Search

Discontinuation from randomised treatment is a common intercurrent event in clinical trials. When the target estimand uses a treatment policy strategy to deal with this intercurrent event, data after cessation of treatment is relevant to estimate the estimand and all efforts should be made to collect such data. Missing data may nevertheless occur due to participants withdrawing from the study and assumptions regarding the values for data that are missing are required for estimation. A missing-at-random assumption is commonly made in this setting, but it may not always be viewed as appropriate. Another potential approach is to assume missing values are similar to data collected after treatment discontinuation. This idea has been previously proposed in the context of recurrent event data. Here we extend this approach to time-to-event outcomes using the hazard function. We propose imputation models that allow for different hazard rates before and after treatment discontinuation and use the posttreatment discontinuation hazard to impute events for participants with missing follow-up periods due to study withdrawal. The imputation models are fitted as Andersen-Gill models. We illustrate the proposed methods with an example of a clinical trial in patients with chronic obstructive pulmonary disease.

Effective treatment of inflammatory diseases is often challenging owing to their heterogeneous pathophysiology. Understanding of the underlying disease mechanisms is improving and it is now clear that eosinophils play a complex pathophysiological role in a broad range of type 2 inflammatory diseases. Standard of care for these conditions often still includes oral corticosteroids (OCS) and/or cytotoxic immune therapies, which are associated with debilitating side effects. Selective, biological eosinophil-reducing agents provide treatment options that improve clinical symptoms associated with eosinophilic inflammation and reduce OCS use. Mepolizumab is a humanized monoclonal antibody that binds to and neutralizes interleukin-5, the major cytokine involved in eosinophil proliferation, activation, and survival. Mepolizumab is approved for the treatment of severe eosinophilic asthma, eosinophilic granulomatosis with polyangiitis and hypereosinophilic syndrome. Additionally, the efficacy of add-on mepolizumab has been observed in patients with severe chronic rhinosinusitis with nasal polyposis and chronic obstructive pulmonary disease with an eosinophilic phenotype. Here, we review the development, approval, and real-world effectiveness of mepolizumab for the treatment of patients with severe eosinophilic asthma, from the DREAM to REALITI-A studies, and describe how knowledge from this journey extended to the use of mepolizumab and other biologics across a broad spectrum of eosinophilic diseases.

In the past, many clinical trials have withdrawn subjects from the study when they prematurely stopped their randomised treatment and have therefore only collected 'on-treatment' data. Thus, analyses addressing a treatment policy estimand have been restricted to imputing missing data under assumptions drawn from these data only. Many confirmatory trials are now continuing to collect data from subjects in a study even after they have prematurely discontinued study treatment as this event is irrelevant for the purposes of a treatment policy estimand. However, despite efforts to keep subjects in a trial, some will still choose to withdraw. Recent publications for sensitivity analyses of recurrent event data have focused on the reference-based imputation methods commonly applied to continuous outcomes, where imputation for the missing data for one treatment arm is based on the observed outcomes in another arm. However, the existence of data from subjects who have prematurely discontinued treatment but remained in the study has now raised the opportunity to use this 'off-treatment' data to impute the missing data for subjects who withdraw, potentially allowing more plausible assumptions for the missing post-study-withdrawal data than reference-based approaches. In this paper, we introduce a new imputation method for recurrent event data in which the missing post-study-withdrawal event rate for a particular subject is assumed to reflect that observed from subjects during the off-treatment period. The method is illustrated in a trial in chronic obstructive pulmonary disease (COPD) where the primary endpoint was the rate of exacerbations, analysed using a negative binomial model.

Assessment of efficacy in important subgroups – such as those defined by sex, age, race and region – in confirmatory trials is typically performed using separate analysis of the specific subgroup. This ignores relevant information from the complementary subgroup. Bayesian dynamic borrowing uses an informative prior based on analysis of the complementary subgroup and a weak prior distribution centred on a mean of zero to construct a robust mixture prior. This combination of priors allows for dynamic borrowing of prior information; the analysis learns how much of the complementary subgroup prior information to borrow based on the consistency between the subgroup of interest and the complementary subgroup. A tipping point analysis can be carried out to identify how much prior weight needs to be placed on the complementary subgroup component of the robust mixture prior to establish efficacy in the subgroup of interest. An attractive feature of the tipping point analysis is that it enables the evidence from the source subgroup, the evidence from the target subgroup, and the combined evidence to be displayed alongside each other. This method is illustrated with an example trial in severe asthma where efficacy in the adolescent subgroup was assessed using a mixture prior combining an informative prior from the adult data in the same trial with a non‐informative prior.

Background: Assessment of efficacy in trials in important subgroups, e.g. as defined by sex and age, is typically performed using separate analysis of the specific subgroup. Bayesian dynamic borrowing is an innovative statistical technique that uses relevant information from the complementary subgroup to provide a more informative estimate of efficacy in the subgroup of interest. Methods: MENSA was a randomised, double-blind trial in severe asthma with an eosinophilic phenotype which compared mepolizumab with placebo in 25 adolescents (age 12-17) and 551 adults (age ≥18). The primary endpoint was the rate of clinically significant exacerbations, analysed using a negative binomial model.The prior for this post-hoc Bayesian analysis of adolescents was a weighted combination of a sceptical prior and the adult treatment effect distribution, with the weight representing the degree of belief in the similarity of adolescent and adult efficacy. Results: The estimated exacerbation rate reduction in adolescents based on a prior weight of 70% on the adult data was 49% (95% credible interval: 1% to 63%). The posterior probability of a positive treatment effect was >97.5% which is analogous to a one-sided p-value of Conclusions: For mepolizumab, given the similarity of disease presentation and pharmacological response between the adolescent and adult subgroups, a 70% weight on the relevance of the adult information to adolescents supported a conclusion of efficacy under reasonable assumptions. This innovative Bayesian dynamic borrowing approach may be especially helpful for evaluation of efficacy in paediatric patients where recruitment is challenging. Method: GSK [MEA115588; NCT01691521]

