Controversial and exciting talks - a report from the Crypto meets Longevity Symposium

Brought to you by the @Aging_Scientist on twitter

This will be the last post in my coverage of the "VitaDAO Crypto meets Longevity Symposium" (see here for part 1). I will summarize several talks from the VitaDAO symposium, although fewer than I aimed for, because it took way too long to write these as it is. And as always I will provide my comments on the state of the art. All errors are mine.

My personal highlight from the symposium was the talk by Savva Kerdemelidis from Crowd Funded Cures who introduced the pay-for-success approach as a solution to market distortions in the pharma business. Finally, the talk by Vittorio Sebastiano inspired me to write about the abuses of the word "healthspan". I think it is essential that we keep biogerontology honest and stop confusing laypeople by claiming that healthspan is somehow magically different from lifespan because it is not. These two are, in most cases, inseparable. More on this below!

Prof. Marco Demaria - University of Groningen
Heterogeneity in Senescence: from Mechanisms to Interventions

The goal is to delay multiple diseases by targeting aging. Telomeres shorten and cells grown in culture reach the Hayflick limit, similar things happen in vivo. We see activation of p53 and p16 expression leading to a stable generally irreversible state that is called senescence. Not all senescence (sen) is bad, because it is of course also a tumor suppressor mechanism. The factors secreted by senescent cells (SASP), are growth stimulatory, angiogenic, and pro-inflammatory. This is also called sterile inflammation, which is inflammation when no pathogen is present.

We see more p16 staining in aging skin and colon, while in the brain, for example, we see elevations of GATA4 and p16.

Then Demaria talks about the, by now, classic p16 reporter mouse model where p16 was tagged with an RFP, luciferase, HSV-Tk cassette. This allows to visualize senescent cells by FACS sorting via the RFP protein, through visible light by luminescence and to destroy them by ganciclovir (GCV). 6 months of GCV treatment improves running distance and grip strength.

Senescent cells accumulate in most tissues and they are at the center of several aging mechanisms. Treatments can be senolytics vs senomorphic, i.e. either kill the cells or reduce their harmful SASP secretion. Many drugs exist but senescent cells are heterogeneous e.g. dependent on tissue of origin, stress type etc. so not all drugs will be universal.

Using an overlapping transcript signature from the different sen models, his team tried to predict drugs that hit all of the pathways (the candidates are proprietary). One of them administered in 3x 3-week cycles decreases luminescence expression and improves grip strength.

Senescent cell accumulation is also seen in the cochlea of aging mice and ablation leads to amelioration of hearing loss.

If high oxygen partial pressure promotes sen, can hypoxia interfere with sen? 20% oxygen is the standard cell culturing protocol whereas in the body cells are exposed to between 1 and 9% O2, with a mean around 5%. Thus for his experiments 1% was considered hypoxia and 5% normoxia.

1% vs 5% although cells do enter sen they do not produce the SASP, and this effect is mediated via mTOR. Roxadustat, an EPO inducer, also reduces the SASP but the effect looks more modest to me. In mice the drug inhibits mTOR and increases grip strength.

During the discussion Demaria mentions that luminescence only shows senescent cells near the body surface, and their abundance is estimated at 3-5% by RFP FACS sorting. In fat it can be higher. Human sen in blood is hard to measure, usually only really feasible via biopsy or in autopsy study. As of now it seems the number of sen cells does not correlate with physiologic function in situ.

Ideally one would like to copy the oxygenation levels of the original tissues when doing experiments but this is expensive to do. Other issues in experiments are too much glucose and 1D culture.

Finally, we should keep in mind there are beneficial components of SASP, while NFKB-driven targets may be bad, others may promote immune surveillance or tissue repair and we do not want to mess this up chronically. In vitro early SASP is the good one, late one is bad, but this seems like an over-simplification.

Prof. Irina Conboy - UC Berkley
Rejuvenation by dilution
The goal is to boost regeneration rather than just preventing damage. In old animals regeneration is somehow inhibited, especially in the brain. You can see that parabiosis promotes regeneration in the brain, e.g. looking at BrdU or Ki67+ staining as markers of neurogenesis, which would be decreased in old animals. Conboy already found this in 2003, she says, but was then rejected by Nature. Later she collaborated with Wyss-Coray who replicated all her findings and then it was published some 7 years later.

