Use of ICS in COPD: From Blockbuster Medicine to Precision Medicine

ABSTRACT

Chronic obstructive pulmonary disease (COPD) is a major cause of mortality worldwide, whose burden is expected to increase in the next decades, because of numerous risk factors, including the aging of the population. COPD is both preventable and treatable by an effective management including risk factor reduction, prevention, assessment, and treatment of acute exacerbations and co-morbidities. The available agents approved for COPD treatment are long-acting or ultra-long-acting β2-agonists (LABAs) and long-acting muscarinic antagonists (LAMAs) bronchodilators, as well as inhaled corticosteroids (ICS) in combination with LABAs. ICS use has been restricted only to selected COPD patients by the most recent documents, mainly based on the risk of exacerbations. However, several observational studies showed a high rate of prescription of ICS in COPD, irrespective of clinical recommendations, questioning the efficacy of these compounds in unselected patients with COPD and leading to possible increase risk of side effects related to ICS use. After examining the low levels of adherence in primary care and in the clinical settings to national and international recommendations for the treatment of COPD in different countries, the most common drivers of the prevailing use of ICS are critically reviewed here by examining their pros and cons, aimed at identifying evidence-based drivers for a proper selection of patients who may benefit from the proper use of ICS.

KEYWORDS:

• COPD
• inhaled corticosteroids
• prescribing
• drivers

Introduction

Chronic obstructive pulmonary disease (COPD) is a chronic disease and a neglected global epidemic, ranking 4th among the mortality causes worldwide (1–3). The prevalence of COPD may be frequently underestimated because the disease is commonly diagnosed only when it becomes clinically evident (4,5).

Patients with COPD have persistent, irreversible, and commonly progressive airflow limitation with recurrent exacerbations, impacting on patient quality of life (6). COPD frequently associates with extrapulmonary co-morbidities, including metabolic and cardiovascular diseases that contribute to the overall disease severity (6).

COPD is a preventable and treatable disorder, provided that an effective management is implemented at all levels of the health care system. The aims of a correct management are risk factor reduction, prevention, assessment, and treatment of acute exacerbations and co-morbidities (6). The currently available bronchodilator therapies approved for COPD are either LABA (long-acting or ultra-long-acting β-2 agonists: formoterol, salmeterol, indacaterol, olodaterol, and vilanterol) or LAMA (long-acting muscarinic antagonists: tiotropium, glycopyrronium, ipratropium, aclidinium, and umeclidinium), as well as inhaled corticosteroids (ICS) in combination with LABA. Bronchodilators, in monotherapy and/or in combinations, are the mainstay of COPD treatment for all COPD patients.

Several randomized controlled studies (RCT) showed that ICS in combination with LABA reduces symptoms and exacerbation rates (7–11), leading to specific indications for ICS/LABA combinations in patients with COPD. The GOLD Report 2011 indicated ICS/LABA combination for patients diagnosed with COPD who have an increased risk of frequent exacerbations (i.e., patients with severe/very severe airflow obstruction and/or 2 or more exacerbations in the previous year or with at least ≥ 1 severe COPD exacerbations leading to hospital admission) irrespective of the magnitude of symptom severity (i.e., Group C and Group D, according to the ABCD assessment tool) (12). The 2017 revised version of the GOLD Report proposed a modified “ABCD” assessment tool. In particular, the spirometry evaluation is now considered mandatory for diagnosis, prognosis, and follow-up, but not for therapeutic decision. The latter is mainly driven by the magnitude of symptoms and history of exacerbations. Based on the new proposed approach, the ICS use in combination with LABA or LABA/LAMA is considered an alternative option in patients that continue to experience exacerbations despite being treated with long-acting bronchodilators. Concomitant history of asthma and airway/blood eosinophilic inflammation are also proposed but still largely debated as an indication for ICS use in COPD (13–16).

