FDA Panel Will Weigh Mixed Efficacy Data For GSK’s Breo Ellipta In COPD
This article was originally published in The Pink Sheet Daily
Executive Summary
Agency says data supporting the relative contribution of the combination drug’s corticosteroid component, fluticasone, are “less robust” than data supporting the bronchodilatory effects of the long-acting beta-agonist component, vilanterol.
FDA’s Pulmonary-Allergy Drugs Advisory Committee will have to sort through inconsistent efficacy data during its April 17 review of GlaxoSmithKline PLC’s Breo Ellipta (fluticasone furoate/vilanterol inhalation powder) for both bronchodilation and reduction of exacerbations in patients with chronic obstructive pulmonary disease.
The committee will be asked to consider the efficacy data on each of Breo Ellipta’s individual components as well as the combination. Although data from the pivotal studies support the bronchodilatory effect of the long-acting beta-agonist component, vilanterol, the data to support the relative contribution of fluticasone, the inhaled corticosteroid component, “are less robust,” FDA said in an agency briefing document released on April 15.
Furthermore, only one of two exacerbation trials met the level of statistical significance for efficacy, although both showed a similar reduction in number of exacerbation events per year with the combination compared to vilanterol alone, thereby suggesting a benefit with the fluticasone component.
Bad weather forced an almost six-week postponement of the advisory committee review, leaving GSK extra time to prepare but FDA little time to act.
FDA is asking the committee to vote on the sufficiency of the efficacy evidence and the risk/benefit profile for each of GSK’s two requested claims: long-term maintenance treatment of airflow obstruction and reduction of exacerbations.
The agency also presents a voting question on the sufficiency of the safety evidence to support once-daily use for the proposed indications. Although there had been concerns about cases of pneumonia in the clinical development program, FDA’s review suggests the adverse events are in line with those seen in other long-acting beta-agonist combination product studies.
The advisory committee was slated to review the Breo Ellipta NDA on March 7. However, that meeting was postponed in the face of an approaching winter storm that shut down federal government offices in the Washington, D.C. area (Also see "GSK’s Breo Ellipta Gets “Snowquestered” As FDA Delays Advisory Cmte" - Pink Sheet, 5 Mar, 2013.). The delay gives FDA just a little over three weeks between the advisory committee review and the product’s May 12 user fee date.
A Departure From Precedent
GSK, which is developing Breo Ellipta (formerly known as Relovair) in partnership with Innoviva Inc., is hoping to make the novel combination product a successor to its top-selling asthma medicine Advair (fluticasone/salmeterol inhalation powder) before the latter faces generic competition. However, the clinical data that have been publicly disclosed have suggested a mixed bag in terms of efficacy, including a failure to demonstrate statistically significant differences in some trials at certain doses and failure to demonstrate superiority to Advair in non-pivotal studies (Also see "Relovair Data Raises Respiratory Stakes For GSK And Theravance" - Pink Sheet, 16 Jan, 2012.).
Two ICS/LABA products are currently approved for relief of airflow obstruction in COPD patients, Advair Diskus and AstraZeneca PLC’s Symbicort (budesonide/formoterol), both of which are dosed twice daily. Advair Diskus also carries a claim for reduction of COPD exacerbations.
GSK is seeking approval for a once-a-day formulation that combines fluticasone furoate (FF) 100 mcg and vilanterol (VI) 25 mcg for COPD. Neither component is currently marketed as single-ingredient inhalation product. Vilanterol is a new molecular entity. Fluticasone is available as a nasal spray suspension approved for the treatment of allergic rhinitis and is marketed by GSK under the brand name Veramyst.
In a Feb. 7 memorandum, FDA Division of Pulmonary, Allergy and Rheumatology Products Medical Team Leader Susan Limb highlighted the novelty of the NDA filing and the combination drug’s development program.
“The sequence and scale of the FF/VI development program differ from prior precedent,” Limb said. “Previous ICS/LABA development programs were based on the initial development of the individual ICS and LABA monotherapies followed by the combination product in asthmatics, a patient population that is presumably more sensitive to both bronchodilators and corticosteroids.” Furthermore, the LABAs salmeterol and formoterol were developed and marketed for monotherapy use in COPD before development of combination products with inhaled corticosteroids.
In contrast, the Breo Ellipta program has been conducted concurrently with development of the individual components in both COPD and asthma, Limb said, adding that GSK has told the agency it has no plans to market vilanterol monotherapy.
“In many respects, the development program for FF/VI is an umbrella program that encompasses the individual development programs for FF and VI and spans two disease indications,” Limb said. “GSK was asked to provide data to support the following: 1) the nominal dose and dosing frequency for each of the components, including evidence of efficacy and safety for FF alone in asthma and VI alone in COPD; 2) data demonstrating the efficacy contribution of VI to the FF/VI combination; and 3) data demonstrating that FF/VI confers a treatment benefit over VI alone in COPD (the contribution of FF).”
