Frequency Therapeutics — Hearing Loss Regeneration

[RB2014]

In hindsight, there were a few times in the past where Frequency Therapeutics was interviewed and they said they were unable to validate a multi-dose strategy with an animal model. Probably a bigger red flag than many of us realized.

However, multi-dosing being untested and ultimately needing more research doesn't mean the drug is a 100% flop. Showing those open-label trial results should be more revealing.

I'm curious if they could do a multi-dose open label trial? Just recruit groups into cohorts based on a 2x dose schedule length? Like a 2-week, 1-month, 2-month group?

They need to test a dose schedule at this point. They were really caught off guard with the trial results. Pursuing the single dose really doesn't get anyone anywhere at this point unless they can figure out why the word scores doubled. We have already discussed the benefits and I've said it before, 5 dB to 10 dB at the single frequency of 8 kHz is a nothing burger even if it is true.

Now that they are a publicly traded company they have to keep going. Either they figure this out or they run out of money. In either case, even if they can figure this out, we all need to prepare as the timeline has been moved much further.

 

[FGG]

I stand by my assertion that there is a lot of faith-based argumentation going on here, shrouded behind intellectual rationalization as it may be. We obviously want this to work so there's no way for us to be completely objective.

No one in the world is 100% completely objective about anything and anyone who says they are probably doesn't recognize their blind spots.

However, seeing a hearing signal which results in multiple people doubling their word scores in Phase 1 and using that to try to understand is not "faith based" and the reason you think it is is that (and you even said this), "the audiogram measures hearing" in a general sense.

Going back and forth on this is a waste of time but it's clear you think the doubling is insignificant for some reason.

 

[RB2014]

No one in the world is 100% completely objective about anything and anyone who says they are probably doesn't recognize their blind spots.

However, seeing a hearing signal which results in multiple people doubling their word scores in Phase 1 and using that to try to understand is not "faith based" and the reason you think it is is that (and you even said this), "the audiogram measures hearing" in a general sense.

Going back and forth on this is a waste of time but it's clear you think the doubling is insignificant for some reason.

Agreed... At this point there is no reason to doubt the Phase 1 data on the word scores and possibly the small gains on the audiogram. Phase 2 has proved that any manipulation or errors will come through in the study.

The real question now is where do they go from here.

 

[Guywithapug]

Would have been great to also see doses administered every 4/6 weeks, we may have acrually seen a stratified effect based upon Phase 1 findings, negating test variance/accuracy.

 

[ajc]

No one in the world is 100% completely objective about anything and anyone who says they are probably doesn't recognize their blind spots.

However, seeing a hearing signal which results in multiple people doubling their word scores in Phase 1 and using that to try to understand is not "faith based" and the reason you think it is is that (and you even said this), "the audiogram measures hearing" in a general sense.

Going back and forth on this is a waste of time but it's clear you think the doubling is insignificant for some reason.

What was it again, 2 people or something doubled their word scores in Phase 1? It is insignificant because the sample size is small and they might not have tried their best in the first test. After they got the injection, they most likely put extra effort into recognizing the words. Or the audiologist administering the word recognition test after the injections could have affected the results if they subconsciously assisted the test subjects recognize the words. There are many possibilities. A small sample size, a somewhat unreliable word recognition test, and the mental state of the patients could all have played a role.

 

[Diesel]

They need to test a dose schedule at this point. They were really caught off guard with the trial results. Pursuing the single dose really doesn't get anyone anywhere at this point unless they can figure out why the word scores doubled. We have already discussed the benefits and I've said it before, 5 dB to 10 dB at the single frequency of 8 kHz is a nothing burger even if it is true.

Now that they are a publicly traded company they have to keep going. Either they figure this out or they run out of money. In either case, even if they can figure this out, we all need to prepare as the timeline has been moved much further.

I agree up to the audiogram part. There are other measures of hearing for a reason. They all need to look good eventually; a significant improvement in just a single measure is more than is available today.

 

[Zugzug]

Showing those open-label trial results should be more revealing.