Now they are exploring neutral blood exchange, with “blood” that is neither old nor young. Old mice explore more after NBE and become more curious, beta-galactosidase staining decreases as well. When they did proteomics they noticed there is a 1st wave during which old-associated proteins are decreased followed by a 2nd wave when there is an increase of young proteins normally suppressed during aging. As far as brain neurogenesis is concerned the senolytic ABT263 is less effective than the blood exchange. Conboy also saw less fibrosis and more regeneration in the liver. Right now they are exploring other hallmarks of aging to see if these are affected by blood exchange.

This is a promising field because apheresis in humans is already approved and it sidesteps many issues aging research faces, i.e. we do not need to fully understand what is happening with thousands of proteins as long as the blood exchange works. Some team members are working on innovative exchange fluids not just albumin, to improve the protocol. However, so far there is no long term data testing whether these proteomic changes are sustained.

Conboy mentions she is now working for Rejuvenation Research as an editor (please submit your manuscripts) and that, in her opinion, lipids do not change much with age as do proteins (addressing Max’ question). Although clearly some lipids do change in human plasma during aging.

Prof. Vittorio Sebastiano – Stanford University
Repurposing the Principles of Reproduction and Embryonic Development to Stall and Reverse Aging

The idea is to hijack our knowledge of embryology and development to reverse cell aging. The species itself does not age, so it should be possible to figure out why. This is important because there will be 2 billion people by 2050 at risk of age related diseases and frailty.

 Sebastiano discloses that he is the Cofounder of Turn Biotechnologies who works on “epigenetic reprogramming of age (ERA™)”. This is something to keep in mind that I want to point out, and I am sure it applies to Conboy just like everyone else these days. All of the big names work in companies that are selling or developing products related to their research. Everyone is biased in some way, so we should never take all their words as gospel.

Old cells have a disorganized epigenetic program. Embryonic stem cells and germ cells undergo massive reprogramming during development which includes a lot of demethylation and histone changes.

Apparently cloned sheep and other animals reach a normal age, which is proof that the somatic mutations were not limiting to lifespan. Although here I would like to comment about the reasons why the species itself does not age. Clearly, something profound is happening, that, if we could harness it, would forever revolutionize aging research. This has not escaped our notice, but can we ever harness it and is it really due to reprogramming? The other candidate rejuvenating mechanism is selection, which is my preferred mechanism (1).

Sebastiano continues, asking if we can separate loss of identity from epigenomic rejuvenation? He thinks we can. His lab uses mRNA as a platform for delivery of reprogramming factors because mRNA-based reprogramming of fibroblasts to iPSCs is well-established in his lab. He characterized reprogramming at different stages (it takes around 12d) to figure out how long it would take to do partial reprogramming. When they reprogrammed human cells from old donors for around 4 days several markers of aging improved e.g. autophagosome formation and membrane potential. As a control, administering just GFP mRNAs does not help. A youthful transcriptome was restored and the epigenetic clock suggested the cells were younger.

Using a skin explant burn wound model in mice they seem improved wound healing after after 8d of doxycycline induced reprogramming (prelim data). Proliferation of keratinocytes seems to be responsible here.

Can they also rescue stem cells? The experiment goes roughly as follows: Keep isolated stem cells in culture for 48h, divide into two groups, reprogram, transplant back into an injured muscle. Contralateral injection of untreated cells serves as a control that shares the same systemic environment (a nod to Irina Conboy). Regenerative capacity of treated cells was as good as that of young stem cells. No sign of tumorigenesis but more and thicker muscle fibres.

Similar experiment, 30d after injury they measured muscle force and the reprogrammed stem cells allowed regeneration that led to 40% more muscle force than control stem cells. Works the same with human stem cells in immunocompromised mice.

They are using the 4 Yamanaka factors plus 2 more, i.e. nanog and lin28. Interestingly, this is different from the Altos Labs publication that was using 4 and from the Sinclair publication with just 3. Still trying to figure out which one is most important. Ratio matters but the reprogramming seems to work with all cell types. Sebastiano argues that c-myc is not always bad because it silences retrotransposons and drives mitochondrial biogenesis.