Four Pivotal Trials
The NDA, which seeks approval only for COPD, is supported by four pivotal studies: two 24-week confirmatory lung function trials (2206 and 2007) and two 52-week exacerbation trials (2871 and 2970).
The confirmatory lung function studies were placebo-controlled and similar in design. Primary efficacy endpoints were the weighted mean FEV1 (a measure of lung function) 0-4 hours post dose on treatment day 168, which was intended to assess the effect of vilanterol, and the change from baseline in trough FEV1 on treatment day 169, which was intended mainly to assess fluticasone’s effect.
Study 2206 assessed two doses of the fluticasone/vilanterol combination (50/25 mcg and 100/25 mcg), fluticasone 100 mcg, vilanterol 25 mcg and placebo. On the endpoint of weighted mean FEV1, vilanterol monotherapy was statistically significantly better than placebo, and the 100/25 mcg combination was significantly better than fluticasone monotherapy, reflecting the relative contribution of vilanterol to the combination product.
Trial 2207 assessed fluticasone/vilanterol in two strengths (100/25 mcg and 200/25 mcg), two doses of fluticasone (100 and 200 mcg), vilanterol 25 mcg and placebo. Statistically significant results were observed for comparisons between vilanterol 25 mcg and placebo, as well as for the 100/25 mcg combination versus fluticasone 100 mcg and the 200/25 mcg combination versus fluticasone 200 mcg.
On the second primary endpoint of trough FEV1, although all combination treatment arms in the two studies showed a numerical benefit over vilanterol alone, none reached the level of statistical significance and no apparent dose response was observed.
The primary endpoint in the two exacerbations trials was the annual rate of moderate and severe COPD exacerbations, and this endpoint is presented as an alternative demonstration of fluticasone’s contribution to the combination product.
“In both trials, the prespecified statistical analysis plan required statistical significance at the 0.05 level for the comparison of FF/VI 200/25 to VI 25 prior to the testing of lower doses. As a result, a statistically significant result for FF/VI 100/25 is observed in Trial 2970, while the p-value reported for the same comparison in Trial 2871 … is a nominal p-value,” Limb said.
In both exacerbation trials, “the mean rate of moderate to severe exacerbation in the VI 25 arm was approximately 1 exacerbation per year; the mean reduction observed with FF/VI was about a quarter to a third of an event in one year with FF/VI 100/25. Analyses of the time to exacerbation and exacerbations requiring systemic corticosteroids were also supportive.”
Safety In Line With Other Combos
In the COPD program, there were eight deaths during the treatment period and three in the post-treatment follow-up periods; they were generally evenly distributed across active treatment arms and placebo. “There was no apparent dose effect in terms of the total number of fatal cases or specific causes cited, with the exception of pneumonia, which appeared to occur most frequently in the FF/VI 200/25 arm,” Limb said.
As in other COPD programs involving high doses of inhaled corticosteroids, an increased risk of pneumonia was seen with Breo Ellipta, most prominently in the exacerbation trials, which were longer in duration and enrolled a sicker population at baseline. The rates of pneumonia ranged from 3%-8% across treatment arms of the exacerbation studies. However, FDA suggests these rates are comparable to those seen in studies of other ICS/LABA products and that rates of bone fracture in the Breo Ellipta program were lower.
Summary safety data from the asthma program did not show an increased risk of severe asthma-related adverse events.
The NDA also included a summary of safety data from the Breo Ellipta program in asthma. Labeling for currently approved LABAs describes an increased risk of severe asthma-related adverse events, although a similar safety concern has not been raised specifically for the COPD population.
“The summary included an adjudicated assessment by an independent, blinded committee of a composite safety endpoint for asthma-related hospitalizations, intubations, and deaths, which did not suggest an increased risk of a severe asthma-related AE associated with VI alone or in combination with FF,” Limb said. “A total of three deaths were reported in the program (1 in FF/VI 100/25, FF 100, and placebo arms each), but none were adjudicated as asthma-related.”
“The nature of the adverse events identified for FF/VI appears generally consistent with the general safety profile of similar combination products,” Limb said, summarizing the safety findings. “In particular, a dose-related risk of respiratory infections and a lesser risk of fractures were identified. While a direct, head-to-head comparison of long-term safety with other approved ICS/LABA combination products is not available, safety data for other ICS/LABA products relative to the corresponding LABA monotherapies is available. This information provides some context for the relative safety of FF/VI 100/25 compared to VI 25 alone. Safety information from the parallel asthma development program provides secondary support, including support for the selection of an appropriate VI dose.”