I'm curious if they could do a multi-dose open label trial? Just recruit groups into cohorts based on a 2x dose schedule length? Like a 2-week, 1-month, 2-month group?

Meh. The open label does very little for me. How do we know their trial design didn't encourage unconscious bias to get into this trial as well? It's so much worse because the placebos and treaters wouldn't offset. All of the responders would be treaters.

We have to be consistent about this. The remaining Phase 1b trials will hopefully paint a different story than Phase 2a.

 

[Jurger]

No one in the world is 100% completely objective about anything and anyone who says they are probably doesn't recognize their blind spots.

However, seeing a hearing signal which results in multiple people doubling their word scores in Phase 1 and using that to try to understand is not "faith based" and the reason you think it is is that (and you even said this), "the audiogram measures hearing" in a general sense.

Going back and forth on this is a waste of time but it's clear you think the doubling is insignificant for some reason.

Wasn't there also one responder who went down in WR in the follow up? How is that possible if IHC regeneration is the cause for these improvements? The hair cells are either there or they aren't there. I find it hard to believe they would die off after 6-12 months.

 

[FGG]

What was it again, 2 people or something doubled their word scores in Phase 1? It is insignificant because the sample size is small and they might not have tried their best in the first test. After they got the injection, they most likely put extra effort into recognizing the words. Or the audiologist administering the word recognition test after the injections could have affected the results if they subconsciously assisted the test subjects recognize the words. There are many possibilities. A small sample size, a somewhat unreliable word recognition test, and the mental state of the patients could all have played a role.

3 patients in Phase 1 and more patients in Phase 1b doubled word scores as well (where Frequency Therapeutics haven't reported the same discrepancies).

You can have small changes from test to test but *doubling*, especially with multiple patients has not historically been possible.

The only other explanation to this would be "discrepancy" between historical values and baseline but Frequency Therapeutics hasn't reported in these cohorts, so assuming it's genuine (and not a conspiracy in the company, the patients or audiologists), this is not chance, not placebo and not "trying harder." You don't literally double your word scores by trying harder or hearing issues would be treated with life coaches.

 

[Jurger]

Would have been great to also see doses administered every 4/6 weeks, we may have acrually seen a stratified effect based upon Phase 1 findings, negating test variance/accuracy.

If they really believe the grass analogy they should go back to multiple dosing with a new schedule at some point.

 

[GlennS]

Going back and forth on this is a waste of time

I don't really know what you're trying to assert anymore. You're in the minority if you think FX-322 still has much of a future, and if that's not what you're after, then I don't find squabbling over the post-mortem minutiae particularly useful.

 

[Jurger]

3 patients in Phase 1 and more patients in Phase 1b doubled word scores as well (where Frequency Therapeutics haven't reported the same discrepancies).

You can have small changes from test to test but *doubling*, especially with multiple patients has not historically been possible.

The only other explanation to this would be "discrepancy" between historical values and baseline but Frequency Therapeutics hasn't reported in these cohorts, so assuming it's genuine (and not a conspiracy in the company, the patients or audiologists), this is not chance, not placebo and not "trying harder." You don't literally double your word scores by trying harder or hearing issues would be treated with life coaches.

I'm not writing off the IHC regeneration theory entirely. We see strange things in patients with sudden hearing loss. The audiogram doesn't improve, but the WR score does. I myself went from 0% WR to 64% in 6 months. The audiogram/PTA? Unchanged. My doctor says the brain adapts.

 

[Zugzug]

What was it again, 2 people or something doubled their word scores in Phase 1? It is insignificant because the sample size is small and they might not have tried their best in the first test. After they got the injection, they most likely put extra effort into recognizing the words. Or the audiologist administering the word recognition test after the injections could have affected the results if they subconsciously assisted the test subjects recognize the words. There are many possibilities. A small sample size, a somewhat unreliable word recognition test, and the mental state of the patients could all have played a role.

I agree with you in principle, but not all studies are created equally.

Just as an example, depression, almost by definition, is something that is inextricably tied to outlook (of course, it's a disease so the outlook is based on neurochemistry).

