Hearing loss specialists at Frequency pin the blame for PhIIa flop on trial design — but they couldn't stem a freefall in share price
When Frequency Therapeutics scored $80 million in cash from Astellas to jumpstart an alliance, the biotech cautiously pointed to Phase I/II data bolstering hope that they might indeed have a groundbreaking regenerative medicine approach to restoring hearing, leaning on research done by Bob Langer and Jeff Karp out of MIT around progenitor cell activation.
Newly released Phase IIa results are now keeping them away from that lofty goal.

At the interim analysis, four weekly injections of FX-322 failed to improve things for patients with mild to moderately severe sensorineural hearing loss, whether they looked at word recognition, words-in-noise, pure tone audiometry and additional exploratory or other measures.
Frequency execs, though, believe the blame lies not with the drug but with the trial design. Armed with preliminary data from a separate Phase I/II trial, they are all but giving up on the four-dose regimen and going back to take a single-shot approach through further evaluation instead — with lessons taken from the latest mid-stage trial.
“I can tell you that we have taken a meaningful step clinically, though not in the way we anticipated,” CEO David Lucchino said in an investor call.
Cowen analyst Phil Nadeau was sympathetic to the argument, noting that while the topline readout brings the biologic effect of the drug into question, Frequency may have a point in continuing its development:
We agree that the replication of the initial single dose data suggest that ‘322’s profile should continue to be explored. Though there are numerous caveats to single arm studies (most notably the contralateral ear in study ‘111 need not have WR deficits at baseline) the consistency of effect in the two trials provides a signal of efficacy for the single injection regimens.
Investors are less cheery, tanking the shares 77.98% Tuesday afternoon.
The first of two main issues, Lucchino explained, was that the four weekly injections seemed to “temporarily overwhelm the ear.”
“This is a bit of an oversimplified analogy, but one that can possibly fit: When you reseed the lawn, you need to stay off the grass,” he said. “In other words, while we can’t see the precise mechanisms in play, multiple weekly injections led to conditions that did not allow FX-322 to achieve its intended effect.”
Secondly, he implied that some patients may have exaggerated the extent of their word recognition deficit in order to get into the study, resulting in an unexpectedly high level of hearing benefit in the placebo group — “outliers” that his team had never seen before.
In the trial, all 95 participants received four injections total, with different combinations of drug and placebo. None of the treatment groups had discernible benefit compared to placebo.
The reason Frequency is holding out hope comes mostly from preliminary results in an open-label study in which a single dose of FX-322 was given to 33 patients with severe SNHL. With the untreated ear as the control, investigators reported that 34% of participants saw a 10% or greater improvement in word recognition scores. A subset even more than doubled their WR scores.
Two other early-stage, placebo-controlled trials on the single injection are ongoing, focusing on age-related hearing loss and severe SNHL, respectively. Results should be in later this year — when Frequency also expects to report final numbers from the flopped Phase IIa.
“We recognized when we started and continue to recognize today that achieving such a bold goal would not be without unexpected turns along the way,” Lucchino said. “We believe in the technology, we believe in the data, and we will follow the science.”