Prof. Vittorio Sebastiano – discussion of safety and the desire for immortality (or lack thereof)

During the discussion Irina Conboy mentions that complete reprogramming gives 100% teratoma formation. Sounds scary. How to make it safer? Timing, dosage, ratio. Plus, mRNAs are safer than other approaches because they have a 16-24h half-life.

Finally, Sebastiano states he is not interested in making people immortal. On top of that he is not even interested in defeating the biological limits of lifespan. He just wants to impact healthspan. Now that is a very disappointing statement because it is scientifically untenable. On the bright side, we can learn that being a bigshot at Stanford does not confer immunity from the naturalistic fallacy. I really wish the interviewer/moderator at least offered some pushback here. Because aren’t humans also animals, and thus part of the natural world? And did we not already triple our natural lifespan? What exactly is the “natural” limit of lifespan anyhow and does Sebastiano disagree with the use of man-made antibiotics?

Not only is the statement vague and contradictory as demonstrated above, it also flies in the face of decades of research. It is impossible to significantly extend healthspan without extending lifespan. If you do not want to believe me because I am an outsider or perhaps a shrill “extremist”, will you believe Rich Miller who said exactly the same thing inthe interview I shared in my other post? All robust models of healthspan extension are also long-lived.

For those who do not know him. Miller is a white man with a white beard who has worked at the National Institute of Aging for decades. It does not get more “establishment” than that. In addition, he is one of the best interventional mouse gerontologists in the world. Rich Miller has authored and co-authored more mouse aging papers than a small European country. He is the grand wizard of animal welfare and good mouse husbandry, as his labs have consistently produced some of the healthiest mice and most rigorous mouse data we have ever seen.

I know there is a push towards healthspan in aging research and, to a large extent, it is totally valid. However, just like many other biogerontologists I know that this push has happened for the wrong reasons; to gain political favor with the elites because many among us believe that the word “lifespan” is too scary for politicians and laypeople. I disagree. Instead, I think we need to champion healthspan on its own merits, without distortions, fallacies and half-truths. Without pitting healthspan as the opposite to lifespan. Way too often biogerontologists have turned into enemies fighting about words like lifespan, healthspan and biological immortality, instead of fighting the common enemy which is aging and age-related diseases. We want to study healthspan because it is a correlate of longevity that can be measured quickly and easily, which could allow us to save money, test more drugs, running smaller but more efficient studies, not because of what is “natural”.

Given the evidence, I am not scared to say that we aim to extend lifespan and healthspan as much as we can. They go hand in hand.

For a more in-depth discussion you can read one of my older posts where I try to debunk the myth of "bad lifespan extension" that does not extend healthspan (The Symmetry Error). There are also a couple of posts I wrote criticizing the conservative mindset of present-day biogerontology (Let us talk about Immortality, Revamping biogerontology for a new decade).

Savva Kerdemelidis - Crowd Funded Cures
Pay-for-success x IP-NFT Pilot: A New Financial Model for Repurposing Off-Patent Rapamycin

Kerdemelidis goal is to develop medicines where the patent system has failed e.g repurposing (finding a new use for an old drug). To do so he funded the Crowd Funded Cures Charity. It certainly is a pressing issue as US healthcare costs went from 4 to 16% in the 20th century (private and public). There is also Erooms law, an observation suggesting that drug development becomes more and more expensive. Patents cause distortions for drugs that allow monopoly pricing. Sometimes this is good, sometimes it fails.

1.5T$ in cost savings over 10y from repurposing should be possible but there exists no mechanism to capture this. Similar to the tragedy of the commons. Diets and supplements also have these issues as do antibiotics and neglected diseases. Then he gives a couple of examples. For example, paricalcitol was “shown effective against late-stage pancreatic cancer”, but pharma will not fund large studies to prove it. I suppose he forgot to add a “possibly” to this sentence, because as it stands this is an oxymoron. The talk might have benefitted from more precise language here.

Actually, he says, there are mechanisms to support repurposing, e.g. the FDA allows exclusivity of repurposed drugs but this is hard to enforce. Propranolol was shown effective for anxiety, while normally it is used for hypertension. Doctors can just prescribe it off-label or somehow make the diagnosis fit hypertension.

This leads to so-called financial orphan drugs.

