In the last post, I explained why I think SENS Research Foundation is probably a great choice for its cost-effectiveness, and I announced that I would interview its chief science officer, Aubrey de Grey, for a further evaluation of this charity. This is the interview, preceded by a small glossary.
SENS: Strategies for Engineered Negligible Senescence. SENS divides aging into seven categories of damage, each with its own corresponding line of research. This categorization is similar to the one described in The Hallmarks of Aging, which represents the current scientific consensus.
SRF: SENS Research Foundation.
Damage repair approach: Repairing the damages of aging, as opposed to geriatrics, which only alleviates disease before death, and old-style biogerontology, which is more focused on studying and tweaking metabolism.
LEV: Longevity Escape Velocity, which is the minimum rate of medical progress such that individual life expectancy is raised by at least one year per year if medical interventions are used. This does not refer to life expectancy at birth; it refers to life expectancy calculated from a person's statistical risk of dying at any given time. This is equivalent to saying that a person's expected future lifetime remains at least constant despite the passing years.
ALT: Alternative mechanisms for telomere lengthening that about 10% of all cancers use to achieve uncontrolled replication.
Macular degeneration: Age-related disease that results in the loss of the central visual field.
NIH: National Institutes of Health, the primary medical research agency of the U.S.
NIA: National Institute on Aging. A division of the NIH focused on aging.
Senescent cells: Cells that have stopped replicating and play a pathological role in aging.
Senolytics: Drugs that eliminate senescent cells.
How do you choose the staff to hire for intramural research projects?
No differently than any other lab would - we look for track record, domain expertise, references, the usual. We don’t particularly prioritise people who specifically want to dedicate their lives to defeating aging, not least because people who join our team without that mindset rather often acquire it over time. The one thing we do try to avoid is hiring people whose longer-term goal is something else, such as a standard faculty position: we like people to stick around, and sure enough, we have very, very low staff turnover.
How do you choose what intramural research to do or extramural research to finance?
It all needs to be within the general theme of repair of aging damage. Other than that, the main metric is that we prioritise areas that we feel are being overly neglected for their importance to the ultimate goal of comprehensive rejuvenation.
To what extent does your funding overlap with (i.e., fund the same organizations as) government funding? Private-sector funding?
The organisations we fund extramurally are all highly prestigious institutions such as Yale, the Buck Institute, Scripps, etc, which of course are all also funded by government grants, and most if not all by the private sector. But, of course, the specific projects are not: they are projects that would not be done without us, because both government and private funding sources impose immense pressure to work on low-hanging fruit rather than high-risk, high-reward work.
How do SRF projects differ from what is currently done at NIA?
There is essentially no overlap. First of all, we take the view that there is no biologically meaningful distinction between so-called "age-related diseases" and "aging itself", whereas the NIH is set up in a way that entrenches that distinction; the result is that it is incredibly hard to get work funded by the NIH that straddles that boundary, which much of our work does; what the NIA does, by and large, sits firmly on one side of it, while what the NCI, NHLBI etc do sits on the other side. (There is the anomaly that Alzheimer’s research falls within the NIA, but in a separate division than "aging biology".) Secondly, there's the issue of low-hanging fruit versus high-risk, high-reward: there are programs at the NIH that nominally address that, such as the Pioneer Awards, but their total budget is a tiny proportion of the NIH budget. And thirdly, even though the NIH's name sounds like it would be focused on translational work in contrast to NSF, in reality, the overwhelming majority of the work funded by the NIA's division of aging biology is curiosity-driven rather than goal-directed.
What portion of your organization's expenses are devoted to funding research as opposed to other activities? What are the other activities?
Really, it's all research (unless you count admin as "activities" - and we perform pretty well in keeping those expenses down; advocacy is very inexpensive). We have a thriving education arm, of which we are very proud, but what it does is research: we fund a dozen or so undergraduate interns (at our lab and labs we work with) each summer and also half a dozen just-graduated youngsters for a gap year before their Masters or whatever, and those people almost all spend that time at the bench.