So if a trial demonstrated 3 people massively improving mood because they temporarily believed in something, this would mean absolutely nothing to me. I think what everyone was honed in on here (everyone saw the small percentages) was picturing how someone would double their word scores legitimately.

Where you are right though, is we do have to go into these situations assuming that a real placebo effect is possible. It doesn't matter how sure we are that it's not possible. It's a painful employment of discipline, but has to be done. It absolutely sucks because I know if a drug improved my hyperacusis, it wouldn't be from a placebo. Sadly, a clinical trial would have to act like it could. For a successful clinical trial, it's on the data to disprove this.

In all likelihood, it's some combination of unknown factors and poor trial design, along with some odd thing that makes people become super responders. I guess we will see.

 

[FGG]

Wasn't there also one responder who went down in WR in the follow up? How is that possible if IHC regeneration is the cause for these improvements? The hair cells are either there or they aren't there. I find it hard to believe they would die off after 6-12 months.

Yes, but that wasn't one of the "doublers". That patient initially went from 29 to 38 and then back down (below baseline).

I'm not sure what happened with that patient, though it's possible he had degenerative hearing since only OHCs/audiograms were used to assess "stability".

 

[RB2014]

I don't really know what you're trying to assert anymore. You're in the minority if you think FX-322 still has much of a future, and if that's not what you're after, then I don't find squabbling over the post-mortem minutiae particularly useful.

@FGG is right. We need to stop the going back and forth. We did this when the Phase 1 results originally came out. Rehashing this won't get us anywhere. We really need to accept the data in all of the trials and move on. If people want to keep questioning the data just go back and reread all of those sections in this thread. It is all there.

 

[Koz]

Last seen one minute ago.

Flattered you had to stalk and amused that's all you got out of it.

Clearly logged in to see if there was a rational response to me. There wasn't

Ciao ciao.

 

[FGG]

I don't really know what you're trying to assert anymore.

I understand that. And that's why you think it's "faith based."

 

[Gb3]

I don't really know what you're trying to assert anymore. You're in the minority if you think FX-322 still has much of a future, and if that's not what you're after, then I don't find squabbling over the post-mortem minutiae particularly useful.

You are really coming off as ignorant.

 

[HootOwl]

then I don't find squabbling over the post-mortem minutiae particularly useful.

This is silly imo. Science is heavily concerned with minutiae. Imagine if after the flawed launch of the Hubble Space Telescope, the engineers and scientists on the team said "well, the whole thing is busted. Now it's space trash."

But they didn't. They diagnosed the issue, and proceeded to fix it. And in their case it was actually a tiny issue causing massive problem.

Like @FGG has said, some sort of hearing signal is being generated. We don't fully understand what this signal is, since it does not seem to correspond with OHC regeneration (as shown from the un-compromised audiogram data in Phase 1b), which has lent rise to the IHC Preference theory.

Make no mistake, I would argue the drug is still quite flawed at the moment. Very few from this thread are arguing otherwise. Penetration still remains a huge issue, and now we have the possibility of IHC preference.

To some this means "the drug is dead". To others (including myself) it means "the drug is flawed. So let's fix it if we can".

 

[Diesel]

What was it again, 2 people or something doubled their word scores in Phase 1? It is insignificant because the sample size is small and they might not have tried their best in the first test. After they got the injection, they most likely put extra effort into recognizing the words. Or the audiologist administering the word recognition test after the injections could have affected the results if they subconsciously assisted the test subjects recognize the words. There are many possibilities. A small sample size, a somewhat unreliable word recognition test, and the mental state of the patients could all have played a role.

So how do we explain that those that doubled their word score consistently showed an improvement from retesting at day 15, 30, 60, and 90? I would get if it was just a test at day 0 and then another at 90, how we could point to errors in the testing practice itself.

Furthermore, how do we then explain the follow up test scores 18 months later that still were significant by the same participants?

What happened if it wasn't drug?

