Frequency Therapeutics — Hearing Loss Regeneration

[Artem]

You're right there are most sophisticated ways to perform audiograms, however audiograms are primarily testing the outer hair cells whereas frequency is primarily interested in regenerating the inner hair cells for clarity.

I think the aforementioned analogy is fine if you consider the difference purposes the hair cells have. Outer hair cells amplify sounds and can be tested via the audiogram whereas the inner hair cells are for clarity and can be tested via word scores or word in noise tests. @Chad Lawton is right if you had to choose between clarity and loudness you should almost always prefer an increase in clarity. Hearing aids can help with subsets of outer hair cell dysfunction however there is no equivalent for clarity.

Hmm, I'm sorry. I believed that the outer sensory cells are responsible for high tones, and the inner ones for low tones. The audiogram can cover the entire hearing range from 100 Hz to 20 kHz. But it looks like it's not so... Strange, but very interesting...

Anyway, thanks for the clarification and optimism.

Веst regards,
Artem

 

[DimLeb]

@Artem, "the inner hair cells convert sound vibrations from the fluid in the cochlea into electrical signals that are then transmitted via the auditory nerve to the brain whereas the outer hair cells amplify low-level sounds that enter into the fluids of the cochlea".

So for each frequency, low or high (20 Hz - 20 kHz), I imagine it's both the two layers (inner and outer hair cells) that are responsible to pick it up.

 

[Street Novelist]

Does FX-322 regrow both inner and outer hair cells, or just inner? What if you have outer hair cell damage, but not inner? Would this drug still work for you?

 

[Akk]

To clarify, outer and inner hairs cells are paraller in "tube". First cells, near ear drum, react to high frequencies and more deeper you go more lower frequencies cells react there.

https://basicmedicalkey.com/special-senses-2/

 

[Street Novelist]

Their exclusion criteria makes my head hurt. If you have mild to moderate hearing loss, FX-322 should be able to use your progenitor cells to regrow completely new hair cells, theoretically restoring hearing. I noticed on their exclusion criteria, they list:

"Any conductive hearing loss of greater than 15 dB at a single frequency or greater than 10dB at two or more contiguous octave frequencies in the study ear."

But, what if your hearing loss is sudden sensorineural due to loud music? That's not conductive, right? So, you should be good to go. Can someone look at my audiogram and tell me what's what?

 

[SD7]

Does FX-322 regrow both inner and outer hair cells, or just inner? What if you have outer hair cell damage, but not inner? Would this drug still work for you?

Theoretically both, however practically it heavily prefers the inner hair cells.

 

[ColinUK]

With all these additional trials has there been any mention of tinnitus at all. Before with the first trial with just a handful of people they said that they had some anecdotal reports of tinnitus improvement. Now after all these trials with hundreds of patients and where they actually opted to test for tinnitus improvement there has been no updates. Or did I miss them?

I mean in light of the poorer than expected hearing signal improvements you would think they would be shouting from the rooftops if there had been improvement of people's tinnitus.

So I can only assume for tinnitus it's a total dud? Or did I miss something?

 

[Diesel]

With all these additional trials has there been any mention of tinnitus at all. Before with the first trial with just a handful of people they said that they had some anecdotal reports of tinnitus improvement. Now after all these trials with hundreds of patients and where they actually opted to test for tinnitus improvement there has been no updates. Or did I miss them?

I mean in light of the poorer than expected hearing signal improvements you would think they would be shouting from the rooftops if there had been improvement of people's tinnitus.

So I can only assume for tinnitus it's a total dud? Or did I miss something?

They're still testing tinnitus outcomes using the TFI in Phase 2B.

However, at this point, with the FDA giving them a clear path to approval with word score / speech recognition measures, I don't see why they'd bother mentioning tinnitus.

 

[ColinUK]

They're still testing tinnitus outcomes using the TFI in Phase 2B.

However, at this point, with the FDA giving them a clear path to approval with word score / speech recognition measures, I don't see why they'd bother mentioning tinnitus.

But surely an improvement in tinnitus would be proof that something positive is happening in the ear. I mean if someone said after their injection they had a 3 dBa improvement and an elimination of tinnitus I would be more impressed with the elimination of tinnitus.

I think the only reason to omit it is if no one is reporting any positive affect? Which unfortunately sucks.

 

[Gb3]

Ya, they would have reported improvement in tinnitus if it happened. They needed some more good news.

 

[Diesel]

But surely an improvement in tinnitus would be proof that something positive is happening in the ear. I mean if someone said after their injection they had a 3 dBa improvement and an elimination of tinnitus I would be more impressed with the elimination of tinnitus.

I think the only reason to omit it is if no one is reporting any positive affect? Which unfortunately sucks.

They have to see statistically significant improvements in the TFI from the participant's baseline. I highly doubt that the Severe group (assuming considering the 3 dB SIN reference) would see tinnitus improvement if they didn't see significant word score improvement. It's also a problem of getting enough participants that have tinnitus to be significant. Some do have it, some do not.

Given how heterogeneous tinnitus is "experienced" on top of how heterogeneous acquired SNHL hearing loss is, it's likely that greater coverage of the cochlea may should better results for tinnitus improvement from baseline (FX-345).

