This lengthy post walks through the process of setting up and running a self-experiment - a trial of one - of candidate senolytic drugs capable of removing some portion of the senescent cells that accumulate with age to cause aging and age-related disease. Metrics are assessed beforehand and afterwards in order to shed some light on whether or not it worked, in the sense of producing some degree of rejuvenation, turning back specific measures of age-related decline.
The outline here is optimized for simplicity, cost, and ability to conduct the experiment without much outside assistance, rather than for maximal effectiveness. There are better candidate pharmaceuticals and better metrics that those settled on here, at least from the point of view of likely effectiveness, fewer side-effects, and relevance to the task at hand, but they require more work, more funds, more complicated logistics, and the assistance of laboratories and physicians. Given these self-imposed constraints, this does mean that the outline here ends up focused on repurposed chemotherapeutics, which make up the majority of the current senolytic drug candidates. That in turn means that side-effects and related risks to health are an important consideration.
The purpose in publishing this outline is not to encourage people to immediately set forth to follow it. If you come away thinking that you should do exactly that, and as soon as possible, then you have failed at reading comprehension. This post is intended to illustrate how to think about self-experimentation in this field: set your constraints; identify likely approaches; do the research to fill in the necessary details; establish a plan of action; perhaps try out some parts of it in advance, such as the measurement portions, as they never quite work as expected; and most importantly identify whether or not the whole plan is worth actually trying, given all that is known of the risks involved. Ultimately that must be a personal choice.
- Why Self-Experiment with Senolytics?
- Caveats in More Detail
- Choosing Senolytic Drug Candidates
- Establishing Dosages
- Obtaining Senolytic Pharmaceuticals
- Storage of Senolytics
- Validating the Purchased Senolytics
- Ingestion Logistics for Powders
- Establishing Tests and Measures
- Guesstimated Costs
- Schedule for the Self-Experiment
- Where to Publish?
- Final Thoughts: Why Not Wait?
Senolytic therapies are those that selectively destroy senescent cells. The build up of senescent cells is one of the causes of aging. So obviously, one hope is to benefit personally from such a therapy sooner than would otherwise be the case, balancing that against incurring some unknown degree of risk of failure or harm. The first human trials, those that establish numbers for that risk, will take another few years to wind through to robust conclusions, and further years beyond that will be required for the medical community to become willing to prescribe senolytics generally. Further, those trials will almost all test only a single candidate therapy, and the evidence to date in mice suggests that different senolytics with different mechanisms are tissue-specific in their effects on senescent cells. Multiple different compounds may be more effective than one - but that won't be discovered in the formal trial process. Lastly, well run self-experimentation carried out by a number of people, where the results are published, can help to guide the direction of later, formal studies.
All of these reasons must be balanced against a sober assessment of the risks involved in obtaining and using pharmaceutical compounds, and an acceptance of personal responsibility for consequences should one choose to run those risks.
There are two areas of personal responsibility to consider here. Firstly, this involves taking chemotherapeutic pharmaceuticals with known side-effects. One should read the relevant papers on their effects, side-effects, and dosages, and make an individual decision on risk and comfort level based on that information. This is true of any pharmaceutical, whether or not approved for use. Do not trust other opinions you might read online: go to the primary sources, the scientific papers, and read those. Understand that where the primary data is sparse, it may well be wrong or incomplete in ways that will prove harmful. Also understand that older physiologies can be frail and vulnerable to the side-effects of specific chemotherapeutic pharmaceuticals in ways that do not occur in younger people and that are not well covered by the studies; pharmacokinetic studies necessary to establish side-effects and tolerances don't tend to be carried out in very old humans, and most cancer trials have participants that almost entirely fall into the 50-80 age range.