The development of mepolizumab, an anti-IL-5 monoclonal antibody for the treatment of severe eosinophilic asthma, is an example of a clinical development program that evolved over time based on sound, basic scientific principles. Initial clinical data on the effects of mepolizumab on lung function in a general asthmatic population were disappointing. However, it became clear that mepolizumab may be more effective against other clinical endpoints, particularly asthma exacerbations, in patients with more severe disease. Furthermore, a developing understanding of asthma disease pathobiology led to the identification of an appropriate target population and predictive biomarker for mepolizumab treatment: patients with severe eosinophilic asthma and blood eosinophil count. Mepolizumab use provides clinically meaningful benefits in this target population, fulfilling an unmet need. This Clinical Commentary Review describes the clinical development of mepolizumab and details of how this program informed the development of other biologic therapies in patients with severe asthma. This account highlights how a personalized approach toward treatment of patients with severe eosinophilic asthma, supported by a large body of scientific evidence, ultimately led to new and effective treatments and improved patient outcomes.

Subgroup analysis in clinical trials for regulatory and reimbursement purposes can be confirmatory or exploratory in nature. Confirmatory subgroup analysis requires pre-specification of the proposed analysis and appropriate control of the type I error rate. Exploratory subgroup analysis is a feature of Phase III clinical trials. Examination of the results by sex, age and race is required by FDA and submissions for regulatory approval typically involve numerous further analyses by baseline characteristics such as disease severity. For efficacy these exploratory analyses are often directed at providing reassurance that the overall estimated treatment effect translates into benefit for each of the subgroups and for safety to investigate the existence of signals in more vulnerable subgroups. For reimbursement purposes, extensive analysis is required to try to identify those groups experiencing most benefit and for whom the medicine is therefore most cost-effective.

Mepolizumab reduces exacerbations of severe eosinophilic asthma. However, even with mepolizumab treatment, exacerbations still occur in this population. We have previously shown that exacerbations on mepolizumab are associated with lower sputum eosinophil counts and lower decrements in symptoms measured by the visual analogue scale compared to placebo. To further characterise exacerbations on mepolizumab, we carried out a post-hoc comparison of exacerbations on treatment with mepolizumab or placebo in three previously reported placebo-controlled trials. We investigated whether exacerbations in each group differ with respect to change in lung function and symptoms in the period before and after starting oral corticosteroid (OCS) rescue treatment. Methods Diary card data was reviewed from the 3 studies; DREAM, a 52-week study of 3 doses of mepolizumab (75, 250 or 750 mg IV 4 weekly) versus placebo; MENSA, an 32-week study of 2 doses of mepolizumab (75 mg IV or 100 mg s/c 4 weekly) versus placebo; and MUSCA, a 24-week study of mepolizumab 100 mg s/c 4 weekly versus placebo. All studies recruited patients with severe eosinophilic asthma and a history of 2 or more exacerbations in the previous year. Mepolizumab dose groups were combined for analysis. Patients completed a daily diary card including a 6 point symptom score assessing asthma symptoms in the previous 24 hours and a best-of-three morning peak expiratory flow (PEF). Exacerbations requiring rescue OCS with at least 20 days of diary data in the period from 14 days prior to starting OCS (Day -14) to 14 days (Day 14) after starting OCS were included in the analysis. Results 1026 exacerbations were analysed. 476 occurred in 248 subjects on placebo and 550 occurred in 338 subjects on mepolizumab. Exacerbations on placebo were associated with a larger drop in PEF (-41.0 L/min [95% CI -47.3, -34.7]) compared to mepolizumab (-26.9 L/min [-32.7, -21.1]) over the 14 days prior to starting OCS. Exacerbations on placebo also tended to have a larger increase in daily symptom score compared to mepolizumab (0.81 points [0.68, 0.94] vs 0.65 points [0.54, 0.76] respectively). Conclusion Exacerbations that occur on mepolizumab are less severe in terms of worsening in PEF and symptom scores.

For many years there has been a consensus among the Clinical Research community that ITT analysis represents the correct approach for the vast majority of trials. Recent worldwide regulatory guidance for pharmaceutical industry trials has allowed discussion of alternatives to the ITT approach to analysis; different treatment effects can be considered which may be more clinically meaningful and more relevant to patients and prescribers. The key concept is of a trial “estimand”, a precise description of the estimated treatment effect. The strategy chosen to account for patients who discontinue treatment or take alternative medications which are not part of the randomised treatment regimen are important determinants of this treatment effect. One strategy to account for these events is treatment policy, which corresponds to an ITT approach. Alternative equally valid strategies address what the treatment effect is if the patient actually takes the treatment or does not use specific alternative medication. There is no single right answer to which strategy is most appropriate, the solution depends on the key clinical question of interest. The estimands framework discussed in the new guidance has been particularly useful in the context of the current COVID-19 pandemic and has clarified what choices are available to account for the impact of COVID-19 on clinical trials. Specifically, an ITT approach addresses a treatment effect that may not be generalisable beyond the current pandemic.

In drug development, we ask ourselves which population, endpoint and treatment comparison should be investigated. In this context, we also debate what matters most to the different stakeholders that are involved in clinical drug development, for example, patients, physicians, regulators and payers. With the publication of draft ICH E9 addendum on estimands in 2017, we now have a common framework and language to discuss such questions in an informed and transparent way. This has led to the estimand discussion being a key element in study development, including design, analysis and interpretation of a treatment effect. At an invited session at the 2018 PSI annual conference, PSI hosted a role-play debate where the aim of the session was to mimic a regulatory and payer scientific advice discussion for a COPD drug. Including role-play views from an industry sponsor, a patient, a regulator and a payer. This paper presents the invented COPD case-study design and considerations relating to appropriate estimands are discussed by each of the stakeholders from their differing viewpoints with the additional inclusion of a technical (academic) perspective. The rationale for each perspective on approaches for handling intercurrent events is presented, with a key emphasis on the application of while-on-treatment and treatment policy estimands in this context. It is increasingly recognised that the treatment effect estimated by the treatment policy approach may not always be of primary clinical interest and may not appropriately communicate to patients the efficacy they can expect if they take the treatment as directed.

