The US Food and Drug Administration’s (FDA’s) Antimicrobial Drugs Advisory Committee voted on January 11 by a large margin (12 no, 3 yes, 1 abstain) not to recommend ciprofloxacin dispersion for inhalation (Cipro DI; Linhaliq, Aradigm) for the treatment of non-cystic fibrosis bronchiectasis in patients with chronic lung infections with Pseudomonas aeruginosa.
“The inconsistency between the two studies, the nature of the endpoint being imprecise and needing a better endpoint…. I remain concerned about the resistance beyond I, in the [gastrointestinal] tract, with other organisms in the respiratory tract, with [nontuberculous mycobacteria], because one can rob Peter to pay Paul, and the burden is to make sure we minimize that concern for those patients,” said committee chair Lindsey R. Baden, MD, director of clinical research, Division of Infectious Diseases, Brigham and Women’s Hospital; director, Infectious Disease Service, Dana-Farber Cancer Institute; and associate professor, Harvard Medical School; all in Boston, Massachusetts.
During the open public hearing portion of the meeting, 11 individuals spoke to ask the committee to recommend Cipro DI. Some were patients and some spoke on behalf of patients who could not be there. A number of patients had received the drug during clinical trials and experienced life-altering improvement in their symptoms.
There are currently no approved therapies for non-cystic fibrosis bronchiectasis, an orphan disease.
Several panel members said they wished they could have voted “yes.”
The panel’s vote follows discussion of data from two phase 3 trials, ORBIT-3 and ORBIT-4. In both studies, researchers randomly assigned patients in a 2:1 ratio to receive Cipro DI or placebo once daily in six cycles of 56 days (28 days on/28 days off treatment) each, during a period of 48 weeks. ORBIT-3 added a pharmacokinetics and safety substudy in the open-label phase.
In ORBIT-3, the full analysis population included 278 randomly assigned patients who received one or more doses of the study drug (183 in the Cipro group and 95 in the placebo group). In ORBIT-4, the full analysis population included 304 patients (206 in the Cipro group and 98 in the placebo group).
Efficacy
The discordance in efficacy between ORBIT-3 and ORBIT-4 was a sticking point for many of the panel members.
The primary endpoint for both studies was time to pulmonary exacerbation (PE). In ORBIT-3, 59% of patients in the Cipro group and 57% of patients in the placebo group experienced one or more PEs by week 48. The difference between the groups in median times was 78 days. In the unweighted, stratified model that included prior PEs and sex, the overall hazard ratio (Cipro/placebo) was 0.99 (95% confidence interval [CI], 0.71 – 1.38; P = .974), a difference that was not statistically significant.
In ORBIT-4, 55% of patients in the Cipro group and 65% of patients in the placebo group experienced one or more PEs by week 48. The difference between the groups in median times was 72 days (hazard ratio, 0.71; 95% CI, 0.52 – 0.97; P = .032), a difference that was statistically significant.
The difference in the proportion of patients who had one or more PEs (Cipro-placebo) was 2.17% in ORBIT-3 and −9.97% in ORBIT-4.
“I have real concerns with the primary endpoint choice for this study and recognize the impact that guidance from the FDA had on these decisions, but I also recognize that this is not the agency’s fault, and reflects the thinking at the time of the design of the study,” voting committee member Michael Green, MD, MPH, professor of pediatrics, surgery and clinical & translational science, University of Pittsburgh School of Medicine, Division of Infectious Diseases, and director, Antimicrobial Stewardship & Infection Prevention, and codirector, Transplant Infectious Diseases, Children’s Hospital of Pittsburgh, Pennsylvania.
“I wanted to vote in support of approval, but I recognize that this vote would not strictly be in response to the FDA’s specific question on the evidence of impacting time to event,” Dr Green added.
Secondary efficacy endpoints included frequency of pulmonary exacerbations, number of severe pulmonary exacerbations, and change in Respiratory Symptoms Domain Score of Quality of Life-B from baseline to week 48.
In ORBIT-3, the mean number of PEs by week 48 was 1.09 in the Cipro group and 1.30 in the placebo group, with an estimated incidence rate ratio of 0.852. The reduction in frequency of PEs in the Cipro group compared with placebo was 14.8% (95% CI, −12.3% to 35.3%) when stratifying on prior PEs and sex, which was not statistically significant.
In ORBIT-4, the mean number of PEs at week 48 was 0.98 in the Cipro group and 1.47 in the placebo group, with an estimated incidence rate ratio of 0.631. The reduction in frequency of PEs in the Cipro group compared with placebo was 36.9% (95% CI, 17.9% – 51.5%; P = .0006) when stratifying on prior PEs and sex, a difference that was statistically significant. “The 48-week difference of 0.5 PEs ([95% CI,] 1.47 – 0.98) can be interpreted as a reduction in one PE every two years, when assuming a constant risk over time,” according to the FDA briefing document.
Safety
The rates of treatment-emergent adverse events were balanced across the Cipro and placebo groups overall. Treatment-emergent adverse events included cough, dyspnea, wheezing, fatigue, and decreased forced expiratory volume.
“Serious adverse events…were mainly manifestations of the underlying disease and, in general, reflected hospitalization events and serious events mostly associated with pulmonary exacerbations that required intravenous antimicrobial treatment,” according to the FDA’s briefing document.
The most common serious adverse events were hemoptysis (Cipro, 3.8%; placebo, 1.1%) and pneumonia (Cipro, 3.8%; placebo, 7.3%) in ORBIT-3 and pneumonia (Cipro, 3.9%; placebo, 0%) in ORBIT-4.
More deaths occurred in ORBIT-3 (eight [3.2%]; Cipro, five deaths; placebo, three deaths) than in ORBIT-4 (six [2%]; Cipro, two deaths; placebo, four deaths). Most (64%) of the adverse events that led to death were in the Respiratory, Thoracic and Mediastinal Disorders System Order Class.
By the end of the sixth on-cycle, the total number of ciprofloxacin-resistant P aeruginosa isolates in sputum increased in the Cipro groups from 21% to 43% in ORBIT-3 and from 15% to 46% in ORBIT-4. The percentages of resistant isolates remained more stable in the placebo groups.
“I voted ‘yes’ in spite of the fact that ORBIT-3 did not meet the primary endpoint. The reasons I voted ‘yes’ were there was a positive trend in the patients who had the most severe disease and there was a positive trend in all the more clinically relevant secondary endpoints, and ORBIT-4 met the primary and secondary endpoints. It does appear to be safe overall, in a very sick population,” said voting committee member Jonathan Honegger, MD, assistant professor of pediatrics, The Ohio State University College of Medicine, Division of Infectious Diseases, and Center for Vaccines and Immunity, Nationwide Children’s Hospital, both in Columbus, Ohio.
Dr Honegger said he was influenced by the compelling unmet need, and he feels that rather than voting “no,” which would require additional studies before approval, he would prefer to see product labeling state that the drug is not indicated to prevent first infection and the secondary endpoint of frequency of exacerbation. He added that he would like to see postapproval studies in children, including pharmacokinetic studies, and safety trials, particularly in children.
For more news, join us on Facebook and Twitter
Tidak ada komentar:
Posting Komentar