there was an attempt to write two posts per month
Why bet on these outcomes? Background Information.
The 20th century brought unprecedented prosperity and well-being for many. Average life expectancy saw a dramatic increase doubling globally from 32 to 66 years, settling somewhere around 80 to 90 years in developed nations. Indeed, life expectancies in developed nations are so high now that the rate of intrinsic aging has become the limiting factor for human health- and lifespan. This, however, poses a tremendous challenge for both the pension system and disease-centered biomedical research.
Most diseases of the elderly are directly caused by aging and due to competing age-related risks for these various diseases the eradication of any one disease would only have surprisingly small benefits, i.e. during aging risk of death increases in parallel for most diseases.
This phenomenon is also known as Taeuber paradox and its importance to modern biomedical research cannot be overstated. Put in the simplest way, if cancer does not kill you so will cardiovascular disease (Olshansky et al. 2016; Keyfitz 1977).
Therefore treatments that can decelerate the rate of aging are the only available path to significantly increase human lifespan going forward. Before drugs and treatments can be considered for larger clinical studies in humans, pilot trials in rodents must be performed as a necessary prerequisite. This is why the National Institute on Aging established the Interventions Testing Program which “is a multi-institutional study investigating treatments with the potential to extend lifespan and delay disease and dysfunction in mice”.
So far it is the largest systematic effort to test potentially life-extending compounds in mice that has been performed. This is an attempt to give the biogerontology community, interested experts and laypeople an opportunity to predict outcomes of this important experiment.
There are two good reasons to establish a prediction market. First of all, prediction markets may outperform expert opinion. Second of all, as of now we have no way to know the public (expert) opinion. It would be valuable to figure out what the biogerontology community thinks about the different compounds even if in the end they were unable to correctly predict the outcome of the lifespan studies.
- Cohort 12: C2016
- for a historical comparison: nicotinamide riboside, candesartan cilexetil, geranylgeranylacetone, and MIF098 all failed. 17-a-estradiol & Canagliflozin failed because they are not robust across genders (male only). If we were to stipulate that the prediction markets work the first four compounds should have received a low score and the second two a somewhat higher score.
- Cohort 13: C2017
- Will 1,3-butanediol significantly extend the lifespan of UM-HET3 mice tested by the NIA’s Interventions Testing Program? (Cohort 13: C2017)
- Will Captopril significantly extend the lifespan of UM-HET3 mice tested by the NIA’s Interventions Testing Program? (Cohort 13: C2017)
- Will L-Leucine significantly extend the lifespan of UM-HET3 mice tested by the NIA’s Interventions Testing Program? (Cohort 13: C2017)
- Will PB125 significantly extend the lifespan of UM-HET3 mice tested by the NIA’s Interventions Testing Program? (Cohort 13: C2017)
- Will Sulindac significantly extend the lifespan of UM-HET3 mice tested by the NIA’s Interventions Testing Program? (Cohort 13: C2017)
- Will Syringaresinol significantly extend the lifespan of UM-HET3 mice tested by the NIA’s Interventions Testing Program? (Cohort 13: C2017)
- Will Phase II - Rapamycin/Acarbose significantly extend the lifespan of UM-HET3mice tested by the NIA’s Interventions Testing Program? (Cohort 13: C2017)
Further reading and notes
I decided against splitting the bets by gender to make the number of bets more manageable. This is acceptable because the key outcome is robust lifespan extension across both genders which - to be fair - is a high bar, so far only passed by rapamycin. Phase II tests will be included because there is potential for negative interaction. Also I decided against numerical predictions simply because I could not find a decent prediction market that allows this and because I do not see the benefits of the added complexity. This post or future posts will be updated with more recent cohorts. Results from Cohort 13 should be published soon so this one is my priority for now.
I am considering writing a companion post about the different drugs being studied by the NIA ITP and the rationale behind them. May be updated.
Olshansky, S. Jay. "Articulating the case for the longevity dividend." Cold Spring Harbor Perspectives in Medicine 6.2 (2016): a025940.
Keyfitz, Nathan. "What difference would it make if cancer were eradicated? An examination of the Taeuber paradox." Demography 14.4 (1977): 411-418.
Nadon, Nancy L., et al. "NIA interventions testing program: investigating putative aging intervention agents in a genetically heterogeneous mouse model." EBioMedicine 21 (2017): 3-4.
Peer-reviewed publication or official pre-print from the NIA ITP. Maximum lifespan will be used and Mean lifespan only should the NIA ITP team not publish max LS data (highly unlikely). If p-value for maximum lifespan change for the pooled cohort is not provided the following criteria will be used for resolution: significant MLS increase in both genders OR <0.01 in one gender and a non-significant numerical increase of at least 5% in the other gender. We relax criteria because the analysis split by gender will obviously have lower power than a pooled analysis. The cut-off is based on the rapamycin 4.7 ppm data, which we want to define as "barely making the cut", see also Table 1: