Abstract
Background
Asthma is a chronic illness that is characterized by a morbidity and mortality that significantly impacts upon socio-economic resources [National Heart, Lung, and Blood Institute, 2002]. Inhaled corticosteroids (ICSs) are recognized as the most effective class of drugs and are recommended by international guidelines as first-line treatment [Frois et al. 2009]. Unfortunately, often ICSs are not sufficient in preventing exacerbations, hospitalizations and sometimes deaths, and thus combination therapy with a bronchodilator is often necessary [Global Initiative for Asthma, 2006]. This combined treatment option allows improvement in symptoms and lung function and reduced exacerbations, and in addition, it is usually well tolerated and makes it possible to reduce the dose of ICSs [Global Initiative for Asthma, 2009].
Patients with severe asthma have inadequate control of their disease, despite the multidrug optimal therapy they follow, and represent the patient subpopulation that consumes the majority of healthcare resources [National Heart, Lung, and Blood Institute, 2002; Loddenkemper et al. 2003]. Drugs which provide improved control of this disease are urgently required. Omalizumab is the first biologic drug available for the treatment of severe asthma [Humbert et al. 2005]. A number of clinical trials have been performed with the aim of assessing the effectiveness and safety of omalizumab. Results from these trials have shown that this molecule was able to significantly improve control of asthma and quality of life (QoL), with an excellent safety profile [Humbert et al. 2005; Rodrigo et al. 2011; Walker et al. 2006]. Real-life data have led to the same conclusions, while the cost-effectiveness assessment varied according to the analytical methods [Oba and Salzman, 2004; Wu et al. 2007; Campbell et al. 2010] and the country-specific setting. With regard to its long-term effectiveness, few studies are available and have limited case histories [Pace et al. 2010; Nopp et al. 2010]. The aim of this study was to evaluate the long-term stability (4-year follow up) of the effectiveness of omalizumab as add-on treatment in patients with severe persistent allergic asthma, to evaluate the safety profile and to analyse the economic savings in terms of use of healthcare resources.
Methods
Study design
This retrospective study was designed to analyse a series of patients with severe allergic asthma treated with omalizumab. Patients were initially enrolled as part of the CIGE025A2425 international multicentre, open-label, parallel-group clinical trial (November 2005/September 2008) (Table 1).
Inclusion and exclusion criteria.
The primary objective was to evaluate the persistency of the response to the treatment with omalizumab administered for 32 weeks as add-on therapy to optimized asthma therapy, in patients who have inadequate asthma control, despite treatment according to Step IV of GINA 2002. All patients signed the informed consent and were enrolled between November 2005 and June 2006.
At the end of the 32-week period, 11 responsive patients went on to complete the study and continued omalizumab treatment until June 2010. The monitoring visits coincided with the timescales planned for administering the drug (every 2 or 4 weeks) and for the follow up, as indicated by Italian Drug Agency (AIFA).
This study was approved by local ethics committee (Protocol Number 2011/0006211/03-01-2011).
End points
The primary end point was to evaluate the persistency of benefit provided by omalizumab as add-on treatment beyond 32 weeks and up to 4 years from the start of the treatment, compared with the same parameters assessed during the 12 months prior to initial drug administration.
Secondary end points were as follows:
safety and tolerability during the follow-up period;
economic impact of treatment in terms of the use of healthcare resources, number of exacerbations and potential reduction in drug consumption
Evaluation of effectiveness
The Global Evaluation of Treatment Effectiveness scale (GETE) was used to evaluate the effectiveness of omalizumab at both 32 weeks and at 4 years after initial drug administration [Global Initiative for Asthma, 2009; Lloyd et al. 2007]. This evaluation was performed independently by both investigator and patient using the same five-point scale. A good or excellent response was used to define a patient who has responded to treatment. The evaluation of QoL was conducted through the Juniper Asthma-Related QoL Questionnaire (AQLQ) [Juniper et al. 1993].
The AQLQ is composed of 32 questions which cover four domains: activity limitation, symptoms, environmental stimuli and emotional function. Patients recall their experiences during the previous 2 weeks and score a number of asthma-related problems on a seven-point scale from 1 (maximum impairment) to 7 (no impairment). We used an overall summary index, which is the mean of the responses to the 32 items (total score). This questionnaire was found to be valid, reproducible and responsive to change over time and a change in the score of 0.5 or more points has been determined to be the minimal clinically important difference.
Serious or mild exacerbations during the 12 months preceding the study were documented, along with those at 32 weeks and at 4 years, together with any hospital admissions or emergency department (ED) visits. Serious exacerbations were classified as those which required systemic steroids, hospitalization or visits to ED. Mild to moderate exacerbations of asthma were classified as those which required treatment at home or in the physician’s office.
