Frequency Therapeutics — Hearing Loss Regeneration

I'll be the brain. I've got the wrinkles to carry it off.

 

I hope the 90-day data at the end of March is satisfactory. Looking forward to its release. Take away my pain.

 

I am suspecting the exact same thing. I was looking at the chart @Aaron91 shared (below) showing the difference in LGR5 proliferation between a single molecule approach and the FX-322 pair of molecules. I started to wonder, how would they know with FX-322 isn't causing transdifferentiation and mitosis? How would they know that in the high-synergy reading that there isn't some transdifferentiation going on from just the VPA hitting cells?

However, when we look at the flow diagram, it appears that VPA flows into the cochlea at the same rate as CHIR, it just dissipates/is absorbed more quickly. You'll notice that both reach the 2 kHz line at about the 2 hour mark. However, most of the VPA has been absorbed at that point, and since there wasn't any notable improvement on the audiogram at 4 kHz or 2 kHz in the Phase 1/2, one roughly assume that there wasn't enough drug there to cause transdifferentiation or mitosis. (I am assuming a transdifferentiated cell, if it were to happen with FX-322 would produce some type of improvement on the audiogram).

I definitely agree with the prior posts from @Aaron91 and @FGG that the amount of VPA used in the current dosing was intentional either to control the mitosis to produce a desired result without created an unwanted outcome (transdifferentiation), and/or to not cause VPA to create another adverse side effect.

In another note: Carl LeBel mentioned (I believe it was on the Tinnitus Talk Podcast) that it's possible that each successive dose in the Phase 2A might go "10% deeper." I wonder if they suspect that VPA will flow more deeply into the cochlea because it isn't being "absorbed" by LGR5 cells that are now replaced with fresh new hair cells. Since there's plenty of CHIR at what seem to be sufficient concentration, that 2x and 4x doses might get far enough to reach (and show measurable improvement at) roughly, 4 kHz or maybe even 2 kHz on the audiogram. May not be 30+ dB, but a modest 10 dB at the outer limit.

So I guess my thoughts now are:

- For the Phase 2A outcomes, I am leaning much more towards it more likely to be a "Meh" Scenario.
- We'll see really strong improvements in the EHF range.
- There will be this long tail of improvement out to possibly 4 kHz at like 10 dB.
- Ex: In the EHF Range: The 4x dose will see near full recovery at the EHF, and at 8 kHz, 30 dB, 6+4 kHz, 10 dB.

What that might mean in terms of reformulation / dose:
- The reformulation probably needs to be a slow release of a higher concentration of VPA. (FX-322 "2.0")

AND/OR

- The dosing needs to be closer together (could be done in Phase 2B/3).

 

The more I'm reading about this, the more I really don't think transdifferentiation on its own will happen in the *cochlea* with just VPA because the mechanisms for that are switched off in the cochlear cells unlike in other areas of the body, like the liver study @Aaron91 showed.

I really don't think that's it. In order for progenitor cells to transdifferentiate there, they would definitely need more than a HDAC drug like VPA because their regen pathways are switched off (unlike the liver with always has a great deal of proliferative/regen ability but you can enhance that by messing with notch).

If it's deliberate, I think it's likely for some other reason. Would be nice to ask them.

 

I'm sure this has been discussed but I cannot find it. If you have cochlear dead regions, will FX-322 help? It seems all my EHF have gone from where I could hear them to nothing at any volume.

 

From a common sense perspective, it would be a odd for a company that's so conservative about safety to brag about how their drug combo specifically does not lead to transdifferentiation, but this fact relies on the VPA not acting alone. I'm not a chemist so maybe the risk is negligible if they go super conservative on VPA compare to CHIR, but it still seems unlikely.

I wonder if actually you all (@FGG, @Diesel , @Aaron91 ) are right in that there's no evidence that VPA leads to transdifferentiation in the cochlea, but to be extra safe they went low on the dose. Almost like a double safety guard. To me, only one safety guard (low VPA knowing it leads to transdifferentiation in the cochlea) is too unsafe.

 

I have no idea what you guys are talking about but I was put on Epilim which is Sodium Valproate for 3 months or so in 2019-2020. Does this mean it has interfered with my ear and may affect how FX-322 works for me?

 

The people at Frequency Therapeutics are making history in medicine, their formula is amazing.

 

It's believed that progenitor support cells still exist in the cochlea up to the severe threshold range on the standard audiogram. However at the EHF, the loss gauges used in to the standard audiogram may not necessarily apply directly. For example, a loss between 20-30 dB is considered “mild” on the standard audiogram up to 8 kHz, but may actually be more normal as the testing progresses to 16 kHz.

I’m actually working on an assessment of this based on what we know from the Phase 1/2, I will share more later.

 

If I could hear to 12 kHz again and other frequencies normally I would be thrilled.

 

Probably not.

