Frequency Therapeutics — Hearing Loss Regeneration

[Daniel Lion]

From the document:

Quick summary:

- The Planned Phase 2a clinical trial will also explore the efficacy of the drug regarding tinnitus

- About 30% improved word recognition for treated patient versus about 5% for the placebo group (only 23 patients)

- About 20% improved word in noise recognition for treated patient versus about 5% for the placebo group

- Apparently no statistically significant improvements on the audiograms

- Worked well on mice

Comment from Reddit:

Some huge problems.

First, no change in audiogram is major. If the treatment worked, that is, if it regenerated hair cells, that is exactly where you would expect to see the difference.

Speech tests are difficult to interpret. First, they only saw significant changes in 4 patients. I also couldn't find what speech test they used and how many items. This is critical. If you are performing in the mid percentages on word rec, a 30% change is likely not significant and is just test-retest reliability (see the Thornton and Raffin tables). So this result does not excite me at all.

Finally, their biggest claims are from post hoc statistical tests. Which, ew. Also, they aren't explicit about their statistical models, but they should have used mixed effects for this which it doesn't appear they did. So it's likely that some of the variability their model "explained" is incorrectly assigned to a fixed rather than random effect, artificially driving down their p values.

If mouse models translated to human models we'd all be basically cancer free.

Still an interesting company, but these results aren't promising.

I wouldn't say not promising, although the read was very humbling indeed.
Humbling for sure. We have a long way to go.

The patients only had 1 injection. Let's remain optimistic and keep hope that the results could be very different when a patient receives four as opposed to one.

Arghhhhhhh...

Wait and see, at least it's still moving forward.
The hurdles will get higher and higher, but so far so good.

After a quick read, I thought it looked promising.
Thank you for posting ajc, much appreciated.

Waiting for annV, JohnAdams, ChrisBoyMonkey, RB2014, FGG, d'Wooluf, and more insights from ajc and everybody else who has contributed to this thread that's good at statistics and science.

Basically my brain power is lacking, so I defer to the smart guys and gals on this. Waiting to see what you think.

 

[ChrisBoyMonkey]

From the document:

Quick summary:

- The Planned Phase 2a clinical trial will also explore the efficacy of the drug regarding tinnitus

- About 30% improved word recognition for treated patient versus about 5% for the placebo group (only 23 patients)

- About 20% improved word in noise recognition for treated patient versus about 5% for the placebo group

- Apparently no statistically significant improvements on the audiograms

- Worked well on mice

Comment from Reddit:

Some huge problems.

First, no change in audiogram is major. If the treatment worked, that is, if it regenerated hair cells, that is exactly where you would expect to see the difference.

Speech tests are difficult to interpret. First, they only saw significant changes in 4 patients. I also couldn't find what speech test they used and how many items. This is critical. If you are performing in the mid percentages on word rec, a 30% change is likely not significant and is just test-retest reliability (see the Thornton and Raffin tables). So this result does not excite me at all.

Finally, their biggest claims are from post hoc statistical tests. Which, ew. Also, they aren't explicit about their statistical models, but they should have used mixed effects for this which it doesn't appear they did. So it's likely that some of the variability their model "explained" is incorrectly assigned to a fixed rather than random effect, artificially driving down their p values.

If mouse models translated to human models we'd all be basically cancer free.

Still an interesting company, but these results aren't promising.

@Daniel Lion the thing with these negative comments is that they are forgetting this was only a phase 1 safety trial. Safety, it was only 1 injection! They just happened to see pretty good results from it. Phase 2 is where they really wanna see if it works or not, and thus will be doing more than only 1 injection.

Truth is we only have reason to be positive about it and wait to see for now. These results are pretty good so far because we can see that this drug is already doing something positive from only one injection. Phase 2 and 3 is where the real tests are. This is just shaking the 8 ball and it saying "outlook is good".

 

[JohnAdams]

First, no change in audiogram is major. If the treatment worked, that is, if it regenerated hair cells, that is exactly where you would expect to see the difference.

Yep.

If mouse models translated to human models we'd all be basically cancer free.

Yep.

More doses are probably needed.

 

[Auron]

First, no change in audiogram is major. If the treatment worked, that is, if it regenerated hair cells, that is exactly where you would expect to see the difference.

I don't know nearly as much as other's here, but, if true... Well, that took the wind out of my sails...

I had high hopes for FX-322. Let's see what happens when they increase the dosage. If that fails as well... Well, I don't know where I would go from there.

 

[ChrisBoyMonkey]

Well, I'll invest once I have more money. Anything to support tinnitus research.