Sorry... should include source:

Each day data point on the WR graph below represents a re-test...

https://patentscope.wipo.int/search/en/detail.jsf?docId=WO2020210388&tab=DRAWINGS

 

[kiki]

In fact, a 5 dB or 10 dB improvement at a particular frequency is not significant for a single injection.

However, if it is possible to repeat multiple "non-short-term, non-intensive" injections at appropriate intervals, that is, "repeat a single injection"...

The total dB improvement can be significant.

The advantage of FX-322 is that it does not deplete support cells.

Many a little makes a mickle.

 

[Street Novelist]

I don't really know what you're trying to assert anymore. You're in the minority if you think FX-322 still has much of a future, and if that's not what you're after, then I don't find squabbling over the post-mortem minutiae particularly useful.

If you really believe FX-322 doesn't work, stop posting. We don't need your negative energy.

 

[Zugzug]

During the teleconference call, when pressed by an investor on what mechanical or molecular effect could be causing the multi-dosing to dampen the effect, they had no answer. When subsequently asked whether there was any pre-clinical work to inform what might be happening, or why they pursued multi-dosing in the first place, again they had no answer. In their defence, they said it was impossible to try multi-dosing in vivo in animals as it would be too traumatic to the cochlea. I can only infer from that that any pre-clinical work in ex vivo human cochlea wouldn't serve to inform anything meaningful. Whatever way you look at it, Phase 2 was a giant experiment in multi-dosing that backfired spectacularly.

Anyone listening in to that webcast could see a deer having been caught in the headlights. I think it's fair to say they know multi-dosing screwed things up, but not being able to explain it, let alone not being able to justify why they pursued multi-dosing without any pre-clinical work to inform making that decision the first place, makes for seriously bad optics. Chris Loose and David Lucchino strike me as very confident and the latter made it very clear they were not going to rest on their laurels and just wait for Phase 1b results to come out and see what happens. It sounded to me as if they're already drawing up some kind of contingency plan. The reality is that this is a new company testing a new platform in an area where there is an unmet need. The odds are stacked against them.

All true. The only area where I will defend the company is that they are working against a populace that doesn't understand nor care about hidden hearing loss.

Some think hidden hearing loss is like witchcraft and just an excuse for a lack of audiograms.

Imagine a world where everyone, including me, was perfectly okay with incremental change. They wouldn't have been so pressured to push it deeper to light up audiograms.

At the same rate, it's also possible that egos simply won out. Maybe I'm giving them too much credit and their leadership wanted to rock the world. I'm not sure, but there's a lot of Monday morning quarterbacking. People who don't even understand phase 1/2 are saying it's all a wash. Kind of a microcosm of some ot the motives they were working against.

All of this said, this is mostly on them. They totally messed up the trial design. But in the future, people have to be cool with incremental change.

 

[FGG]

I'm not writing off the IHC regeneration theory entirely. We see strange things in patients with sudden hearing loss. The audiogram doesn't improve, but the WR score does. I myself went from 0% WR to 64% in 6 months. The audiogram/PTA? Unchanged. My doctor says the brain adapts.

Interesting. A few questions:

1) Did you notice the drastic improvements in clarity from the 0% to 64% or did it surprise you it went up quite that much?

2) I'm assuming this change was acute. OHCs/audiograms can dramatically recover acutely after inflammation is reduced so I imagine it could be the same with IHCs.

What was your cause?

 

[Diesel]

At the same rate, it's also possible that egos simply won out. Maybe I'm giving them too much credit and their leadership wanted to rock the world. I'm not sure, but there's a lot of Monday morning quarterbacking. People who don't even understand phase 1/2 are saying it's all a wash. Kind of a microcosm of some ot the motives they were working against.

All true. The only area where I will defend the company is that they are working against a populace that doesn't understand nor care about hidden hearing loss.

Some think hidden hearing loss is like witchcraft and just an excuse for a lack of audiograms.

Imagine a world where everyone, including me, was perfectly okay with incremental change. They wouldn't have been so pressured to push it deeper to light up audiograms.