However, despite the above-mentioned specific indication, several real-world observational studies in general practice COPD patients and physicians (17,18), as well as the evaluation of COPD guidelines of European countries and Russia (19,20), revealed the frequent and widespread prevailing use of ICS combined with bronchodilators in patients at lower exacerbation risk.

This review will critically examine the drivers leading to the prescription of ICS, from diagnosis to clinical practice, aiming at identifying clear and validated criteria for the identification of subpopulations of patients who may benefit from including ICS in their treatment regimen.

Use of ICS-containing treatments in COPD: The real-life scenario

Real life analysis conducted in more than 3,000 European and US patients showed that ICS were prescribed in 38.8% and 51.8% of patients classified as Group A and Group B, respectively (21), i.e., patients for whom ICS treatment, according to international recommendations, is not a therapeutic option. These data are corroborated by market research analysts who estimated that more than 50% of patients with a diagnosis of COPD receive ICS/LABA as initial therapy and that more than 70% of patients with COPD use these medications in Europe and United States (22).

The OSMED report from the Italian Medicines Agency (AIFA) on the use of medications in Italy in more than 1 million patients revealed that 55.6% of patients with COPD treated with ICS have no risk of exacerbation and this ratio increased to 63.4% when patients receiving ICS for the first time were considered (23,24).

The present scenario of widespread ICS prescribing raises concerns in terms not only of treatment appropriateness but also of health costs and of patients’ safety (25).

Appropriate management of COPD starts with a correct diagnosis

Diagnosis of COPD can represent a challenging issue, particularly in the elderly and in patients with clinical characteristics overlapping asthma and COPD (e.g., partial airflow reversibility, active smoking habit but history of asthma) (26,27). The insecurity of the diagnostic process may lead to the overuse of ICS as monotherapy or in association with LABA, irrespective of their symptoms and risk of exacerbations (18) and even more generally of unappropriated inhalation treatment. This is a crucial issue that needs to be considered particularly in the population of COPD patients characterized mainly by elderly/fragile patients with multiple co-morbidities (6).

Overuse of ICS may cause adverse events (28), including the increased risk of developing pneumonia (17), particularly in patients with COPD, as shown by the large TORCH and INSPIRE clinical trials (29–31) and other studies (17,32). For the increased risk of non-fatal pneumonia in patients with COPD, underscored by the most recent GOLD report (6), the existence of a class effect has been hypothesized (32,33). In addition, chronic use of high-dose ICS may carry the potential to induce systemic side effects, such as cataracts, glaucoma, diabetes, osteoporosis, and fractures. These ICS-related adverse events are more prevalent in elderly patients because of the coexistence of extra-pulmonary diseases, particularly cardiac diseases, as well as and age-related behavioral and cognitive issues. Price et al. recently reported that ICS use was associated with an increase in the risk of incident diabetes in the COPD population (34).

Spirometry assessment is mandatory for COPD diagnosis. The guidelines recommend including the evaluation with spirometry in symptomatic patients with risk factors in the diagnostic process. A recent international survey revealed that most European specialists and GPs use spirometry, when available, during a COPD diagnosis (20). However, a remarkable percentage of GPs, especially in France and Italy, reported that no spirometry was available in their setting (20,35). The under-utilization of spirometry for COPD diagnosis is more frequent in other world regions, as demonstrated by a large web survey among GPs from Asia-Pacific, Eastern Europe, and Latin America (6,20,36,37). On the other hand, there is a general consensus that the assessment of the disease severity and the definition of prognosis should include further elements beside pulmonary function, such as symptoms (cough, expectoration, shortness of breath), risk factors for COPD (smoking, air pollution, infections and genetic predisposition), limitation of patient activity and history of exacerbations (6,20,38–41). In the attempt of overcoming the underutilization of lung function assessment in the clinical settings, the collaboration between GPs and pulmonologists should be strongly encouraged and implemented. A pro-active approach to the individuals attending the primary care offices should be possibly used, followed by an in-house intervention by specialists. Expert systems specifically designed to assist the GP in the diagnostic process could also open new strategies to unveil undiagnosed chronic obstructive lung diseases.

