FDA Advisors Question Effectiveness of CF Drug, Bronchitol (Mannitol Powder)
High drop-out rates cited (Jan. 29)
The FDA’s Pulmonary–Allergy Drugs Advisory Committee (PADAC) plans to meet on Jan. 30 to discuss the new drug application (NDA) for mannitol inhalation powder (proposed trade name: Bronchitol; Pharmaxis Pharmaceuticals) 400 mg twice daily for the management of cystic fibrosis (CF) in patients aged 6 years and older to improve pulmonary function.
In an online briefing document, however, the committee has questioned the drug’s clinical efficacy in view of the high number of drop-outs in two 26-week trials.
Both studies (301 and 302) were randomized, double blind, parallel-group trials designed to assess the efficacy and safety of 26 weeks of treatment with inhaled dry powder mannitol (DPM) 400 mg twice daily versus control (50 mg inhaled mannitol, which was felt to be an ineffective dose) in a total of 300 patients aged 6 years and older. The double-blind phase was followed by an open-label phase of up to 52 weeks and 26 weeks for trials 301 and 302, respectively. In both studies, the primary efficacy endpoint was the absolute change from baseline in the forced expiratory volume in 1 second (FEV1) at week 26.
Approximately 66% of enrolled patients completed the 26-week double-blind portion of study 301 and 85% in study 302. Early discontinuation occurred more often in the DPM-treated group (37% in study 301 and 17% in study 302) than in the control group (28% in study 301 and 12% in study 302) in each study. The primary reasons for premature discontinuation were adverse events (including CF exacerbations) and patient withdrawal.
According to the briefing document, the drug’s developer responded to the disproportionate numbers of treatment-related early discontinuations in the DPM groups by creating a “modified” intent-to-treat (MITT) population, which included only ITT patients who attended the week-6 study visit. As a result, patients who dropped out before week 6 in either study were excluded from efficacy analyses. In study 301, only 88% (156/177) of DPM-treated patients were included in the MITT analysis, compared with 95% (112/118) of control patients. In study 302, 96% (174/184) of DPM-patients and 99% (120/121) of control patients were included in the MITT population.
The briefing document also identifies additional missing data as a result of differential drop-outs at weeks 14 and 26, when efficacy assessments (FEV1 determinations) were made. For example, in study 301, at week 26, only 66% (116/177) of DPM-treated patients compared with 77% (89/116) of control patients had observed data, whereas in study 302, 85% (157/184) of DPM-patients and 92% (111/121) of control patients had observed data.
“It is notable that there is inconsistency with regard to the efficacy results when analyses are conducted with and without inclusion of missing data as a result of differential patient drop-out,” the document notes. For example, the results for study 301, which had the greatest differential drop-out, went from demonstrating a statistically significant increase in FEV1 for the MITT population to “not significant” when missing data were accounted for.
The briefing document concludes that “given the difference in results when data for missing patients are included in the analyses along with the patients with observed data, from a statistical perspective, a replicated statistically significant effect of DPM on the primary efficacy endpoint has not been demonstrated and, as such, the overall effect of DPM in CF patients in terms of the change from baseline in FEV1 in the ITT population cannot be confirmed.”
The PADAC says that the appropriateness of using modified study populations and the resulting differences in study results will be “a significant topic of discussion” in the upcoming review meeting.
Source: FDA; January 29, 2013.