Frequency Therapeutics — Hearing Loss Regeneration

[Whice]

If you cherry pick word scores to eliminate the ceiling effect seen by a few in Phase 1 entirely (which is what we are talking about here), this doesn't say anything about the drug's effectiveness over the larger population.

This might be the case but need a large Phase 3 — with all comers — to see who it benefits. And based on that the FDA makes recommendations.

 

[Aaron91]

I used to do research for a living at least part time. In smaller studies — like Phase 2s — you are not recruiting a lot of participants (typically). To find a signal in a smaller study — keep in mind Frequency Therapeutics found a signal in their Phase I study with speech recognition — you want to maximize effect size and limit your heterogeneity within the study population. They did this by — I think — picking subjects with the most to gain. My speech-in-noise is about 85%. Once you get above 90% or so you are approaching normal. If your speech-in-noise is 25 - 30%, you can see how much room there is to gain without approaching the envelope of normal (think asymptote) — like they saw in Phase I. I also wouldn't be surprised if they defined some composite endpoints too — like improved, audiogram + speech-in-noise. There is NOTHING wrong with this approach in a Phase 2 study. And it helps Frequency Therapeutics get an idea of where they're going to go in Phase 3 (which BTW the nurse recruiter told me to call back and reapply when it started) which will have a lot more subjects. The greater the number of subjects — as in a Phase 3 study — tend to limit heterogeneity between groups and makes the signal stand out from the noise.

Agreed. It's just that the bear thesis says if you look at the distributions of patients from the Phase 1 study, all the placebo patients had much less headroom to improve than those who received FX-322, hence why there were no sizeable gains and why it's unfair to compare. This was really unfortunate, but it means this is a legitimate point, albeit one I don't agree with for reasons me and others have already given. Just to help people visualise this, look what happens if you draw a red line at where some of us may suspect Frequency Therapeutics are drawing the maximum WR baseline score. There's only one placebo patient who falls under it. Yes, that patient worsened while all treatment patients improved, but the number is still too small to make any group-level inferences from.

So I would imagine in Phase 2 we will be seeing placebo patients with similar baselines to those who improved from FX-322 in Phase 1. It then really becomes a very simple question: do you believe placebo patients with WR scores of say, on average, 25/50, will show a group trend of improving more than say 10% and if so, how much do you think the treatment group will improve over and above 10%? This is where the bears come in again and say things like the cleaning of the round window by the gel is enough to do that. Quite the stretch, but I think to really shut the doubters up we're gonna have to see gains across the treatment group of much more than 10%, even though 10% still translates to QoL improvements.

For me, the pre-clinical work is just too strong to ignore, even if the Phase 1 data remains somewhat cloudy. I'm confident Frequency Therapeutics will deliver, although I have some reservations as to how the COVID-19 situation may have affected things. I have been wondering if people being home all the time means they're engaging in louder activities more frequently, such as listening to music or sitting in 8-hour zoom calls and how much that could affect newly developing hair cells. Then again, you have to offset that against all the other noisy activities people take part in a non-COVID-19 world: noisy restaurants, bars, clubbing, public transport etc.

 

[Diesel]

We also need to note that the two Phase 1bs for Age-Related and Severe hearing loss may also be rolled into an "all encompassing" Phase 3.

If so, the wide spectrum of SNHL conditions (mild-severe SSNHL/NIHL, mild-moderate ARHL) and probably 500+ participants internationally that can then be included in the Phase 3 would provide ample data to understand side effects and additional efficacy outcomes.

 

[Diesel]

Agreed. It's just that the bear thesis says if you look at the distributions of patients from the Phase 1 study, all the placebo patients had much less headroom to improve than those who received FX-322, hence why there were no sizeable gains and why it's unfair to compare. This was really unfortunate, but it means this is a legitimate point, albeit one I don't agree with for reasons me and others have already given. Just to help people visualise this, look what happens if you draw a red line at where some of us may suspect Frequency Therapeutics are drawing the maximum WR baseline score. There's only one placebo patient who falls under it. Yes, that patient worsened while all treatment patients improved, but the number is still too small to make any group-level inferences from.

