Three speakers shared their opinion on the underappreciated opportunities in the aging industry and work that is currently not well incentivized but could dramatically advance progress. Mike West discussed the need for treatments of chronic degenerative diseases associated with aging and the potential of partial reprogramming for induced tissue regeneration. Karl Pfleger described the need for better “debuggers for biological blackboxes”, also the need for trialing combinations of therapies and aiming for robust mouse regeneration, as well as the overlooked opportunity of the professional sports market for longevity. Sonia Arrison focused on immune system regeneration, regenerative medicine and organ repair, as well as more focus on the brain in respect to aging, enabled by a general uptake in focus on brain research. Finally, we discussed a range of interesting ideas for instrumentation and tooling that could advance progress in the field, and risks that the community needs to be wary of, with a few concrete examples of what the field can do to prevent them.
- The largest unmet need is for treatments of chronic degenerative diseases associated with aging: They account for over 80% of the $3 trillion in healthcare expenditures in the US. Focusing directly on lifespan extension instead of chronic degenerative diseases is not in the best service of mankind right now. Degenerative diseases are by far the most strategic target economically and in terms of alleviating human suffering, and framing longevity as such in terms of quality of life improvements and economic benefits is our best bet.
- The most powerful interventions will come from the technologies addressing chronic diseases, which will be able to induce tissue regeneration: Current typical modalities like small molecule therapeutics are ineffective in that regard, they cannot regrow your heart or cartilage. But early in human development, our bodies can regenerate scarlessly, so it is possible, we just lose that ability with age. That ability is now within our reach to induce in adult humans again through reprogramming technologies, using transcription factors taking cells back in time into regenerative state. What is sometimes called partial reprogramming is a very powerful but underexplored modality and technology promising the possibility of inducing tissue regeneration in the context of chronic diseases of aging.
- Partial reprogramming probably needs significant advances in synthetic biology to work in situ: The problem is that biodistribution is not flat in situ, there are exponential differences in the distribution curves so even a lowest common denominator approach is difficult. For example the liver is going to take orders of magnitude more of the particle than the next tissue (that was precisely what Ocampo found in 2016 – the liver takes most of the reprogramming). And then the states of cells are also massively heterogenous, so we will need a really complicated control logic to solve for that. And that control logic is really a domain of synthetic biology. So the delivery is what keeps this technology in a dish right now. The typical approach is to first target a life threatening compartmentalized disease, so not a full body systemic treatment in the first clinical trial, which seems like a likely approach here as well.
- There will almost certainly be a huge concern about cancer with reprogramming technologies: Because the same pathways used for reprogramming are also often used by cancer cells, which is not yet well known but will be published in the near future. However, that’s also an opportunity, because reprogramming seems to behave differently based on the senescent cell load in the tissue (P53 pathway which induces senescence can induce apoptosis if it’s in that regenerative state). Anyway, the scientific community needs to focus on demonstrating that reprogramming can be done in a safe way. And that it reverses aging without increasing the risk of cancer, since that will be the first major concern of the FDA.
- Question is how large fraction of the damage accumulated with aging will partial reprogramming be able to address: Since it most likely doesn’t clear the extracellular matrix and proteins and carbohydrates that exist outside of the cells and that don’t get turned over that much in the organism. However the jury is still out, we still know very little about all the effects of reprogramming. Case in point is the most recent shocking fact about reprogramming: in the process, it turns on autophagy and adjusts mTOR.
- Companies working on partial reprogramming: Turn.bio, Iduna therapeutics (part of Life Biosciences), Reverse Bioengineering (AgeX subsidiary), Oxstem
- We need better assays and debuggers to look under the hood inside the blackboxes of biological systems on the most granular level: Most work on biomarkers right now is on the level of blackbox whole organism biomarkers (like composite methylome clocks), and the area is relatively well resourced and researched at the moment. Slightly more under the hood are breakdowns according to some specific fields, like biomarkers or clocks actually assigned to specific hallmarks or SENS areas, or other kinds of more low level categorizations. Work is starting to be done on that (measuring senescent cell load for example). But we need to go even one more level deeper. We need better assays, ways to track what’s actually going on inside. To see the intermediate variables in the metabolic processes, or long chain of chemical reactions important for aging, to verify if things work as they should and as we expect them to work, and to be able to debug them if not.