Secondly, obtaining and using pharmaceuticals in the manner described here is illegal: choosing to do so would be a matter of civil disobedience, as is the case for anyone obtaining medicines outside the established national system of prescription and regulation. People are rarely prosected for doing so for personal use in the US - consider the legions of those who obtain medicines overseas for reasons of cost, despite the fact that doing so is illegal - but "rarely" is not "never." If you believe that the law is unjust, then by all means stand up against it, but accept that doing so carries the obvious risks of arrest, conviction, loss of livelihood, and all the other ways in which the cogs of modern society crush those who disagree with the powers that be.
Lastly, senolytics is a fast-moving field. This post will become outdated quite rapidly in its specifics regarding candidate pharmaceuticals, as new knowledge and new candidate therapies arrive on the scene. Nonetheless, the general outline should still be a useful basis for designing new self-experiments involving later and hopefully better compounds, as well as tests involving more logistical effort.
The criteria for choosing senolytic drug candidates for the purposes of this outline are: (a) it must be taken by mouth, rather than through injection, as the logistics for assembling materials and carrying out injections are considerably more complicated; (b) it must have shown senolytic effects in animal studies, not just in cell studies, as there are all too many failures to make the leap from cell to mouse in pharmaceutical development; (c) there must be enough human data to determine the effects and side-effects of doses used. The more human data, the better, in fact. Finding a list of senolytics and assessing them against these criteria involves research: dig through PubMed in search of senolytic studies and review articles, and then follow chains of references to find other papers. Carefully check the magnitude and other details of the results claimed in animal studies: some senolytics are better than others. Armed with the names of drug candidates, then look up studies of dosage and effects for cancer and other trials in both mice and humans. Not all papers are open access. Where they are not, taking advantage of the efforts of the copyright heretics of Sci-Hub is the best approach to obtain a copy.
At the present time, the criteria above narrow the field to a few repurposed chemotherapeutics, one of which was shown to synergize with the flavonoid quercetin. These are: (a) dasatinib in combination with quercetin, while noting that the data shows it isn't worth trying either one on its own, (b) navitoclax / ABT-263, and (c) alvespimycin / 17-DMAG. Navitoclax, however, has side-effects that are common enough and severe enough to want to avoid it; it causes a loss of platelets in the blood to the point of producing noticable medical consequences, and did so in about half of cancer trial participants.
There is also venetoclax / ABT-199 to consider, however, a modified form of navitoclax intended (and shown) to reduce the worst of its platelet-destroying side-effects, but lacking animal data for senolytic effects. Navitoclax and venetoclax are both BCL-2 family inhibitors, operating on roughly the same mechanism, and another member of this family of pharmaceuticals, ABT-737, has been shown to have senolytic effects. So while venetoclax lacks mouse data for senolytic effects, at first glance it makes some sense to include it in a test: the trade-off is a matter of losing some of the unpleasant side-effects and gaining uncertainty in whether or not the senolytic effects will carry over. There is a very helpful paper from a few years back that covers the relationships between these BCL-2 family inhibitors, and that does a good job of explaining why venetoclax is a favored alternative to navitoclax, at least for the cancer research community.
Of the other compounds we might consider, A1331852, A1155463, piperlongumine, and fisetin are ruled out for lacking published animal data on senolytic effects. FOXO4-DRI is ruled out for being injected and for lacking any human data on dosage or side-effects - though in principle, it should be the best of all the drug-like options so far discovered, if the animal data carries over into human tests. ABT-737 is ruled out for being injected - unlike other BCL-2 family inhibitors it doesn't really interact usefully with mammalian biochemistry if ingested.
The only definitive way to establish a dosage for a pharmaceutical in order to achieve a given effect is to run a lot of tests in humans. Testing in mice can only pin down a likely starting point for experiments to determine a human dose, but the way in which you calculate that starting point is fairly well established for most cases. That established algorithm is essentially the same for most ingested and intravenously (or intraperitoneally in small animals) injected medicines, but doesn't necessarily apply to other injection routes. The relationship between different forms of injection, dosage, and effects is actually a complicated and surprisingly poorly mapped topic, and we'll set that to one side here. Some compounds - as always - are exceptions to the rule, and the only way that scientists discover that any specific compound is an exception is through testing at various doses in various species.