View attachment 44158

So I would imagine in Phase 2 we will be seeing placebo patients with similar baselines to those who improved from FX-322 in Phase 1. It then really becomes a very simple question: do you believe placebo patients with WR scores of say, on average, 25/50, will show a group trend of improving more than say 10% and if so, how much do you think the treatment group will improve over and above 10%? This is where the bears come in again and say things like the cleaning of the round window by the gel is enough to do that. Quite the stretch, but I think to really shut the doubters up we're gonna have to see gains across the treatment group of much more than 10%, even though 10% still translates to QoL improvements.

For me, the pre-clinical work is just too strong to ignore, even if the Phase 1 data remains somewhat cloudy. I'm confident Frequency Therapeutics will deliver, although I have some reservations as to how the COVID-19 situation may have affected things. I have been wondering if people being home all the time means they're engaging in louder activities more frequently, such as listening to music or sitting in 8-hour zoom calls and how much that could affect newly developing hair cells. Then again, you have to offset that against all the other noisy activities people take part in a non-COVID-19 world: noisy restaurants, bars, clubbing, public transport etc.

If all participants were of a homogeneous group, all having similar moderate SNHL conditions, then the placebo group would clearly continue to have poor hearing conditions at day-90. Based on the 6 responders, those 24-72 that got FX-322 would likely show substantial gains when compared to placebo.

The placebo group in the trial already had pretty bad hearing if they recruited like they said the have. So, we might see a +/- 10% group average for hearing measures for the placebo group at 90-days.

For a homogeneous single dose group, the gains should look like 24x of those 6 original responders. So, the gain average should like 80%+ for word score etc. it should be distinct to placebo. The 2x and 4x should be even more distinct in average from placebo. Not sure how much they will be from each other.

The cleaning of the round window is absurd. Let's be honest. It's not like either drug acts like an all-purpose cleaner.

QoL measures will definitely be an additional differentiator for drug v. placebo. I can't speak for anyone with really poor hearing, but I feel like if you suddenly noticed more clarity in that ear, you would know.

I want to add... We have to keep in mind, that an entrant to the Phase 2a that looked like one of the 6 Phase 1/2 responders would probably make a great "non-responder" if they got randomized into the placebo group.

 

[Zugzug]

The way I see it, the group comparisons have to be stronger in Phase 2 than Phase 1. If they are applying filters to create homogeneity, there is a price to pay. Also, with larger sample sizes across all cohorts -- even just think about the fact that the total sample size in Phase 1 was n=23 (and really when you only count WR <= 90%, it was n=10) -- which is roughly the size of half of each of the 4 cohorts: Placebo, 1x, 2x, and 4x, where the combined sample size is 95.

The probability of imbalanced data is also much lower. I don't think the filter is some underhanded, cherry-picked thing. While my expectations are low for massive difference amongst cohorts, I am expecting something. My low expectations are that when comparing the ~23 Placebos to the ~72 FX-322 subjects, there should be obvious group dominance. Then when you break down the FX-322 subjects, there should be some stratification between groups, even if it's not statistically significant. In my opinion, the sample sizes are big enough to demonstrate this.

 

[Diesel]

The way I see it, the group comparisons have to be stronger in Phase 2 than Phase 1. If they are applying filters to create homogeneity, there is a price to pay. Also, with larger sample sizes across all cohorts -- even just think about the fact that the total sample size in Phase 1 was n=23 (and really when you only count WR <= 90%, it was n=10) -- which is roughly the size of half of each of the 4 cohorts: Placebo, 1x, 2x, and 4x, where the combined sample size is 95.

The probability of imbalanced data is also much lower. I don't think the filter is some underhanded, cherry-picked thing. While my expectations are low for massive difference amongst cohorts, I am expecting something. My low expectations are that when comparing the ~23 Placebos to the ~72 FX-322 subjects, there should be obvious group dominance. Then when you break down the FX-322 subjects, there should be some stratification between groups, even if it's not statistically significant. In my opinion, the sample sizes are big enough to demonstrate this.

I wouldn't think of the filtering as underhanded, it's a matter of conducting good science. They want to show that the starting point for all patients was as close to the same as possible. It creates another control, and minimizes variables as more doses are added. They're creating a "standard" / benchmark patient.