Background: In METREX and METREO, mepolizumab reduced exacerbation rates in patients (pts) with eosinophilic COPD. In severe asthma, log-transformed screening blood eosinophil count (sBEC) is a strong predictor of mepolizumab efficacy; however, it is not known whether this is true in COPD. Objective: Assess the effect of sBEC on moderate/severe exacerbation rates with mepolizumab vs placebo. Methods: METREX and METREO: Phase 3, placebo-controlled, randomised, double-blind, multicentre trials enrolled pts aged ≥40 years with COPD, ≥2 moderate/≥1 severe exacerbations (prior yr) despite ICS-based triple maintenance therapy. Pts received placebo or mepolizumab SC (100mg [METREX], 100/300mg [METREO]) every 4 wks for 52 wks as add-on therapy to ICS/LABA/LAMA. Log-transformed eosinophil count was a poor fit for predicting exacerbation rates in COPD so fractional polynomial modelling was used in a post hoc analysis; all mepolizumab doses were combined. Results: Predicted exacerbation rates increased with increasing sBEC in the placebo group and were relatively stable for mepolizumab (placebo rate/yr: 1.44, 1.63, 1.77, mepolizumab rate/yr: 1.45, 1.41, 1.38 at 150, 300 and 500 cells/µL respectively). Predicted exacerbation rate reductions with mepolizumab vs placebo increased with higher sBEC: 13% (95%CI 1–24%) at 300 cells/µL and 22% (95%CI 7–35%) at 500 cells/µL. Conclusions: sBEC is a useful predictor of mepolizumab response in COPD. These data support the role of eosinophils in COPD exacerbations and the utility of mitigating eosinophilic inflammation to improve COPD outcomes in frequent exacerbators despite being on optimal ICS-based triple maintenance therapy. Funding: GSK (207577; 117113/NCT02105961; 117106/NCT02105948)

The draft addendum to the ICH E9 regulatory guideline asks for explicit definition of the treatment effect to be estimated in clinical trials. The draft guideline also introduces the concept of intercurrent events to describe events that occur post-randomisation that may affect efficacy assessment. Composite estimands allow incorporation of intercurrent events in the definition of the endpoint. A common example of an intercurrent event is discontinuation of randomised treatment and use of a composite strategy would assess treatment effect based on a variable that combines the outcome variable of interest with discontinuation of randomised treatment. Use of a composite estimand may avoid the need for imputation which would be required by a treatment policy estimand. The draft guideline gives the example of a binary approach for specifying a composite estimand. When the variable is measured on a non-binary scale, other options are available where the intercurrent event is given an extreme unfavourable value, for example comparison of median values or analysis based on categories of response. This paper reviews approaches to deriving a composite estimand and contrasts the use of this estimand to the treatment policy estimand. The benefits of using each strategy are discussed and examples of the use of the different approaches are given for a clinical trial in nasal polyposis and a steroid reduction trial in severe asthma.

Background Mepolizumab significantly reduces exacerbations in patients with severe eosinophilic asthma. The early identification of patients likely to receive long-term benefit from treatment could ensure effective resource allocation. Objective To assess potential continuation rules for mepolizumab in addition to initiation criteria defined as 2 or more exacerbations in the previous year and blood eosinophil counts of 150 cells/mL or more at initiation or 300 cells/mL or more in the previous year. Methods This post hoc analysis included data from 2 randomized, double-blind, placebo-controlled studies (NCT01000506 and NCT01691521) of mepolizumab in patients with severe eosinophilic asthma (N [ 1,192). Rules based on blood eosinophils, physician-rated response to treatment, FEV1, Asthma Control Questionnaire (ACQ-5) score, and exacerbation reduction were assessed at week 16. To assess these rules, 2 key metrics accounting for the effects observed in the placebo arm were developed. Results Patients not meeting continuation rules based on physician-rated response, FEV1, and the ACQ-5 score still derived long-term benefit from mepolizumab. Nearly all patients failing to reduce blood eosinophils had counts of 150 cells/mL or less at baseline. For exacerbations, assessment after 16 weeks was potentially premature for predicting future exacerbations. Conclusion There was no evidence of a reliable physician-rated response, ACQ-5 score, or lung functionebased continuation rule. The added value of changes in blood eosinophils at week 16 over baseline was marginal. Initiation criteria for mepolizumab treatment provide the best method for assessing patient benefit from mepolizumab treatment, and treatment continuation should be reviewed on the basis of a predefined reduction in long-term exacerbation frequency and/or oral corticosteroid dose.

Rationale: The PH3 mepolizumab (mepo) MENSA (M) study included patients (pt) with a history of ≥2 exacerbations in the year (yr) preceding randomization, receiving high dose ICS + ≥1 additional asthma controller, and blood eosinophils (eos) ≥300 cells/µL in the past yr or ≥150 cells/µL immediately prior to treatment. Recently, there has been interest in the sub-group of patients with a history of ≥3 exacerbations in the past yr and ≥300 cells/µL at baseline (sub-pop). Methods: This post-hoc analysis was conducted using a negative binomial regression model. The 100mg SC and the 75mg IV dose groups were combined for analysis. Results: In M, 191 and 385 pt received placebo (pbo) or mepo respectively, compared with 61 and 128 in the sub-pop. The exac rate/yr in M was 1.74 and 0.88 in pbo and mepo respectively, compared with 2.62 and 0.89 in the sub-pop. The rate ratio mepo v pbo (95% CI), was 0.50 (0.39 0.65), a 50% reduction for mepo pt, and 0.34 (0.23 0.51), a 66% reduction in the sub-pop. The ED/hospitalization rate/yr in M was 0.20 and 0.11 in pbo and mepo respectively, compared with 0.41 and 0.18 in the sub-pop. The rate ratio mepo v pbo (95% CI) was 0.52 (0.28 0.96), a 48% reduction for mepo pt, and 0.43 (0.17 1.07), a 57% reduction in the sub-pop. Mepo was well tolerated in the sub-pop with a safety profile similar to pbo and comparable to the M profile. Discussion: These results show increased efficacy of mepo in a sub-population with greater morbidity as compared with the overall M population. This observation is not unexpected as it has been previously reported that blood eos counts and exac history are positively correlated with treatment benefit with mepo. (Funding GSK: NCT01691521).

