Gene therapy to treat cystic fibrosis patients was associated with stabilization — but not improvement — in lung function, according to the results of a randomized, double-blinded phase IIb trial of the therapy in the U.K.

In a per protocol analysis, Professor Eric W.F.W. Alton, of Imperial College in London, and colleagues found a significant, but modest, treatment effect in cystic fibrosis patients treated with the nonviral, chemically designed gene-liposome pGM169/GL67A compared with those treated with a placebo of 0.9% saline (3.7%, 95% CI 0.1-7.3, P=0.046).

The primary endpoint of the study was defined as relative percent change in forced expiratory volume (FEV1) after 12 months. The authors achieved that because while treatment with pGM169/GL67A did not improve lung function, it did not worsen it either.

Modest Benefit

“This study proves for the first time that copies of the normal CF gene delivered by aerosol inhalation can have a measurable beneficial effect on lung function, compared with placebo, in patients with cystic fibrosis,” senior co-author Dr. Alastair Innes, of Western General Hospital in Edinburgh, Scotland, told MedPage Today. He added that while the effect was statistically significant, he also described it as “modest.”

A small number of patients in both groups did experience improvements in lung function. The authors note that a post-hoc analysis showed 18% of patients (15 in the treatment group and six in the control group) showed an improvement in percent predicted FEV1 of 5% or more of their initial baseline values. By contrast, overall treatment effect in the 65 patients in the treatment group and 56 in the control group was 3.6% (95% CI 0.2-7.0,P=0.039). Of the 20 patients who did not complete the full treatment of one dose per 28 days for 12 months, they received a mean 3.7 doses (SD 1-9).

Patients were randomized into a number of stratified subgroups, but the authors attributed any treatment effect to a greater decline in FEV1 from the placebo group as opposed to greater improvement from pGM169/GL67A. Stratifying by baseline predicted FEV1 (<70% versus ≥70%) found that patients with a more severe disease (FEV1 49.6%-69.2% predicted) had a treatment effect of 6.4% (95% CI 0.8-12.1). By contrast, those with less severe disease (FEV1 69.6%-89.9% predicted) had a 0.2% treatment effect (CI -4.6 to 4.9, P interaction=0.065).

The authors also cited the post-trial and pre-trial changes in the placebo group (-4.9%) compared with the treatment group (1.5%) as contributing to the treatment effect. There were no differences observed by age, sex, or CTFR mutation.

The study also had a number of secondary outcomes, which achieved mixed results. Patients in the treatment group experienced greater improvements in forced vital capacity (FVC) and CT gas trapping, or the inability to exhale completely (P=0.031 andP=0.048, respectively) than the control group. But authors observed no treatment effect for other measures of lung function, imaging, and quality of life. Similar to the primary analysis, they did note that secondary outcomes tended to be more favorable for those with more severe disease.

The authors commented that patients with more severe disease seemed to experience an enhanced treatment effect, and saw this as an opportunity for further research.

“A larger trial with a stratified trial entry design, powered to assess subgroups, and that addresses the mechanisms of response heterogeneity will be important to verify or refute these data,” they wrote.

A total of six serious adverse events were recorded from the pGM169/GL67A group. The committee judged that they were unrelated to the treatment, though one may have been related to a trial procedure (bronchoscopy), the authors said. Two patients total discontinued treatment; one in the placebo group due to fatigue and one in the treatment group due to flu-like symptoms. There were no deaths during the study, and the authors saw no clinically relevant changes in patients throughout the study.

This randomized, double-blinded, placebo-controlled trial consisted of two cystic fibrosis centers in London and Edinburgh at 18 sites in the U.K. from June 12, 2012, to June 24, 2013. Participants were eligible if they were ages ≥12 years, had a FEV1 of 50%-90% predicted and had any combination of CFTR gene mutations. Of the 140 patients, 78 received pGM169/GL67A and 62 received a placebo. There were 116 patients (83%) completing the treatment and included in the per protocol analysis.

Limitations

The most important limitation the authors cite is that the mean difference is at the lower end of clinical trials for gene therapy in patients with cystic fibrosis, mainly due to the reduction in FEV1 volume in the placebo group. They suggest several reasons for this, such as optimal respiratory health for patients at time of trial entry, enthusiasm for the trial leading to improvements in lung function during the recruitment period, and that the trial included all available data, even if the patients were unstable, while registry data only contains measurements from an annual review. They also note the trial’s heterogeneous response and that the fact that changes may be the result of a “non-specific response” to the pGM169/GL67A treatment.

The authors describe their conclusions as a “proof of concept” for nonviral CFTR gene therapy, calling it “another step along the path of translational cystic fibrosis gene therapy.”

Innes said that the efficiency of the gene uptake needs to be improved before the therapy is applicable to clinical practice, adding that the UK Gene Therapy Consortium is engaged in pursuing several lines of research.

“We are exploring whether increased or more frequent dosing would increase benefit, the possible additional benefit of combining gene therapy with other basic treatments which help the CF ion channel to remain open, and novel viral gene therapy vectors which may increase the efficiency of gene transfer,” he said.