Pharmaceuticals Corporation (East Hanover, NJ). This study was supported by Novartis Pharma AG (Basel, Switzerland) and Genentech Inc.
- Received April 9, 2003.
- Accepted November 20, 2003.
- Address correspondence to Philip E. Silkoff, MD, National Jewish Medical and Research Center, 1400 Jackson St, Denver, CO 80206. E-mail:
The authors have not made any financial arrangement whereby the value of the compensation could be influenced by the outcome of the study, have not received significant payments of other sorts from the sponsors (excluding the costs for conducting the study), do not have a proprietary or financial interest in the test product such as patent, trademark, copyright, or licensing agreements, and do not hold significant equity interest in the sponsors of the study. Dr Gupta, holds a permanent position with Novartis.
- ↵Milgrom H. Is there a role for treatment of asthma with omalizumab? Arch Dis Child.2003;88 :71– 74
- ↵Soler M, Matz J, Townley R, et al. The anti-IgE antibody omalizumab reduces exacerbations and steroid requirement in allergic asthmatics. Eur Respir J.2001;18 :254– 261
- ↵Milgrom H, Berger W, Nayak A, et al. Treatment of childhood asthma with anti-immunoglobulin E antibody (omalizumab). Pediatrics.2001;108(2) . Available at: www.pediatrics.org/cgi/content/full/108/2/e36
- ↵Saleh D, Ernst P, Lim S, Barnes PJ, Giaid A. Increased formation of the potent oxidant peroxynitrite in the airways of asthmatic patients is associated with induction of nitric oxide synthase: effect of inhaled glucocorticoid. FASEB J.1998;12 :929– 937
- ↵Balint B, Kharitonov SA, Hanazawa T, et al. Increased nitrotyrosine in exhaled breath condensate in cystic fibrosis. Eur Respir J.2001;17 :1201– 1207
- ↵Jatakanon A, Lim S, Kharitonov SA, Chung KF, Barnes PJ. Correlation between exhaled nitric oxide, sputum eosinophils, and methacholine responsiveness in patients with mild asthma. Thorax.1998;53 :91– 95
- ↵Little SA, Chalmers GW, MacLeod KJ, McSharry C, Thomson NC. Non-invasive markers of airway inflammation as predictors of oral steroid responsiveness in asthma. Thorax.2000;55 :232– 234
- ↵Barnes PJ, Kharitonov SA. Exhaled nitric oxide: a new lung function test. Thorax.1996;51 :233– 237
- ↵Silkoff PE, Milgrom H, Tran ZV, et al. Exhaled NO (ENO) and anti-inflammatory effects of a recombinant humanized monoclonal antibody to IgE (rhumab-E25) in pediatric asthma [abstract]. Chest.2000;118 :101S
- ↵Fick RBJ. Anti-inflammatory activities of omalizumab (Xolair) a recombinant humanized monoclonal antibody binding IgE. In: Fick RBJ, Jardieu PM, eds. Lung Biology in Health and Disease: IgE and Anti-IgE Therapy in Asthma and Allergic Disease. New York, NY: Marcel Dekker Inc; 2002:265–282
- ↵Djukanovic J, Wilson SJ, Kraft M, Jarjour N, Steel M, Chung KF. Effect of treatment with anti-IgE antibody (omalizumab) on airway inflammation in mild atopic asthma [abstract]. Am J Respir Crit Care Med.2003;167 :A703
- ↵Ho LP, Wood FT, Robson A, Innes JA, Greening AP. The current single exhalation method of measuring exhales nitric oxide is affected by airway calibre. Eur Respir J.2000;15 :1009– 1013
- Copyright © 2004 by the American Academy of Pediatrics