Add-on omalizumab in children with severe allergic asthma: a 1-year real life survey

Introduction

The management of severe asthma in children remains a challenge, with its prevalence being estimated at 5% of the asthmatic population [1]. The term “problematic severe asthma” has been recently preferred to describe these school-aged children who have poorly controlled asthma despite maintenance therapy with ≥400 μg of fluticasone per day or equivalent of inhaled corticosteroids (ICSs) plus inhaled long-acting β-agonist (LABA) or leukotriene receptor antagonist or theophylline [2]. Those children are characterised at various degrees by daily symptoms, severe or frequent exacerbations, high rates of healthcare utilisation, persistent lung function alteration and altered quality of life. Problematic severe asthma is therefore a costly disease in childhood despite its fairly low frequency. Omalizumab is a recent and expensive treatment designed as an add-on therapy in difficult-to-treat patients with allergic asthma. Children included in clinical trials have been part of a wide age group including mainly adults in the first clinical trial (age range 12–79 years) [3], and then of specific paediatric trials [4–8]. These studies have shown a significant decrease in exacerbations and healthcare use, and an improvement in quality of life, but nonetheless a modest effect on symptoms and lung function tests (LFTs). The French Health Authority has limited the utilisation of omalizumab to allergic children with uncontrolled persistent asthma and/or severe exacerbations, severe airway obstruction evaluated by LFTs when aged >12 years, despite being administered high doses of ICS or oral corticosteroids in association with at least LABA. In France, omalizumab has been available since 2006 for children aged >12 years and with total IgE ≤700 kU·L⁻¹ and since 2009 for those aged >6 years and with total IgE ≤1500 kU·L⁻¹. A few real-life studies have been conducted in adults [9, 10]. The latter have not only confirmed but also extended previous drug efficacy and safety observed in randomised trials. Conversely, knowledge on omalizumab utilisation and safety has remained limited in childhood [11, 12]. This was marked in adolescents (12–18 years) who were diluted in the large primary study [3]. Moreover, the effect of omalizumab on disease control has not yet been assessed as an efficacy outcome [4–7].

This 1-year observational study reports the real-life efficacy and safety of add-on treatment with omalizumab in a large group of children with allergic severe asthma. The primary objective was to evaluate the effect of add-on omalizumab on asthma control, as mentioned in step 5 of the Global Initiative for Asthma (GINA) guidelines [13]. We also assessed other outcomes of efficacy (exacerbation rate, healthcare utilisation, ICS sparing effect and change in lung function), as well as safety.

Results

Descriptive data

104 children were included, all fulfilling the criteria of severe asthma (table 1). 44% of them had been hospitalised during the previous year, with 20 children requiring more than one admission, leading to 87 stays, eight of which were in the ICU. Six (5.8%) children required continuous oral corticosteroid therapy. This atopic population was characterised by: 1) very high IgE levels, with a value above the threshold of 700 kIU·L⁻¹ in 57 (55%) children, IgE levels between 1500–3000 kU·L⁻¹ in 20 children and >3000 kIU·L⁻¹ in seven children; 16 of these 27 children with the highest IgE levels were aged <12 years; 2) polysensitisation: 66% had allergic sensitisation to at least three allergens; 3) a high frequency of food allergies (35%); and 4) a high frequency of allergic rhinitis (85%) (table 1 and fig. 2). Lung function tests revealed airway obstruction with FEV₁ <80% pred in 36%, or FEF_25–75% <60% pred in 51%, FEV₁/FVC <0.8 in 53% and <0.7 in 26% of children.

View this table:

• View inline
• View popup

Table 1– Characteristics of the total population^#

• Download figure
• Open in new tab
• Download powerpoint

Figure 2–

Allergen sensitisations detected at baseline in the overall population (n=104). Food allergies, Alternaria and cockroach sensitisations were overrepresented in these severe asthmatic children.

