Bronchitol AdComm Shows Difficulty Of Following In Kalydeco’s Shadow

Comparisons between the robustness of the efficacy data for dry powder mannitol and Vertex’s ivacaftor were difficult for FDA to avoid during the Pulmonary-Allergy Drugs Advisory Committee’s Jan. 30 review of Pharmaxis’ cystic fibrosis drug.

The FDA advisory committee review of Pharmaxis Ltd.’s cystic fibrosis drug Bronchitol (mannitol inhalation powder) shows the difficulty faced by a new drug moving through the regulatory process shortly after a breakthrough treatment’s approval in the same disease.

The Pulmonary-Allergy Drugs Advisory Committee’s Jan. 30 review of dry powder mannitol (DPM) came almost one year to the day FDA approved Vertex Pharmaceuticals Inc.’s Kalydeco (ivacaftor), a personalized medicine approved in less than four months that has been hailed as a game-changing treatment for some CF patients.

The memory of Kalydeco’s robust efficacy only served to highlight the weaknesses in the Bronchitol clinical data.

Comparing Bronchitol to Kalydeco is like comparing apples to oranges. As a treatment targeting the symptoms of CF, mannitol is intended to improve pulmonary function by increasing mucociliary clearance. In contrast, ivacaftor targets the underlying genetic cause of the disease in approximately 4% of CF patients.

Despite their very different mechanisms of action and target populations, it was perhaps inevitable that comparisons between the robustness of the efficacy data supporting the drugs would be drawn, which only served to highlight the weaknesses in Pharmaxis’ clinical program.

The advisory committee recommended against approval of Bronchitol for both efficacy and safety reasons, although several panelists suggested the drug’s risk/benefit profile would have been more favorable had the proposed indication been limited to adults rather than extending to patients as young as six years old (Also see "Pharmaxis’ CF Drug Bronchitol Gets Thumbs Down From FDA Panel" - Pink Sheet, 30 Jan, 2013.).

Efficacy Questioned

Submitted in May 2012, Pharmaxis’ NDA is based on data from two similarly designed studies, 301 and 302. Both were randomized, double-blind, controlled trials designed to assess the efficacy and safety of 26 weeks of treatment with DPM 400 mg twice daily. Patients who successfully completed a mannitol tolerance test were randomized to DPM 400 mg or control (50 mg inhaled mannitol) twice daily using a breath-actuated handheld dry powder inhaler.

Advisory Committee Votes

• 11-3 that there is not substantial evidence of efficacy for DPM at a dose of 400 mg twice daily for improvement of pulmonary function in CF patients ages six years and older
• 11-3 that DPM’s safety profile for maintenance treatment is not sufficient to support approval
• 14-0 that efficacy and safety data do not provide substantial evidence to support approval in patients six years and older

Evaluations were made at screening to assess for eligibility, and following randomization at baseline, week 6, week 14 and week 26 during the double-blind phase. The primary efficacy endpoint was absolute change from baseline in forced expiratory volume in one second (FEV₁), a measure of lung function, at week 26. Secondary endpoints included pulmonary exacerbations, quality of life, rescue antibiotic use and hospitalization days due to exacerbations.

Under Pharmaxis’ analysis, DPM demonstrated a statistically significant improvement in Study 301 and a favorable effect in 302 that fell just short of statistical significance.

However, FDA’s clinical and statistical reviewers raised a host of efficacy questions about the data. Among these, Study 301 was marked by differential drop-out rates in the active and control arms, leading the agency to assert that the sponsor’s pre-specified statistical analysis was not appropriate (Also see "Pharmaxis’ Bronchitol Will Face Efficacy Questions At FDA Panel Review" - Pink Sheet, 28 Jan, 2013.).

FDA presented a continuous responder analysis, saying this represented a true intent-to-treat population and accounted for missing data from both treatment and control groups. Under this analysis, the DPM group had a numerically higher proportion of subjects who achieved an increasing change from baseline in FEV₁ thresholds. However, the differences between treatment groups across a range of thresholds were not statistically significant for either study, FDA's briefing document said.

Secondary Endpoints Show No Benefit

At the advisory committee meeting, agency reviewers said various sensitivity analyses suggested a range of effect on FEV1. They also questioned the clinical relevance of the treatment effect, particularly given the lack of a statistically significant benefit on various secondary endpoints.

Mannitol is believed to increase hydration of mucus and facilitate clearance of secretions that otherwise are difficult for CF patients to expel and can lead to recurrent pneumonia and lung damage. Consequently, the drug would be expected to improve pulmonary outcomes, clinical reviewer Kimberly Witzmann said.

“In this case, FEV₁ is being used as a measure for overall improvement in pulmonary function,” she said. “In the context of cystic fibrosis, we would expect that meaningful improvement should carry over to other non-spirometric endpoints that better reflect overall pulmonary function, such as fewer infections, hospitalizations and exacerbations, and a better quality of life.”