How important do you think the education programs are compared to research? Have they brought any noticeable benefit (good researchers that you hired, researchers who founded companies, made discoveries…)?
I believe that the education arm of our activities has fantastic value. Several of our alumni have gone on to join companies or labs in the field, such as Longevity Fund, Underdog and OpenCures, and I can confidently say that all of them gained a thorough understanding of what rejuvenation biotechnology is all about, which will be influencing their career choices going forward.
What is your biggest achievement obtained with intramural research? What is the best project you financed extramurally? What makes these the greatest?
I don't think I can single out one best project in each group, but let me instead highlight one in each. Our intramural work on macular degeneration, begun by one of our first funded PhD students over a decade ago in Phoenix and later brought in-house for several years, became one of our first spin-out companies, Ichor. Ichor has grown to a >50-person company with a number of other activities, and the AMD work is now pursued by a division named LysoClear; their progress has continued at pace, and they expect to enter clinical trials this year. On the extramural side, a good highlight would be Revel, which was created last year out of work we funded at Yale, focused on restoring the elasticity of the extracellular matrix, with potential benefits to age-related hypertension, presbyopia, and even wrinkles. That work would definitely never have been done without us, and there is nothing like it elsewhere.
What are the biggest research contributions that SRF spin-off companies have made?
The contributions aren't particularly impressive yet, because we've only been spinning out companies for three or four years thus far. I’d say that the progress that Ichor/LysoClear made over and above what we had done internally are pretty good though.
In many interviews, you stated what you would use at least ten times your current spending. How would you use it, and what research projects would you finance? Would you put all of it in research, or would you also scale up education and advocacy?
Advocacy occupies very little of our budget as of now, and although I would certainly ramp it up, I can’t see it costing all that much even if we became a $50M/year organisation. Education I would definitely expand a lot: we are literally 30-40 times oversubscribed for our summer internship program each year, which means we could absolutely add another digit to its budget with no appreciable reduction in the quality of the people we recruit. The gap-year program is going the same way (already 10-fold oversubscribed, even though it's only in its second year of operation). As for research projects per se, we would divide any increased funding between increases in personnel for existing projects and initiation of new projects. Historically, I have felt that the existing projects would merit most of the money, but in the past year or so, there has been a clear improvement in the quality and attractiveness of the project proposals we receive from external researchers, so that’s progressively changing.
How likely do you think it is that someone will close your funding gap in 2 years? 5 years? 10 years? Why?
I’m hesitant to say, because in 2006 when Peter Thiel came in as our first 7-digit donor, I was way overoptimistic - I felt that because he had such a great reputation in the HNW tech community, the other billionaires would be lining up to follow him in donating, and of course, that didn’t happen until quite recently. Accordingly, we’ve tended to make plans based on a range of possibilities and not to flesh out the high-end plans all that much until it becomes more likely that the money will materialise: we only make sufficient plans to avoid any chicken/egg issue of discouraging potential donors because of our lack of plans.
Have you been overconfident/underconfident about the pace of the research you funded in the past? Have you been overconfident/underconfident about how much funding you would have received? If the answer is "yes" to both, were these failed predictions dependent on each other?
No and yes. If I had to identify one thing that gives me the greatest confidence for the future of our work, it is that the research has not hit a single unanticipated major roadblock, in any of the SENS strands, in the entire 15 years since we started funding research (or even the five years before that, starting with my initial "repair is easier than retardation!" Eureka moment in 2000). We have hit the minor obstacles that any pioneering technological effort is bound to hit, but not at an unexpected rate. In summary, I can look back at those years and say with confidence that the progress we have made is at least as much as I would have expected if I had known how much funding would be available. On the funding, I was way overoptimistic in relation to major donations from HNWIs - see my previous answer. I was probably a bit overoptimistic about grassroots donations too, but not significantly - they have always been only a minority of our income, and I never expected otherwise.