Given that, the discussion here should be taken to apply only to orally administered drugs, as that is the deliberately restricted scope of this post. Further, when considering pharmaceutical dosage, it is important to emphasise that more is not better; this cannot be approached in the way people tend to naively approach the (over)use of dietary supplements. The primary goal, if self-experimenting, is to take as little as necessary of any chemotherapeutic, senolytic compound, as they are all toxic in any meaningful dose. I enourage a careful reading of the papers in which the side-effects in patients at chemotherapeutic doses and treatment durations are described, as well as the studies showing aggressive chemotherapy to produce a higher, rather than lower load of senescent cells. That the dose makes the poison is an ancient adage, but no less true today.
The steps to figure out a suitable starting point for a human test of an orally administered senolytic pharmaceutical are as follows: firstly read the mouse studies for the senolytic compound in question, in order to find out how much was given to the mice and for how long. Doses for most ingested pharmaceuticals of interest will usually be expressed in mg/kg. Secondly apply a standard multiplier to scale this up to human doses, which you can find in the open access paper "A simple practice guide for dose conversion between animals and human". Do not just multiply by the weight of the human in kilograms - that is not how this works. The relative surface area of the two species is the more relevant scaling parameter. Read the paper and its references in order to understand why this is the case. Again, note that the result is only a ballpark guess at a starting point in size of dose. The duration of treatment translates fairly directly, however. For the period of treatment, start with the same number of doses, spacing of doses, and duration as takes place in senolytic studies in mice.
If there isn't enough data to do more than guess at a dose, then that is a good indication to write off that particular compound. Wait for more data, or look for different compounds with better existing data.
Dasatinib and Quercetin
In the case of the dastinib and quercetin combination, the mouse study of senolytic effects used a single dose of 5 mg/kg dasatinib plus 50 mg/kg quercetin. For a 60kg human this scales up to a little less than 25mg dasatinib and 250mg quercetin. For comparison, mouse studies of dasatinib as a chemotherapeutic can be found that use 50 mg/kg per day for multiple days to evaluate its ability to kill cancer cells. A very useful study on dose effects and duration for dasatinib in humans used a dose of 100mg in volunteers, and you can find other trials of dasatinib as a cancer treatment at that dose. Quercetin is an established and widely sold supplement, and it would be a real challenge to consume enough of it to cause any ill effects, never mind significant ones, judging by the toxicology data.
Another way of thinking about dosage is to aim at producing the same concentration of the pharmaceutical in blood that was used in cell culture studies, or observed in mouse studies. To do this you will need existing human data on how a dose maps to concentration in blood or tissues. The senolytic mouse study noted above and the useful human study provide sufficient numbers to make an estimate at this. In the cell culture section of the mouse study, 100-200 nM/L (nanoMoles per liter) is the effective concentration - more than that adds no greater benefit. Given the molecular weight of dasatinib, one can convert the observed blood concentration of 104.5 ng/ml (nanograms per milliliter) in the human study for a 100mg dose to get something in the ballpark of 200 nM/L. Why doesn't everyone use the same units? Well, we wouldn't want to make this too easy. The human study also provides results for a 180mg dose if you want to try scaling up or down to estimate the dose needed to hit different concentrations.
So via one method, the single dose for a 60kg human is 25mg dasatinib and 250mg quercetin. Via the other method the single dose is in the vicinity of 75-100mg dasatinib and 750-1000mg quercetin, assuming that we scale up the quercetin to match the dasatinib, and depending on where we are aiming for in the 100-200 nM/L concentration. The second dose is at chemotherapy trial levels, but is a single dose rather than taken daily over weeks or months, so the impact will accordingly be more limited. You can look at the single dose study for a summary of side-effects at this dosage level. Remember that there is no evidence to suggest that dosing with a senolytic treatment like the dasatinib and quercetin combination frequently will achieve any better result than dosing once every few years: the treatment kills the senescent cells it is capable of killing, and until more of those cells are created in significant numbers, then more of the treatment will most likely do nothing helpful. There are apparently senolytic self-experimenters out there taking dasatinib regularly; I think this shows a poor understanding of the situation, and is probably harmful.