Discrepancy between guidelines and clinical practice

National and international guidelines aim at supporting the physicians during the processes of diagnosis and disease management to provide any single COPD patient with the most appropriate medical care, by adhering to evidence-based medicine. Nevertheless, despite the generally good knowledge of guidelines among respiratory specialists (less common among GPs), low rates of adherence to recommendations of the guidelines for the treatment of COPD were observed in different countries and world regions, when the patterns of real-life clinical practice were examined (20,21,37,42). The large survey conducted in 12 countries on the knowledge and the application of GOLD recommendations among physicians, reported that only 58% of the GP population examined were familiar with the GOLD and/or the American Thoracic Society/European Respiratory Society (ATS/ERS) documents. The disparity in this data, compared to the high percentage of specialists who resulted highly familiar with these recommendations (93%), was attributed to the fact that GPs perceive the revised GOLD categories for COPD as hardly applicable to primary care (37). This may have a relevant impact on the initiation of therapy because GPs are more frequently the physicians who are responsible for the initial management of COPD patients. Indeed, the analysis in approximately 25,000 patients from the UK primary-care setting revealed that no treatment was administered to a relevant fraction of patients with symptoms, while most patients were treated with ICS, irrespective of the GOLD criteria for this treatment (17).

However, a discrepancy between guidelines and clinical practice was observed also in the setting of specialist centers. A survey of COPD outpatients managed by pulmonary specialists in Germany revealed that the existing discrepancies between GOLD advices for COPD and the current routine clinical practice included overuse of ICS in patients with moderate COPD, as well as lack of prescription of pulmonary rehabilitation in cases for which guidelines recommend this intervention (moderate COPD or in presence of symptoms and disability) (38). Similarly, an observational study in 4,094 patients treated in 49 pulmonary units across Italy showed that, in accordance with the GOLD recommendations effective at the time of the study (43), a considerable proportion (62.1%) of COPD patients for which there is no specific indication for ICS use (i.e., patients with mild to moderate airflow obstruction) received this treatment suggesting an over-prescription/overuse of ICS in COPD (44).

More common drivers of prevailing use of ICS

A large body of literature on the randomized controlled studies of the last 20 years demonstrated the efficacy of ICS in patients with COPD, especially in combination with a LABA, in reducing the rate of exacerbation and even of all-cause mortality (41). These results made available to clinicians a drug class that can be systematically used for the management of COPD in the more advanced disease, as indicated by the common guidelines. As a consequence, a considerably higher number of patients have been potentially exposed to the adverse events associated with ICS prolonged use without a specific indication for the use of this medication (25).

Several drivers determining the uncoupling between the evidence-based recommendations and the real-life practice will be discussed here.

Diagnostic uncertainty. The above-mentioned inaccurate diagnosis is one of the drivers that may lead to inappropriate treatment choice. This occurs especially in patients who have concomitant clinical features of asthma, as demonstrated by the high rate of misdiagnosis found by the UK CADRE study on a large population of patients in primary-care setting, where asthma was overdiagnosed and COPD underdiagnosed. Therefore, since ICS are the gold standard for the management of asthma, ICS are frequently overused in patients with COPD (17). Inaccurate diagnosis may arise from the unavailability of accurate spirometry devices (20), because in absence of reliable diffusion data, GPs may be prompted to prescribe ICS based on the assumption that this medication is harmless. In summary, the diagnostic uncertainty may be tolerated if it is a temporary condition that needs to be solved to achieve a full and accurate characterization of the patient, who shall receive the most appropriate treatment then.

We will discuss other major drivers for the prevailing use of ICS, by examining the evidence in favor (pro) and against (con) the use of ICS in COPD.