The outcomes were available in 101 patients at V1 and 92 patients at V2 (fig. 1). At V1, treatment was discontinued in eight children by the physician. One more patient was lost to follow-up between V1 and V2.

Asthma control

Control of asthma clearly improved over the year of treatment (fig. 3). Asthma control levels were good in 0%, partial in 18% and poor in 82% of the population at treatment initiation and improved to 53%, 30% and 17% at week 20, and to 67%, 25% and 8% at week 52, respectively (p<0.0001). 20 (20%) patients with partially controlled (n=16) or uncontrolled (n=4) asthma at V1 became in good control at V2. Conversely, six children improved at V1, and then lost control between V1 and V2. In addition, 11 (10.5%) children with poor asthma control at V0 moved to partial control at V1 and V2. Thus, only 14% of the children did not improve over the year. Mean baseline IgE levels were not different between patients with good control and without control (partial or poor control) at V1 or V2 (V1: 1214 (385–1797) kU·L⁻¹ versus 1040 (380–1445) kU·L⁻¹, respectively, p=0.32; V2: 1173 (365–1714) kU·L⁻¹ versus 891 (276–1061) kU·L⁻¹, respectively, p=0.21).

• Download figure
• Open in new tab
• Download powerpoint

Figure 3–

Asthma control levels at baseline (n=104), at 20±4 weeks (n=101) and at 52±4 weeks (n=92) after the initiation of add-on omalizumab treatment.

Exacerbations

A significant decrease in the number of exacerbations was observed under treatment, when compared with the previous year (fig. 4). The mean rate of severe exacerbations decreased from 4.4 (3.7–5.2) per patient during the previous year to 1.25 (0.55–1.95) during the year of treatment (p<0.0001). This represented a reduction of 72% over the year. A low rate of exacerbation (0.66 (0.33–0.99) per patient) was early observed over the period V0–V1, and was maintained at 0.57 (0.23–0.93) per patient over the period V1–V2. In addition, the percentage of children requiring hospitalisation decreased from 44% in the past year to 6.7% (n=7, 10 admissions), with none necessitating a stay in ICU during the year on treatment (p<0.001). Hospital admissions were reduced by 88.5%.

• Download figure
• Open in new tab
• Download powerpoint

Figure 4–

Change in exacerbation rates over a period of 52 weeks with add-on omalizumab treatment in children with uncontrolled severe allergic asthma. Data are presented as mean (95% confidence interval).

Lung function tests

FEV₁ was available in 78 and FEF_25–75% in 64 children over the year of follow-up (fig. 5). The mean improvement of lung function assessed at V2 was 4.9% pred (95% CI 0.69–9.19; p=0.023) and 9.5% pred (95% CI 3.7–15.2; p=0.002) for FEV₁ and FEF_25–75%, respectively.

• Download figure
• Open in new tab
• Download powerpoint

Figure 5–

Changes in mean (95% confidence interval) pre-bronchodilator forced expiratory volume in 1 s (FEV₁) and forced expiratory flow at 25–75% of forced vital capacity (FEF_25–75%) from baseline during the treatment with omalizumab.

Inhaled steroid sparing effect

Over the year on treatment, the ICS sparing effect was -212 (-284– -140) μg, equivalent to a 30% reduction in ICS dose (fig. 6). The mean administered dose of FP or equivalent was 703 (642–764) μg at V0, 592 (528–656) μg at V1 and 481 (412–551) μg at V2 (p<0.0001). 46.7% of patients achieved ≥50% reduction in the administered dose, and only 5.4% maintained unchanged ICS doses between V0 and V2. Oral corticosteroids were withdrawn in all six children with this maintenance therapy at baseline.

• Download figure
• Open in new tab
• Download powerpoint

Figure 6–

Changes in daily inhaled corticosteroid (ICS) dose (μg·day⁻¹ fluticasone propionate-equivalent) from baseline during the treatment with omalizumab. Data are presented as mean (95% confidence interval).