Drug development is cheap, R&D effort is 3 billion dollars to bring a new drug to the market and it usually takes 15 yrs. In contrast, repurposing costs on average 30 millions and takes 3 yrs.

Examples include low-dose naltrexone for pain, which is now off-patent for addiction. Simvastatin for multiple sclerosis. Dichloroacetate (DCA) and IV vitamin C for some cancers. Also, impressively, fluvoxamine for COVID (this one has decent evidence IIRC). I noticed, interestingly many of these are also (in)famous drugs that are often promoted by cranks and self-proclaimed maverick doctors, which may be a state that is fueled by promising phase II trials with no follow-up.

Then another famous example: Psilocybin mushrooms for depression and other mental illnesses.

Prescribing requires big trials and RCTs, off label prescribing rarely is done based on smaller studies. What can we do about this?

90% of clinical trials are funded by private capital because government and academia do not want to take the risk.

However, with already approved drugs the investor has no way to recoup costs after the trial due to off-label use. In his proposed system, the insurer could act as a “success payer” who would offer to pay for a successful trial demonstrating benefit. In principle this is very simple. The insurer just has to calculate cost savings from the approval and payout part of these future savings.

As a proof of principle he wants to fund a rapamycin plus exercise phase II pilot study with the main outcome being change in the 30s chair stand test. I did not fully get how this is going to work, though.

Discussion and comments: what can we do about financial orphan drugs?

As mentioned by Kerdemelidis, most drugs are off-patent already so we have a nice treasure trove at our disposal. I love the idea. It is not the first time I thought about market inefficiencies in pharma and how to solve them but I would have never come up with such an ingenious idea. My suggestion would have been clumsy, albeit workable. The government could impose a tax on certain classes of drugs which is directly used in order to fund drugs that are not economically viable. For example, income from a tax on mainstream antibiotic sales could be paid out to companies that developed new antibiotics. Or an exponential tax on me-too drugs could be used to fund more novel drugs, etc.

The solution to get more pharma investment for aging is to reclassify aging as a disease or as a treatable condition as per the FDA and EMA. This will get us new drugs. However, even once this is resolved we will still want to promote repurposing studies for old drugs that could be effective for aging. This is where his pay-for-success model shines.

Finally, I remain a bit worried about first-mover (dis)advantage. Since all the insurers reap the benefits from the approval and the published studies, each individual institution benefit from waiting for the other insurers to act as a “success payer” first. This should be solvable by a consortium, one would hope, though.

It was fun to write about the science that I love.

1. Several selection-based mechanisms operate to ensure the viability and fitness of the (future) fetus and child. For example, on the father's side we have competition between sperm which will select for cells that are able to produce ATP efficiently via aerobic glycolysis, capable of efficient chemotaxis to find the egg cell, expression of the right enzymes to penetrate the zona pellucida etc. On the mother's side, I speculate, there are similar mechanisms to select for a healthy egg cell to mature. In addition, we have the well-described mitochondrial bottleneck which allows a single mitochondrial genotype to become fixed, thus allowing evolution to act on this. Act, it does. Evolution is cruel and indifferent, but effective. The rate of spontaneous abortions rises to 75% in pregnant women 45yo of age (Andersen et al. 2000) and this does not count pre-implantation failure which is also very high and post-partum selection that would act, e.g. on aneuploid, weak, sick or unfit individuals, in the wild.

The level of pre-implantation selection can be naively estimated from fertility statistics:

"What are the chances of pregnancy as a woman gets older? For healthy couples in their 20s and early 30s, around 1 in 4 women will get pregnant in any single menstrual cycle. By age 40, around 1 in 10 women will get pregnant per menstrual cycle."

If we assumed the couple had sex every 3 days this would give a failure rate of 39 in 40 for young and 99 out of 100 encounters for older women. This should give us an intuition for the severity of selection. Given the importance of fertility for species survival the high failure rates likely serve some purpose, although this remains speculation. Hopefully, I can write something more rigorous on this topic in the future.

see e.g. Zaidi, Arslan A., et al. "Bottleneck and selection in the germline and maternal age influence transmission of mitochondrial DNA in human pedigrees." Proceedings of the National Academy of Sciences 116.50 (2019): 25172-25178.