Because there are no published senolytic studies using venetoclax, coming to some kind of ballpark human dose for that purpose involves analogy and educated guesswork. The approach is to compare cancer studies of navitoclax and venetoclax, of which there are many, and then scale the venetoclax cancer study dose down in accordance with the difference between the cancer and senolytic study doses of navitoclax. This is far from ideal, but I'm including this discussion here to point out exactly why one should only choose pharmaceuticals with animal and human senolytic data; as soon as any of that data is absent, there is all too much trial and error and guesswork involved. It is far better to wait rather then venture into the complete unknown, given that more data and better alternative senolytics will emerge in the years ahead.
Firstly, the senolytic dosage for navitoclax in mice from the 2015 study is 50 mg/kg daily for two periods of 7 days spaced 14 days apart - one thing you'll notice fairly quickly in all of this data is that BCL-2 family inhibitors compare unfavorably to dasatinib in terms of the amount needed and duration of treatment. That translates to a 60kg human dose of around 250mg.
A good place to start researching comparative dosages for venetoclax and navitoclax is the 2015 summary paper "ABT-199 (venetoclax) and BCL-2 inhibitors in clinical development". From there, and the references, the chemotherapeutic human dose of navitoclax was settling to somewhere in the 200-300 mg per day range for 14 to 21 days before it was discarded in favor of other tools, with the upper end of that dose range producing the aforemention ugly side-effects related to platelet loss. Dosage for human cancer trials of venetoclax is, on the other hand, all over the map: doses range from 200mg to 1200mg daily carried out over a period of a few weeks to a month, with the dose given cycling in sometimes complex ways. To keep things simple, one point of comparison is to look at the trials versus chronic lymphocytic leukemia for navitoclax and for venetoclax, as they are quite similar. For navitoclax the tested doses ranged from 100mg to 300mg - essentially the same as the senolytic dosage. For venetoclax, the tested doses were 200mg to 1200mg. In both cases, these were daily doses taken for a period of a week or more.
So if one were forced to put a pin into the map based on these numbers, the senolytic human dosage of venetoclax would probably be in the 400-600 mg/day range, every day for a week. Note that this is firmly in chemotherapy with side-effects territory, and there is no direct supporting evidence for effectiveness in mammals whatsoever. All said and done, the rough back of the envelope estimation comes to a result that looks very unattractive, given that we expect better options in the future.
Alvespimycin is an HSP90 inhibitor, and this class of pharmaceutical may produce senolytic effects through less direct BCL-2 family inhibition. Certainly the effects and side-effects look broadly similar to those of navitoclax and venetoclax. One thing to note when researching this compound is that it is used in both injected and oral forms, so one has to be careful to work with only the papers that cover the oral delivery mode, at least in the present context. The senolytic study in mice used an oral dose of 10 mg/kg provided 3 times every other day. This scales up to something like 50mg with the same dosage schedule for a 60kg human.
For comparison, looking at a mouse cancer study, the researchers there used dosages in the 5-15 mg/kg, with a variety of daily and intermittent schedules. A human cancer study also used a wide range of doses, and is a good resource if you are interested in reading up on the potential side-effects. Based on their data, the authors recommended either 40mg every other day or 20mg daily for a period of four weeks of every six weeks for the follow on study.
From these numbers, a human senolytic treatment of three 50mg doses on alternate days once again sounds like something that veers into full blown chemotherapy territory, just not to the same degree as venetoclax above. Anyone considering this would have to make their own decision about risk, as there just isn't enough information out there to talk sensibly about risk and side-effects resulting from a much shorter exposure than was carried out in the human cancer studies.