At the same rate, it's also possible that egos simply won out. Maybe I'm giving them too much credit and their leadership wanted to rock the world. I'm not sure, but there's a lot of Monday morning quarterbacking. People who don't even understand phase 1/2 are saying it's all a wash. Kind of a microcosm of some ot the motives they were working against.

All of this said, this is mostly on them. They totally messed up the trial design. But in the future, people have to be cool with incremental change.

I agree with your points here. It also seems that those living with hearing loss are a tough crowd to work and engage with, for good reason. There's obviously already a ton of frustration with doctors and products, and general outlook.

On one end, extreme terms are swung around like "cure" and "scam." And the expectations are that the 1st product to be any good must be a sure fire hit on all measurements.

I don't know if this is unique to hearing loss, or elsewhere in the medical space.

I think if we compare where Frequency Therapeutics is at with FX-322 as a product, parallels can be drawn to their industries. In some cases, ultimately the iterative approach to product release became successful over products that ended up vaporware by trying to meet customer expectations in the first go-round.

From my own experience, I recall seeing a fully disassembled Tesla Roadster in '07 at a previous employer who developed gas engines. Based on the ideas of what a made a good gas engine/car, almost nobody at the time thought that niche sports car company could ever get into the mainstream. There was one engineer on the team that was a little more eccentric than the others, and acted as if this was the future, and gas cars might as well be dead. He was discounted, but I bet he is laughing today about it.

I suck at analogies, and I'm sure to get some responses from the haters on this post. But I think if Frequency Therapeutics now sees FX-322 as a single-dose product, it is in many ways their Roadster.

 

[Ed209]

So how do we explain that those that doubled their word score consistently showed an improvement from retesting at day 15, 30, 60, and 90? I would get if it was just a test at day 0 and then another at 90, how we could point to errors in the testing practice itself.

Furthermore, how do we then explain the follow up test scores 18 months later that still were significant by the same participants?

What happened if it wasn't drug?

Sorry... should include source:

Each day data point on the WR graph below represents a re-test...

https://patentscope.wipo.int/search/en/detail.jsf?docId=WO2020210388&tab=DRAWINGS

View attachment 44314

I'll play devil's advocate here. Phase I/II was primarily testing the safety of FX-322, and the cohorts were small, totalling 23 people. I'd like to have seen their baseline measurements at various intervals also. I don't think there's enough to go by to draw any meaningful conclusions.

 

[Zugzug]

So how do we explain that those that doubled their word score consistently showed an improvement from retesting at day 15, 30, 60, and 90? I would get if it was just a test at day 0 and then another at 90, how we could point to errors in the testing practice itself.

Furthermore, how do we then explain the follow up test scores 18 months later that still were significant by the same participants?

What happened if it wasn't drug?

Sorry... should include source:

Each day data point on the WR graph below represents a re-test...

https://patentscope.wipo.int/search/en/detail.jsf?docId=WO2020210388&tab=DRAWINGS

View attachment 44314

My intuition aligns with yours. But looking at this from a 100% neutral supposition and ignoring all of the other facts that we know, there is a huge problem in this study with data imbalance.

We have the super responders, which we all have many opinions on. But all of the placebo patients dealt with more of a ceiling effect. If we look at the percentage improvement graph that you posted, it all looks similar except for the few with lower starting points.

I'm just being honest. The groupwide statistics mean less to me than making sense of the super responders. This data, by chance, was destined for the treaters to beat the placebos. I know that's counterintuitive because as a rule of thumb, averages matter more than outliers. However, this is an unusual situation with an unusual thing being tested.

In hindsight, I find the lack of PTA in Phase 2a not surprising at all. There wasn't even groupwide dominance in PTA in the Phase 1/2 study, despite the data imbalance favoring the treaters.

We keep coming back to the same thing, but the autopsy has to be focused on determining how people double word scores. How is it humanly possibly to do that if you did the study honestly and the person wasn't treated with the drug?

This is painful to propose, but is there any chance that someone could have like a diagnosed focusing problem that leads to more variation in results? Maybe on top of hearing problems, they have cognitive disabilities that are undiagnosed? Trust me, I don't think that's what's going on here. But the data has to disprove this consideration.