Studies 301 and 302 showed numerically positive trends, but not statistically significant changes, in any of the chosen secondary endpoints, including incidence of or time to first pulmonary exacerbation, rescue antibiotic use, days in the hospital due to exacerbation or quality-of-life scores, she said.

“It’s also possible that the small change in absolute FEV₁ seen is too small to impact more clinically meaningful outcomes in the 26-week study period,” Witzmann said, suggesting a longer double-blind study period may have been needed.

Meeting The “Substantial Evidence” Standard

In discussing the regulatory framework for “substantial evidence” of efficacy to support approval, Division of Pulmonary, Allergy & Rheumatology Products Clinical Team Leader Anthony Durmowicz drew a contrast between the strength of the efficacy evidence supporting the mannitol NDA and ivacaftor’s approval. Durmowicz was the cross-discipline team leader on the ivacaftor NDA, and Witzmann was the medical reviewer.

Ivacaftor has become a shining example of good drug development in the eyes of regulators. In review documents and speeches, FDA officials have been enthusiastic about the drug’s efficacy and the quality of Vertex’s clinical program (Also see "FDA Liked What It Saw In Kalydeco’s Development Program" - Pink Sheet, 19 Jun, 2012.). The drug also was the recipient of the first two “breakthrough therapy” designations granted under the FDA Safety and Innovation Act (Also see "FDA’s First Breakthrough Designations Continue Vertex Kalydeco’s Regulatory Successes" - Pink Sheet, 7 Jan, 2013.).

Durmowicz said that substantial evidence typically has meant “replicate, well-designed, well-controlled studies demonstrating an efficacy finding. This means two studies, studying an appropriate endpoint, both winning statistically and clinically. Endpoint FEV₁ as a surrogate endpoint for improved lung function fits into this category. One positive study does not meet that bar.”

However, there are times when a single positive study may be enough, he said. “An excellent design study showing highly reliable, statistically strong evidence for an important clinical benefit, such as survival, may suffice. Also, a single study itself that demonstrates statistically and clinically meaningful benefit in multiple unrelated endpoints can also suffice.”

Ivacaftor fits into this latter category, Durmowicz said. “The approval of that drug was based primarily on one adequate, well-controlled trial in adults and adolescents, and that trial showed statistically strong and clinically meaningful benefit on multiple endpoints,” including FEV₁, pulmonary exacerbations and weight gain. “So, that could be an example of the type of study you would need to fit into that category.”

Ivacaftor was approved on the basis of a 48-week, randomized, double-blind, placebo-controlled study in 161 adults and adolescents 12 years and older, as well as a second similarly designed study in 52 patients ages 6-11 years old. In both trials, ivacaftor was associated with a statistically significant improvement in FEV₁. In contrast to the Pharmaxis trials, the dropout rate was low and almost all key secondary endpoints statistically significantly favored ivacaftor in both studies (Also see "Secondary Endpoints Backed Kalydeco’s Efficacy, Despite Lack Of Correlation With Pulmonary Function" - Pink Sheet, 19 Jun, 2012.).

Suggestions Of Efficacy In Adults

In their discussion and votes on the questions, several committee members said they found the efficacy data generally underwhelming and the effect size not particularly large.

Temporary voting member John Connett, University of Minnesota, said it was interesting that in Study 301 there was good efficacy but problems with missing data, whereas in Study 302 there was less of a problem with missing data but “the efficacy went away. I think there might be a message in that.”

“There’s a borderline effect on FEV₁ that’s in the 2%-4% range,” temporary voting member Robert Castile, Ohio State University, said. “Beyond that, there’s no evidence at all of any clinical effect, and so that’s my summary of the data. I think though in the data there’s a suggestion … that there’s a significant subset that may benefit from the drug, and … this subset probably has an FEV₁ between 40 and 70 and they’re probably adults.”

Several panelists suggested they would have viewed the risk/benefit more favorably had the indication been limited to adults. They did not believe efficacy had been demonstrated in the pediatric group, and they were concerned about the higher rates of hemoptysis (bloody sputum) among these younger patients.

Temporary voting member Jeffrey Wagener, University of Colorado, said that although the statistical evidence for mannitol’s efficacy does not meet the usual standard for approval, an exception may be warranted.

Mannitol would be the first FDA-approved drug in CF that improves mucociliary clearance, he said, also questioning whether FEV₁ was the most appropriate endpoint to measure. “I think there is evidence of efficacy. It may not be based on some of the statistics that we’ve used historically, but in a situation where you have a life-threatening disease with no other drug in class, I would be willing … to stretch the definition of efficacy beyond the straightforward, simple purist statistics.”