The main objective is the hearing loss but yes they are officially looking at tinnitus now. Not that you can profit from it since they're not IPOs, Dr. Shore's device, Dr. Tzounopoulos drug approach, and Dr. Bao's TNF-alpha drugs are potential treatments being researched specifically for tinnitus.

 

[Daniel Lion]

@Daniel Lion the thing with these negative comments is that they are forgetting this was only a phase 1 safety trial. Safety, it was only 1 injection! They just happened to see pretty good results from it. Phase 2 is where they really wanna see if it works or not, and thus will be doing more than only 1 injection.

Truth is we only have reason to be positive about it and wait to see for now. These results are pretty good so far because we can see that this drug is already doing something positive from only one injection. Phase 2 and 3 is where the real tests are. This is just shaking the 8 ball and it saying "outlook is good".

That's heartening... thanks for your comment.

My interpretation was similar to yours... wait and see and remain "cautiously optimistic".

4 patients did show a 10 decibel shift at day 90, out of like 13 people at 8,000Hz.
The drug passed the first safety trial. The word recognition tests were positive. Certainly not doom and gloom or "disappointing".

Bet it helps with tinnitus.
See you all in late 2020 on this thread...

 

[JohnAdams]

 

[Auron]

The thing with these negative comments is that they are forgetting this was only a phase 1 safety trial. Safety, it was only 1 injection!

You are 100% correct.Thanks for reminding me.They always stated that Phase 1 was for safety test only. Phase 2 was to increase Dosage/Injection count. As you said, we will actually find out then if it works or not. Now that I took a step back and thought about it a little more, I feel any improvement, no matter how small, is only a good sign, considering this was always JUST A SAFETY TEST.

 

[Badger19]

I guess it wasn't the case here, but normally phase 1 is tested on healthy people who don't even have the condition it attempts to treat.

 

[jeff W]

Wait a second, does the cochlear fluid in the cochlea also circulate into the vestibular system? If so, then this may cure vertigo.

So if one could obtain VHIT data from the trial (which they are probably passing around already as it's not the targeted pathology) it may provide inarguable evidence of functional peripheral recovery. If I was an investor, I would sure like to see that.

 

[FGG]

So the reddit doom and gloomer thankfully seems to have downplayed a few points. Namely that improvements in the audiogram at 8000 Hz were seen by 4 patients (but not other frequencies)--see screenshot below. Though they may not have been statistically significant in aggregate, i find that pretty remarkable for a single injection.

The document even discusses the probable reason for this (diffusion distance is shorter at high frequencies).
It's also telling to me that the design of phase 2 includes audiogram up to 16000 Hz. This means that they know it works better at higher frequencies.

I suspect the higher frequency improvements help with word recognition. There is a small amount of speech information in those frequencies.

Just my thoughts from one quick read.

 

[Coleoptere]

I would not say that 30% / 20% improvement in word recognition is disappointing after 1 injection...

I may be wrong but as far as there being a mention of audiogram in the document it refers to a check to confirm if the patients have stable hearing loss before they enter the trial:

"....meaning their hearing function at study entry was not significantly different based on a documented audiogram from at least six months prior to the study".

 

[d'Wooluf]

The IPO thing surprised me. You float a company to fund expansion or make scads of money from speculative investors. Frequency Therapeutics have raised about $200 million so far? Last I looked there were about 20 people to pay, premises to put them in, and clinical trials to fund. I would have thought that they had plenty of money to fund all that for a few years. If you really believe in your product you don't dilute ownership unless you really need to. It means less profit for you when the money from sales starts rolling in.

Do they believe in FX-322 or not?

 

[d'Wooluf]

So 4 out of 6 participants with moderate to moderately severe hearing loss had significant improvement in word recognition scores. 4 out of of those same six participants showed a 10 dB improvement in audiometry at 8000 Hz. The same four? That would be interesting to know. Those same four (improved word recognition scores) had non-statistically significant improvements in speech-in-noise.

I'm not sure how you can pick out the 4 best results and call them statistically significant. Statistical significance applies to populations from my dim memories of statistics. Clinically significant maybe. Having said that I'd probably be pretty happy if I were one of those four subjects.

 

[JohnAdams]

I wonder how they actually administered the drug. Did they just do the injection and send them off on their way? Did they have them lay there with their head tilted sideways for 1-2 hours like I did in Korea? Did they advise the patients to not talk or swallow? All of that could have had a major impact on how well the drug entered into the cochlea.

 

[FGG]

The IPO thing surprised me. You float a company to fund expansion or make scads of money from speculative investors. Frequency Therapeutics have raised about $200 million so far? Last I looked there were about 20 people to pay, premises to put them in, and clinical trials to fund. I would have thought that they had plenty of money to fund all that for a few years. If you really believe in your product you don't dilute ownership unless you really need to. It means less profit for you when the money from sales starts rolling in.