 

[Survivor234]

They're still testing tinnitus outcomes using the TFI in Phase 2B.

However, at this point, with the FDA giving them a clear path to approval with word score / speech recognition measures, I don't see why they'd bother mentioning tinnitus.

I have a question. Why are they focusing on word score/speech recognition and not high frequency hearing? Wouldn't the latter be more indicative of any sort of successful hair cell regeneration?

I have a feeling that regenerating hair cells, and by extension eliminating tinnitus, is probably going to be a multi-treatment process. I feel like we lose more than we gain from the competitiveness and the close-guarding of advancements and information among firms.

Hopefully, once the treatments are on the market, the curtain will rise and people will begin to experiment with different treatments (I know for certain that tinnitus sufferers will take as much treatments as they can).

 

[Diesel]

Why are they focusing on word score/speech recognition and not high frequency hearing? Wouldn't the latter be more indicative of any sort of successful hair cell regeneration?

Their strategy has shifted to focus on it.

1. FX-322 has shown the most statistically significant improvements in word score / speech recognition amongst trial participants.
2. There is data available on patients with word score / speech recognition deficits, making recruiting more accessible in a reasonable timeframe in the US. HF audiograms for patients are not common at all in the US.
3. The FDA gave Frequency Therapeutics the "all clear" to focus on word score / speech recognition as an acceptable outcome for FX-322. So, why bother deviating?

 

[Survivor234]

Their strategy has shifted to focus on it.

1. FX-322 has shown the most statistically significant improvements in word score / speech recognition amongst trial participants.
2. There is data available on patients with word score / speech recognition deficits, making recruiting more accessible in a reasonable timeframe in the US. HF audiograms for patients are not common at all in the US.
3. The FDA gave Frequency Therapeutics the "all clear" to focus on word score / speech recognition as an acceptable outcome for FX-322. So, why bother deviating?

But why isn't there a statistically significant improvement in high frequency hearing?

 

[Diesel]

But why isn't there a statistically significant improvement in high frequency hearing?

One can only speculate.

 

[ColinUK]

From the Tinnitus Talk Podcast interview eons ago:

"Carl: Again, we don't have data. Certainly there is anecdotal reports as patients have come back and visited with ENTs when they have had conversations with them about how they are doing. Some of them have offered that they have had improvements in tinnitus"​

Makes no sense why they would keep it secret. Their first trial with a handful of patients they were happy to report it. Why after 100s injected have none come forward to report an improvement and if they had, why on earth would Frequency Therapeutics, who so happily told us the anecdotes before, not mention anything. What possible advantage could there be to not report people's perception of their tinnitus having improved?

I mean, if they found out it only gave a small hearing boost but fixed tinnitus, I think they would be happy to get the drug approved as a tinnitus drug?

 

[kiki]

But why isn't there a statistically significant improvement in high frequency hearing?

I think this is because in Phase2a, there was an improvement in the placebo group that canceled out the significant improvement.

 

[Survivor234]

I think this is because in Phase2a, there was an improvement in the placebo group that canceled out the significant improvement.

Do you have any evidence of this? Are the Phase 2a results publicly available anywhere to see?

 

[Diesel]

From the Tinnitus Talk Podcast interview eons ago:

"Carl: Again, we don't have data. Certainly there is anecdotal reports as patients have come back and visited with ENTs when they have had conversations with them about how they are doing. Some of them have offered that they have had improvements in tinnitus"​

Makes no sense why they would keep it secret. Their first trial with a handful of patients they were happy to report it. Why after 100s injected have none come forward to report an improvement and if they had, why on earth would Frequency Therapeutics, who so happily told us the anecdotes before, not mention anything. What possible advantage could there be to not report people's perception of their tinnitus having improved?

I mean, if they found out it only gave a small hearing boost but fixed tinnitus, I think they would be happy to get the drug approved as a tinnitus drug?

They aren't keeping it a secret.

The first trial had anecdotes, but no data to validate improvement from placebo. No TFI.

In the phases that followed, TFI was added was a secondary outcome. Phase 2A data was effectively worthless and inconclusive, so no way to know if TFI data showed improvements. The other two placebo-controlled Phase 1Bs (Age-Related, Severe), weren't effective at the primary outcomes, so it's highly likely that TFI as an experimental outcome didn't provide significant data either.

There may be a patient or two that did have an improvement in their tinnitus, but now that their data is aggregated with a pool of 100s, it's probably not significant at a group-level. EX: Tinnitus = YES. Therefore, they can't reasonably say, "hey a couple people's tinnitus got better, but it's all anecdotes." That would only introduce unnecessary scrutiny, and take focus off of where the drug IS showing an effect.

 

[Diesel]

Do you have any evidence of this? Are the Phase 2a results publicly available anywhere to see?

It's in their investor deck, and they do a video of it on the R&D day page. On their website.

 

[Gb3]

They aren't keeping it a secret.

The first trial had anecdotes, but no data to validate improvement from placebo. No TFI.

