Real-life effectiveness of mepolizumab in severe asthma: a systematic literature review

Abstract Objective The efficacy and safety of mepolizumab in patients with severe eosinophilic asthma in randomized controlled trials is well established. Following approval of mepolizumab as add-on therapy for severe eosinophilic asthma in multiple regions worldwide, it is now important to determine its impact in real-world settings in which patients are not subject to stringent eligibility criteria. This systematic literature review assessed published evidence of clinical outcomes, safety, and healthcare resource use among patients with severe asthma receiving mepolizumab in real-world settings. Data sources Searches were conducted in Embase, MEDLINE, and MEDLINE In-Process via Ovid. Study selections Eligible studies were observational, and enrolled ≥10 patients with asthma who received mepolizumab 100 mg subcutaneously. Data extracted included annualized exacerbation rate, mean daily oral corticosteroid (OCS) dose, proportion of patients using OCS, several measures of lung function, patient-reported asthma control and health-related quality of life (HRQoL), safety, and economic burden. Results Twenty-three articles (22 unique studies; 2,040 patients with severe asthma on mepolizumab) were identified. Mepolizumab use was associated with a reduction in annualized exacerbation rates (requiring OCS) of 54–97% (p < 0.05 in all studies), reduced mean/median daily OCS doses, and OCS discontinuation during follow-up (27–84% of patients). Improvements in lung function, asthma control, and HRQoL were also observed. The most commonly reported adverse events included headache and arthralgia; discontinuation of mepolizumab due to adverse events occurred in 0–10.6% of patients. Conclusion Findings show that patients with severe asthma consistently demonstrate clinically relevant benefits with mepolizumab treatment in a real-world setting. Supplemental data for this article is available online at at www.tandfonline.com/ijas .


Introduction
Asthma is a heterogeneous respiratory condition affecting more than 339 million people globally (1). Patients with severe asthma typically have poorly controlled disease and experience frequent asthma exacerbations, impaired lung function and reduced health-related quality of life (HRQoL), despite standard of care therapy, which consists of medium to high doses of inhaled corticosteroids (ICS) plus one or more additional controller medications (such as a long-acting beta agonist [LABA]) (2)(3)(4). Patients whose symptoms persist despite optimization (i.e. good adherence and correct inhaler technique) of medium to high dose ICS/LABA treatment should be considered for add-on therapy such as biologics that target the inflammatory signaling pathways seen in asthma (4).
A subset of patients with severe asthma have severe eosinophilic asthma, characterized by elevated levels of eosinophils in the blood and/or airways (3). Mepolizumab is a humanized monoclonal antibody that binds to and neutralizes interleukin (IL)-5, thus inhibiting IL-5 signaling and blocking eosinophil proliferation, activation, and survival (5,6). It is approved as an add-on therapy for patients with severe eosinophilic asthma and eosinophilic granulomatosis with polyangiitis in multiple regions worldwide, and for patients with hypereosinophilic syndrome in the US (7)(8)(9). Randomized controlled trials (RCTs) (10)(11)(12)(13) and real-world studies (14)(15)(16) have demonstrated that compared with placebo, mepolizumab reduces exacerbations and oral corticosteroid (OCS) dependence, in addition to improving lung function, asthma control and HRQoL in patients with severe eosinophilic asthma. As real-world use of mepolizumab increases, the population of patients with access to mepolizumab is more diverse than those included in the mepolizumab RCTs, which have strict eligibility criteria that often lead to the exclusion of patients with respiratory comorbidities and other characteristics, such as a recent smoking history and lack of adherence to controller medications, frequently found in the real-world severe asthma population (11,13,(17)(18)(19). It is therefore of clinical interest to fully understand the extent to which the clinical benefits associated with mepolizumab are observed in routine clinical practice.
The aim of this systematic literature review (SLR) was to assess the published evidence relating to a broad range of clinical outcomes, safety, and healthcare resource utilization (HCRU) among patients with severe asthma receiving add-on therapy with mepolizumab in a real-world setting.

Methods
This SLR was conducted and reported according to Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines (20). All SLR methodology followed the established best methods used in systematic review research (21,22).

