Frequency Therapeutics — Hearing Loss Regeneration
- Thread starter RB2014
- Start date
MemberI don't think they will lie. A lot of well known people left their old jobs to join Frequency Therapeutics believing FX-322 works. I still believe in it because Phase 1b for both trials showed improvement in word scores which has never been done before.
The reason why Phase 2a failed was multiple dosing of FX-322 weekly, as McLean stated, it takes 12 days to regrow hair cells.
- Aug 5, 2019
- 1,852
- Tinnitus Since
- 05/2019
- Cause of Tinnitus
- Autoimmune hyperacusis from Sjogren's Syndrome
To be honest, and I don't say this to sound conceited but rather to add substantively, I don't know why they didn't calculate the p-values like I did. I actually wonder if they just sort of said "well, Thornton and Raffin margin of errors are used all the time. Let's go to the table and find the values." Granted, there are some (definitely not that farfetched) assumptions in the modeling approach, but those p-values help to illuminate the rarity of the situation. Besides, if you're going to use Thornton Raffin confidence intervals anyways, it's an immediately logical thing to calculate the p-values under the same setting. In other words, a bear could beat up the use of a Binomial Model, but whatever their critique is, it would apply to the confidence intervals and the p-values. If you're going to live by these clinical guidelines, at least add more evidence for your arguments.
I don't think most people understand just how ridiculous it is for the same person with the same hearing to go from 40% to 78%. And then you had three people like this. Where you and I agree is that there's not a doubt in my mind that there's no way those 3 were standard situations of getting lucky. At least 1 person was helped by the drug or at least 1 person had deflated scores at baseline. Even as a bear, I accept that.
A natural question is then, what's the probability that the 3 superstars landed in the FX-322 group?
In a sample of n=23, with 15 in FX-322 and 8 in placebo, it ends up being 25.7%. That's not that low. In other words, if we worked under the assumption that all of the 3 superstars were baseline fakers, there's still a 25.7% chance that they all landed in the FX-322 group.
Sadly, even Phase 1/2 can be questioned, but the questioning is not "was it chance?"
But FREQ has reviewed those trials, as has the FDA. Surely if there was such a discrepancy after the unblinding, either or both of them would have caught them.
In total agreement. It's not placebo and not chance. But we can't *totally* rule out shenanigans for reasons I can't even begin to speculate on (since depressed word scores were not a Phase 1 requirement).
This won't be enough for the FDA though (you and I agree on this). Frequency Therapeutics can't just say "our company is made up of people with good reputations and there is no obvious motive at all for Phase 1 participants to have deflated their baseline, therefore we should go to pivotal" (even with good age-related and/or severe group results).
The hearing effect needs to be reproduced in a larger sample. They need a better designed Phase 2, hopefully with a thorough analysis of why they got the results they did.
That's a good point. They probably did go back and look at the records of Phase 1 participants. If they didn't before, they certainly would have in the last week before releasing data.
- Aug 5, 2019
- 1,852
- Tinnitus Since
- 05/2019
- Cause of Tinnitus
- Autoimmune hyperacusis from Sjogren's Syndrome
That's a good thought. However, a lot of it reduces back to my 25.7% calculation of the 3 superstars landing in the FX-322 group.
The investigation of the Phase 2a trials is quite different because presumably, they hope that after unblinding themselves, they can see superstar individuals in the placebo group. Then it makes sense to wonder how they became superstars under a placebo.
The issue is that there are no superstar placebos in Phase 1/2 so maybe they think it's conceivable that the superstars were helped by the drug.
I guess for Phase 1/2, they could try to look for consistent word scores >=6 months before the trial, at screening, and at baseline, but you would think if they were tracking it, they would have done a better job with Phase 2a.
What a mess. Seems like what they should have done, in hindsight, is required longer stability than >=6 months. Maybe make the time frame so long that it's unreasonable to believe the person took a faulty word test years ago, anticipating that they would be on the verge of landing in a clinical trial.
Besides, the superstars in Phase 1/2 had disease durations of 21.05, 5.23, and 1.70 years, respectively.
It seems like in all likelihood, a lot of the Phase 2a participants easily exceeded the >=6 month stability. That makes it that much more incriminating to have a trial bias since they mostly had to just do a better job of comparing screening word scores to word scores from years ago when they weren't lying.