It's reasonable that the 72 FX-322 vs 24 Placebo hearing performance should be distinct.

I think when comparing the 1x, 2x, 4x groups, it may be a mixed bag of results at the group average. For example, if we know word scores average up to 35-40 with 1x dose, 2x or 4x might not do any better. But, WIN or specific audiogram bands may look slightly better for 2x and 4x compared to 1x.

I think the most notable comparisons will be between Placebo, 1x and 4x.

 

[FGG]

There are two points to unpack here. The first point is one on efficacy and FDA approval, your other point has to do commercialisation risk.

On your first point: the FDA will view it quite favourably, in fact. Remember, half the bear thesis (not put forward by bears, but by bulls with concerns) is that Frequency Therapeutics may not be able to successfully "cherry-pick" the population well enough to produce statistically significant results that will allow them to move on to Phase 3 in the first place. The FDA are not asking:

"Does this drug treat hearing loss of any and every kind and can that be shown at a statistically significant level in all groups"?

Instead, they are asking:

"Does this drug treat hearing loss for at least some kind of population at a statistically significant level in any group"?

Just because there is an unmet need in all groups doesn't mean a drug won't be approved only because it still meets some kind of unmet need.

There's also a wider point to be made here and that is people can get really, really bogged down in the maths sometimes, especially the bears. Just because there were 6/15 "responders" doesn't mean those 9 other patients didn't experience some kind of quality of life improvement that wasn't necessarily reflected in WR/WIN/audiogram tests. As Frequency Therapeutics have said in their KOL presentation, just a 10% improvement can translate to a huge quality of life improvement. A case in point here is Sarepta and their DMD drug Exondys 51 (thanks to @FGG for bringing this to my attention). If I remember correctly, the drug could only be used in 14% of cases of DMD and if you looked at the maths and trial design even their results could have been described as somewhat spurious, especially compared to Frequency Therapeutics. And yet when you looked at patient testimonies and how the drug translated to real life, you had examples such as the one @FGG described of a kid that should have been in a wheelchair but was skiing! So all this to say that quality of life improvement as a measure is a real thing and that is, I suspect, one of the reasons why a secondary endpoint is the Quality of Life questionnaire just to cover all bases. I'm going to go out on a limb and say that in the event we do not see statistically significant improvements in the mild hearing loss category, that does not necessarily mean we won't see a statistically significant QoL improvement in that same category.

All this brings me to your second point about commercialisation risk. I think I speak for a lot of us here when I say that just because some of us may not fall within the group parameters that showed statistical significance, that doesn't mean we are not willing to try this drug when it gets to market. We shouldn't underestimate how far a person will go when they are desperate, particularly when their health is on the line. This thread is a case in point: over 500 pages of debate and research. I come back to a quote I quite like: "a healthy person has a thousand wishes, a sick person only one". Assuming this drug costs something in the region of a hearing aid (which only has a 25% uptake rate due to stigma/aesthetics/crudeness of technology/ineffectiveness with regards to clarity), I think most people with hearing loss of any kind would be willing to give that a shot.

I quite liked @Diesel's take on this. I think it's very plausible that a selection criterion could have been tinnitus in the treated ear and not the other. How difficult that would have been to recruit for though I have no idea, especially if we also assume the other boxes they would have wanted ticked, such as asymmetric hearing loss etc. I suppose it's possible though that if you have asymmetric hearing loss, chances are that if you unilateral tinnitus, it's more likely to be in the bad ear than the good ear.

More on the Sarepta story, there were 6 minute walk tests that had not previously ever been achieved in any historical controls at the tested age (DMD is a degenerative condition) and the Bear thesis was they just tried harder because they believed.

I loved Ricky Tsang's response to that (he tragically passed before the drug was released). He said something like "when I was that age I wanted to walk so bad that I deliberately froze my hands to a pole in winter to force myself to stand to try to strengthen my muscles. The idea that any of us sufferers just weren't trying hard enough before is absurd."

Anyway, I just don't think it's possible for the placebo effect to ever double word scores. Believe me, people with bad hearing spend all their mental energy trying to hear better.

 

[Zugzug]

Anyway, I just don't think it's possible for the placebo effect to ever double word scores. Believe me, people with bad hearing spend all their mental energy trying to hear better.