Search strategy
The SLR was designed to identify studies reporting data on the effectiveness and safety of mepolizumab in patients with asthma in a real-world setting. Systematic literature searches were conducted in Embase, MEDLINE and MEDLINE In-Process via Ovid, using a combination of free-text terms and medical subject headings for asthma combined with terms for mepolizumab (Supplementary Table 1). Searches were limited to articles published in the English language and publications reporting the results of clinical trials were excluded. A hand search of conference proceedings for several key 2018, 2019, and 2020 meetings (as available) was also conducted. This included the following congresses: American Academy of Allergy, Asthma, and Immunology (2018,

Study selection and data extraction
The eligibility criteria for abstracts and full-text articles were based on the Population, Intervention, Comparison, Outcomes, and Study (PICOS) design framework, and are described in full in Supplementary  Table 2. Briefly, eligible publications were published prior to 10 July, 2020 (with no lower limit applied to the publication date) and reported data from observational studies (prospective and retrospective cohort studies, registry and claims database analyses, cross-sectional studies) that enrolled ≥10 patients ≥6 years of age with asthma who had been treated with the approved dose of mepolizumab (100 mg, administered subcutaneously). Eligible publications also had to include data on at least one outcome of interest, including clinical outcomes (such as exacerbation frequency or lung function), disease control or HRQoL outcomes, or economic outcomes (Supplementary Table 2). The data could be from the overall study population, or for subgroups of patients stratified by baseline blood eosinophil count. Other subgroups (such as patients deemed to be treatment responders/ non-responders) were not included, owing to inconsistent definitions of treatment response across studies.
Any duplicate publications identified across databases were removed from the initial search results and all remaining unique publications were screened using Distiller Systematic Review software. Titles/abstracts and then full-text publications were screened for inclusion by two independent investigators and any discrepancies between investigators were resolved by a third investigator. Once all abstracts and full-text articles meeting the eligibility criteria were identified, data from these publications were extracted by a single investigator into a data extraction template. The data were then independently validated by a second, senior investigator.

Data synthesis
The results of the SLR were summarized qualitatively and no statistical analyses were planned. With regards

Ethics
Only publicly available summary-level data were included in this SLR; ethical approval was therefore not required.

Exacerbations
A total of 16 studies (three prospective and 13 retrospective) reported annualized exacerbation rates. Although the definitions of an exacerbation varied across the studies, common factors across the definitions used included worsening asthma symptoms requiring emergency department (ED) or hospital visits, treatment with systemic corticosteroids, and/or increases in mOCS doses. Three studies (28,32,34) did not provide a definition for exacerbations. The exacerbation definition(s) adopted by each study as well as changes in exacerbation rate with mepolizumab are described in Table 2.
Three prospective studies, all with 12 months of follow-up data, reported statistically significant (p < 0.001) reductions in OCS-requiring exacerbations with mepolizumab. These studies reported reductions from baseline to follow-up of 66% (4.16 vs 1.41 events/year) (24), 68% (2.80 vs 0.90 events/year) (25) and 73% (3.14 vs 0.85 events/year) (15), with one study also reporting a 54% reduction in the annualized rate of exacerbations requiring hospitalization (0.57 vs 0.26 events/year at baseline vs follow-up; p < 0.001) (24) (Figure 2A and 2B). The findings of the prospective studies were supported by data from the 13 retrospective studies, which included between 6 and 24 months    Figure 2B). Two of the retrospective studies presented results in subgroups according to baseline blood eosinophil count (29,40). In both studies, exacerbation rates were reduced with mepolizumab across all blood eosinophil count groups; of note, the majority of patients from both studies (80% and 93%) were receiving mOCS at baseline. The proportion of patients with exacerbations (any definition) before and after mepolizumab initiation was reported by one prospective (15) and three retrospective (14,30,35) studies ( Table 2). The prospective study reported an 85% relative risk reduction from baseline to follow-up in the proportion of patients with ≥1 exacerbation (85% in the 12 months preceding vs 13% in the 12 months following mepolizumab initiation). The three retrospective studies all reported reductions from baseline to follow-up in the proportion of patients with ≥1 exacerbation, ranging from 24% (14) to 85% (15); one of these also showed a 65% relative risk reduction in the proportion of patients with an exacerbation requiring hospitalization (14).
Three studies (two prospective (24,26) and one retrospective (35)) showed that mepolizumab had a statistically significant (p < 0.001) beneficial impact on Asthma Quality of Life Questionnaire (AQLQ) or mini-AQLQ score. Mean baseline scores ranged from 3.48 (26) to 4.05 (35) points; the prospective studies reported AQLQ improvements of 1.5 (24) and 1.24 (26) points, and the retrospective study reported a 0.86-point improvement (35) in mini-AQLQ score (Supplementary Table 7). One retrospective study (33) also reported Sino-nasal Outcome Test-22 scores among 26 patients with severe eosinophilic asthma, and showed a statistically significant improvement of 16.9 points (from a mean baseline score of 40.5 points) following mepolizumab treatment, almost double the MCID of 8.9 points (44) (Supplementary Table 7).