Clinical phenotype. Treatments of choice preferred by physicians may be based on the different clinical characteristics of the patients. However, differences among the national guidelines exist and it is debated whether patients’ features are to be considered as distinct phenotypes (i.e., mostly emphysematous or mostly bronchitic) and how useful the phenotype-based classification may be used as a driver for treatment strategies (20).

Pro: In favor of the phenotype-based identification as main decision driver, the pharmacological approach to ICS therapy has been proposed for COPD, according to the following classification of the prevalent underlying disease: panlobular emphysema (no ICS prescribed), intermediate phenotype of chronic bronchiolitis with centrilobular emphysema (low dose of ICS), and chronic bronchiolitis (high/low dose of ICS) (45). Notably, patients with emphysematous phenotypes are associated with a rapid annual decline in FEV1, while those characterized by the eosinophilic phenotype display a slower decline or are even non-decliner (46).

Con: On the other hand, it has been shown that the fraction of patients with a prevalent emphysematous phenotype is very low in the population of patients examined (approximately 3.7%) (47,48). Therefore, the phenotype appears as having a poor impact on clinical practice as a driver for the therapeutic decision (48).

Frequency of exacerbations. “COPD exacerbations are defined as an acute worsening of respiratory symptoms that result in additional therapy” (6). Exacerbation rate negatively impacts on the natural history of the disease. Indeed, patients with COPD who have frequent exacerbation (>2 episode/year), display poorer quality of life, faster lung function decline, and worst long-term survival. Reducing exacerbation rate is one of the main clinical outcomes in COPD treatment and management (6). The rate of occurrence of exacerbations in patients with COPD has been suggested as a useful prognostic marker, driving the decision toward the use of ICS. However, whether patients with frequent exacerbation have a better response to ICS treatment compared to patients without frequent exacerbation remains to be specifically evaluated yet.

Pro: Several clinical studies, like ISOLDE (7), TRISTAN (8), TORCH (9), as well as a Cochrane meta-analysis (49), and a meta-analysis (10) on 10 and 9 randomized studies respectively, revealed a remarkable clinical benefit of ICS/LABA combinations also in terms of reduced frequency of exacerbations (41). Interestingly, a very recent analysis performed in a large real-life database showed that elevated blood eosinophil counts may predict COPD exacerbation risk suggesting a biological substrate for ICS treatment in COPD patients with frequent exacerbations (50).

Con: A clear evaluation of the changes in the frequency of exacerbations is difficult to achieve due to multiple factors including patient selection and inter-countries differences in hospital accessibility (11). The additional favorable effect on bronchodilation obtained by addition of ICS to bronchodilator drugs occurred commonly in selected subpopulations of patients (with moderate to severe COPD) (7,10,49). Moreover, dual bronchodilator therapy (LABA/LAMA combinations) has been outlined as a valid alternative to the ICS combination with a bronchodilator drug (usually LABA) (51). This therapeutic option is supported by the WISDOM study, demonstrating that the frequencies of exacerbations were not modified upon withdrawal of ICS from patients with severe COPD treated with a LABA/LAMA regimen (52), and by the large randomized FLAME study, showing the superiority of a LABA/LAMA combination (indacaterol/glycopyrronium) versus an ICS/LABA regimen (fluticasone/salmeterol) in preventing COPD exacerbations (53) in COPD patients characterized by severe airflow obstruction and mainly no history of frequent (≥2) exacerbations.

These results altogether suggest that the frequency of exacerbations should be used as a driver only for tailored inhaled therapies in selected subpopulations of patients with COPD, rather than being considered as a generalized tool for making therapeutic decisions.

Respiratory function. The spirometry evaluation of FEV1 is used to routinely verify lung function. The decline in lung function with age is a physiological event occurring in healthy subjects. Patients with COPD, especially if smokers and with exacerbations, experience a faster decline in their respiratory functions. The decline in lung function is an independent risk factor for hospitalization and mortality (54). It has been suggested that the pharmacological treatment with ICS may improve the respiratory function or may reduce the decline in FEV1, even if the amplitude of these effects and their clinical relevance have been questioned (41).