The p-values that I calculated for these people were insanely small (and doubled under a two-sided scheme), under the assumption of the null hypothesis.

Something I look for, carefully, with these methods are the assumptions built in. Also, the errors in approximations. With this being said, even if we say word tests don't follow a Binomial Distribution perfectly and that applying the Central limit theorem to n=50 word tests and accounting for error in approximating the variances, it certainly doesn't make up for a p-value of p=0.00000047.

I would basically go as far as say that if the drug did nothing for this patient, there would have to be a mistake made by the experimenters, either with regards to the randomization or somehow having the patient learn the list. Yet this happened with 3 ears that had similar p-values. The Placebos look exactly like ears who were untreated.

Somewhat off topic, but I think I understand why everyone is so insistent on differentiating stem cells from progenitors. It's not just semantics; it's because progenitors start off closer to efficient and safe activation. I think the people who don't know the science don't know this, but the people who do think it makes the PCA process seem more iffy (by calling it stem cell activation).

 

[Diesel]

Let's be honest. It seems most here that have done their homework on the primary outcomes are pointing to positive/favorable outcomes on WR + WIN on the Phase 2A.

The standard audiogram isn't going to look groundbreaking below 4 kHz, and that is something the bears will continue to focus on. The PTA4 score probably isn't going to budge.

Setting expectations.

 

[FGG]

Let's be honest. It seems most here that have done their homework on the primary outcomes are pointing to positive/favorable outcomes on WR + WIN on the Phase 2A.

The standard audiogram isn't going to look groundbreaking below 4 kHz, and that is something the bears will continue to focus on. The PTA4 score probably isn't going to budge.

Setting expectations.

I have yet to read a Bear (on Twitter or anywhere else) that has a good understanding of hearing anatomy and physiology, so a lot of completely incorrect conclusions are made (e.g., the EHF audiogram is outside the "speech banana" therefore it is not possible word scores increased without large audiogram changes.)

The Bulls I read, in contrast, are really well versed in it.

 

[Kendra]

Very low chances of 5 dB causing your tinnitus...

I had a sudden 5 dB drop and it caused tinnitus for me. So it is possible.

 

[Diesel]

I have yet to read a Bear (on Twitter or anywhere else) that has a good understanding of hearing anatomy and physiology, so a lot of completely incorrect conclusions are made (e.g., the EHF audiogram is outside the "speech banana" therefore it is not possible word scores increased without large audiogram changes.)

The Bulls I read, in contrast, are really well versed in it.

Good point.

 

[Ed209]

Have you checked the pitch of your own tinnitus using any kind of tone generating software? If so, does it sound similar to a 4 kHz tone, or is it much higher? My hunch, as many others suspect too, is that the pitch of your tinnitus would correspond to the area where you need the treatment.

Most people are not very good at pitch-matching. It's common to see octave errors and other miscalculations amongst those who are not musically trained. What some believe is a match at 4 kHz is actually closer to 8 kHz or 2 kHz, believe it or not.

 

[FGG]

Most people are not very good at pitch-matching. It's common to see octave errors and other miscalculations amongst those who are not musically trained. What some believe is a match at 4 kHz is actually closer to 8 kHz or 2 kHz, believe it or not.

Agree. @HootOwl is exceptionally good at this and tried to help me pitch match but I fall into the "most people" here.

 

[Keith Handy]

Most people are not very good at pitch-matching. It's common to see octave errors and other miscalculations amongst those who are not musically trained. What some believe is a match at 4 kHz is actually closer to 8 kHz or 2 kHz, believe it or not.

How would we know this without another person being able to hear the tinnitus sound?

 

[Ed209]

How would we know this without another person being able to hear the tinnitus sound?

Because we can use two sounds that everyone can hear to assess the candidates skill level. I do this for a living, including ear training, and most people are really bad at it to begin with.

However, my previous post wasn't based on my experiences, it's mentioned in some studies.