The last decade has seen the approval of several new biologics for the treatment of severe asthma-targeting specific endotypes and phenotypes. This review will examine how evidence generated from the mepolizumab clinical development program showed that blood eosinophil counts, rather than sputum or tissue eosinophil counts, evolved as a pharmacodynamic and predictive biomarker for the efficacy of treatment with mepolizumab in patients with severe eosinophilic asthma. Based on the available evidence and combined with clinical judgement, a baseline blood eosinophil threshold of 150 cells/μL or greater or a historical blood eosinophil threshold of 300 cells/μL or greater will allow selection of patients with severe eosinophilic asthma who are most likely to achieve clinically significant reductions in the rate of exacerbations with mepolizumab treatment.

Exploratory subgroup analyses are an increasing source of controversy as part of the interpretation of the results of clinical trials. In this article, we review the major challenges of multiplicity, statistical methods available to assess consistency of effect, and the part appropriate design plays in mitigating the risk of false conclusions from subgroup analyses. We discuss the problems associated with using definitions of consistency based on effect sizes in specific subgroups. We argue that what is required is a return to basic statistical principles, including more use of modeling techniques.

BACKGROUND: Many patients with severe asthma require regular treatment with oral glucocorticoids despite the use of high-dose inhaled therapy. However, the regular use of systemic glucocorticoids can result in serious and often irreversible adverse effects. Mepolizumab, a humanized monoclonal antibody that binds to and inactivates interleukin-5, has been shown to reduce asthma exacerbations in patients with severe eosinophilic asthma. METHODS: In a randomized, double-blind trial involving 135 patients with severe eosinophilic asthma, we compared the glucocorticoid-sparing effect of mepolizumab (at a dose of 100 mg) with that of placebo administered subcutaneously every 4 weeks for 20 weeks. The primary outcome was the degree of reduction in the glucocorticoid dose (90 to 100% reduction, 75 to less than 90% reduction, 50 to less than 75% reduction, more than 0 to less than 50% reduction, or no decrease in oral glucocorticoid dose, a lack of asthma control during weeks 20 to 24, or withdrawal from treatment). Other outcomes included the rate of asthma exacerbations, asthma control, and safety. RESULTS: The likelihood of a reduction in the glucocorticoid-dose stratum was 2.39 times greater in the mepolizumab group than in the placebo group (95% confidence interval, 1.25 to 4.56; P=0.008). The median percentage reduction from baseline in the glucocorticoid dose was 50% in the mepolizumab group, as compared with no reduction in the placebo group (P=0.007). Despite receiving a reduced glucocorticoid dose, patients in the mepolizumab group, as compared with those in the placebo group, had a relative reduction of 32% in the annualized rate of exacerbations (1.44 vs. 2.12, P=0.04) and a reduction of 0.52 points with respect to asthma symptoms (P=0.004), as measured on the Asthma Control Questionnaire 5 (in which the minimal clinically important difference is 0.5 points). The safety profile of mepolizumab was similar to that of placebo. CONCLUSIONS: In patients requiring daily oral glucocorticoid therapy to maintain asthma control, mepolizumab had a significant glucocorticoid-sparing effect, reduced exacerbations, and improved control of asthma symptoms. (Funded by GlaxoSmithKline; SIRIUS ClinicalTrials.gov number, NCT01691508.).

Introduction: Blood eosinophil count is a determinant of disease severity and anti-IL5 treatment response in patients with Severe Asthma. Use of oral corticosteroids (OCS) suppresses blood eosinophils in the presence of concomitant inhaled corticosteroids (ICS). There is, however, no published dose response for the extent of additional suppression by OCS in this population. Objectives: Evaluate the effect of OCS dose on blood eosinophil count in a population of patients with Severe Eosinophilic Asthma taking high dose ICS therapy. Methods: Baseline blood eosinophil count from two GSK-funded mepolizumab studies (DREAM, MENSA) was analysed post-hoc using linear and non-linear models for the relationship to maintenance OCS dose with adjustment for multiple covariates (study, region, weight, and past exacerbation history). Model predictions were corroborated using independent data from a third GSK-funded OCS-sparing Phase 3 study, SIRIUS. Results: In these studies, 28% of patients were taking maintenance OCS (median 10mg, IQR 7-20mg), there was a dose-ordered reduction of blood eosinophil count with increasing OCS dose group (prednisolone-equivalent dose >0-8, >8-16, >16-32 and >32mg): respective ratios 1.05, 0.68, 0.63 and 0.77 compared with no OCS. Analysis using a dose response (Emax) model suggested maximal inhibition of 44% and half maximal OCS dose of 12mg. An observed increase in blood eosinophil count of 33% for placebo patients in study SIRIUS corroborated model predictions. Conclusions: In patients with Severe Eosinophilic Asthma, OCS use on background maximum ICS therapy leads to additional dose-dependent reduction of blood eosinophils.