Pro: A meta-analysis including 8 long-term studies suggested that ICS led to a substantial decrease in the rate of FEV1 deterioration when used for at least 2 years (55). A post-hoc analysis of the TORCH study established that in patients with COPD, where ICS was added to a bronchodilator regimen, the rate of lung function decline was slower compared to placebo (56). Moreover, lung function decreased more during the final step of the withdrawal of glucocorticoids than in conditions of continuous glucocorticoid therapy along the 52-week period of the WISDOM study, with no relevant effects on exacerbations and on the modified Medical Research Council (mMRC) score (52). A post-hoc analysis of the WISDOM study after complete ICS withdrawal revealed that loss of pulmonary function was stabilized in the entire 12-month period (57). Finally, the recent large SUMMIT multicenter and multinational study revealed that the rate of decline in FEV1 was considerably reduced by ICS/ultra-LABA combination therapy and even by ICS (fluticasone furoate) monotherapy (58).

Con: In the post-hoc analysis of the TORCH study (56), the reduction in decline of lung function was observed in all treatment arms (fluticasone propionate/salmeterol, fluticasone propionate, and salmeterol) compared to placebo, but no differences were observed among the treatments, making more difficult to ascertain the actual role of ICS in FEV1 reduction (41,59). Similar results were obtained by two replicated parallel-group randomized studies, where no differences in lung function between treatment arms (fluticasone propionate/vilanterol versus vilanterol) were observed, in presence of a relevant difference observed in both groups versus placebo (60). Moreover, a large Cochrane meta-analysis on long-term use of ICS in 16,154 patients with a diagnosis of COPD reached the conclusion that no beneficial effect on FEV1 was attributable to continued use of ICS (49). Finally, it must be emphasized that the favorable results of the SUMMIT study were obtained in a highly selected population of patients with moderate or severe COPD and with concomitant history, or at increased risk, of cardiovascular disease (58).

In conclusion, the slowing down of FEV1 decline over time is beneficial for the patient, in terms both of quality of life and of disease control. However, the evidence from the studies on the effect of ICS on the decline of the respiratory function indicates that the effects of inhaled corticosteroids are questionable or, when present, are relatively small and poorly relevant in clinical terms.

Eosinophilia. COPD is a disease characterized by persistent and usually progressive airflow limitation, associated with enhanced chronic inflammatory response in the airways and the lungs following exposure to noxious particles or gasses (6). Moreover, exacerbations of COPD are associated with increased airway inflammation. While the predominant component of inflammation in COPD patients is neutrophilic, it has been demonstrated the existence of a subset of COPD patients with eosinophilic airway inflammation. Interestingly, high “baseline” eosinophils are more likely associated with eosinophilic exacerbations (61). In these patients, blood eosinophils levels above 2% have been considered a surrogate of airway eosinophil levels above 3%.

Pro: The presence of eosinophilic inflammation has been proposed as a marker of response to ICS treatment, because of a possible higher effectiveness of ICS on this inflammation profile (14,15,62). Notably, a reduced or even zero risk of ICS-related pneumonia was shown in patients with high level of eosinophil counts (14), suggesting that this subgroup of COPD patients might be eligible for a safe ICS treatment.