I almost wish there could be an independent person looking for discrepancies between screen and baseline. The problem though is that however unlikely, confidence intervals only describe a high likelihood of being in them. One can break them by chance, by assumption. Yeah, deep down, we're all thinking this can't happen with hearing, but they have to operate under the assumption that it can. It's hard to remove someone from a trial just because they broke a confidence interval. Maybe it was a (assumed) 5% chance that occurred.
Yeah, assuming they did go back and review Phase 1 records carefully (I think that is a totally reasonable assumption), then Frequency Therapeutics did not find the same discrepancy there or they omitted it but instinct says Frequency Therapeutics is a company with integrity. These are long time industry veterans. I would be pretty shocked if it was fraud.
I'm going with there is a hearing effect for sure if that's the case (and there were definitely responders in Phase 1b for word score as well. Btw, since the subjects were mild to severe in this study, it would be really interesting to compare between individuals here).
- Aug 5, 2019
- 1,852
- Tinnitus Since
- 05/2019
- Cause of Tinnitus
- Autoimmune hyperacusis from Sjogren's Syndrome
I'm still not clear on the entry process. The clinical trials say >=6 month stability, no more than 10 dB changes at one frequency or no more than 5 dB changes at two consecutive frequencies.
But there's nothing about word scores. How does one even cheat? Do they really not use the Thornton and Raffin confidence intervals? Maybe even raise the alpha level to be extra rigorous. For example, if you had 80% confidence intervals comparing their >=6 month WR to their screening WR, it would be easier for there to be statistically significant differences. If there are, they get excluded from the trial. Tough luck, at least it's a rigorous standard. If someone gets falsely excluded, big deal.
Is there a sheer recruiting numbers problem with this approach? Do they not know how to calculate the 80% confidence intervals from the Thornton and Raffin paper? Maybe I'm ignorant here, but I don't see this as the world's hardest problem.
There's no way Frequency Therapeutics are frauds, but they definitely didn't just "get burned" by cheaters.
Street Novelist
Member
- Apr 2, 2018
- 290
- Tinnitus Since
- February 2018
- Cause of Tinnitus
- Listening to loud music through headphones
I agree they didn't go back through the medical records of Phase 1 at the time because, if they thought to do that, they would have had better measures in place and they could have prevented the "discrepancy" problem in Phase 2a.
But I think it's a really good assumption they did it in the last few weeks when they had a Phase 2a "oh shit" moment.
I think an independent person looking for these discrepancies between screen and baseline would have prevented this. I had the thought, maybe more blinding is not always better.
Correct me if I'm wrong, but did I see somewhere that the severe group was only double blinded? Vs quadruple?
Actually, was Phase 1 also quadruple blinded?
If the two ongoing trials for age-related and severe hearing loss show improvements in word scores, it will be definite that FX-322 works and that weekly dosing of FX-322 kills the regrowing process and they need to space out the dosing even longer e.g. monthly.
We have had two Phase 1b trials showing improvements in word scores with a single dose of FX-322.
- Aug 5, 2019
- 1,852
- Tinnitus Since
- 05/2019
- Cause of Tinnitus
- Autoimmune hyperacusis from Sjogren's Syndrome
To anyone curious, here is a very good video on the WR test from an audiologist:
Captions are available. Here's a summary of what it talks about:
For those of you wondering if effort or a true placebo can be the reason why someone genuinely improves dramatically, it doesn't make any sense since the words are played at a perfectly comfortable level. If you don't hear a word, it's because your inner hair cells or synapses are broken. Nothing can be done for it.
Of course, this doesn't change anything. We still have the problem of "Responded to the drug or deflated baseline scores without the drug helping," but it's interesting at least.
Captions are available. Here's a summary of what it talks about:
- WR has nothing to do with volume. In fact, the test is given to see if one is a good hearing aid candidate. The volume of the words is chosen in such a way that the patient will have no issue hearing the words in volume. Hearing aids won't help people with this problem because hearing aids only amplify (OHC).
- Some incompetent audiologists may have used live words instead of recorded words. The test-retest stability is only reliable if it's calibrated the same both times.
- 50 questions per test is more reliable than 10 or 25 questions (duh)
For those of you wondering if effort or a true placebo can be the reason why someone genuinely improves dramatically, it doesn't make any sense since the words are played at a perfectly comfortable level. If you don't hear a word, it's because your inner hair cells or synapses are broken. Nothing can be done for it.
Of course, this doesn't change anything. We still have the problem of "Responded to the drug or deflated baseline scores without the drug helping," but it's interesting at least.