The testing protocol involved testing for hearing thresholds, loudness matches, and pitch matches. These three different tasks were "intermingled" according to the following procedural algorithm. At each test frequency, the hearing threshold was obtained to the nearest 1 dB, followed by a loudness match to the nearest 1 dB. An adaptive two-alternative forced-choice (2AFC) procedure was used to present pairs of tones from which the individual chose the one that was "closest in pitch" to his or her tinnitus. Except for octave-confusion testing (see paragraph after next), tone pairs for 2AFC pitch-matching were always separated in frequency by 1/3 octave.

Testing started at 3,180 Hz, followed by testing at 4,000 Hz. When loudness matches had been obtained at these two frequencies, the 2AFC procedure was used to identify which of these two tones was closest in pitch to the person's tinnitus. (The computer screens presented to subjects during pitch-match testing with the Binary method are shown in Figure 1.) This selection bracketed the tinnitus pitch to either above 4,000 Hz or below 3,180 Hz (binary bracketing). The computer algorithm then directed testing to, respectively, an octave above or an octave below the initial pair of test frequencies. A new pair of loudness matches was obtained, and the 2AFC procedure again identified whether the tinnitus pitch was above or below the pair of frequencies. Testing occurred in this fashion until the tinnitus pitch was bracketed to within 1 octave. Testing then occurred for tone pairs in the octave-frequency range, starting with the lowest frequency and progressing toward the highest.

When a final pitch match had been selected, "octave-confusion" testing was done. Octave confusion is thought to be a source of error when patients incorrectly identify their tinnitus pitch by selecting a frequency that is an octave above or an octave below their actual tinnitus pitch. Graham and Newby first reported this phenomenon, and testing for this error was later advocated as an integral part of any tinnitus-testing battery. For the present study, octave-confusion testing was done an octave above and an octave below the pitch match whenever such frequencies were available. Completion of the octave-confusion task identified the "octave-confirmed" pitch match, which was obtained only one time with the Binary procedure.

https://www.rehab.research.va.gov/jour/04/41/2/henry.html​

This is where I originally read about it:

General Comments about Pitch Matching

Although the pitch of tinnitus may seem to be one of the easiest attributes of tinnitus to quantify, in fact, there are some traps for the unwary. Inexperienced subjects may tend to confuse pitch and loudness, causing them to reject a tentative frequency match for the tinnitus on the basis of loudness differences or to accept an inappropriate match because its loudness is equivalent to the tinnitus (Vernon and Meikle, 1981). It is important, therefore, during pitch matching to adjust the loudness of comparison stimuli very carefully to make them equal to the loudness of the tinnitus (Fowler, 1940;Vernon & Meikle, 1981).

Another potential source of error is the possibility of "octave confusion," in which the sub- ject may at first select tones an octave above or below the frequency that they later identify as the "best" tinnitus pitch match. Octave confusions occur quite commonly even for normal-hearing people making comparisons between external tones. The fact that pitch matches for tinnitus might be affected by octave confusion came to the attention of clinicians working with tinnitus patients in the 1960s (Graham and Newby, 1962) and was emphasized in the 1980s by Vernon et al (1980a). Subsequently, the tinnitus testing procedures recommended in 1981 by the CIBA symposium on tinnitus included a test for octave confusion as an integral part of the test battery (Evered and Lawrenson, 1981).

https://www.audiology.org/sites/default/files/journal/JAAA_11_03_03.pdf​

 

[spacehunter]

I had a sudden 5 dB drop and it caused tinnitus for me. So it is possible.

So your lowest "dip" is 5dB? It is just very surprising to me, that such a low dip can cause tinnitus. The only explanation I could have is that hidden with the loss in the audiograms is a much more significant synaptic hearing loss. Or maybe a very sharp dip, which could not be resolved within an audiogram (so basically the audiogram just shows some side part but not the absolute bottom of the pitch).

 

[Whice]

Let's be honest. It seems most here that have done their homework on the primary outcomes are pointing to positive/favorable outcomes on WR + WIN on the Phase 2A.

The standard audiogram isn't going to look groundbreaking below 4 kHz, and that is something the bears will continue to focus on. The PTA4 score probably isn't going to budge.

IF this is the case, it's fantastic for all of us — the science works — and we move forward to Phase 3.

