Frequency Therapeutics — Hearing Loss Regeneration

@Lucifer @patorjk

Here is the trial schedule for FX-322 that I was given when I applied (and was not accepted):

 

Can I ask, post-injury (let's assume for simplicity's sake we all have a one-time event or catalyst that induced our hearing loss / tinnitus), is there any part of the ear for which regenerative therapies are time-sensitive? Let's exclude steroid intervention - I mean while we wait for therapies, is there part of the ear which is deteriorating in a way that is irreversible.

As far as I can tell, hair cell regeneration isn't time-sensitive so long as you don't reach a profound level of hearing loss. Likewise, synaptic damage isn't time-sensitive so long as the spiral ganglions stay alive once detached, which also shouldn't be an issue as it sounds like they remain okay for years.

That being said I'm trying to sanity check assumptions as this seems like a critical issue to be aware of if it exists.

 

There is no reason to believe it's time sensitive and if it helps, "length of hearing loss" was not an exclusion parameter for any of the regenerative therapies so clearly they (regen biotechs) don't think so either.

 

Frequency Therapeutics has made no mention of a "therapeutic window" that you're referring to. Also, if a therapeutic window were a concern, it would have likely been a recruiting criteria in the Phase 2A. To note: During the JP Morgan Q&A, a question was asked regarding how long some of the participants had hearing loss in the trial. Chris Loose replied that its likely that some of the participants had been living with hearing loss for many many years prior to receiving FX-322.

 

It was also stated that Frequency Therapeutics would consider compassionate access and usually this means a release post phase 2. If their next trial is the pivotal one, then there is a reasonable level of feasibility that this will mean a release after this trial.

 

I think that there has not been a therapeutic window mentioned in not only the information released relating to FX-322 but also actually in relation to any of the other treatments too. Thus this tends to indicate that regeneration can consequently happen at any time.

I think that this no limit on regeneration time has actually been already shown to be true with treatments like OTO-413 which demonstrate that this can happen whenever.

 

Where and when was this stated?

 

In one of their talks. They said that compassionate use after a completion of phase 2 might be granted if it is looking like the treatment is promising. Obviously they couldn’t commit to it definitively because they have a really keen interest in not making unverifiable statements on trials or any things which will get them in trouble with the FDA. Fairly likely that this comment was simply a hint that this would be something they would look at and also be something that they are interested in.

 

I heard many times that FX-322 is more likely to improve higher frequencies. But what frequencies specifically are we talking about? 2kHz+, 8kHz+, 15kHz+?

 

Generally speaking, I don't think compassionate use of FX-322 is going to occur, for the most part, and should not be expected. Frequency Therapeutics are focused, I believe, on the successful completion of the clinical trial process before anyone has access other than trial participants.

Compassionate use with pharmaceuticals is primarily based on life-threatening or very serious conditions with unmet needs. While it may be easy to disagree, the diagnosis of conditions relating to the ear may not satisfy the coordinates of such need, with an exception to Meniere's Disease which FX-322 is not geared toward. The number of individuals out there are just too great.

 

FX-322 can theoretically improve all frequencies. It is just an issue of getting it to work throughout the cochlea across all frequencies. This is why in the current trials Frequency Therapeutics are looking at multi dosing and may also look at a different dosing technique to improve drug delivery.

The only reason FX-322 is being touted as being more likely to treat higher frequencies is because of the fact that the higher frequencies are hit first when it gets injected and secondly because of what was seen in the first clinical trial results.

 

This current phase will answer that question more. Results next spring.

 

In the Tinnitus Talk Podcast, Carl LeBel was asked specifically about this.

 

We won't have data on this until the end of Phase 2a. In Phase I/II, 4 of 13 patients saw a 10dB increase at 8kHz [1], however, this was deemed statistically insignificant. It's theorized that the frequencies above 8kHz saw greater improvements. In Phase 2a Frequency is testing the extended high range frequencies [2] (most likely up to 20kHz), so we'll have a better idea of what the drug can do.

[1] https://investors.frequencytx.com/static-files/6d161090-16f5-49f4-9606-8caceb5a88a1 (slide 25)
[2] https://clinicaltrials.gov/ct2/show/NCT04120116

 

Firstly, the testing is only being done to 16000 Hz in the current trials.

Secondly, where did you get the evidence from that 10 dB at 8000 Hz was statistically insignificant? It was also 4/6 of the big dose patients who had improvement at 8000 Hz.

 

It was statistically significant. The placebo patients and control ears did not see this improvement.

 

The range wasn't listed on the clinical trial site so I used a number I got from Google (which is why I said "most likely" rather than stating it as a definite). However, digging deeper, you are correct, 16kHz is the upper range they are testing [1].

Source?

From what I've seen, Frequency Therapeutics has not stated that this was statistically significant. They don't state anything about statistical significance on the Pure Tone slide (slide 31), while they do state it on the WR slide (slide 22). I'd love to be proven wrong though.