Background Findings from previous studies showed that mepolizumab significantly reduces the rate of exacerbations in patients with severe eosinophilic asthma. To assess the relationship between baseline blood eosinophil counts and efficacy of mepolizumab we did a secondary analysis of data from two studies, stratifying patients by different baseline blood eosinophil thresholds. Methods We did a post-hoc analysis of data, which was completed on Sept 25, 2015, from two randomised, double-blind, placebo-controlled studies of at least 32 weeks duration (NCT01000506 [DREAM] and NCT01691521 [MENSA]) done between 2009 and 2014. In these studies, mepolizumab (DREAM: 75 mg, 250 mg, or 750 mg intravenously; MENSA: 75 mg intravenously or 100 mg subcutaneously) versus placebo was given at 4-week intervals in addition to standard care (high-dose inhaled corticosteroids plus ≥1 additional controller with or without daily oral corticosteroids) to patients aged 12 years or older with a clinical diagnosis of asthma, a history of at least two exacerbations in the previous year that required systemic corticosteroid treatment, and evidence of eosinophilic airway inflammation. The primary endpoint in both studies was the annual rate of clinically significant exacerbations (defined as worsening of asthma that required the use of systemic corticosteroids, or admission to hospital, or an emergency-room visit, or a combination of these occurrences). In our analysis, the primary outcome was the annualised rate of exacerbations in patients stratified by baseline eosinophil counts (≥150 cells per μL, ≥300 cells per μL, ≥400 cells per μL, and ≥500 cells per μL) and baseline blood eosinophil ranges ( Findings Of 1192 patients, 846 received mepolizumab and 346 received placebo. The overall rate of mean exacerbations per person per year was reduced from 1·91 with placebo to 1·01 with mepolizumab (47% reduction; rate ratio [RR] 0·53, 95% CI 0·44–0·62; p Interpretation Our analysis has shown a close relationship between baseline blood eosinophil count and clinical efficacy of mepolizumab in patients with severe eosinophilic asthma and a history of exacerbations. We noted clinically relevant reductions in exacerbation frequency in patients with a count of 150 cells per μL or more at baseline. The use of this baseline biomarker will help to select patients who are likely to achieve important asthma outcomes with mepolizumab.

Background: Studies show that mepolizumab can reduce the frequency of clinically significant exacerbations in patients with severe eosinophilic asthma, compared with placebo. However, important events such as hospitalizations and emergency room visits are rare and difficult to characterize in single studies. Objective: We sought to compare hospitalization or hospitalization and/or emergency room visit rates in patients with severe eosinophilic asthma treated with mepolizumab or placebo in addition to standard of care for at least 24 weeks. Methods: This study was conducted and reported in accordance with the Preferred Reporting Items for Systematic reviews and Meta-Analyses statement. PubMed and the GSK Clinical Study Register were searched for suitable studies. The primary end points were the rate of exacerbations requiring hospitalization and the rate of exacerbations requiring hospitalization/emergency room visit. The proportion of patients with 1 or more event was also assessed. All mepolizumab doses were combined and individual patient-level data were analyzed. Results: Four studies (n = 1388) were eligible for inclusion. Mepolizumab significantly reduced the rate of exacerbations requiring hospitalization (relative rate, 0.49; 95% CI, 0.30-0.80; P = .004) and hospitalization/emergency room visit (relative rate, 0.49; 95% CI, 0.33-0.73; P < .001) versus placebo. Significant reductions of 45% and 38% were also observed for the proportion of patients experiencing 1 or more hospitalization and hospitalization and/or emergency room visit, respectively. Conclusions: Mepolizumab approximately halved exacerbations requiring hospitalization and/or emergency room visits compared with placebo in patients with severe eosinophilic asthma. This treatment addresses a key outcome in a patient population with a high unmet need (GSK Study 204664).

BACKGROUND: Some patients with severe asthma have recurrent asthma exacerbations associated with eosinophilic airway inflammation. Early studies suggest that inhibition of eosinophilic airway inflammation with mepolizumab-a monoclonal antibody against interleukin 5-is associated with a reduced risk of exacerbations. We aimed to establish efficacy, safety, and patient characteristics associated with the response to mepolizumab. METHODS: We undertook a multicentre, double-blind, placebo-controlled trial at 81 centres in 13 countries between Nov 9, 2009, and Dec 5, 2011. Eligible patients were aged 12-74 years, had a history of recurrent severe asthma exacerbations, and had signs of eosinophilic inflammation. They were randomly assigned (in a 1:1:1:1 ratio) to receive one of three doses of intravenous mepolizumab (75 mg, 250 mg, or 750 mg) or matched placebo (100 mL 0·9% NaCl) with a central telephone-based system and computer-generated randomly permuted block schedule stratified by whether treatment with oral corticosteroids was required. Patients received 13 infusions at 4-week intervals. The primary outcome was the rate of clinically significant asthma exacerbations, which were defined as validated episodes of acute asthma requiring treatment with oral corticosteroids, admission, or a visit to an emergency department. Patients, clinicians, and data analysts were masked to treatment assignment. Analyses were by intention to treat. This trial is registered with ClinicalTrials.gov, number NCT01000506. FINDINGS: 621 patients were randomised: 159 were assigned to placebo, 154 to 75 mg mepolizumab, 152 to 250 mg mepolizumab, and 156 to 750 mg mepolizumab. 776 exacerbations were deemed to be clinically significant. The rate of clinically significant exacerbations was 2·40 per patient per year in the placebo group, 1·24 in the 75 mg mepolizumab group (48% reduction, 95% CI 31-61%; p

Background: Limited data describe the association between the frequency of asthma exacerbations and the decline in lung function in severe asthma. Objective: To determine whether asthma exacerbations are associated with enhanced decline in lung function. Methods: Changes in lung function were analyzed retrospectively using data from the DREAM and MENSA studies of mepolizumab intervention in patients with severe asthma. Patients were either nonsmokers or former smokers. A linear regression model was used to analyze the relationship between the number of exacerbations and decline in FEV1 across treatment groups. Results: In a combined post hoc analysis, 57% (n = 572) of patients had no exacerbations and experienced an improvement in postbronchodilator FEV1 of 143 mL. In contrast, in patients who experienced 3 or more exacerbations, there was a decrease in postbronchodilator FEV1 of 77 mL in the combined analysis. The linear modeling analysis estimated that for each exacerbation seen during the observational period, there was a decrease of 50 mL in FEV1 (P < .001). Conclusions: A direct relationship between the number of exacerbations in patients with severe eosinophilic asthma and decline in lung function was observed. Repeated exacerbations may be associated with accelerated loss of lung function.

Intent-to-treat (ITT) analysis is viewed as the analysis of a clinical trial that provides the least bias, but difficult issues can arise. Common analysis methods such as mixed-effects and proportional hazards models are usually labeled as ITT analysis, but in practice they can often be inconsistent with a strict interpretation of the ITT principle. In trials where effective medications are available to patients withdrawing from treatment, ITT analysis can mask important therapeutic effects of the intervention studied in the trial. Analysis of on-treatment data may be subject to bias, but can address efficacy objectives when combined with careful review of the pattern of withdrawals across treatments particularly for those patients withdrawing due to lack of efficacy and adverse events.