In the phases that followed, TFI was added was a secondary outcome. Phase 2A data was effectively worthless and inconclusive, so no way to know if TFI data showed improvements. The other two placebo-controlled Phase 1Bs (Age-Related, Severe), weren't effective at the primary outcomes, so it's highly likely that TFI as an experimental outcome didn't provide significant data either.

There may be a patient or two that did have an improvement in their tinnitus, but now that their data is aggregated with a pool of 100s, it's probably not significant at a group-level. EX: Tinnitus = YES. Therefore, they can't reasonably say, "hey a couple people's tinnitus got better, but it's all anecdotes." That would only introduce unnecessary scrutiny, and take focus off of where the drug IS showing an effect.

C'mon man! I had a lot of faith in this drug too but obviously there was no improvement in tinnitus. Faking word scores wouldn't mess up the TFI data.

 

[Lucifer]

But why isn't there a statistically significant improvement in high frequency hearing?

If you're talking about audiograms in the Phase 1b trials, there were a few patients that had an improvement at the highest frequency which was tested at 8 kHz.

Now with Phase 2b trial they are testing up to 16 kHz so we should see more patients with an improvement in audiograms.

 

[Diesel]

C'mon man! I had a lot of faith in this drug too but obviously there was no improvement in tinnitus. Faking word scores wouldn't mess up the TFI data.

If you choose to go into a trial and fake word scores, who's to say you wouldn't underweight other data?

 

[Gb3]

If you choose to go into a trial and fake word scores, who's to say you wouldn't underweight other data?

Are you saying they inflated their TFI?

 

[SD7]

Are you saying they inflated their TFI?

I think Diesel's point is that it's tainted regardless.

 

[Survivor234]

If you're talking about audiograms in the Phase 1b trials, there were a few patients that had an improvement at the highest frequency which was tested at 8 kHz.

Now with Phase 2b trial they are testing up to 16 kHz so we should see more patients with an improvement in audiograms.

So, was the lack of improvement because patients could already hear at 8 kHz or were the patients incapable of hearing at 8 kHz and there was no improvement?

Furthermore, if they could already hear at 8 kHz, how did they get any improvement?

Hopefully it's the former but your phrasing allows for the latter meaning.

Also, isn't FX-322 done with Phase 2 by this point? Is Phase 2b new or a redo?

 

[Diesel]

So, was the lack of improvement because patients could already hear at 8 kHz or were the patients incapable of hearing at 8 kHz and there was no improvement?

Furthermore, if they could already hear at 8 kHz, how did they get any improvement?

Hopefully it's the former but your phrasing allows for the latter meaning.

Also, isn't FX-322 done with Phase 2 by this point? Is Phase 2b new or a redo?

The audiogram doesn't measure whether or not a patient can hear a tone in the binary sense. It measures how much sound power in decibels are required at a specific frequency for a patient to hear a tone. For example, at 8 kHz, if a patient can hear the tone with 10 dB of power, it is considered within the normal range. If 50 dB are required, it is considered a moderate loss at that frequency.

The challenge with the audiogram as a measurement tool is as follows:

1. The tests produce a margin of error of up to +/- 10 dB at any given frequency. So, a patient could come in one day and detect 8 kHz at 20 dB one day, and 15 dB the next, and that's considered normal.

2. Decibels are a logarithmic measurement of sound power. A tone produced at 20 dB is 10x more powerful than the same tone produced at 10 dB. The same tone produced at 30 dB is 10x more powerful than one at 20 dB, which is 100x more powerful than the original produced at 10 dB. When looking at the audiogram, which measures from roughly 0 dB to 90 dB, it's not sensitive enough to show changes on the low or high end of the power spectrum.

3. The biology of the cochlea doesn't match the output of the audiogram completely. To a human ear, each increase in tone by 10 dB, represents about 3x-4x improvement, not 10x as per the decibel scale. There's also some level of redundancy in the cochlea hair cells, as up to 50% of outer hair cells need to be depleted before losses consistently show up in the audiogram. The human biology doesn't take into account frequencies, those are man-made. So, the preset tones in the audiogram may not reflect the optimal biological ability of the cochlea.

 

[Eliot Martin]

Only 6 dB improvement and they call it "clinically meaningful". Who is going to pay hundreds of dollars for a slight yet not even noticeable improvement in hearing?

 

[Diesel]

Here's where we get into trouble with FX-322.

If a patient is at / near a "normal" ceiling in terms of decibels, and regeneration takes place. Let's say 20 dB at 8 kHz. What is the likelihood that a regenerated cell or cells in that region are going to show an effect? There is an obvious ceiling effect and greater complication due to the margin of error with the audiogram as a measurement tool.

If we consider that a range of patients with various hearing loss levels at 8 kHz were recruited into the program. Where there are normal (10-20 dB), all the way to as much as 70 dB needed to hear the tone. Then, consider the potential for a ceiling effect for those in the normal range. It makes it difficult to show a significant improvement vs placebo for a cohort that got drug.

 

[Gb3]

I think Diesel's point is that it's tainted regardless.

Well if they said there was improvement in tinnitus, he wouldn't be saying that. Listen, I was very hopeful about this drug too but I'm just following the outcomes of the latest trials. It doesn't seem that promising.