- Aug 5, 2019
- 1,852
- Tinnitus Since
- 05/2019
- Cause of Tinnitus
- Autoimmune hyperacusis from Sjogren's Syndrome
- Aug 5, 2019
- 1,852
- Tinnitus Since
- 05/2019
- Cause of Tinnitus
- Autoimmune hyperacusis from Sjogren's Syndrome
Haha, I'm not sure, that's why I didn't comment. It says the masking is quadruple, but the titles are "Double-blind." I would think it would be quadruple blind, but I'm not sure.
- May 27, 2020
- 556
- Tinnitus Since
- 2007
- Cause of Tinnitus
- Loud music/headphones/concerts - Hyperacusis from motorbike
I have spoken to a patient from the Phase 2 trial. More below.
I think many of you by now know me as a member that often does deep dives and goes down rabbitholes. Ever since I first read the pre-clinical work, there was no doubt in my mind this drug was the real deal, especially after the Phase 1/2 study. For me, it was simply a question of optimising delivery and dosage. However, for the last 72 hours, I have been questioning everything I previously believed to be true. It has been a very unnerving experience to say the least, as I'm sure it has been for others. So I decided to go back and read all the preclinical work again, as well as other literature. This time I wasn't satisfied, because I realised peer-reviewed papers were no longer good enough in face of the current situation. I had planned on taking a break from the forum, but I felt the only way I could settle this one way or another, at least in my mind, was to find someone who took part in the trial.
I can confidently say that I have tracked down two patients from the clinical trial. I am in touch with one of them and waiting to hear back from the other. I went in prepared with questions and made a lot of notes from the conversation with the first patient. I shall summarise them below.
Patient X Background
Age: 55-65 (I'm avoiding giving the exact number as not to dox the patient)
Presentation: Unilateral hearing loss and bilateral tinnitus
Cause: Unknown
Treatment history: 30 injections of ITT steroids and HBO treatment
Time of onset: 2 years
Hearing aid: Yes
Patient X - FX-322 trial
Baseline WR score: 40%
Post-treatment WR score: 50%
Post-treatment audiogram: 10 dB gains in three frequency bands - two in the extended audiogram and one in the standard audiogram
Post-treatment tinnitus: 90% improvement in both ears (despite only one ear being treated) after 3.5 months
Summary
As you will infer from the above, the patient had a sudden onset of hearing loss in a single ear. This was preceded by tinnitus two weeks before onset. Following 30 injections of ITT steroids and HBO treatment, they were able to go from, for all intents and purposes, deaf to a 40% WR score baseline in the affected ear. They describe their hearing pre ITT steroids and HBO treatment as garbled and sounding like chipmunks. Although this helped regain some hearing, they still required the use of hearing aids moving forward. This was their presentation going into the trial.
Following participation in the trial, the patient saw a 10 point gain in their WR score from baseline and 10 dB gains across three different frequency bands. They started to notice their hearing improvements one month after their last injection. Interestingly, the reduction in bilateral tinnitus, came 3.5 months after the last injection - outside the scope of the interim study. They describe their stress levels induced by tinnitus as having been dramatically reduced and that their general hearing in noisier environments is "better but not perfect". Examples of sounds they say are now easier to discern include the chime on the dryer when the laundry is dry, hearing a waitress's voice in a restaurant (higher pitched female voices), soft and squeaky toys. They also have a better sense of directionality. They still, however, require the use of a hearing aid.
The patient remains blinded and unaware as to whether they received placebo or FX-322 at this moment in time, although they have kindly agreed to letting me know once they know. Similar to another anecdotal experience posted here, they described a burning sensation during 2 of the four injections. They experienced no other discomfort associated with the regularity of the injections.
When asked whether they were surprised the trial had failed, the patient said they were "shocked".
Conclusion
I'm sure some of you have questions as to how I was able to track down a patient from the trial. The answer is I can never know for sure whether this really is a patient from the trial, but what I can confirm, for whatever my word is worth on here, is that Patient X has a publicly documented history of hearing loss and a publicly documented declaration of having applied to (and entered) the trial. I don't wish to say much more than that, but I think if you read between the lines here you'll be able to figure out what I've done and I'm saying. I'm already mindful of the fact that this person has been bombarded with a lot of questions from a complete stranger (yours truly), so if there is something you want to ask, I think the best course of action would be for me to get back in touch with them directly.
I'm about to sleep so won't be able to answer any questions until tomorrow at the earliest, although I still intend to take a break for a week or so from the forum (although I am keeping general tabs on discussions). I will report back if and when I hear from the second patient.