 

I don't think so at all, for reasons I went into in the Pipeline Therapeutics thread (because now this discussion is taking place on two separate threads...).

 

Will FX-322 work for me? Asking because I have hearing loss in my low frequencies.

 

Theoretically, anyway, you might hear up to ~20 kHz again, which many of us haven't heard since we were children.

 

What I do wonder is, if this drug does indeed work for the severe range and not profound due to lack of support cells, does the drug still regenerate hair cells in the profound range if there are enough support cells in "better hearing" areas of the cochlea? Or will it only regenerate the ones with support cells present and the profound regions remain totally dead?

I understand fully profound people won't benefit from this, but I do wonder if the benefit for those with severe-to-profound would see partial improvement only for the non-profound ranges or a joint improvement because there are enough support cells in other non-profound frequencies.

 

You would be a good candidate for a hearing aid. Those frequencies are critical for speech.

 

The current consensus is that the current delivery method will likely work better for higher frequencies, since not enough of both molecules (but especially VPA) reaches down into the cochlea. My guess is that already 8 kHz will be borderline working, but in the end only the results of the study will show for sure.

 

I don't think the support cells migrate to other areas of the cochlea during this process if that's what you are asking.

 

So after spending most of my day on this, I think @FGG is right.

Until about an hour ago, my stance was still somewhat somewhat agnostic in that, while I definitely agreed that we were looking at different types of cells in different areas of the body, it wasn't immediately obvious to me why VPA would not transduce hair cells in the cochlea when used alone, especially given that studies in birds and zebrafish (both of which regenerate their hearing naturally) would suggest otherwise. So I decided to go back and watch one of my favourite talks by Will McLean which is no longer publicly available but a kind soul did upload here.

For those of us with hyperacusis, here's what Will had to say at around 11:25:

"Primary cells are very important because context is important. Turning notch on in one system may induce proliferation but in another system may induce transdifferentiation"​

So although this isn't a clear statement of VPA not inducing transdifferentiation when used alone in the cochlea, it certainly backs up what I believe to be the crux of @FGG's argument, which is that just because VPA induces transdifferentiation in one part of the body doesn't mean it will transduce a cell in another part. If I am reading correctly, FGG's argument is that, in the case of the cochlea, this is because the regenerative genes are not switched on in the first place. This sounds plausible, but I felt this wasn't something that we ourselves could either prove or disprove, because it would seem to me that you can't actually test VPA's ability to transdifferentiate a cell in whatever part of the body it may be unless the cells in that part of the body have some kind of transdifferentiation potential in the first place, but maybe I'm wrong.

So I was thinking for a while that the only way we could come to some kind of conclusion about this was if we found a study of VPA being used in a human cochlea where transdifferentiation was observed (I haven't found one, but if anyone knows of one I'd like to see it). I had previously found studies in both zebrafish and birds showing that VPA depletes support cells without affecting hair cell numbers, suggesting that support cells continue to have the gene expression needed to become hair cells but only via transdifferentiation and not asymmetric division, but neither of these were adequate examples because both zebrafish and birds can regenerate their hair cells anyway.

As I continued to watch the video, Will McLean made some comments on the graphs I originally posted in the Pipeline Therapeutics thread. He said that while CHIR had some effect on LgR5 proliferation, VPA had no effect on LgR5 proliferation at all - zero. The numbers did not go up, so there was no proliferation, but the numbers did not go down, suggesting there was no transdifferentiation either. I must say the graphs are a bit confusing. The y-axis says "count", but the x-axis shows the number of proliferated cells (as measured by their green fluorescence), so it's unclear to me what the y-axis is actually measuring. If anyone could shed some light on this, I would be grateful.

In any case, if VPA alone caused transdifferentiation, I feel like this would have been something Will would have mentioned, but it's still something that I would like us to ask Frequency Therapeutics for peace of mind when we next have them on the Tinnitus Talk Podcast. What I don't fully understand though is that there are other single-molecule notch inhibition approaches out there attempted by other companies, such as Regain and Pipeline Therapeutics, that DO cause transdifferentiation and deplete support cells. So the question then becomes: what role is VPA actually playing here, and how is it different from other notch inhibition approaches in that, when used in isolation, it doesn't affect progenitor cells?

About 10.5 minutes into the video, you will see Will explaining that signal pathways are highly complex (no kidding lol), and that a combination of different signals can lead to very different outcomes. As Will shows, in the case of the intestine, if you take all the possible ON/OFF combinations of Wnt/Notch (e.g. Wnt on/Notch off, Wnt off/Notch on, Wnt on/Notch on etc.) each one of those combinations will give you a different cell outcome to the extent that you have all the cells you need for the intestine - pretty crazy stuff, but as we know, nature is an amazing thing.