I can't really put my hand on my heart and call it "reasoning" - it is highly subjective, for sure (and it's gradually coming down as we spin projects out - it’s probably only 2 now as against 3 a decade ago). But it arises from my day-to-day exposure to the ongoing research, where I see breakthroughs being made using approaches that were delayed in favour of other approaches when with a bigger budget, we would have done both in parallel, and also from the number of projects (either externally proposed or thought up by us) that we have to turn down or defer because of lack of funds even though they meet our scientific criteria.
In one interview with LEAF, you state that 1 billion donated to SENS would bring longevity escape velocity nearer by 10 years and save approximately 400M lives. This means that one (more than 1000 years long) life would be saved with $2.50. What are the details of the reasoning that made you arrive at the "10 years" figure?
My estimate of how far off LEV is with 50% probability started out at 25 years 15 or so years ago, and is now 17 years, so let’s use round numbers and say 20 years. Those estimates have always been explicitly "post-money", though - in other words, when I say the money would make 10 years of difference, I mean that without the money, it would be 30 years. I think $1B is enough to remove that factor of 2-3 that you mentioned in the previous question, i.e. to take it down to around 1, because it would add a digit to our budget for 20 years. That factor is already coming down, and I expect that it will continue to do so as further progress is made at the bench, which is why I average the benefit out to a factor of 1.5 (i.e. 30/20).
Let's say you start to fund the most neglected project in aging research. After 8 years of work, you estimate that you are 2 years away from completion. However, at that time, a new technology, completely independent from your research, comes along, and it enables researchers to find a solution in just 1 year. So your prioritization 8 years earlier has become useless because you could have started 9 years later without any loss of impact. How likely is a scenario similar to this?
We have had such cases where we were "overtaken" by other groups. The best example is our work at the U of Arizona on rejuvenating the immune system: that consisted largely of trying to regrow the thymus, which is something that other groups (initially a group in Europe) started to have real success on when we were still not clearly achieving much. However, this is a rare event - in fact I can only think of two other cases in our entire 15 years of research funding, and they are less clear-cut (i.e. we stopped when other groups were not conclusively ahead of us, and we'd probably have kept going but for budget constraints, and we probably wouldn’t have regretted doing so). Also, the work can bear fruit anyway, for indirect reasons: in the above case, just last year (at least five years after we stopped funding), the group reported some important findings explaining why the approach back then hadn't worked, and motivating a whole new line of research (which we might well be helping fund right now if we had the money) that would not have been identified at all if not for the foundational work that we funded.
Is it possible that solving one aging damage would completely solve another one? How likely is this? If yes, how does this impact your reasoning on what to finance?
It’s pretty much impossible. The amount of "crosstalk" between the various SENS strands may turn out to be greater than I (or anyone, I'd say) would have predicted a decade ago - in particular, removal of senescent cells seems to have broader benefits than expected - but I’m sure that that crosstalk will in no case rise to "complete": we definitely need to address all of them individually.
How likely is it that the SENS approach, before coming to fruition, will be rendered useless by a single general solution that has nothing to do with SENS? Is this kind of thing even on the horizon?
I'd be delighted, but nothing has emerged that gives me the slightest reason to think that there is any possibility of that.
How high do you think the probability of LEV happening at all is? How much is this probability improved by SRF?
It’s absolutely certain to happen eventually, with or without SRF, so our role is purely to hasten it. The reason it's certain is because (a) the human body is a machine whose function is determined by its structure, so the challenge is only one of complexity, and (b) the further we get towards comprehensiveness of damage repair, the longer it takes for the residual damage that we can't yet fix to reach pathogenic levels, i.e. the actual required rate of progress that I call LEV actually declines over time, making it more and more implausible that we will fail to maintain it.