The patients with COPD having eosinophilic inflammation were shown very responsive to ICS treatment in terms of FEV1 improvement (13,62). Moreover, sputum eosinophil counts increased in COPD patients with exacerbations (61) and exacerbations were reduced in patients treated with ICS by titrating corticosteroid doses (63). A secondary analysis of data from two parallel randomized controlled studies in patients treated with ICS/ultra-LABA or ultra-LABA alone showed that eosinophil counts higher than 2% were not associated with an increased risk of pneumonia in patients treated with corticosteroids compared to bronchodilator therapy alone (14). A post-hoc analysis of the WISDOM study revealed the association of blood eosinophil count with exacerbations of COPD after ICS withdrawal. In patients treated with LABA/LAMA the increase in the exacerbation rates upon ICS withdrawal occurred at baseline eosinophil counts ≥ 2% and was proportional to the cut-off level, being important at 4% and above (57). The combination of a high baseline eosinophil count with the history of exacerbation allowed the identification of a subgroup of patients at high risk of exacerbations upon ICS withdrawal (64). Eosinophil counts data are quite easy to obtain since for all COPD patients at least one blood count per year is available.

Therefore, the potential use of high level of eosinophils as a biomarker of eosinophilic airway inflammation, as a predictor of future exacerbations (50), and as a predictor of response to ICS became an attracting perspective.

Con: One problem of this approach is the selection of the cut-off value of the eosinophil count. A cut-off of 2% (regarded as within the normal range) has been chosen by several studies. However, the reduction of exacerbation frequency was observed both in the groups with baseline eosinophil counts below 2% and in patients with baseline eosinophils counts above this cut-off (14,65), suggesting that this percentage value is not a valid threshold for the prediction of exacerbations. The total eosinophil count cut-off of 0.34 × 10⁹ cells per liter has been used recently, associating a relevant increase in severe exacerbation rates with cell counts above this level. However, these interesting results have been obtained only for severe COPD exacerbations requiring hospitalization, while for moderate or moderate exacerbations the effects were weak and inconsistent (65). The recent post hoc analysis of the WISDOM study evaluated subgroups of patients with eosinophil counts above 4% of the total white blood cell count at study onset and found that the 4% cut-off (or 0.3 × 10⁹ cells per liter) better identified the unfavorable effect of ICS withdrawal, in term of increase of the exacerbation rate (57). Notably, eosinophilia is more an unspecific marker of inflammation rather than a specific biomarker for steroid action and it has been reported that eosinophil counts of single COPD patients were either stable or variable, spanning the 2% cut-off independently from the exacerbation patterns. This suggests that the fluctuation of eosinophilia is more patient-specific rather than linked to the exacerbation rate (66).

Moreover, even if in subgroups of patients with eosinophil counts above 4% fewer exacerbations were observed with continued use of ICS, most of these patients showed a clear-cut reduction of exacerbations while taking a LABA plus LAMA alone. In conclusion, the management of most patients with stable severe COPD is attainable with a suitable therapy based on inhaled bronchodilators (57).

Conclusions

Airflow limitation is a common feature of patients with either COPD or asthma referred to primary care. This, along with other confounding factors (age, smoking, and airflow reversibility), contributes to diagnostic uncertainty. The inaccurate diagnosis of COPD might result in the prevailing prescribing of ICS in monotherapy or associated with LABA (26). ICS therapy is a first-line option for asthma, but according to current documents on the management of COPD, the association of ICSs with long-acting bronchodilators should be considered for severe patients at high risk of exacerbations and/or with persistent exacerbations despite regular treatment with bronchodilators (Groups C and D) (6) or patients with disease history suggestive of asthma-COPD overlap (6). Misdiagnosis and overuse of ICS may cause adverse events and the class-effect of increased risk of pneumonia (32,33).

The availability of biomarkers predicting ICS effectiveness shall make the drivers guiding this selection univocal and more evidence-based. Unfortunately, none of the biomarkers currently proposed (e.g., eosinophil count) has been clearly validated yet, neither were the most common drivers in use (patient phenotype, frequency of exacerbations, respiratory function decline). Indeed, a unique parameter/driver can hardly be considered conclusive, in the typical COPD scenario of variability and complexity. In daily clinical practice, where some tools may be hardly accessible, exacerbation could be set at the top of the hierarchy, followed by symptoms and lung function (67).