Lung function was evaluated by measurement of initial forced expiratory volume in one second (FEV1) and then FEV1 values at 32 weeks and at 4 years.
Cost analysis
In order to estimate the economic impact on the use of health services (hospital admissions, visits to ED, exacerbations and drugs), a monetary value was assigned to each event recorded in both the 12 months prior to the beginning of the study and in the follow-up period.
More specifically:
– For admissions, the average value of the diagnosis-related group (DRG) 96 and 97 (bronchitis and asthma with and without complications) was considered; in the event of admissions for respiratory insufficiency, the DRG 87 [Ministero della Sanità, 2006] was used.
– The economic cost for a visit to ED was obtained by calculating the fees charged [Ministero della Sanità, 1996] for each individual routine service conducted during the stay in ED (general checkup, pulmonary checkup, blood gas analysis, blood chemistry tests and chest X-ray).
– Moderate exacerbations were evaluated based on the fees charged [Ministero della Sanità, 1996] for the individual services generally used (e.g. specialist checkup and weekly treatment with oral steroids, long-acting beta agonists [LABAs] and antibiotics). No cost was attributed to mild exacerbations.
– The daily cost for each class of drugs was obtained from information available on drug price for the public and the national effective use of these drugs during the period January–August 2010. In particular, for the class of LABA, formoterol and salmeterol were considered; for anticholinergics, tiotropium; for ICSs, beclomethasone dipropionate, budesonide and fluticasone propionate; for antileukotrienes, montelukast and zafirlukast; for oral steroids, prednisone and methylprednisolone; for antibiotics, penicillins, cephalosporins, macrolides and fluoroquinolones, assuming an average treatment duration of 6 days (except 3 days for azithromycin). The cost of fixed combinations was not included as, in the case that the patient had used a fixed combination instead of a free combination LABAs + ICSs, the daily cost could have been slightly underestimated.
– The PRE–POST treatment comparison was obtained by comparing the annual pretreatment costs with an annual average of the 4-year posttreatment period costs.
Statistical analysis
Data are summarized by the primary measures of central tendency and dispersion, in addition to frequencies and percentages. The t-test and Wilcoxon test for paired data were used to verify the significance of trends observed from the main parameters of lung function and QoL indicators. A value of p < 0.05 was considered statistically significant. Data were analysed using SPSS v.15.0 (Chicago, IL, USA).
Results
Baseline
A total of 11 white patients were examined in the present study and their demographic characteristics and smoking status are presented in Table 2.
Baseline clinical characteristics.
All patients were affected by severe allergic asthma which was not controlled by optimized asthma therapy and daily symptoms leading to a significant impairment in QoL, represented by the low AQLQ score at baseline (median 2.8; range 1.2–3.6). The median value of baseline IgEs (ImmunoCap-Phadia, Sweden) was 256 IU/l (range 31–687.6). The dosage of omalizumab was calculated using dosage tables, the monthly average being 437.5 mg (range 150–750 mg). Seven patients (63.6%) received the drug every 2 weeks and four patients (36.4%) every 4 weeks. During the 12-month period preceding the treatment, all patients had at least one mild–moderate exacerbation, seven (63.6%) at least one hospitalization and four (36.3%) at least one emergency visit to ED (without subsequent hospitalization). Approximately half of patients (45%) had between four and five exacerbations and an average number of visits to ED of 2.5 (range 1–4).
The average value of the baseline FEV1 was 58.6% (range 42–75) (Table 3).
Effectiveness of omalizumab treatment.
The Δ% shows the percent variation from baseline
AQLQ, Asthma Quality of Life Questionnaire; FEV1, forced expiratory volume in one second; GETE, Global Evaluation of Treatment Effectiveness; IgE, Immunoglobulin E.
All patients used ICS and LABA inhalers at the maximum dose, eight patients (72.7%) antileukotrienes, 10 patients (90.9%) oral steroids, one patient (9.1%) theophylline, five patients (45.5%) anticholinergics and eight patients (72.7%) short-acting bronchodilators (Table 4).
Drug therapy before and after omalizumab treatment.
The percentage is calculated from total number of patients (n = 11).
ICS, inhaled corticosteroid; LABA, long-acting beta agonist; SABA, short-acting beta agonist.
Evaluation of effectiveness
Follow up at 32 weeks
Eight patients (72.7%) showed a good/excellent GETE and three patients (27.3%) showed a moderate GETE (Table 3). The median value of AQLQ was 4.60 points. Eight patients (72.7%) showed an increase in the total AQLQ score (>0.5 points). Two patients (25%) showed a moderate increase (>1.0 points) and six patients (75%) showed an excellent increase (>1.5 points) in AQLQ score. Overall, the average increase was by 1.4 points (Table 3).