Use Small Test Doses Prior to Any Study
Near all studies of chemotherapeutics start with low doses, a tenth of the expected study dose. Near all studies report a couple of patients who experience enough of a reaction to the chemotherapeutic at those low doses to drop out or require adjustment of the protocol. If risking chemotherapeutics yourself for senolytic purposes, even single doses or doses for a short time only, it is still important to first test a low dose at a tenth of the desired level of so, to help ensure that there is no adverse reaction. As is true for all of the rest of the considerations here, if you try this, it is entirely your own responsibility to identify, understand, and manage the risks involved.
Verify All of the Above
Assume that anything written anywhere other than the primary materials might be incorrect or misleading. Do not take my word for any of the above information; chase down the primary sources, run the numbers, and make the judgement calls yourself. Is it foolish to self-experiment with chemotherapeutics rather than waiting for better information from human trials or some better form of treatment to emerge? Only you can answer that question, and only you are responsible for any consequences resulting from the answer.
For individuals without suitable connections, the easiest way to obtain pharmaceuticals is to order them from manufacturers in China or other overseas locations. As noted at the outset of this post, this is illegal - it would be an act of civil disobedience carried out because the laws regarding these matters are unjust, albeit very unevenly enforced. Many people regularly order pharmaceuticals from overseas, with and without prescriptions, for a variety of economic and medical reasons, and all of this is illegal. The usual worst outcome for individual users is intermittent confiscation of goods by customs, though in the US, the FDA is actually responsible for this enforcement rather than the customs authorities. Worse things can and have happened to individuals, however, even though enforcement is usually targeted at bigger fish, those who want to resell sizable amounts of medication on the gray market, or who are trafficking in controlled substances. There is a fair amount written on this topic online, and I encourage reading around the subject.
Open a Business Mailbox
A mailbox capable of receiving signature-required packages from internal shipping concerns such as DHL and Fedex will be needed. Having a business name and address is a good idea. Do not use a residential address.
Use Alibaba to Find Manufacturers
Alibaba is the primary means for non-Chinese-language purchasers to connect to Chinese manufacturers. The company has done a lot of work to incorporate automatic translation, to reduce risk, to garden a competitive bazaar, and to make the reputation of companies visible, but it is by now quite a complicated site to use. It is a culture in and of itself, with its own terms and shorthand. There are a lot of guides to Alibaba out there that certainly help, even if primarily aimed at retailers in search of a manufacturer, but many of the specific details become obsolete quickly. The Alibaba international payment systems in particular are a moving target at all times: this year's names, user interfaces, and restrictions will not be the same as next year's names, user interfaces, and restrictions.
Start by searching Alibaba for suppliers of the senolytic pharmaceuticals of interest. There are scores of resellers and manufacturing biotech companies in China for any even somewhat characterized pharmaceutical or candidate pharmaceutical. Filter the list for small companies, as larger companies will tend to (a) ignore individual purchasers in search of small amounts of a compound, for all the obvious economic reasons, and (b) in any case require proof of all of the necessary importation licenses and paperwork. Shop around for prices - they may vary by an order of magnitude, and it isn't necessarily the case that very low prices indicate a scam of some sort. Some items are genuinely very cheap to obtain via some Chinese sources.
Many manufacturers will state that they require a large (often ridiculously large) minimum order; that can be ignored. Only communicate with gold badge, trade assurance suppliers with several years or more of reputation and a decent response rate. Make sure the companies exist outside Alibaba, though for many entirely reliable Chinese businesses there are often sizable differences between storefronts on Alibaba, real world presence, and the names of owners and bank accounts. Use your best judgement; it will become easier with practice.