The TORCH (Towards a Revolution in COPD Health) trial has highlighted some important issues in the design and analysis of long term trials in chronic obstructive pulmonary disease. These include collection of off-treatment exacerbation data, analysis of exacerbation rates and the effect of inclusion of patients receiving inhaled corticosteroids (ICS) prior to randomisation. When effective medications are available to patients who withdraw, inclusion of off-treatment data can mask important treatment effects on exacerbation rates. Analysis of on-treatment data avoids this bias but it needs to be combined with careful analysis of withdrawal patterns across treatments. The negative binomial model is currently the best approach to statistical analysis of exacerbation rates, while analysis of time to exacerbation can supplement this approach. In the TORCH trial, exacerbation rates were higher among patients with previous use of ICS compared to those with no prior use on all study treatments. Retrospective subgroup analysis suggests ICS reduced exacerbation rates compared with placebo, regardless of prior use of ICS before entry to the study. Factorial analysis provides an alternative analysis for trials with combinations of treatments, but assumes no interaction between treatments, an assumption which cannot be verified by a significance test. No definitive conclusions can yet be drawn on whether ICS treatment has an effect on mortality.

Various statistical methods have been used to measure the impact of treatment on chronic obstructive pulmonary disease (COPD) exacerbations. Poisson regression has recently been recommended as the appropriate method but the model does not satisfactorily account for variability between patients. In contrast, use of a negative binomial model, which corresponds to assuming a separate Poisson parameter for each patient, offers a more appealing approach. The present paper reviews analysis methods, with particular focus on the negative binomial model. To illustrate the differences that arise from using different analysis methods, we have reanalysed data from two large studies which, among other objectives, investigated the effectiveness of inhaled corticosteroids in reducing COPD exacerbation rates. Using the negative binomial model to reanalyse data from the TRISTAN and ISOLDE studies, the overall estimates of exacerbation rates on each treatment arm are higher and the confidence intervals for comparisons between treatments are wider, but the overall conclusions of TRISTAN and ISOLDE regarding reduction of exacerbations remain unchanged. The negative binomial approach appears to provide a better fit to the distribution of the data than earlier methods and is currently the method of choice. Research needs to continue on further methods to improve the analysis of exacerbation data.

For trials with repeated measurements of outcome, analyses often focus on univariate outcomes, such as analysis of summary measures or of the last on-treatment observation. Methods which model the whole data set provide a rich source of approaches to analysis. For continuous data, mixed-effect modelling is increasingly used. For binary and categorical data, models based on use of generalized estimating equations account for intra-subject correlation and allow exploration of the time course of response, as well as providing a useful way to account for missing data, when such data can be maintained as missing in the analysis. The utility of this approach is illustrated by an example from a trial in influenza. Copyright © 2004 John Wiley & Sons Ltd.

Nucleoside analogues against herpes simplex virus (HSV) have been shown to suppress shedding of HSV type 2 (HSV-2) on genital mucosal surfaces and may prevent sexual transmission of HSV.We followed 1484 immunocompetent, heterosexual, monogamous couples: one with clinically symptomatic genital HSV-2 and one susceptible to HSV-2. The partners with HSV-2 infection were randomly assigned to receive either 500 mg of valacyclovir once daily or placebo for eight months. The susceptible partner was evaluated monthly for clinical signs and symptoms of genital herpes. Source partners were followed for recurrences of genital herpes; 89 were enrolled in a substudy of HSV-2 mucosal shedding. Both partners were counseled on safer sex and were offered condoms at each visit. The predefined primary end point was the reduction in transmission of symptomatic genital herpes.Clinically symptomatic HSV-2 infection developed in 4 of 743 susceptible partners who were given valacyclovir, as compared with 16 of 741 who were given placebo (hazard ratio, 0.25; 95 percent confidence interval, 0.08 to 0.75; P=0.008). Overall, acquisition of HSV-2 was observed in 14 of the susceptible partners who received valacyclovir (1.9 percent), as compared with 27 (3.6 percent) who received placebo (hazard ratio, 0.52; 95 percent confidence interval, 0.27 to 0.99; P=0.04). HSV DNA was detected in samples of genital secretions on 2.9 percent of the days among the HSV-2-infected (source) partners who received valacyclovir, as compared with 10.8 percent of the days among those who received placebo (P0.001). The mean rates of recurrence were 0.11 per month and 0.40 per month, respectively (P0.001).Once-daily suppressive therapy with valacyclovir significantly reduces the risk of transmission of genital herpes among heterosexual, HSV-2-discordant couples.

Suissa [1] provides the event-based number needed to treat (NNT) for a reduction in exacerbations with fluticasone/salmeterol compared with placebo for the TORCH (Towards a Revolution in COPD Health) trial by period of follow-up (0–1 years after treatment, 1–2 years after treatment and 2–3 years after treatment) [2]. These NNT data are misleading as they fail to recognise that in chronic obstructive pulmonary disease (COPD) trials, patients who exacerbate are more likely to withdraw from the trial than patients with no exacerbations. Keene et al. [3] show that for the TRISTAN (TRial of Inhaled STeroids ANd long-acting β2 agonists) trial, the exacerbation rate is more than three per year among placebo patients withdrawing prior to 1 year compared with an exacerbation rate of one per year for patients completing a year of placebo treatment. In the TORCH trial, 25% of the placebo group withdrew compared with 15% of the fluticasone/salmeterol arm during the first year of follow-up; therefore, patients entering the second year of the trial on placebo and on fluticasone/salmeterol were no longer directly comparable. This is not accounted for in the calculation of exacerbation rates or NNT during years 2 and 3, as presented in table 2 of the study by Suissa [1]. NNT calculations during years 2 and 3 are misleading as they don–t account for differential withdrawal in year 1