Safety and treatment discontinuation
Across 10 studies with available data (three prospective (24)(25)(26) and seven retrospective (16,27,33,35,37,40,41)), the overall rate of mepolizumab discontinuation ranged Darker spheres indicate prospective studies and paler spheres indicate retrospective studies; sphere size is proportional to study patient population size; for studies reporting statistical analyses, statistical significance information was extracted and has been reported in tables 2 and 3 (exacerbation and oCs data), and supplementary tables 5-7 (fEV 1 , blood eosinophil count, aCQ and aCt data).

Economic outcomes: HCRU and cost
Two retrospective studies presented data on HCRU before and after mepolizumab treatment. Both of these studies, one from Italy (29) and one from the US (14), reported a statistically significant decrease in the mean number of hospitalizations in the 12 months following versus the 12 months preceding mepolizumab initiation. The US study also reported a significant reduction from baseline to follow-up in mean exacerbation-related costs per patient ($5,178 vs $2,383; p < 0.001) and a numerical reduction in the mean number of OCS pharmacy claims during the follow-up period (3.9 claims) compared with the baseline period (5.5 claims) (14).

Findings from the gray literature
Results reported in the gray literature generally supported those of the full-text peer-reviewed publications. Where data were available, overall improvements in exacerbation rates, mOCS use, lung function, blood eosinophil count, and asthma control were observed. Rates of mepolizumab discontinuation reported in the gray literature were also similar to those reported in the full-text peer-reviewed publications.