Arrange Purchase and Shipping via Alibaba
Given the names of a few suppliers, reach out via the Alibaba messaging system and ask for a quote for a given amount of the senolytic pharmaceutical in question. Buy twice what you'll think you need, as some of it will be used to validate the identity and quality of the compound batch, and buy that much from at least two different suppliers present in widely separated regions. Payment will most likely have to be carried out via a wire transfer, which in Alibaba is called telegraphic transfer (TT). Alibaba offers a series of quite slick internal payment options that can be hooked up to a credit card or bank account, but it is hit and miss whether or not those methods will be permitted for any given transaction. Asking the seller for a pro-forma invoice (PI), then heading to the bank to send a wire, and trusting to their honesty is good enough for low cost transactions. It should work just fine when dealing with companies that have a long-standing gold badge.
To enable shipping with tracking via carriers such as DHL, the preferred method of delivery for Chinese suppliers shipping to the US or Europe, you will need to provide a shipping address, email address, and phone number. Those details will find their way into spam databases if you are dealing with more than a few companies, and will be, of course, sold on by Alibaba itself as well. Expect to see an uptick of spam after dealing with suppliers via Alibaba, so consider using throwaway credentials where possible.
Chinese manufacturers active on Alibaba are familiar with international shipping practices, and smaller companies will, on their own initiative, apply whatever description to packages will most likely get it past customs. Since declared pharmaceuticals may well be taken aside and confiscated, the description will therefore not involve pharmaceuticals. This is as much motivated by dealing with customs at the Chinese end as pushing things past the US authorities; it is again a form of widespread civil disobedience that reflects a popular disdain for petty laws and regulators where they act as impediments to useful activity.
Quercetin is a Supplement, Buy it at the Store
Any specialist vitamin store will sell quercetin, or at worst it can be ordered online from any reputable retailer.
Dasatinib, venetoclax, and similar compounds are manufactured as fairly resilient powders and then formed into pills where sold as medications. In powder or pill form, put them in airtight containers in a fridge, and they have a shelf-life of a few years; the specific storage recommendations are easy enough to find online. The same is true of quercetin. This is one of the big advantages of most ingested pharmaceuticals versus injected pharmaceuticals; they are comparatively low-maintenance, stable, and long-lasting. That in turn means less logistical planning and effort.
A senolytic compound may have been ordered, but that doesn't mean that what turns up at the door is either the right nondescript powder or free from impurities or otherwise of good quality. Even when not ordering from distant, infrequent suppliers, regular testing of batches is good practice in any industry. How to determine whether a compound is what it says it is? Run the compound through a process of liquid chromatography and mass spectrometry, and compare the results against the standard data for a high purity sample of that compound. Or rather pay a small lab company to do that.
Obtain the Necessary Equipment
Since this process will involve weighing, dividing up, and shipping powders in milligram amounts, a few items will be necessary: spatulas or scoops for small amounts of a substance; a reliable jeweler's scale such as the Gemini-20; sealable vials; small ziplock bags; labels; and shipping and packing materials. All of these are easily purchased online. The recommended shipping protocol is to triple wrap: a labelled vial, secured within a ziplock bag and tape, and then enclosed within a padded envelope.
Use Science Exchange to Find Lab Companies
Science Exchange is a fairly robust way to identify providers of specific lab services, request quotes, and make payments. Here again, pick a small lab company to work with after searching for LC-MS (liquid chromatography and mass spectrometry) services. Large companies will want all of the boilerplate registrations and legalities dotted and crossed, and are generally a pain to deal with in most other ways as well. Companies registered with Science Exchange largely don't provide their rates without some discussion, but a little over $100 per sample is a fair price for LC-MS to check the identity and purity of the compound.
Work with the Company to Arrange the Service
The process of request, bid, acceptance, and payment is managed through the Science Exchange website, with questions and answers posted to a discussion board for the task. Certainly ask if you have questions; most providers are happy to answer questions for someone less familiar with the technologies used. Service providers will typically want a description of the compounds to be tested and their standard data sheets, as a matter of best practice and safety. It is good enough to provide the name for established pharmaceuticals, as the data sheets, mass spectrometry profiles, and other detailed information are freely available online from databases such as DrugBank.