Do they believe in FX-322 or not?

200 million is not as much as it sounds when you have to float the company through a 5 year FDA trial (and continued research of FX-322 as well as their MS pipeline). Have to also consider future marketing costs, distribution and the cost of the trial itself. All that aside, operating costs alone of even a small biotech could easy run 100 million over 5 years and it may take 8 years plus for the company to be profitable even if the drug is extremely successful.

Additionally, one possible explanation is also that they are preparing to have a smaller market share than previously anticipated since they, thus far, only saw changes in 8000 Hz and higher.

This would definitely not replace hearing aids or implants and, in fact, may only be useful for more specific hearing loss (though might be very effective for many kinds of tinnitus) than originally thought. Unless they can think of a way to get the drug further into the cochlea (surgery eventually? New vehicle? Increased dose?), the initial market will be smaller than originally anticipated and a lot more money will be needed to ride out the biotech initial cash haemorrhage.

One of the things they are specifically looking at in the second trial is tinnitus as well as measuring the audiogram up to 16000 Hz. It may very well be that the drug works, but can only "reach" the very high frequencies.

 

[Mr Mister]

This is a major disappointment, there is no way around it. Majority got no improvement in audiogram and some got 10 dB which is far from curing hearing loss. 10 dB is not statistically significant but rather a typical fluctuation between audiograms over time for someone with at least moderate hearing loss. My audiograms vary 5-10 dB depending on the day, tinnitus at the time, whether I want to guess or be really certain I am hearing the test tone etc. Word score improvements mean nothing.

The fact that they IPO the business tells current owners lack trust in their project. It is essentially an exit for them. Selection of their investors and pharma partners already raised suspicions that something isn't right.

I have no expectations for Frequency Therapeutics based on what I have read here today. I hope that Audion/Regain can actually show material and consistent improvement in audiograms.

 

[JohnAdams]

This is a major disappointment, there is no way around it. Majority got no improvement in audiogram and some got 10 dB which is far from curing hearing loss. 10 dB is not statistically significant but rather a typical fluctuation between audiograms over time for someone with at least moderate hearing loss. My audiograms vary 5-10 dB depending on the day, tinnitus at the time, whether I want to guess or be really certain I am hearing the test tone etc. Word score improvements mean nothing.

The fact that they IPO the business tells current owners lack trust in their project. It is essentially an exit for them. Selection of their investors and pharma partners already raised suspicions that something isn't right.

I have no expectations for Frequency Therapeutics based on what I have read here today. I hope that Audion/Regain can actually show material and consistent improvement in audiograms.

It was just one dose in one ear.

If the next multidose trial doesn't show anything better, then we can poo poo this.

 

[d'Wooluf]

One of the things they are specifically looking at in the second trial is tinnitus as well as measuring the audiogram up to 16000 Hz.

It sounds a bit like tinnitus is their fall-back option. If hearing loss is your problem but not tinnitus, you really don't care about 16000 Hz if you don't get the frequencies below. As you said, multiple and/or increased dosages is the hope now.

you have to float the company through a 5 year FDA trial

How much would it cost to do a medical exam, inject a subject 4 times and do audiometry for 7 months. Throw in admin costs. A grand maybe? And a successful result at each step would trigger a milestone payment from their pharma partner. To me, it looks like they want to lock in some profit.

 

[FGG]

This is a major disappointment, there is no way around it. Majority got no improvement in audiogram and some got 10 dB which is far from curing hearing loss. 10 dB is not statistically significant but rather a typical fluctuation between audiograms over time for someone with at least moderate hearing loss. My audiograms vary 5-10 dB depending on the day, tinnitus at the time, whether I want to guess or be really certain I am hearing the test tone etc. Word score improvements mean nothing.

The fact that they IPO the business tells current owners lack trust in their project. It is essentially an exit for them. Selection of their investors and pharma partners already raised suspicions that something isn't right.

I have no expectations for Frequency Therapeutics based on what I have read here today. I hope that Audion/Regain can actually show material and consistent improvement in audiograms.

IPOing a business does not necessarily suggest lack of confidence in their drug IMO. It might be easier to get continued funding publically (versus private equity) when your profitability is still 5-10 years away.

With a publically traded company, you have "catalysts" that increase shareholder value long before the company is floating in the black.

There is no way Astellas didn't have early access to the same trial data we just received before investing multi-millions to partner. That tells me they see potential.

 

[FGG]

So 4 out of 6 participants with moderate to moderately severe hearing loss had significant improvement in word recognition scores. 4 out of of those same six participants showed a 10 dB improvement in audiometry at 8000 Hz. The same four? That would be interesting to know. Those same four (improved word recognition scores) had non-statistically significant improvements in speech-in-noise.