Discussion
Across the current respiratory treatment landscape, there is an important need for real-world studies with high external validity to support the efficacy results of RCTs (45). This SLR aimed to provide a broad picture of the full range of clinical impacts seen with mepolizumab in a real-world setting, in order to determine whether the clinical benefits shown in RCTs of mepolizumab in severe eosinophilic asthma, namely reduced exacerbation rates, reduced OCS use, improved asthma control, and improved HRQoL (10)(11)(12)(13), are also seen in more diverse, real-world populations. Our findings indicate that patients with severe asthma receiving mepolizumab treatment in real-world clinical practice experience important health benefits, including reductions in exacerbations, OCS use and blood eosinophil counts, plus improvements in lung function, asthma control and HRQoL.
The observational real-world studies identified by the SLR included a heterogeneous patient population, with a broader spectrum of disease severity, comorbidities and concomitant medications compared with those typically enrolled in RCTs. Compared with patients enrolled in previous mepolizumab RCTs (10)(11)(12)(13)46,47), the patients included in real-world studies were generally older, with more severe asthma and a larger number of comorbid conditions. Overall, 40-93% of patients included in this SLR were using mOCS prior to mepolizumab initiation, compared with 23-33% in the RCTs of mepolizumab in patients with severe asthma (where mOCS use was not a requirement for patient eligibility) (11)(12)(13). Although definitions varied, the annualized rate of exacerbations requiring at least OCS treatment was similar in the real-world literature prior to mepolizumab initiation compared with previous RCTs (2.1-5.8 events/year in this SLR vs 2.7-3.8 events/year in RCTs). Only a few of the studies included in this SLR reported on asthma-specific comorbidities; among these, the proportion of patients with comorbidities was high, and generally exceeded the proportion reported in a recent open-label extension study of mepolizumab (22% to 100% in this SLR vs 7% in RCT with nasal polyps; 49% to 100% in this SLR vs 12% in RCT with chronic rhinosinusitis) (47). This is expected, since interventional studies in patients with severe asthma typically exclude those with comorbidities associated with respiratory symptoms, such as chronic obstructive pulmonary disease. The studies identified by this SLR included a wide selection of geographic locations, with publications from the United States, Japan, Australia, and multiple European countries. Although no studies from China (where mepolizumab is not licensed) or South America were identified as meeting the eligibility criteria, the geographical spread of the available data appears to be sufficiently representative of real-world use of mepolizumab.
Despite the range of differences between the real-world populations included in this SLR and the populations included in RCTs of mepolizumab in patients with severe asthma, the clinical benefits shown in real-world settings were consistent with those shown previously in RCTs. These included a significant reduction in asthma exacerbations, which is a primary goal of asthma treatment, and indeed, reductions in exacerbation rates were among the most important efficacy results of the landmark mepolizumab RCTs. The SLR also demonstrated mepolizumab treatment to have a significant OCS-sparing effect in clinical practice, with 24-84% of patients able to discontinue OCS use completely with mepolizumab. Reducing OCS dependence is another key goal of asthma treatment, owing to multiple negative side effects associated with long-term OCS use (48). The real-world data reported here build on the results of a formal OCS-sparing trial, which reported 23% of patients receiving mepolizumab to have reduced their mOCS use by 90-100% by study end (10). It should be noted that the ability to reduce OCS was not at the expense of loss of disease control; moreover, irrespective of their improvements in disease control with mepolizumab, some patients may not have been able to discontinue mOCS entirely due to adrenal insufficiency (49). Finally, the published real-world evidence also indicated that geometric mean absolute blood eosinophil counts were reduced to levels within the identified normal range following mepolizumab initiation (50). Despite variations in the assessment tools used, improvements in asthma control and HRQoL were also observed with mepolizumab therapy. It is important to consider that since mepolizumab specifically targets IL-5, patients who had any respiratory comorbidities that were not associated with eosinophilia may have experienced smaller improvements in their asthma control than those without (24). Together the data from the large number of studies included in the SLR demonstrate that the broad range of clinical benefits shown with mepolizumab in patients with severe asthma during RCTs was mirrored in the real-world population of patients with asthma receiving mepolizumab.
In the SLR studies with available safety data, headache was the most commonly reported AE, followed by injection site reactions and arthralgia. These AEs were reported in the mepolizumab clinical trials, where headache was often also the most commonly reported AE (11)(12)(13)46,47). Mepolizumab discontinuation rates due to AEs were typically low in the studies identified by the SLR, and broadly similar to those reported in RCTs (10,11,47,51).
With regards to the impact of mepolizumab on economic burden, only one study reporting exacerbation-related costs and two studies reporting exacerbation-related HCRU were identified by the SLR. The findings of these studies associated mepolizumab treatment with substantial reductions in exacerbation-related costs for patients with severe uncontrolled asthma, compared with baseline values. In addition, hospitalizations due to asthma exacerbations were significantly decreased following mepolizumab treatment.
There are several limitations that must be considered when interpreting the findings of this SLR. Firstly, differences in study design may limit comparison of the data from real-world studies with those from RCTs. For example, restrictions in OCS tapering protocols may have prevented patients from substantially tapering or discontinuing OCS in the RCT setting; conversely, RCTs with rapid reduction algorithms may have reduced OCS use more quickly than in real-world clinical practice. Second, the absence of a control group in the included studies makes them prone to bias; however, the general consistency of the extracted data does not suggest that the results were confounded. Moreover, results of the single-centre clinical studies included in this SLR (n = 12) are likely to be less generalizable to real-world clinical practice than those of the multi-centre registry/national database studies (n = 5) (52). Third, SLRs are by nature limited to published data (and in this case those published in the English language) only. In addition, the risk of bias within individual studies was not assessed. Finally, we observed considerable heterogeneity in study design, patient cohort, sample size and outcome definitions across the included studies. In particular, the small sample size of some of the studies included in this SLR may have resulted in a lack of power to identify statistical differences in specific outcomes. Furthermore, the definitions of exacerbation varied across the studies, as did the patient reported outcomes instruments used to assess HRQoL. Owing to this between-study heterogeneity, the SLR did not pool results to perform meta-analyses. Nonetheless, the data identified by this SLR consistently demonstrated clinical benefits in patients with severe asthma receiving mepolizumab treatment. Notably, since the search was conducted in July 2020, several additional real-world studies of mepolizumab have been published, all of which report clinical benefits with mepolizumab (53)(54)(55) and therefore support the findings of this SLR.

Conclusions
Despite substantial heterogeneity observed in the available literature, mepolizumab was consistently associated with clinically relevant benefits in eligible patients with severe asthma receiving treatment in real-world clinical practice. It is therefore reasonable to conclude that the benefits of mepolizumab demonstrated in previous RCTs extend to patients with heterogeneous demographics and characteristics, treated in a real-world setting.