Ship the Samples
Measure out 50mg or so from each separate order as a distinct sample, label it carefully, make sure you have a record linking the sample label to the specific supplier, and package it up. More in the sample is better than less, as several attempts might be needed to get a good result out of the machines used, but each attempt really only needs a very tiny amount of the compound. Ship the sample via a carrier service such as DHL, UPS, or FedEx. Some LC-MS service companies may provide shipping instructions or recommendations. These are usually some variety of common sense: add a description and invoice to the package; reference the order ID, sender, and receiver; clearly label sample containers; and package defensively with three layers of packing; and so forth.
Examine the Results
Once the LC-MS process runs, the lab company should provide a short summary regarding whether or not the compound is in fact the correct one and numbers for the estimated purity. Also provided are the mass spectra, which can be compared with the standard spectra for the compound, which can be found at DrugBank or other sources online.
To match to the way in which ingested compounds are taken in most studies, in pill form, it is probably best to make up pill capsules rather than just, for example, taking a measure of the powder in water or wrapped in bread. This is fairly easy to manage, given the tools already obtained for measuring out small powder samples. Specialist vitamin stores, and a range of other vendors, sell empty gelatin pill capsules for supplement enthusiasts, and they will do just fine here. Putting powder into capsules is a fiddly business that only becomes more frustrating with age; I'd suggest trying it out with flour if you haven't done this before. It is a lot harder than one might think. Fortunately, there are a variety of simple, inexpensive tools to help with that; references and video guides are easy to find by searching online. At the very least, unless you happen to have three hands, a capsule holding tray is essential, and I'd recommend some form of small powder funnel.
Unfortunately there is no established, proven, useful test that can directly assess senescent cell level in humans or human biopsies. It is possible to use immunohistochemistry to assess cellular senesence in tissue samples, which is a standard approach in animal studies, but no-one appears to have yet validated that in humans, given biopsies taken from a living individual. Since senescent cells are generated temporarily by wounding, it is quite possible that anything that starts with a biopsy will prove to be unhelpful as a before and after comparison measure for senolytic trials - the levels measured may not bear any resemblance to the normal levels absent a wound.
Without a direct measure, we must fall back on indirect assessments of the detrimental effects of senescent cells. The objective here is a set of tests that anyone can run without the need to involve a physician, as that always adds significant time and expense. Since we are really only interested in the identification of large and reliable effects as the result of an intervention, we can plausibly expect a collection of cheaper and easier measures known to correlate with age to be useful. Once that hill has been climbed, then decide whether or not to go further - don't bite off more than is easy to chew for a first outing.
From an earlier exploration of likely tests, I picked the following items on the basis of a likely connection to the actions of senescent cells, reasonable cost and effort, and ability to carry out the test without a physician's office being involved. Note that this does rule out, to pick one example, the interesting and relevant examination of kidney and liver function, as it would have to be carried out via the radioactive tracer methods of nuclear medicine to obtain decent results. That leaves the tests below quite focused on (a) the cardiovascular system, particularly measures influenced by vascular stiffness, and (b) inflammatory and other markers in the bloodstream:
- A standard blood test, with inflammatory markers.
- Resting heart rate and blood pressure.
- Heart rate variability.
- Pulse wave velocity.
- Biological age assessment via DNA methylation patterns.
The cardiovascular health measures in that list are those that are impacted by changes in the elasticity or functional capacity of blood vessels, such as would be expected to occur to some degree following any rejuvenation therapy that addresses senescent cells, chronic inflammation, or other factors that stiffen blood vessels, such as calcification or cross-linking. Positive change of the average values in most of these metrics are achievable with significant time and effort spent in physical training, so movement in the numbers in a short period of time as the result of a treatment should be an interesting data point.
There exist online services such as WellnessFX where one can order up a blood test and then head off the next day to have it carried out by one of the widely available clinical service companies. Of the set of test packages offered by WellnessFX, the Baseline is probably all that is needed for present purposes. But shop around; this isn't the only provider.