Discussion

We report an observational survey of 104 atopic children and adolescents with severe allergic asthma, who benefited from omalizumab treatment as an add-on therapy to high level maintenance treatment. All children in this series had received long-term follow-up in tertiary care centres, which allowed them to have a precise characterisation of their asthma, and to target the associated co-factors with poor asthma control, such as compliance or environmental exposures. The interest of observational studies is to provide complete real-life data, which might differ from that obtained in clinical trials (table S5). Our survey showed a greater improvement in asthma control, a greater reduction in exacerbation rates and healthcare utilisation and a greater steroid-sparing effect than that reported in efficacy trials. Notably, the children studied appeared to be more severely affected, less controlled and more atopic than those in the previous clinical trials, suggesting a specific clinical profile targeted by omalizumab. These results in childhood are in agreement with those recently observed in other observational studies in adults and adolescents with severe allergic persistent asthma [9, 10]. The primary outcome point of this survey was the number of children who achieved good control level (according to the GINA criteria [13]), thus providing new data of clinical relevance. The clinical trials have definitely shown omalizumab efficacy on exacerbation rates. By contrast, this effect was less marked in the control of symptoms [4–7]. In this survey, control was obtained for half of the children during the first 20 weeks of drug administration. In addition, prolonged treatment of up to 1 year enabled two-thirds of these children to achieve good control. These data suggest that the recommended mean recognised time interval of 4 months for drug response evaluation has to be considered on a case-by-case basis. Taking into account those who moved from poor to partial control, only 14% of the children did not improve. That means that one in six should be considered as nonresponders. A reduction in night symptoms and use on demand of bronchodilators was observed by Milgrom et al. [6], but not by Lanier et al [5]. In less severe asthmatic children, Busse et al. [4] reported a decrease of 25% of symptom score, comprising day and night symptoms, school absenteeism and activity interference, over 2 weeks. These authors observed a significant gain of asthma control, assessed by the Asthma Control Test, only in the 6–11-year-old age range. In our survey, the response was also observed in this age group, but was less pronounced. Young age should not delay anti-IgE treatment when required.

The other main finding was a 72% reduction in the rate of severe exacerbations, with only a few participants needing hospitalisation and none being admitted to the ICU when receiving omalizumab. A reduction in exacerbation rate was the most significant effect attributed to omalizumab and was estimated to lead to a 43–45% reduction in a recent meta-analysis [11]. Our results reinforced the previous 43–50% reduction reported in children with moderate or severe asthma [5, 7]. In addition, the effect on exacerbations was obtained during the first months of treatment, in accordance with the 28- and 16-week studies by Milgrom et al. [6] and Brodlie et al. [8], respectively. In contrast, Kulus et al. [7] observed the highest reduction of exacerbation during the final 6 months in their 1-year trial.

One of the additional benefits observed in this survey of severe allergic asthmatics was the corticosteroid-sparing effect. The drop in oral steroid bursts was an expected consequence of the dramatic reduction in exacerbations. Furthermore, all six children who had previously required daily oral steroid treatment were able to stop taking them, in accordance with a recent report [8]. It was possible to decrease the ICS dose in almost all patients. This contrasts with the modest 4% reduction in ICS dosing reported in children with similar high ICS doses, but less severe asthma [5]. Conversely, Milgrom et al. [6] reported a 100% decrease in the median ICS dose in young patients exposed to a dose five times lower than in this and other series [5]. Although limited, these current data support the fact that the step down was not related to the ICS dose administered at the onset of omalizumab. Busse et al. [4] also reported a significant difference in the ICS dose of 109 µg·day⁻¹ budesonide-equivalents between the omalizumab and placebo groups, which was greater in those children exposed and sensitised to cockroaches and those with a more severe asthma.

Lastly, both large (FEV₁) and small (FEF_25–75%) airway functional parameters improved over the year of treatment, which was not observed in previous trials [4, 6]. Although the increase was small and may be not clinically relevant, it did provide a positive outcome, as a decline in lung function has been described in severe asthmatic children followed up for many years [15]. The persistence of this positive impact on lung function over time may illustrate a modifier effect on long-term disease progression in patients with severe asthma unlike that of inhaled corticosteroids drugs.