- Case in point is Unity’s failed trial for osteoarthritis: We don’t even know whether the drug actually cleared senescent cells or whether it failed to clear enough of senescent cells. Another case in point is stem cell therapy, where it took a long time to figure out that stem cells were dying and that signaling was actually a large part of how the beneficial effect was felt. We need to be able to answer those kinds of questions faster.
- A good example of better assay tooling is Correlia Biosystems: They have an assay that does basically the same as ELISA, but requires much less blood, and therefore enables longitudinal mice studies, because it’s not necessary to kill the mice to get enough blood samples. We need to do longitudinal studies and in mice we end up needing multiple cohorts at different timepoints. Very inefficient and not truly longitudinal. And there’s another advantage of the Correlia’s small sample size: Any subtractive tech (like NaNots or pheresis) must document impact on the microenvironment following target depletion from circulation. This is done using liquid micro biopsies from microenvironment, but it’s very hard to get enough liquid for Luminex.
- We need more work combining different interventions and significant therapeutics attacking multiple hallmarks or sens categories: We need to see how much we can do with what we already know works, when combined together. More push towards robust mice regeneration. It’s not feasible to do those combinational trials in humans first. Maybe shorter lived dogs, or hamsters, if mice are not great for this kind of research. But basically any way to enable more research into combinations. For example SENS Research Foundation are starting combination therapies in earnest, starting with MSCs plus senolytics in mice.
- Nobody is focusing on aging and degenerative processes making life worse in the 40’s and 50’s: There might be a good way to market through athletes and professional sports players – rejuvenate professional athletes whose performance degraded because of the onset of aging to their prime competitive levels. The route through professional sports players could work as a good awareness vehicle for the longevity field as well. Similar for reproductive medicine, that as well as sports medicine tends to bring a lot of corporate incentives and financial attention.
- Immune system regeneration: There is more appetite and opportunity for it because of covid-19. So more people working on thymus regeneration with different strategies is something that would be great. There was a very recent report of good success for the thymus with decell/recell approach (Taylor).
- Regenerative medicine, new organs from stem cells: A great public example that attracted attention was the child with a lab grown bladder who was featured in a famous TED talk. Yet, it seems like the field has stalled a bit so it could make sense to work on the repair approach (growing a new heart or just injecting some growth repair factors into the heart so it heals without scarring), to move the field a good step further before all the systemic therapies being worked on right now get to their prime. Basically try to look for the low-hanging fruit in this field and get it to market.
- There is a fine balance between regeneration and cancer: A lot of stem cell researchers are finding it hard to get that balance right, especially in the heart. There are approaches that people are working on at Imperial College, UK that exploit the regenerative properties without increasing the risk of cancer. For example doing the programming and growing of stem cells ex vivo, putting the cells into patches and then putting the engineered tissue back into the body as a band aid for the heart for example.
- Longevity and the brain: With so much work being done on the brain, like mapping done by Allen Brain Atlas, Connectome Project or The Brain Initiative, having more people focusing on approaches that start to utilize all this new data to look at the brain with relation to longevity and aging might be another overlooked opportunity.
New tools and platforms that could dramatically advance the progress
- AgeMap – Create a map of every RNA transcript, every mRNA by RNA sequencing, and the chromatin structure on a single cell level resolution of aging tissues in the human body. It’s doable with existing technology, and it would be used for millennia. This map could then be plugged into AI and be an invaluable way for researchers to query the map with queries like: “What are the type 2 pneumocytes in the lung doing during aging and when?” If we have that map, we could also plug it into all the companies doing NGS and have this massive database that researchers from all over the world could use. Might be a good idea to be funded by the Allen foundation as well. The idea is kind of adjacent to Human Cell Atlas by Chan Zuckerberg Initiative or AgeX subsidiary LifeMap Sciences.
- An artificial model, something like an organoid system, one which ages faster than normal: A model where we can test 80 years’ lifetime in 8 months. One possibility is an organoid model with knockin progeroid genes, but the problem with that approach is that we are putting something in the organoid that we already think is aging and therefore we’re looking at curing something we’ve caused ourselves and “know the answer” to.
- Virtual human simulation: There are some people working on this like Comp Biomed – mainly on virtual rats or worms due to complexity. But perhaps there could be some kind of hybrid approach that uses organoids or physical models that can be partly simulated. The most obvious problem with that approach is that the complexity of the aging process is and that it might be easier to fight aging itself than model it in its entirety. It is relatively in line with what AgeX are developing – models for reversing or speeding up aging with epigenetic reprogramming.