I think many of you by now know me as a member that often does deep dives and goes down rabbitholes. Ever since I first read the pre-clinical work, there was no doubt in my mind this drug was the real deal, especially after the Phase 1/2 study. For me, it was simply a question of optimising delivery and dosage. However, for the last 72 hours, I have been questioning everything I previously believed to be true. It has been a very unnerving experience to say the least, as I'm sure it has been for others. So I decided to go back and read all the preclinical work again, as well as other literature. This time I wasn't satisfied, because I realised peer-reviewed papers were no longer good enough in face of the current situation. I had planned on taking a break from the forum, but I felt the only way I could settle this one way or another, at least in my mind, was to find someone who took part in the trial.
I can confidently say that I have tracked down two patients from the clinical trial. I am in touch with one of them and waiting to hear back from the other. I went in prepared with questions and made a lot of notes from the conversation with the first patient. I shall summarise them below.
Patient X Background
Age: 55-65 (I'm avoiding giving the exact number as not to dox the patient)
Presentation: Unilateral hearing loss and bilateral tinnitus
Cause: Unknown
Treatment history: 30 injections of ITT steroids and HBO treatment
Time of onset: 2 years
Hearing aid: Yes
Patient X - FX-322 trial
Baseline WR score: 40%
Post-treatment WR score: 50%
Post-treatment audiogram: 10 dB gains in three frequency bands - two in the extended audiogram and one in the standard audiogram
Post-treatment tinnitus: 90% improvement in both ears (despite only one ear being treated) after 3.5 months
Summary
As you will infer from the above, the patient had a sudden onset of hearing loss in a single ear. This was preceded by tinnitus two weeks before onset. Following 30 injections of ITT steroids and HBO treatment, they were able to go from, for all intents and purposes, deaf to a 40% WR score baseline in the affected ear. They describe their hearing pre ITT steroids and HBO treatment as garbled and sounding like chipmunks. Although this helped regain some hearing, they still required the use of hearing aids moving forward. This was their presentation going into the trial.
Following participation in the trial, the patient saw a 10 point gain in their WR score from baseline and 10 dB gains across three different frequency bands. They started to notice their hearing improvements one month after their last injection. Interestingly, the reduction in bilateral tinnitus, came 3.5 months after the last injection - outside the scope of the interim study. They describe their stress levels induced by tinnitus as having been dramatically reduced and that their general hearing in noisier environments is "better but not perfect". Examples of sounds they say are now easier to discern include the chime on the dryer when the laundry is dry, hearing a waitress's voice in a restaurant (higher pitched female voices), soft and squeaky toys. They also have a better sense of directionality. They still, however, require the use of a hearing aid.
The patient remains blinded and unaware as to whether they received placebo or FX-322 at this moment in time, although they have kindly agreed to letting me know once they know. Similar to another anecdotal experience posted here, they described a burning sensation during 2 of the four injections. They experienced no other discomfort associated with the regularity of the injections.
When asked whether they were surprised the trial had failed, the patient said they were "shocked".
Conclusion
I'm sure some of you have questions as to how I was able to track down a patient from the trial. The answer is I can never know for sure whether this really is a patient from the trial, but what I can confirm, for whatever my word is worth on here, is that Patient X has a publicly documented history of hearing loss and a publicly documented declaration of having applied to (and entered) the trial. I don't wish to say much more than that, but I think if you read between the lines here you'll be able to figure out what I've done and I'm saying. I'm already mindful of the fact that this person has been bombarded with a lot of questions from a complete stranger (yours truly), so if there is something you want to ask, I think the best course of action would be for me to get back in touch with them directly.
I'm about to sleep so won't be able to answer any questions until tomorrow at the earliest, although I still intend to take a break for a week or so from the forum (although I am keeping general tabs on discussions). I will report back if and when I hear from the second patient.
I'm taking a stab but maybe it means it's blinded in both directions and masked at the four different points?
Thanks for that. That's kind of my whole point.
My entire IHC-preference-at-first-pass speculation is based on Phase 1 not having the same Phase 2a baseline deflation problem (which wasn't the only issue, since apparently multiple low interval injections also were in a different way) but having an actual genuine hearing effect that does not appear to be (largely over the group, some individuals may have some probably minor changes) OHC based on audiograms. But it's not a subtle effect. One that you could reasonably say is placebo or chance in this case.