So coming back to the role of VPA, we know that it's not causing transdifferentiation, but it's not causing cell proliferation either. We know that Wnt alone does cause some progenitor cell proliferation, but not asymmetric division. So it would appear to me then that VPA, for reasons I can only speculate, enables CHIR to have an effect on gene expression specifically related to asymmetric division that it otherwise does not have when used alone. I think I have a possible explanation for this, but that would require a whole post of its own.

As I'm running out of steam right now, I'll leave that follow up for tomorrow, but if you want a spoiler, the final paragraph of my most recent post on the Pipeline Therapeutics thread goes into a bit more detail about as to what that explanation may be.

Finally, just some fun facts I had forgotten from the video that people may be interested in. It would appear that FX-322 has more than just two small molecules. It also has Vitamin C and Repsox, which is a small molecule TFGβ inhibitor. By adding these, Will McLean says they were able to maximise what they got out of VPA and CHIR, producing 2,000x times more progenitor cells than the field had achieved before compared to traditional growth factors.

 

I ask Frequency Therapeutics not to follow the Dr. Susan Shore policy as an example and delay the results for trivial reasons.

 

Can someone explain "notch" to me in the above analyses?

 

My best tl;dr:

It's a pathway involved in cell to cell signaling. It's complex and has a lot of different functions, one being to push a neighboring cell down one path ("fate"), another being lateral inhibition preventing cells from proliferating on top of each other essentially.

VPA stimulates notch (vs notch inhibitors like Regain's drug) but inhibits part of the cascade so its effects are complex.

You do need to affect cell signaling as well as the regen pathways for regeneration to work.

 

I had no idea that McLean had a lecture. That is helpful to know. This thread has recently been going a million miles an hour on cellular biology so I have some questions that remind me of this meme.



1) What exactly is the difference between LGR5+, progenitor, and support cell? My understanding LGR5+ is a protein receptor in support cells that increases when the WNT pathway is activated and there is inhibition of the notch pathway HDAC, enabling proliferation. Progenitor is just a synonym for support in the ear.

2) How would proliferation work in vivo, and how does it differ from mitosis (which is the same as asymmetric division?)? Am I correct in understanding that "proliferation," as seen on the GFP graphs, is just the proliferation of LGR5+, which allows for mitosis, and prevents transdifferentiation?

 

I will give it a go:

1) LGR5+ cells are a subset of support cells that express the LGR5+ protein. These are the ones that are the targets for the drug. The other support cells are not receptive. A progenitor cell is a "further along" stem cell and the LGR5+ are support cells that are in this niche.

2) Asymmetric division is not exactly mitosis. Mitosis produces two identical daughter cells. In this case, you are producing a copy daughter cell and a hair cell, not two identical daughter cells (which in this case would be two LGR5+ support cells).

Proliferation just refers to an increase in the number of cells.

 

Oh, 1) makes a lot of sense. I was confused because I've seen LGR5 being called a gene, protein, and cell. This is maybe an embarrassing thing to admit, but why not just air it out? I thought that LGR5+ were the only support cells in the cochlea. I have a very simple understanding of this; when I see YouTube videos, in my mind, there are like 5 total object types in the ear.

For 2) I'm still confused, but I think this is a good thing. In the figure below, it has "mitosis" but then clearly there is a LGR5+ cell asymmetrically splitting into what appears to be a copy and a hair cell. Am I to understand this diagram as saying there's mitosis and immediately after, one of the copy cells transdifferentiates into a hair cell? In other words, the whole process of one LGR5+ leading to one LGR5+ and one hair cell holds, but it's a two step process?

Thanks.

 

Is that diagram from avian hair cell regeneration? Their pathway is a bit different from what Frequency Therapeutics is using. It looks what's going on in the drawing you showed is direct transdifferiation with notch inhibition (which is more like Audion's method).

 

I'm not sure. It was from @Aaron91's post #14572 (previous page). I may be spreading fake news. Regardless, from the diagram, at the top, they call the method of Frequency Therapeutics to be "mitotic regeneration." The diagram is supposed to depict two types. The second, involving WNT, is supposed to be the one describing FX-322.

If you Google Image "mitotic regeneration," you get:

 

Yes, I guess the same thing!

The fact right now is FX-322 works in frequencies from 8 kHz - 20 kHz, but that is not enough for the market, knowing that the most people with hearing loss are in the range of 0.20 kHz - 8 kHz!

Of course many would benefit from FX-322 if they have hearing loss in 8 kHz - 20 kHz!

My opinion and hope is that the Phase 2A results show that FX-322 will work in low frequencies too. If not, they will need to start over with another delivery method to reach deeper in the cochlea!

 

At least FREQ is publicly traded so they have to release information. Most of us will probably be 6ft under by the time Shore’s device comes out!

 

Tinnitus might originate at the 8 kHz - 20 kHz range tough, so good for us; the first generation of the formula almost certainly won't get rid of our tinnitus, but I suspect it will give us a good amount of relief.