 

[Diesel]

My intuition aligns with yours. But looking at this from a 100% neutral supposition and ignoring all of the other facts that we know, there is a huge problem in this study with data imbalance.

We have the super responders, which we all have many opinions on. But all of the placebo patients dealt with more of a ceiling effect. If we look at the percentage improvement graph that you posted, it all looks similar except for the few with lower starting points.

I'm just being honest. The groupwide statistics mean less to me than making sense of the super responders. This data, by chance, was destined for the treaters to beat the placebos. I know that's counterintuitive because as a rule of thumb, averages matter more than outliers. However, this is an unusual situation with an unusual thing being tested.

In hindsight, I find the lack of PTA in Phase 2a not surprising at all. There wasn't even groupwide dominance in PTA in the Phase 1/2 study, despite the data imbalance favoring the treaters.

We keep coming back to the same thing, but the autopsy has to be focused on determining how people double word scores. How is it humanly possibly to do that if you did the study honestly and the person wasn't treated with the drug?

This is painful to propose, but is there any chance that someone could have like a diagnosed focusing problem that leads to more variation in results? Maybe on top of hearing problems, they have cognitive disabilities that are undiagnosed? Trust me, I don't think that's what's going on here. But the data has to disprove this consideration.

Interesting point.

I wonder if since there isn't any real good measure to specifically, only isolate IHC performance. If the distribution of missing/damaged IHC played a role in the success of the responders.

So, I am suggesting, much like tinnitus, that there may be a subset of hearing loss patients at every level that happen to have an imbalance of IHC damage that makes them better, there's just no objective test to identify them.

 

[FGG]

My intuition aligns with yours. But looking at this from a 100% neutral supposition and ignoring all of the other facts that we know, there is a huge problem in this study with data imbalance.

We have the super responders, which we all have many opinions on. But all of the placebo patients dealt with more of a ceiling effect. If we look at the percentage improvement graph that you posted, it all looks similar except for the few with lower starting points.

I'm just being honest. The groupwide statistics mean less to me than making sense of the super responders. This data, by chance, was destined for the treaters to beat the placebos. I know that's counterintuitive because as a rule of thumb, averages matter more than outliers. However, this is an unusual situation with an unusual thing being tested.

In hindsight, I find the lack of PTA in Phase 2a not surprising at all. There wasn't even groupwide dominance in PTA in the Phase 1/2 study, despite the data imbalance favoring the treaters.

We keep coming back to the same thing, but the autopsy has to be focused on determining how people double word scores. How is it humanly possibly to do that if you did the study honestly and the person wasn't treated with the drug?

This is painful to propose, but is there any chance that someone could have like a diagnosed focusing problem that leads to more variation in results? Maybe on top of hearing problems, they have cognitive disabilities that are undiagnosed? Trust me, I don't think that's what's going on here. But the data has to disprove this consideration.

As someone with hearing loss, I can spend all of my focus and energy trying to hear the TV better and it makes minimal difference.

 

[Zugzug]

Interesting point.

I wonder if since there isn't any real good measure to specifically, only isolate IHC performance. If the distribution of missing/damaged IHC played a role in the success of the responders.

So, I am suggesting, much like tinnitus, that there may be a subset of hearing loss patients at every level that happen to have an imbalance of IHC damage that makes them better, there's just no objective test to identify them.

For the life of me, I just don't understand why they didn't have word scores as a stability factor. As has been pointed out here, given the concept that someone is likely to have a loss of both OHC and IHC if they have hearing impairment (and actually, OHC are more easily lost), it's hard to recruit people that specifically have word problems. It's so unforgivable to not have a really high standard for getting in.

In all likelihood, the incremental success of this drug will simply be in IHC improvement. Forget about OHC, tinnitus, hyperacusis for now. Stay in the word score lane and do the trials right. High, high standards of repeat performance stability for getting in. If it takes longer to recruit, tough shit. We all have to swallow it and stop whining.