- “Human trials in a dish”: Another approach to speed trials up, make them less expensive and improve success. Companies like Organovo, Viscient and Volumetric Bio (which were funded by Methuselah Fund) are going this way.
- Standard set of measurements for each of the hallmarks: Telomere measurements still aren’t standardized after 30-40 years of research, and harmonizing them is an open research question.
- One barrier here is that there is not enough money from the government to do the systematic basic research on that: Developing good quantification for all hallmarks would be a huge process. And people in the industry jump to what seems the most important at the moment, and telomeres weren’t deemed that important recently, senescent cells taking the priority perhaps.
- Quantification of the hallmarks is so hard because it is so dependent on the cell types, organisms, inducers of the aging process, etc: From the practical point of view, there are so many variables, and each of the hallmarks is basically a field of biology by itself, so you need to be an expert in the field. And since the hallmarks are also vaguely defined, there’s really no one number that can quantify it usually. Perhaps aiming to quantify and standardize all of the hallmarks is an ill specification of the problem and we need to try asking a better question?
- Correlate some internal measure with some physiological or machine sensable telltale: With this approach we can use the massive data collection initiatives of large private companies: We can already see their buy-in and interest in that. Like voice based modalities, pupil dilation, etc. Linking external to internal measures is where we can get enterprise buy-in.
- MALDI imaging: It enables pixel wise determination of the full mass spectrum. So we could effectively quantitate metabolites at any single location in a tissue and then compare that based on age. Similar to what Recursion Pharmaceuticals is doing, but instead of optical imaging with 4 color channels, do it with mass spectroscopy and get thousands of metabolites at a time. That would be a hugely powerful resource. Example use for aging would be determining whether there are some metabolites that pool up in aging organisms, whether or not there’s effectively a metabolic backlog for some mitochondrial metabolites, or see whether some proteins are accumulating in the brain as another example.
Roadblocks to progress
- A broader problem with the development of new tools and platforms is that the economic incentives are broken:
There are big financial gains to be made, but it’s hard to take those
future gains because they will be distributed across a lot of different
areas. That diminishes the opportunities for new companies to come
along, because they’re not going to be able to reap all those future
rewards. So we need some ways to jog that negative feedback loop.
- One counterexample is Illumina: A good action step would be creating some case study of how much of an outlier it is. Is it one in a million odds type company? And then uncover the roadblocks and figure out how to incentivize more of those kinds of companies to startup.
- There seems to be a weird communication gap between people wanting some instrumentation and the people wanting to build such tools: We should create a shopping list for the instrumentation and tools that we can show around to people who want to build that.
- It would be useful not to just say that this is a field that is underinvested, but also figure out the structural reasons why that is the case: Good example is running trials on off-patent drugs – there might be a high social return on them but low private return, so it isn’t incentivized in a good way for private companies to pursue those clinical trials. And then when we figure out such structural reasons, we should try to figure out what are some of the levers for addressing some of the structural barriers.
- Safety concerns about the speed with which these groundbreaking technologies are deployed:
We need to be careful so we avoid technological winters because of
backlash to reckless experiments. Like what happened with genetic
engineering or the regenerative organ scandals
that have pushed the field back, possibly years. With the amount of
misinformation in stem cell and regenerative medicine, can you imagine
what will happen when snake oil salesmen start talking about aging
reversal? That needs to be something we need to focus on to communicate
well. There are multiple concrete things that can be done to try to
avoid this in the longevity field.
- Setting up an industry group which would put its imprint on what it qualifies as real science: Like regenerative medicine industry organization or International Society for Stem Cell Research.
- Something similar to the aviation incident reporting system: An extremely successful system for anonymous reporting of failures in the cockpit. It enables whistleblowing for the benefit of the whole field.
- Apply lessons learned from gene therapy and CRISPR in regards to scientific self-regulation.
- Allow people to publish negative results: Every new paper in Nature has to be the most interesting novel thing ever. But if people hide their negative results data, we don’t have this kind of anonymous reporting of sorts on the academic level.
- Clear code of journalistic ethics serving as a litmus test that helps limit the irresponsibly shared information floating around: Basic requirements like properly researching information, doing due diligence before publishing, disclosing conflicting interests, etc. Lifespan.io is working on a tailored version of journalistic ethics for the longevity field specifically, with some additional things that should be met.