At the intermediary check, two hospital admissions were recorded, with a drop compared with the pretreatment period of 89.5%. The number of mild–moderate exacerbations was reduced by 70.9% (n = 16) compared with the previous 12 months. The average FEV1 at 32 weeks was 78.1% (range 40–100%) (p = 0.006). All patients continued to use ICSs and LABAs, 45.5% antileukotrienes (n = 5), 9.1% oral steroids (n = 1), none theophylline, 18.2% anticholinergics (n = 2), 63.6% short-acting bronchodilators (SABAs; n = 7); see Table 4.
Follow up at 4 years
Nine patients (81.8%) showed a good/excellent GETE and two showed (18.2%) a moderate rating (Table 3). The median value of AQLQ was 5.6 points.
Two patients (18.8%) showed a moderate increase in the total AQLQ score (>1 point) while nine (81.2%) showed an excellent increase (>1.5). The average increase was 2.6 points compared with the baseline value (Table 3).
At the final check, there was only a single visit to ED (–94.7%) and no hospital admissions were observed, compared with the pre-treatment period. The frequency of mild–moderate exacerbations also decreased by 41.8% (Table 3).
The average FEV1 at 4 years was 75.4% (range 39–109), significantly increased from baseline (p = 0.009), with only a slight reduction compared with the 32-week time point.
All patients continued to use LABAs, 10 patients (90.9%) ICS, two patients (18.2%) anticholinergics, and three patients (27.3%) SABAs. No patients were treated with antileukotrienes, oral steroids or theophylline (Table 4).
Laboratory tests
At 4 years, the median value of total IgEs was 287 IU/ml, with a negligible increase (p = 0.131) compared with the pretreatment value (256 IU/ml); see Table 3. No change in levels of blood count, creatinine or liver function parameters were observed.
Adverse events
No patient showed any systemic or local side effects related to omalizumab treatment over the entire follow-up period.
Economic impact
During the follow-up period, the reduction in hospital admissions was 97.3%, the reduction in costs related to ED visits was 97.5% and the reduction in the cost due to mild-moderate exacerbations was 84% (Table 5).
The cost of treatment and drug therapy before and after omalizumab.
Calculations based on the average cost per patient
ED, emergency department; ICS, inhaled corticosteroid; LABA, long-acting beta agonist; SABA, short-acting beta agonist.
With regard to the consumption of drugs, there was an average reduction of 36% compared with the year preceding the use of omalizumab, there were no substantial variations relating to the use of LABAs + ICSs, the consumption of antileukotrienes reduced by 61%, that of anticholinergics by 55%, SABAs by 33%, theophylline by 92% and oral steroids by 87% (Table 5).
The average annual cost of omalizumab was €12,850 per patient. When the reduction in healthcare costs was considered, the annual cost of omalizumab was reduced by 27% (median –22%, range –83% to +7%) resulting in a net cost of €9880 per patient. The analysis did not consider the savings associated with the reduction of complications and long-term comorbidity associated with oral steroids.
Discussion
For patients with severe uncontrolled allergic asthma despite optimal therapy, the monoclonal anti-IgE antibody is a safe and effective add-on treatment which, in suitable patients [Rodrigo et al. 2011], allows them to obtain better control of their asthma with an excellent safety profile [Humbert et al. 2005; Walker et al. 2006], but with an increase in direct costs.
Within the European Union, the annual cost of asthma treatment is approximately €17.7 Billions, associated with an estimated loss of working capacity of €9.8 billion, with the majority of costs attributed to patients with poorly controlled asthma [Loddenkemper et al. 2003].
The purpose of the present study was to evaluate the long-term effectiveness of omalizumab in addition to evaluating data available from previous published studies to determine the economic impact of this treatment, assessing the impact upon health services, exacerbations and use of other drugs.
The limitations of this study are determined primarily by the small amount of case histories, initially selected based on the restrictive criteria of an international clinical trial, by the origin and ethnicity of the patients, all whites of Italian origin treated in a single centre. Such factors may represent bias to be taken into consideration when interpreting the results.
The main findings from baseline evaluation of parameters revealed an inadequate control of the disease, which resulted in poor QoL. Furthermore, the majority of patients were treated with traditional drugs at maximum dosage, including chronic use of systemic steroids with their known medium and long-term side effects. The frequent use of health services and emergency visits resulted in high costs for the national health service (NHS) and directly impacted upon working activity. Overall, findings from the present study show a superior effectiveness of omalizumab compared with previous published studies regarding exacerbations and QoL, with some improvement in the control parameters at 4 years compared to the 32 initial weeks [Humbert et al. 2005, 2008; Brusselle et al. 2009; Cazzola et al. 2010].