Phase II trials have two aims: (1) to provide a proof of concept that the drug is able to affect the disease it is designed to affect without overt safety problems and (2) to define the dose(s) to be taken forward to phase III. Trials that specifically address these issues are often referred to as phase IIa and phase IIb trials, respectively. Therefore, phase II trials provide the initial test of efficacy and reassurance on safety in patients and, if successful, define the dose or doses to be carried forward to be studied in the wider populations and longer-term phase III trials. Keywords: Phase II trial; dose response; proof of concept

Background: In routine practice, influenza is diagnosed on clinical grounds. When influenza is known to be circulating locally, this is usually accurate, but awareness of the markers of influenza severity could be used by physicians to aid diagnosis and begin appropriate treatment early. This analysis compared baseline symptom profiles of subjects with severe and non-severe influenza. Methods: Symptoms of influenza-like illness (ILI) were recorded from 2235 subjects in six clinical trials of inhaled zanamivir 10 mg bd. ILI was defined as fever (37.8 °C) or feverishness plus at least two of the following: headache, myalgia, cough, sore throat, weakness, nasal congestion, loss of appetite. Results: Thirty percent (474/1572) of influenza-positive subjects were classified as having severe influenza and had a higher mean baseline temperature of 38.5 vs. 38.1 °C (P

Statistical analyses of recurrent event data have typically been based on the missing at random assumption. One implication of this is that, if data are collected only when patients are on their randomized treatment, the resulting de jure estimator of treatment effect corresponds to the situation in which the patients adhere to this regime throughout the study. For confirmatory analysis of clinical trials, sensitivity analyses are required to investigate alternative de facto estimands that depart from this assumption. Recent publications have described the use of multiple imputation methods based on pattern mixture models for continuous outcomes, where imputation for the missing data for one treatment arm (e.g. the active arm) is based on the statistical behaviour of outcomes in another arm (e.g. the placebo arm). This has been referred to as controlled imputation or reference-based imputation. In this paper, we use the negative multinomial distribution to apply this approach to analyses of recurrent events and other similar outcomes. The methods are illustrated by a trial in severe asthma where the primary endpoint was rate of exacerbations and the primary analysis was based on the negative binomial model.

Influenza infection rates are higher in children than in other age groups. This study evaluated the efficacy, safety and tolerability of a 5-day course of twice daily inhaled zanamivir, 10 mg, compared with placebo in the treatment of symptomatic influenza A and B viral infections among children 5 to 12 years of age.This double blind, randomized, placebo-controlled, parallel group, multicenter study conducted in the Northern Hemisphere during the 1998 and 1999 influenza season enrolled 471 patients with influenza-like symptoms foror = 36 h. Patients were randomly assigned to zanamivir (n = 224) or placebo (n = 247). Symptoms were recorded on diary cards twice daily during treatment, for 9 days after treatment and for 14 additional days (if still reporting moderate/severe cough and/or taking relief medication).A total of 346 (73%) patients were influenza-positive by culture, serology or polymerase chain reaction (65% influenza A, 35% influenza B). Zanamivir reduced the median time to symptom alleviation by 1.25 days compared with placebo among patients with confirmed influenza infection (P0.001). Zanamivir-treated patients returned to normal activities significantly faster and took significantly fewer relief medications than placebo-treated patients. Zanamivir was well-tolerated, demonstrating adverse event profiles similar to those of placebo and no clinically significant changes in laboratory findings. Viral susceptibility testing revealed no zanamivir-resistant strains of influenza A or B.Zanamivir was effective in shortening the duration and severity of influenza symptoms and was well-tolerated among children 5 to 12 years of age.

Zanamivir, a potent, highly selective inhibitor of influenza virus A and B neuraminidase, has been evaluated in seven, similarly designed, placebo-controlled studies of the treatment of influenza. Patients with typical influenza symptoms were recruited when influenza was known to be circulating in the community. Six of these studies included a zanamivir 10 mg inhaled bd (for 5 days) treatment arm, the dose regimen submitted to regulatory agencies. Pooled analyses were conducted to evaluate efficacy more precisely in terms of the alleviation of symptoms in population subgroups and for secondary endpoints. Median time to alleviation of symptoms, the primary endpoint, was reduced from 6.0 days in the placebo group (n = 1102) to 5.0 days in the zanamivir group (n = 1133), P < 0.001. In febrile, laboratory-confirmed, influenza-positive (IP) patients, time to alleviation was reduced from 6.5 days to 5.0 days, a treatment benefit of 1.5 days (P < 0.001). A larger treatment benefit (3 days) was seen in IP patients who had severe symptoms at entry (n = 474, P < 0.001), compared with 1 day in patients whose symptoms were not severe (n = 1098, P < 0.001). Similarly, a 3 day treatment benefit (P = 0.003) was observed in IP patients aged

The performance of different partial AUCs, including partial AUC from zero to t(max) of the reference formulation (AUC(r)) and partial AUC from zero to tmax of test or reference formulation, whichever occurs earliest (AUC(e), as indirect measures of rate of absorption have been evaluated using simulated experiments. The performance of these metrics relative to C(max), t(max) and C(max)/AUC(infinity) was further assessed using the results of actual studies involving a Glaxo drug. The normalised metrics AUC(r)/AUC(infinity) and AUC(e)/AUC(infinity) have also been evaluated. Our provisional conclusions were: (1) AUC(r)/AUC(infinity) and AUC(e)/AUC(infinity) had greater statistical power than C(max) and the non-normalised partial AUCs at detecting true differences in rate of absorption. Using real data, the performance of AUC(e)/AUC(infinity) was poor, however, the performance of AUC(r)/AUC(infinity) was good; (2) C(max)/AUC(infinity) was more precisely estimated than AUC(r)/AUC(infinity) or AUC(e)/AUC(infinity) and may be a superior metric for assessing absorption rates of highly variable drugs.

There are a number of different approaches to the assessment of dose proportionality, but most of these are directed toward hypothesis testing. The analysis of dose proportionality studies, however, requires estimation rather than significance testing in order that the pharmacokinetic and clinical significance of any nonproportionality can be assessed.The methods of statistical analysis of these studies have been reviewed. An empirical model relating the log of the pharmacokinetic parameter linearly to the log of the dose (the “power model”) provides a readily interpreted measure of the degree of nonproportionality. The potential utility of this model has been demonstrated using a number of observed dataseis from different drugs. If there is doubt over the goodness of fit of the power model then analysis of variance (after log transformation), together with pairwise comparisons between doses, provides a suitable alternative approach.