This improvement in asthma control was observed despite the fact that nearly half of patients might be considered as receiving a suboptimal dose of omazilumab, as the European dosing table at the time of study was limited to an IgE level of <700 kU·L⁻¹. Nonetheless, response to treatment depends on the reduction of free IgE levels [16], which is not related to measured IgE levels and cannot be routinely evaluated. Finally, response to treatment was not modified by administration frequency. This supports the proposal to decrease the frequency of the injections every 4 weeks when the total dose is ≤600 mg every 4 weeks.

We cannot definitely exclude that improvement may be linked to the subcutaneous mode of omalizumab administration, which improved compliance in comparison with that of inhaled drugs. However, our population was comprised of children with poorly controlled asthma, followed up over many years in tertiary care centres where compliance was routinely assessed [2, 14]. This suggests that most of them belonged to a true therapy-resistant asthma profile. Compared with the children included in US paediatric trials [4–7], the patients included in our study had more severe asthma and with higher rates of exacerbation needing oral steroids (4.4 per year versus 1.9–2.7 per year), hospitalisations (74% versus 18–52%) and requiring high rates of healthcare utilisation. In fact, European omalizumab approbation was more restrictive than US approbation, limiting administration in poorly controlled severe asthma treated with high doses of ICS or oral steroids in association with at least a LABA. As is often the case in severe childhood asthma, the associated alteration of lung function remained moderate, but with a wide range of variation [15, 17]. Half of the children, however, displayed a diminished FEF_25–75%, which appeared to be the most sensitive marker of functional alteration in severe asthma [1, 18].

These children featured a peculiar severe atopic phenotype. Total IgE concentration was more than twofold that previously reported in omalizumab trials [5, 6], two-thirds of the children were polysensitised, one-third had food allergy, and rhinitis comorbidity was nearly constant. An elevated production of total IgE was demonstrated to be a marker of asthma severity in children [19] and was related to greater healthcare utilisation, altered lung function and airway hyperresponsiveness [20, 21]. In contrast, such a severe allergic profile has not yet been clearly individualised as a risk group in problematic severe asthma either in previous cohorts [15, 17] or in recent guidelines [2]. A polysensitisation to three or more allergens has nevertheless been associated with severe [22, 23] or uncontrolled asthma [24]. This is in accordance with recent findings, which defined a specific phenotype of severe asthma in children with multiple sensitisations not only to inhaled but also to food allergens [25]. Food allergies have been determined as a risk factor of life-threatening asthma exacerbations [26], and also appear to be involved in cases of difficult asthma. Lastly, a sensitisation to Alternaria, an allergen previously associated with severe asthma [27], was also overrepresented within this cohort. The rate was eightfold higher than the 2.8% recently determined in a large French study [28].

The proportion of systemic reactions or general symptoms was comparable to that recently detailed by Milgrom et al. [29] in a safety review, where no difference in the number or categories of AEs between the drug and control groups was revealed in the previous trials. Pain or local reaction after injection was not only the most frequent but also the most expected phenomenon, given the viscosity of the product at the time of the survey. These AEs are currently diminishing with the recent modification of the drug composition. Compared with this safety review [29], we report fewer cases of urticaria, headaches and only one related anaphylaxis. Six patients presented with serious adverse effects and discontinued treatment administration. According to the benefit response to omalizumab, two patients were successfully retreated at hospital and no adverse effect occurred. This suggests that the benefit/risk balance should be reassessed in children with therapy resistant asthma.

In conclusion, omalizumab is an effective and safe add-on therapy in uncontrolled severe asthmatic and allergic children. Those characterised by high IgE production, polysensitisations and/or food allergy were revealed to form a subpopulation of true highly allergic severe asthma, and responded well to omalizumab.