These data not only show an effectiveness that remains stable over time, but also which progressively improves over time in the majority of patients. This is supported by evidence from the absence of hospital admissions following the intermediate clinical visits. The two hospitalizations (which in both cases occurred within the first 16 weeks of treatment) may be explained by the fact that the effectiveness of the drug was not yet fully optimized, as they did not occur in the remaining follow-up period. The effectiveness of omalizumab during the follow-up period is further demonstrated by the significantly reduced use of asthma medications such as oral steroids, leukotriene modifiers and SABAs.
Another important finding is the marked improvement of spirometric values. The difference between our findings and those observed from other studies [Humbert et al. 2005, 2008; Cazzola et al. 2010] may be due to various factors, but we believe that the selection criteria of patients according to CIGE025A2425 (i.e. FEV1 reversibility test +12%) would have provided ample capacity of improvement in these patients.
The evaluation of QoL using AQLQ showed a clear increase even at the first check, which further improved up to the end of the study, when the majority of patients showed an excellent increase in their scores.
In addition to the reduction in the cost of healthcare resources, decreasing the use of drugs has also allowed significant cost savings.
The concern of NHSs related to the cost of omalizumab could have led the national health authorities to restrict access to the treatment. In order to assess a new therapeutic strategy, especially in the case of biologic treatments, it is fundamental to evaluate its benefits in terms of real-life effectiveness and QoL improvement beyond the overall financial cost implications. From this wider perspective, any incremental therapeutic cost represents an investment in the patients’ health improvement as demonstrated in this real-life study and in other cost-effectiveness evaluations. Several positive cost-effectiveness evaluations have been conducted in recent years, each demonstrating very different results, possibly due to the variations in methods used for calculating the clinical–economic data and differences in reimbursement policies [Wu et al. 2007; Brown et al. 2007; Dewilde et al. 2006].
Our findings show a marked decrease in the use of healthcare services, with an almost complete reduction in the costs of ED visits and hospital admissions and a 36% reduction in the cost of other treatments. On average, these avoidable costs would offset 27% of the annual cost of omalizumab. The net cost per exacerbation (excluding mild exacerbations) avoided using omalizumab is €2273 and may be an acceptable value to avoid a worsening of the patients’ health status. These additional costs must be compared with the level of effectiveness and the patients QoL achieved at the same time of treatment. In this regard, there is evidence available showing that the cost-effectiveness profile of omalizumab is often considered positive according to national and international parameters [Oba and Salzman, 2004; Brown et al. 2007; Dewilde et al. 2006]. These figures provide justification for the use of omalizumab in healthcare resources.
It must also be noted that our economic analysis was conservative as it did not take into consideration the following elements:
– the costs avoided from complications caused by systemic steroid use (cataracts, steroid-induced diabetes, osteoporosis, etc.) and the long-term cost avoided due to an improvement of health status;
– the cost of omalizumab and other treatments, as the full (not discounted) price was used and hospitals can purchase the drug at a lower cost;
– the social costs, as our evaluation did not include the impact of omalizumab on working activity considering that the average age of patients is 47.5 years.
Conclusions
The overall evaluation has confirmed the effectiveness of omalizumab with evidence of persistency in long-term results, which even tend to improve over time in all of the parameters considered, as shown by few studies about long-term treatment [Pace et al. 2011]. We would like to remind the reader, however, that patients were initially selected as part of a clinical trial; because of that, the inclusion criteria are therefore more restrictive than those contained in the data sheet of the drug. This initial selection bias, however, could help identify people who may be more responsive to omalizumab in the long run; of course, to remove this bias would require studies with larger series. The safety profile was also shown to be optimal, as no patient showed any systemic or local side effects. A slight increase in total IgE was observed; it was lower than that found by other authors [Hamilton et al. 2005]. Concerning this matter, it is well known that patients treated with omalizumab may exhibit a reduction of serum free IgE levels with increased total IgE due to the formation of IgE anti-IgE small immune complexes, which have a longer half-life than free IgE [Hamilton et al. 2005; Hayashi et al. 2006]. However, this condition does not have any pathological significance, and it has not been detected in all studies; for example, a paediatric study showed a decrease of total IgE after omalizumab treatment [Steiss et al. 2008].
The estimated increase in direct costs is therefore significantly offset by the medium- and long-term savings made by healthcare services. These savings are also independent of quantifying the improvement in work and school performance, the probable reduction in comorbidity and long-term complications linked to the chronic use of systemic steroids and the significant improvement in QoL.
Footnotes
Acknowledgements
We thank Dr Irene Colangelo (Health Economic Head, Novartis Pharma, Italy) for his valuable advice on the economic analysis.
This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
Francesco Menzella and Luigi Zucchi participated in the clinical trial and contracted research for Novartis.