As indirect measures (metrics) of rate of drug absorption, Cmax is confounded by extent of drug absorption and tmax is a discrete variable, dependent on blood sampling frequency; therefore, there is a need for improved metrics of rate. Building upon the work of Endrenyi et al. (1) and Chen (2), different metrics have been compared using simulated single dose bioequivalence studies of immediate release (IR) dosage forms. Those investigated included: Cmax/AUC∞, partial AUC from zero to tmax of the reference formulation (AUCr), partial AUC from zero to tmax of the test or reference formulation, whichever occurs earliest (AUCe), and the normalized partial AUCs (AUCr/AUC∞, and AUCc/AUC∞, which like Cmax/AUC∞, are not confounded by extent of absorption) (3,4). Importantly, the performance of these metrics was further assessed using the results of several actual pharmacokinetic studies involving IR dosage forms of Glaxo drugs.For extended release (ER) dosage forms, bioequivalence assessments are carried out at s...

Bioequivalence assessment of extended release (ER) dosage forms is usually carried out at steady-state, using area under the curve (AUC) to evaluate extent of absorption and maximum concentration (Cmax) and % peak trough fluctuation ratio (%PTF) to evaluate rate of absorption. Other metrics such as Cmax/AUC and partial AUCs have recently been proposed as alternatives for assessing the absorption rate of drugs from immediate release (IR) dosage forms under single dose conditions. The performances of these metrics were assessed using the results of two sets of simulated experiments of ER dosage forms at steady-state and 2 actual pharmacokinetic studies involving ER dosage forms of a Glaxo drug. In the first set of simulations there was no difference in bioavailability between the two formulations; in the second set of simulations the test formulation had a 50% greater absorption rate-constant (ka) than the reference formulation. The following conclusions were reached: 1. For ER dosage forms at steady-state, all the metrics, with the exception of %PTF, resulted in much smaller increases than the underlying 50% increase in ka. Although, %PTF gave the largest effect it was also the most imprecisely estimated. 2. In our studies, none of the metrics tested provided reliable information about changes in the underlying rate of absorption from ER dosage forms under steady-state conditions. 3. The current practice of comparing rate of absorption from ER dosage forms using steady-state Cmax is inappropriate due to lack of sensitivity. The use of %PTF may require a widening in the currently accepted 80-125% permissible range set for Cmax and AUC.

The aims were to determine the effect of an oral inhibitor of the signaling mediator p38 mitogen-activated protein kinase (GW856553, losmapimod) on sputum neutrophils, pulmonary function, and blood biomarkers of inflammation in chronic obstructive pulmonary disease (COPD). Three hundred and two individuals with GOLD stage II COPD were randomized to oral losmapimod 7.5 mg twice daily, inhaled salmeterol/fluticasone propionate 50 µg/500 µg combination (SFC), or placebo in a 12-week, randomized, double-blind, double-dummy study (MKI102428/NCT00642148). Neither losmapimod nor SFC had an effect on the primary end point of sputum neutrophils. Losmapimod was well tolerated and reduced plasma fibrinogen by 11% (-0.4 g/L, ratio of effect of losmapimod/placebo 0.89; 95% confidence interval, 0.83-0.96; P = .002) with nonsignificant reductions in interleukin-6, interleukin-8, and C-reactive protein. There was evidence of improvement in hyperinflation with losmapimod compared with placebo (overall P = .02). Inhaled SFC significantly improved lung function and reduced serum CC-16 (ratio of effect of SFC/placebo 0.87; 95% confidence interval, 0.82-0.93; P < .001). It was concluded that oral losmapimod significantly reduced plasma fibrinogen in patients with COPD.

The following viewpoint from PSI and EFSPI regarding the current level of statistical resource in the European regulatory agencies was first presented as a position paper to a meeting of the EU Heads of Agencies in July 2009, and was endorsed by EFPIA.

To the Editors: In his recent editorial, Suissa 1 has again expressed criticism of the TORCH (Toward a Revolution in Chronic Obstructive Pulmonary Disease (COPD) Health) trial 2. The focus this time is the analysis of rate of decline in lung function 3, and in particular he questions whether the reduction in rate of decline in forced expiratory volume in one second (FEV1) observed with the active treatments (salmeterol, fluticasone or the combination of both treatments) compared with placebo is due to “regression to the mean” caused by missing data. Missing data for lung function and exacerbations are unfortunately inevitable in any long-term trial in COPD. Suissa 1 states that “… measurements were missing so that the pure intent-to-treat analysis was not possible”. Such a “pure intent-to-treat” analysis of lung function or exacerbation data is not possible for any long-term COPD trial, as this would require complete data. In fact, the quantity of missing data for lung function in TORCH is similar to the UPLIFT (Understanding Potential Long-term Impacts on Function with Tiotropium) 4 and OPTIMAL 5 trials quoted in the article by Suissa 1. In TORCH, 4,857 (79%) patients provided lung function data at 48 weeks, OPTIMAL had complete 1-yr lung function data for 322 (72%) patients and, …

Three oral doses of sumatriptan, 100, 200 and 300 mg, given as dispersible tablets, were compared in the acute treatment of migraine in a double-blind, placebo-controlled, parallel-group study of 1,13

A double-blind, randomized, multicentre, parallel-group study was carried out to compare intranasal sumatriptan with placebo in the treatment of migraine. Seventy-four patients (37 in each treatment group) were recruited into the study. Patients received two insufflations of the same treatment (sumatriptan or placebo) 15 min apart. Sumatriptan (20 mg plus 20 mg) was more effective than placebo at relieving headache, defined as a reduction in severity from moderate (grade 2) or severe (grade 3) to mild (grade 1) or none (grade 0), at 60 and 120 min. At 120 min, 75% of patients in the sumatriptan group reported headache relief, compared with 32% of patients in the placebo group (p