 

[Aaron91]

So your lowest "dip" is 5dB? It is just very surprising to me, that such a low dip can cause tinnitus. The only explanation I could have is that hidden with the loss in the audiograms is a much more significant synaptic hearing loss. Or maybe a very sharp dip, which could not be resolved within an audiogram (so basically the audiogram just shows some side part but not the absolute bottom of the pitch).

I vaguely recall reading something about people with predictive brains being more likely to develop tinnitus, I think there may even be a thread on here somewhere about it - hopefully someone who knows more can chime in.

Tinnitus could, of course, be caused by any failure along the auditory pathway, synapses included, but I think it's perfectly reasonable to suggest that some people's brains will respond differently from loss of input to others'. Even though supposedly 85% of all tinnitus patients have some hearing loss, there are cases of severely deaf patients who do not have tinnitus. Equally, there are members on here who have perfect audiograms, who do not struggle with hearing in noise, who have a great sense of directionality and yet still have tinnitus.

 

[FGG]

I vaguely recall reading something about people with predictive brains being more likely to develop tinnitus, I think there may even be a thread on here somewhere about it - hopefully someone who knows more can chime in.

Tinnitus could, of course, be caused by any failure along the auditory pathway, synapses included, but I think it's perfectly reasonable to suggest that some people's brains will respond differently from loss of input to others'. Even though supposedly 85% of all tinnitus patients have some hearing loss, there are cases of severely deaf patients who do not have tinnitus. Equally, there are members on here who have perfect audiograms, who do not struggle with hearing in noise, who have a great sense of directionality and yet still have tinnitus.

Both Will Sedley and Thanos Tzounopoulos have described this idea and it's the model that I believe best fits. For the same level of hearing loss, some people are just more prone to getting tinnitus. Some people have very predictive brains that are extremely sensitive to even minor differences in what the brain "expects to hear" and what is actually heard. For almost anyone with severe (especially sudden) losses though, their brains will respond with tinnitus as it attempts to reconcile these differences in maladaptive ways (ion channel mediated plasticity changes).

This mismatch occurs separately in the visual system, too (the "color changing dress" is a famous example of the brain's prediction changing actual perception). It might be interesting at some point to see if there is a correlation between how the dress looks and tinnitus in minor hearing loss cases. Maybe a thread on this would be interesting but it's possible the visual system genes are different enough.

 

[Paulmanlike]

Good point.

I thought you were "taking a break" until readout of Phase 2A.

You just can't help yourself, can you Diesel....

 

[Aaron91]

In Aaron's words copy pasting this from another chat thread we had:

"1) Cell fate. Unlike stem cells, progenitor cells have a pre-determined cell "fate". If you watch the video, you learn that half of Will McLean's research was simply figuring out which progenitor cell becomes a hair cell (the LGR5 cell). Unlike a stem cell, a progenitor cell CANNOT become any other kind of cell.

2) Stem cell otology approaches in previous attempts over the years have failed because they did not have the right gene expression. The hair cells did not assimilate into colonies or "organise" themselves. This is what I mean when I say there is no "retrofitting" in the case of progenitor cells. If this was a pure stem cell approach I would not be touching this thing with a large bar pole.

3) Delivery. Progenitor cells are already there in the cochlea. It's really, really difficult to deliver a stem cell to the cochlea even if you could make it integrate.

4) Safety. Stem cells always run the risk of uncontrolled proliferation and therefore cancer. All the pre-clinical data suggests that progenitor cells have a self-limitation to their proliferation, in line with how they proliferate in utero during development. Once there are enough hair cells, the regeneration stops. I even remember reading somewhere that they gave FX-322 to healthy cochlea mice and fuck all happened. It's not as if the mice got super-sonic hearing or something. Literally nothing happened."​

TL;DR:

 

[FGG]

I thought you were "taking a break" until readout of Phase 2A.

You just can't help yourself, can you Diesel....

The "peak" of analysis has a downslope, though. No?

 

[Diesel]

I thought you were "taking a break" until readout of Phase 2A.

You just can't help yourself, can you Diesel....

 

[Cernuto]

Some of us have tinnitus with multiple pitches in relatively close proximity and intensity. This makes it extremely difficult, if not impossible to discriminate the exact frequency for each pitch.