In conclusion, bronchodilators are the mainstay of COPD treatment. Adding ICS to LABA and LAMA regimen can be a therapeutic option for some patients. The last version of the international clinical recommendations suggests ICS-containing treatments for patients for COPD patients at high risk of exacerbations and/or who develop further exacerbations, despite being regularly treated with bronchodilators. Post-hoc analyses from several trials suggest that eosinophil counts in sputum and blood may serve as a biomarker to predict the efficacy of ICS, regarding exacerbation prevention in particular. History of asthma has been proposed as another possible element able to predict response to ICS (6). Therefore, it appears as mandatory that a selection of patients is properly made to promote a tailored ICS therapy and to minimize risks.

Declaration of interest

MC reports grants from GlaxoSmithKline and Chiesi, and personal fees from Chiesi, AstraZeneca, Boehringer Ingelheim, Novartis, Menarini, Mundipharma, Almirall, and Zambon.

AGC has received payments for board membership, income for lectures, or support for research in respiratory disease from the following organizations: AstraZeneca SpA, Boehringer Ingelheim Italia SpA, GlaxoSmithKline SpA, Grifols Italia SpA., Meda Pharma, Novartis Farma SpA., Stallergenes Italia Srl, CSL Behring SpA, A. Menarini Industrie Farmaceutiche Riunite Srl.

PS participated as a lecturer, speaker, and advisor in scientific meetings and courses under the sponsorship of AstraZeneca, Biofutura, Boehringer Ingelheim, Chiesi Farmaceutici, GlaxoSmithKline, Menarini, Malesci, Guidotti, Mundipharma, Novartis, and Zambon.

FDM has received honoraria for lectures at national and international meetings from Almirall, AstraZeneca, Boehringer Ingelheim, Chiesi Farmaceutici, Dompe, Guidotti/Malesci, GlaxoSmithKline, Menarini, Novartis, and Zambon. He has served as a consultant for AstraZeneca, Chiesi Farmaceutici, Novartis, and Zambon. He has received financial support for research from Novartis.

FB participated as a lecturer, speaker, and advisor in scientific meetings and courses under the sponsorship of AstraZeneca, Biofutura, Boehringer Ingelheim, Chiesi Farmaceutici, GlaxoSmithKline, Menarini, Malesci, Guidotti, Lallemand, Mundipharma, Novartis, Zambon, and Dompè.

PR participated as a lecturer, speaker, and advisor in scientific meetings and courses under the sponsorship of Almirall, AstraZeneca, Biofutura, Boehringer Ingelheim, Chiesi Farmaceutici, GlaxoSmithKline, Menarini Group, Mundipharma, and Novartis. Her department was funded by Almirall, Boehringer Ingelheim, Novartis, and Zambon.

LC has participated as an advisor in scientific meetings under the sponsorship of Boehringer Ingelheim and Novartis, received non-financial support by AstraZeneca, received a research grant partially funded by Boehringer Ingelheim, Novartis, and Almirall, and is or has been a consultant to Edmond Pharma, Zambon and Verona Pharma. His department was funded by Almirall, Boehringer Ingelheim, Novartis, and Zambon.

NS received honoraria for lectures at scientific meetings from Astra-Zeneca, Boehringer Ingelheim, Chiesi Farmaceutici, Guidotti/Malesci, Menarini, and Novartis. He has served as a consultant for AstraZeneca, Boehringer Ingelheim, Chiesi Farmaceutici, Novartis, and Zambon. He has received financial support for research from Astra Zeneca, Chiesi Farmaceutici, Guidotti/Malesci, and Novartis.

Acknowledgments

Editorial assistance for the manuscript was provided by Dr. Luisa Granziero, on behalf of Health Publishing & Services srl and supported by an unrestricted grant from Boehringer Ingelheim. Dr. Granziero declares there are no potential conflicts of interest relating to her assistance.