That leaves synapses or IHCs and since all their pre-clinical work pointed to hair cell restoration and not synapses on healthy cells, it would be IHCs in this case imo. Especially since IHC regrowth is not being something you could see on an audiogram.
This is the only way I can make sense of the data if the company and the Phase 1 participants were legit (and Frequency Therapeutics would probably know about Phase 1 by now).
I think the entire problem is similar to how protein structures were solved. With better testing.
Hell vitamin B12 wasn't even known to have a chelated metal until recently, with it being proved by X-ray crystallography.
Hearing tests are the bloody leaches of the dark ages for scientific rigor. Using that to base your cutting edge science on, looking for significant results when your variance can be so wild is precisely the reason that NO MATTER THE DRUG or MECHANISM it will FAIL TRIALS because your testing foundational methods are TOO INACCURATE to put it up to anything.
If we had hearing tests measuring antibiotic effectiveness, we'd still be using penicillin based on an accident.
You'd literally have to have something not make an incremental or series of small incremental improvements, but be a home run in performance, given you can only work with what you can work with in the cochlea, that won't happen!
It's the difference between looking at something change smell or colour in chemistry versus NMR, obviously you can tell something is happening but for any real detail the former simply isn't good enough! It's simply night and day to which hearing tests need a contrasting measure to.
Perhaps they should have measured otoacoustic emissions as hair cells don't lie if they come back, but even that resolution is pretty poor, but it's better to have two factors of verification of your results (or lack of) regardless.
I'm sure the scientific community is aware of this, but to be blunt all the focus is spent on treating hearing loss as opposed to measuring the variety of ways it can occur, which if you're approaching a problem is completely ass backwards no matter how brilliant your treatment regime.
Even the salicylate rat tinnitus method is a complete freaking joke. You need objective direct proof you run a linear regression and calibration curve against religiously. Full stop.
Hell vitamin B12 wasn't even known to have a chelated metal until recently, with it being proved by X-ray crystallography.
Hearing tests are the bloody leaches of the dark ages for scientific rigor. Using that to base your cutting edge science on, looking for significant results when your variance can be so wild is precisely the reason that NO MATTER THE DRUG or MECHANISM it will FAIL TRIALS because your testing foundational methods are TOO INACCURATE to put it up to anything.
If we had hearing tests measuring antibiotic effectiveness, we'd still be using penicillin based on an accident.
You'd literally have to have something not make an incremental or series of small incremental improvements, but be a home run in performance, given you can only work with what you can work with in the cochlea, that won't happen!
It's the difference between looking at something change smell or colour in chemistry versus NMR, obviously you can tell something is happening but for any real detail the former simply isn't good enough! It's simply night and day to which hearing tests need a contrasting measure to.
Perhaps they should have measured otoacoustic emissions as hair cells don't lie if they come back, but even that resolution is pretty poor, but it's better to have two factors of verification of your results (or lack of) regardless.
I'm sure the scientific community is aware of this, but to be blunt all the focus is spent on treating hearing loss as opposed to measuring the variety of ways it can occur, which if you're approaching a problem is completely ass backwards no matter how brilliant your treatment regime.
Even the salicylate rat tinnitus method is a complete freaking joke. You need objective direct proof you run a linear regression and calibration curve against religiously. Full stop.
HOLY SHIT, 90 FUCKING PERCENT IMPROVEMENT!
THANK YOU @Aaron91.
You just gave me my will to live back.
Thank you for this. So he noticed improvements in his hearing about one month after his last FX-322 injection. Therefore this does make sense that weekly dosing of FX-322 kills the regrowing process as it takes about a month to notice the effects of the FX-322 injection at which point his tinnitus seems to have been 90% gone.
Do you know if he has mentioned this to Frequency Therapeutics before the release of the Phase 2a preliminary results? Could he ask them if there were any other patients that experienced the same effects as him?
- Aug 5, 2019
- 1,852
- Tinnitus Since
- 05/2019
- Cause of Tinnitus
- Autoimmune hyperacusis from Sjogren's Syndrome
Yeah, but this is the first time you pulled out an actual rabbit.
Gotta be the bear. Those gains could be from a placebo or the drug. Tinnitus could have improved with time or from help of a placebo.
Probably more likely that the drug barely works and they improved a little, but I'm not sure I've learned much, being critical here.
Thank you. They say there are 5 stages of grief. I'm going to be stuck on stage 1 (aka denial) forever with this FX-322...
Log in or register to get the full forum benefits!
Similar threads