How much do you think that SRF's education, advocacy, and even research are improving the pace at which LEV will spread through the world population once it will be accessible?
I’d say that that’s mainly down to the advocacy work. My advocacy efforts have shifted in emphasis over time, as different audiences have become receptive. Over the past few years, I’ve done far more than previously to reach out to the investor community, and that’s mostly because there now is one. The next phase, which is just starting, is to reach out to policy-makers. My goal is to engender as much horizon-scanning as possible in the corridors of power, as soon as possible, so that disseminating these medicines will be as fast and smooth as possible. I also do a lot of educating other advocates to give more effective input to audiences that may resonate better with their style of communication than with mine.
When will aging research start to help other altruistic interventions to be more effective? Example: If the date of LEV is moved closer by N years, then the timeframe in which preventing a child from dying from malaria (e.g. via bednets) also makes him/her reach LEV widened by N years. Normally, if you save a child's life by preventing him from dying from malaria, you could buy him at most 80 QALYs. However, if you save him in the right time frame (which aging research widened), you might be buying him 1000 QALYs. Therefore, aging research boosts the overall effectiveness of "preventing kids from dying from malaria" by widening that timeframe.
I don’t think I can improve on the answer implied by the LEV timeframe answer - except insofar as maybe the amount we allocate to bednets etc may rise in advance of that, as a result of simple anticipation of the likelihood that aging will be defeated in the relevant timeframe.
Other measures of impact for aging research are: end of life DALYs averted, impact on life satisfaction and "the longevity dividend": economic and societal benefits of improved healthspan. How much do you think SENS research is influencing these more "short term" measures?
Well, I’d say that those measures apply long-term as well as short-term. Short term, neither SENS nor non-SENS aging research really affects them, because it won't get to the clinic any time soon. Long term, SENS research will influence them the most, simply because it will lead to the biggest benefits in the shortest time.
When discussing the impact of aging research, the focus is usually on humans, and animals are almost always overlooked. How large do you think would be the impact of SENS research on non-human animals (example: pets)?
These therapies will certainly be applicable to other species. However, it's important to recognise that it's harder to reach LEV for a shorter-lived species, because research has less time to make the nest step.
Are there any research projects that you had to shut down due to a lack of funding? Would you restart them if you could?
So far, the only example is our work on ALT (alternative lengthening of telomeres). We would probably restart it, but it’s a bit uncertain because others have made significant progress since we shut it down: we would make the decision based on the latest information.
What would things look like if those projects had been actually put forward?
Well, it’s research, so I can only speculate - but if it had continued at the same level of funding, I think we would have had a very good chance of being world leaders in understanding and inhibiting the ALT mechanism, just as we now are with our project aimed at obviating mitochondrial mutations.
What would things look like if you have had all the money you needed from the beginning?
Again, I must stress that I’m only speculating - but basically what you're asking is where would we be with Calico-level funding starting in 2005 or so. I think there is at least an 80% chance that we would have reached what I call "robust mouse rejuvenation" by now - that’s the increase of mouse lifespan by 1-2 years (I have varied my exact numbers because they are chosen in relation to my sense of the willingness of my prominent biogerontology colleagues to start saying serious things about the defeatability of aging) with interventions begun only when they are in middle age. 15 years ago, I said there was a 50% chance that we would get there in 10 years subject to funding, and now I say 5 years. (There’s that factor of 3 again.)
Do you have reports that provide your organization's track record of producing concrete output from your research - for example, analysis of patents, publications, or citations that came from your research?
I’m not sure what you mean by "analysis", but obviously, our publications in the primary literature are accessible via the usual sources, as are the number of times they have been cited. Patents likewise.
Open Philanthropy's questions
In their medium investigation on aging research, Open Philanthropy concludes with some questions. I asked Aubrey de Grey similar questions, based on what I deem more important or difficult to evaluate:
What are the most neglected areas in aging research?