One thing I discovered when I had my past two audiograms performed is that the tones they test with warble in pitch. I imagine this modulation in pitch is designed to stimulate any hair cells in proximity to the median frequency. To me, this doesn't seem like a good way to determine if you have hair cell damage. It seems like a terrible way, actually.

Thinking out 'louder" here, but to determine hair cell damage one way would be to frequency sweep through multiple times and accurately produce a continuous sweeping tone in equal dB across the entire hearing spectrum for each sweep. Performing the test from the threshold of not audible, slowly increasing in dB through each iteration of sweeps, taking note of where the subject is able to detect the test frequency. Then remove the detected pitches from the next sweep iteration that is slightly louder. Perform the iterations until a sufficient data is collected and there's your true audiogram.

 

[Zugzug]

View attachment 44160

"They." Is that what we're calling our tinnitus pitches these days?

 

[tommyd87]

I have yet to read a Bear (on Twitter or anywhere else) that has a good understanding of hearing anatomy and physiology, so a lot of completely incorrect conclusions are made (e.g., the EHF audiogram is outside the "speech banana" therefore it is not possible word scores increased without large audiogram changes.)

The Bulls I read, in contrast, are really well versed in it.

The problem is however that many audiologists also actually believe that the speech banana basically is the only relevant criteria when it comes to one's ability to hear. Hence until there is greater evidence demonstrated that this is not true (many already on here know this to be the case), a lot of people will buy this and audiologists also will continue to push this.

I hope this clarifies your concerns:

View attachment 44154

That picture doesn't show all causes of conductive hearing loss though...

Let's be honest. It seems most here that have done their homework on the primary outcomes are pointing to positive/favorable outcomes on WR + WIN on the Phase 2A.

The standard audiogram isn't going to look groundbreaking below 4 kHz, and that is something the bears will continue to focus on. The PTA4 score probably isn't going to budge.

Setting expectations.

It's probable that FX-322 won't make the gains you mentioned above in the current trial due to the current dosing method/formulation. However it is I think also foolish to conclusively rule anything out when we still haven't received the information as yet or been presented with benefits of multiple doses.

 

[kiki]

If 4 kHz improves by 3 dB with a single injection, does that translate to a 10 dB improvement with 4 injections? Will you get a bigger effect if you inject 4 times and a further 4 times after a while?

Then the cochlea paintings they first showed that worked up to 4 kHz would be true.

At first, they seemed to want to do it in a single injection.

 

[Tezcatlipoca]

I think we're asking too much of a first formulation. I have faith that future generations of FX-322 will get better and better, but what I personally want out of this one is just some relief.

If the first generation can get me to the 'not a problem tinnitus" then I will be perfectly content waiting for the next generations.

 

[dd314]

Both Will Sedley and Thanos Tzounopoulos have described this idea and it's the model that I believe best fits. For the same level of hearing loss, some people are just more prone to getting tinnitus. Some people have very predictive brains that are extremely sensitive to even minor differences in what the brain "expects to hear" and what is actually heard. For almost anyone with severe (especially sudden) losses though, their brains will respond with tinnitus as it attempts to reconcile these differences in maladaptive ways (ion channel mediated plasticity changes).

This mismatch occurs separately in the visual system, too (the "color changing dress" is a famous example of the brain's prediction changing actual perception). It might be interesting at some point to see if there is a correlation between how the dress looks and tinnitus in minor hearing loss cases. Maybe a thread on this would be interesting but it's possible the visual system genes are different enough.

The model with the fewest assumptions is probably the correct one, and that one requires many. What accounts for such variability in the tone and intensity of this gap-filling, if hearing loss is permanent? I suspect the signal generation is coming from the cochlea itself, as in the case of otoacoustic emissions or hair cells that are stuck in the "on" position. Obviously damage can occur at any point along the auditory pathway, but for a condition so common, there are likely more common points of failure.

This was discussed already, but the biggest indicator is that in a study on 151 patients that had their auditory nerve cut, 95.4% of patients achieved a favorable outcome. The author of this particular study is now deceased, but if you trust the data it does indicate the signal generation coming from the cochlea. Its entirely possible that all of the neural correlates observed in fNIRS, fMRI, EEG scans of people's brains are just correlates of the psycho-adaptations of tinnitus.