[1] https://investors.frequencytx.com/static-files/6d161090-16f5-49f4-9606-8caceb5a88a1 (slide 31)

 

I know we're going to be tripping over semantics... but I believe Frequency Therapeutics has called the 10-15 dB improvement at 8000 Hz by 4/6 patients in the Phase 1/2 "Clinically Meaningful."

The P-Value on the Clarity of Sound slide (23) at 0.010 indicates statistical significance. While it can depend on N size (sample population), typically a p value below .05 is considered significant.

In my opinion, and i'm sure others would agree, that a P value below .05 from a N=23 sample size is actually impressive.

What you would expect to see from a small sample set is what they show on Page 24, where P = .211. Its possible that the N-value increasing (more participants) in the Phase 2A might drive down the P-value for hearing-in-noise.

 

Sorry, you are correct.

Statistically and clinically significant in word scores (but not audiogram) and clinically significant for the patients with 10 dB audiogram improvements but not statistically significant across the whole cohort (though likely would have been, based on word score improvements, if they had done an extended audiogram in phase 1).

 

When improvement was seen at frequencies higher than let's say 10 kHz for patients older than 50 years of age, they also included presbycusis.

 

While Frequency Therapeutics is accurate in not stating that the audiogram improvements are statistically significant (and won’t due to their desire to be accurate), I think that this tends to come down to the limits of the trial and also the very fixed and strict rules around statistics. I have noticed that there have been some FX-322 naysayers who have criticised the results of the first trial for providing statistically insignificant benefits but while disingenuously overlooking the facts about the trial size and outcomes.

Thus I agree with others that had Frequency Therapeutics done expanded testing then they would have reached a different outcome.

 

Have they though? I can't find anything where they say that. The closest I can find is that they say "clinically meaningful improvements in hearing function" - which is ambiguous and could refer to the WR score improvements (they also state the clinically meaningful line on the WR score slides).

I don't doubt that the 10dB improvement was a good sign, and I'm sure we'll see bigger improvements at the higher frequency bands, but we should be careful in our interpretation of what they tell us.

As an aside, there is a Facebook group I'm in where they are much more negative on FX-322, and the 10dB improvement was derided as pretty insignificant. I haven't had my hearing tested since 2014 (I had the hearing of an 70-80 year old then, I'm sure it's much worse now), so I don't really have a feel for what 10dB would mean.

 

I wonder why they don't test up to 20 kHz, but just to 16 kHz. I really wanted to know how FX-322 affects those ultra-high skull piercing frequencies, since that's where my hearing loss is. It seems like I have to wait a bit longer then.

 

Because it's super rare for the places to have the capacity to test 20 kHz. The cap at 16 kHz opens the trial up to more testing centers and speeds it along.

It's also unnecessary to test 20 kHz because if you get improvements at 16 kHz, you would have them at 20 kHz.

 

They have. I can paraphrase Carl LeBel saying in an investor presentation about a year ago, that a 10 dB improvement at 8 kHz represents a roughly 3x improvement in hearing sensitivity; in a real world setting he referenced "hearing leaves rustling."

10 dB as a measurement of sound pressure on an A-weighted scale represents a 10X increase in sound. So, another way of looking at that 10 dB improvement is that the treated ear is now able to detect an 8 kHz sound that is 10x LESS powerful than before treatment.

I agree that there isn't a clear definition for what "clinically meaningful" means; in a clinical setting it might be argued that a 10 dB improvement may indicate a patient improving from a moderate hearing loss to mild hearing loss.

As it relates to the Clinical Trial, from what I gather, "Clinically Meaningful" endpoints represent those results that can be reproduced in a clinical setting. These are not derived/proxy endpoints based on study-only metrics. So, for this example; an audiologist should be able to directly measure a 10 dB improvement if a single dose FX-322 were administered in a doctor's office over a 90-day period.

 

I think the assumption is that if patients show improvement in the 8 kHz -> 16 kHz audiogram, it's highly likely that 16 kHz+ are seeing improvement as well. FX-322 enters the cochlea starting at the highest range and works its way down, so it's not likely it can just skip over those ultra-high frequencies. I am also curious of the limitations for the testing facility to test at the ultra-high ranges. The 8 kHz-16 kHz test seems fairly available.

I also learned recently that genetic factors can play a role in ultra-high frequency hearing; so some just wont be able to hear that high up, regardless of hearing health. This would skew the results of the test.

 

Would giving multiple shots at once push the first injection deeper into the cochlea, thereby hitting the lower frequencies?

 

That's a fair point.

I really hope it's going to work. My case is somewhat mild, but I really miss hearing that 16kHz+ ,,air" from music.

 

Can anyone comment on the above?

 

The theory that if you have improvement at 16000 Hz, you will have improvement at 18000 Hz is also actually the same reasonable conclusion which can be drawn about why when we saw improvement at 8000 Hz in the inaugural trial, we could also assume an improvement was seen at those very high frequencies. Frequency Therapeutics cannot and will not declare that has happened however because it would be non-compliant with the FDA requirements.