Frequency Therapeutics is going back to the drawing board. After showing promise in phase 1, four dosing regimens of its hearing loss treatment did no better than placebo in a phase 2a study. The company will press ahead with a single-dose regimen and is waiting on final results from this study as well as readouts from two more phase 1b studies to plot its next steps.
The company's stock fell as much as 72% in premarket trading Tuesday morning.
In the phase 2a study, Frequency tested the drug, FX-322, in nearly 100 patients with mild to moderate sensorineural hearing loss who had suddenly lost their hearing or have a history of chronic noise exposure. The patients received four weekly injections through their eardrum. Some got four placebo shots or four shots of FX-322, while others received one, two or three doses of the drug, with placebo making up the remaining injections.
Early data show all four dosing schedules had “no discernible hearing benefit” over placebo, the company said.
“We can’t discern any differences between any of the dose cohorts and the placebo group. They all look comparable,” said Carl LeBel, Ph.D., Frequency’s chief development officer.
RELATED: Frequency's hearing loss treatment shows long-term promise
Patients in all groups had improved word recognition scores, but those changes were smaller in magnitude than the improvements seen in separate phase 1, single-arm studies of the drug, LeBel said.
“It’s some of the first evidence we’ve got to indicate that maybe hitting the ear once a week for four consecutive weeks with these materials is just dampening the ability of the ear to demonstrate the benefit on could get from FX-322,” LeBel said.
On top of that, the patients in the placebo group did surprisingly well on their hearing tests, showing an “unexpected apparent level of hearing benefit” that was neither seen in earlier trials nor in historical data.
The company chalks the trial’s outcome to bias in its design.
“Patients, our subjects, they desperately need hearing solutions,” said Kevin Franck, Ph.D., senior vice president of strategic marketing and new product planning. “When we communicated there were entry criteria of a depressed word score, we may have induced an unconscious bias in subjects who wanted to get into the trial.”
RELATED: Frequency bags $62M to advance Astellas-partnered hearing loss drug
The company designed the phase 2a trial to learn whether repeat dosing could boost the effects of FX-322, a medicine that harnesses the body’s healing abilities. It works by triggering dormant progenitor cells in the ear to spur the growth of new hair cells that sense and translate sound.
“FX-322 is two small-molecule drugs and once injected, they essentially diffuse into the inner ear. We can imagine a wave of two drugs washing through a particular region of the cochlea,” LeBel said in a previous interview. “One week later, we come back. Maybe that second injection gets a bit further throughout the cochlea, maybe it hits an additional number of progenitor cells and we get a bigger benefit. Or, maybe with additional injections, we will get more people with the same benefit.”
The trial didn’t go the way Frequency thought it would, but the company is looking on the bright side.
“Four injections aren’t going to be a favored approach and I think that is very different than what we thought going into the study … When we’re advancing a drug in a field where we are sort of the leaders, this phase 2a has been really about probing; it allows us to continue to understand the depth of our technology,” Frequency CEO David Lucchino said.
Frequency expects to report full data in the second quarter of this year. Later this year, it plans to read out a pair of phase 1b studies testing FX-322 in patients with severe sensorineural hearing loss and patients with age-related hearing loss.
The company also revealed early results from a phase 1b study in 33 patients with mild to moderate sensorineural hearing loss who received the treatment in one ear. Three months after treatment, one-third of the patients logged a 10% or greater improvement in word recognition scores in their treated ear compared to the untreated ear.
The study was designed to gauge the safety of different dosing conditions and not to measure efficacy.
“But it did give us the confidence to move forward down a path that looks like a single injection. As we work through some bias issues, I think we’re going to be well-positioned,” LeBel said.
“It’s fair to say we have more phase 2 work to do. Some of these things we uncovered out of the four-injection study, we really want to home in on as we begin to think about what a pivotal trial design will look like in the future,” he added.