I'm not sure how you can pick out the 4 best results and call them statistically significant. Statistical significance applies to populations from my dim memories of statistics. Clinically significant maybe. Having said that I'd probably be pretty happy if I were one of those four subjects.

In the document they mentioned not statistically significant in aggregate but that 4 patients who had moderate to severe loss saw the 10 dB improvements. It's very interesting to me that those improvements happened at 8000 Hz and not at the lower frequencies they tested. This at least suggest the improvements are due to the drug but more testing is needed.

 

[Mr Mister]

It was just one dose in one ear.

If the next multidose trial doesn't show anything better, then we can poo poo this.

What a terrible day this is really. This is a worst case scenario come true.

For years people here have been begging that Frequency Therapeutics shares something about the efficacy of their drug. Now we know the reason why they have been quiet: there is nothing interesting to be told. No material impact on hearing.

 

[JohnAdams]

What a terrible day this is really. This is a worst case scenario come true.

I really don't think so. This was only one dose. I was Skyping with Minbo Shim a few weeks ago and we were discussing this and he said that he thinks that FX-322 will need 4-6 doses in a best case scenario, simply because of the small amount of medicine that can diffuse through the round window membrane. They did show that there were some improvements at 8kHz, which means that it did something. They did show that FX-322 grew hair cells in human explants, so I believe that it did in fact show efficacy, even at just one dose. Also, the dB scale is logarithmic so 10 decibels at the level of hearing loss they were testing this on, which was like what, 70 decibels (?) is actually really good. What would that kind of gain do for those of us with "normal" audiograms? The basic question that we've been wanting the answer to is if restoring hearing will alleviate tinnitus, and many if not most of us have so-called perfect hearing tests, when in reality, we have hidden hearing loss, which is characterized by speech in background noise difficulty, and they have demonstrated an ability for this drug, at just one dose, to improve that.
I think that this will just require many doses. Don't let your spirits get crushed, yet. This is not over.

Also, we may learn more when they publish their results on the FDA trials site as of now there are none posted.

 

[FGG]

It sounds a bit like tinnitus is their fall-back option. If hearing loss is your problem but not tinnitus, you really don't care about 16000 Hz if you don't get the frequencies below. As you said, multiple and/or increased dosages is the hope now.


How much would it cost to do a medical exam, inject a subject 4 times and do audiometry for 7 months. Throw in admin costs. A grand maybe? And a successful result at each step would trigger a milestone payment from their pharma partner. To me, it looks like they want to lock in some profit.

They have an entire staffed research facility that is not only working on FX-322 but their future pipeline (they already have an upcoming MS drug candidate).

In addition, there is still ongoing research on a drug during active trials. I'm willing to bet they are working on trying to find a better vehicle for FX-322 so that an updated version could diffuse further. That's not a reason to halt existing trials but it's so much more complicated than just the cost of the procedures. I read that it's up to 100,000 pages of paperwork to get a drug FDA approved. FDA studies are an enormous financial burden on a company.

 

[GregCA]

What a terrible day this is really. This is a worst case scenario come true.

For years people here have been begging that Frequency Therapeutics shares something about the efficacy of their drug. Now we know the reason why they have been quiet: there is nothing interesting to be told.

You haven't heard about efficacy simply because they haven't tested efficacy yet: they've tested safety.

 

[JohnAdams]

And then there's this:

Sorry, this is totally fake news bullshit I just created, but it SHOULD be the reality, or something similar. Where are you at LaGuinn Sherlock? Where is our lobby?

 

[FGG]

To add to this:

https://en.wikipedia.org/wiki/Cost_of_drug_development

Even companies like Biospecifics, which have very few staff members, and a non-expensive drug to manufactor (and I'm guessing injecting a contracted penis requires about the same level of expertise as an intratympanic one) required many millions to run their trial. Unfortunately, that's just the reality if drug development.

 

[scotty03874]

I think that this will just require many doses. Don't let your spirits get crushed, yet. This is not over.

John was this the news you were talking about last week that was expected to be announced? IPO. I can't find the patient results online anywhere...

 

[JohnAdams]

John was this the news you were talking about last week that was expected to be announced? IPO. I can't find the patient results online anywhere...

No, they are giving some speech on the 10th at a symposium or something. I was totally blindsided by this announcement.

 

[Flamingo1]

- Apparently no statistically significant improvements on the audiograms

Uhh... Weren't the participants of this study in such bad shape that they were having their cochlea removed?? It sounds like these people had more problems than just hair cells hence the lack of improvement in their audiogram. Let's hope this is the case and that multiple doses can help to restore SOME hearing.