I think it is clear that damage-repair (i.e. rejuvenation) research remains the most neglected area, despite SRF’s efforts.
What are the most promising unfunded projects in the field?
We maintain an informal list of projects that have nearly but not quite made our funding-dictated cut. The list changes often, but I’m always happy to discuss nearly-funded projects with anyone who may be interested in helping. Also, we are happy to take restricted grants that the donor wants to go to specific research areas.
How likely is it that general-application tools and basic research areas that might not be thought of as part of “aging research” (analogous to epigenetics, stem cells, neuroscience, and drug delivery) will be bottlenecks to accomplishing the core objectives of translational aging research?
Very likely, and that’s why we pay attention to those things too. Our main project in that space at present is the development of a new gene therapy vector that will allow insertion of new genes at a chosen genomic location with uniquely low off-target disruption.
What tools and/or research directions under these headings are most neglected relative to their promise, for the purpose of addressing these bottlenecks?
The reason we have the above project is that gene therapy research suffers just as much as anything from the low-hanging-fruit problem I describe earlier. There's plenty of research, but its focus is the minority of congenital issues where health can be restored by rectifying only a tiny minority of relevant cells.
Would interventions focused on these more basic/general themes have greater or smaller effects on the time by which such objectives might be achieved?
Greater or smaller than interventions focused on specific aspects of aging damage? One can’t really say - we need both - the benefit of the combination will be much more than the sum of the parts.
The following questions were asked in the comment section of the previous post, or were inspired by comments:
Of the treatments currently being developed, is it likely that treatments for multiple hallmarks can be used in parallel?
Sure, of course - that’s the whole idea, that damage repair is a divide-and-conquer approach. Maybe I’m misunderstanding the question?
Are there currently any observed or expected interactions between different treatments?
Not that I’m aware of - but very little coadministration has been done yet, even in mice. Indeed, I can see that becoming a big priority for SRF in the near future, because it may be unattractive by the metrics used by other funding sources (both public and commercial).
Has any effort been made to see if the effects of multiple treatments are additive, in terms of improved lifespan, in a pre-clinical study?
No, and indeed we would not expect them to be additive, because we would not expect any one of them to make a significant difference to lifespan. That’s because until we are fixing them all, the ones we are not yet fixing would be predicted to kill the organism more-or-less on schedule. Only more-or-less, because there is definitely cross-talk between different damage types, but still we would not expect that lifespan would be a good assay of efficacy until we’re fixing pretty much everything.
What side effects have been observed regarding the treatments currently in clinical trials?
I’m not aware of any significant side effects reported in any clinical trials of stem cell therapies for age-related issues, nor in the trials by Unity on senolytics. Other strands of SENS have not reached the clinical trial stage yet, though quite a few will probably do so within the next year or two.
Does the rate of damage accumulation accelerate in the elderly? If yes, do you think the rate could continue to accelerate even after the damage is repaired? How much time do you expect should pass between one SENS treatment and its repetition, considering the rate of damage and its acceleration? Has in-vivo testing that sheds light on these questions been performed?
The rate of damage accumulation does accelerate in the elderly, but only because they have more damage, which impairs the function of the natural systems we have for keeping the rate of damage accumulation as low as it already is. Thus, removing damage will restore those systems to youthful efficiency and the rate of damage creation to youthful rates, just as it will restore all other functions. In other words, no, there will be no acceleration that's independent of the damage load. The interval between treatments will not be one number - some treatments will be ideally administered more often than others, depending on the details of the therapy - but it should not change over time. This is not a question that needs testing - it’s a consequence of the fact that the body is a machine, nothing more. If we elide for a moment the fact that there are multiple different types of damage, i.e. we just describe damage as one number: if the rate of damage accumulation following a damage repair therapy that you think removed X amount of damage is faster than it was when the person originally had X less damage than immediately before the therapy, that just means you didn’t actually remove as much as X.
Crossposted to LessWrong