Resting Heart Rate and Blood Pressure
A simple but reliable tool such as the Omron 10 is all you need to measure heart rate and blood pressure. It is worth noting here a couple of general principles for cardiovascular measures. Firstly, the further away from the center of the body that the measurement is taken, the less reliable it is - the more influenced by any number of circumstances, such as position, mood, stress, time of day, and so forth. Fingertip devices are convenient, but nowhere near as useful as something like the Omron 10 that uses pressure on the upper arm. Secondly, all of the above-mentioned line items also influence every cardiovascular measure, so when you are creating a baseline or measuring changes against that baseline, carry out each measure in the same position, at the same time of day, and make multiple measurements over a week to gain a more accurate view of the state of your physiology. The Omron 10 is solid: it just works, and seems quite reliable.
Heart Rate Variability
Surprisingly few of the numerous consumer tools for measuring heart rate variability actually deliver the underlying values used in research papers rather than some form of aggregate rating derived by the vendor; the former is required for any serious testing, and the latter is useless. Caveat emptor, and read the reviews carefully. As an alternative to consumer products, some of the regulated medical devices are quite easy to manage, but good luck in navigating the system to obtain one. The easiest way is to buy second hand medical devices via one of the major marketplaces open to resellers, but that requires a fair-sized investment in time and effort - which comes back to the rule about keeping things simple at the outset.
After some reading around the subject, I settled on the combination of the Polar H10 device coupled with the SelfLoops HRV Android application. I also gave the EliteHRV application a try. Despite the many recommendations for Polar equipment, I could not convince either setup to produce sensible numbers for heart rate variability data: all I obtained during increasingly careful and controlled testing was a very noisy set of clearly unrealistic results, nowhere near the values reported in papers on the subject. However, plenty of people in the quantified self community claim that these systems work reasonably well, so perhaps others will have better luck than I. Take my experience as a caution, and compare data against that reported in the literature before investing a lot of time in measurement.
Pulse Wave Velocity
For pulse wave velocity, choice in consumer tools is considerably more limited than the sitation for pulse wave velocity. Again, carefully note whether or not a device and matching application will deliver the actual underlying data used in research papers rather than a made-up vendor aggregate rating. I was reduced to trying a fingertip device, the iHeart, picked as being more reliable and easier to use than the line of scales that measure pulse wave velocity. Numerous sources suggest that decently reliable pulse wave velocity data from non-invasive devices is only going to be obtained by measures at the aorta and other core locations, or when using more complicated regulated medical devices that use cuffs and sensors at several places on the body.
Still, less reliable data can be smoothed out to some degree by taking the average of measures over time, and being consistent about position, finger used for a fingertip device, time of day, and so forth when the measurement is taken. It is fairly easy to demonstrate the degree to which these items can vary the output - just use the fingertip device on different fingers in succession and observe the result. All of this is a trade-off. A good approach is to take two measures at one time, using the same finger of left and right hand, as a way to demonstrate consistency. While testing an iHeart device in this way, I did indeed manage to obtain consistent and sensible data, though there is a large variation from day to day even when striving to keep as many of the variables as consistent as possible. That large variation means that only sizable effects could be detected.
DNA methylation tests can be ordered from either Osiris Green or Epimorphy / Zymo Research - note that it takes a fair few weeks for delivery in the latter case. From talking to people at the two companis, the normal level of variability for repeat tests from the same sample is something like 1.7 years for the Zymo Research test and 4.8 years for the Osiris Green tests. The level of day to day or intraday variation between different samples from the same individual remains more of a question mark at this point in time, though I am told they are very consistent over measures separated by months. Nonetheless, as for the cardiovascular measures, it is wise to try to make everything as similar as possible when taking the test before and after a treatment: time of day, recency of eating or exercise, recent diet, and so forth.
An Example Set of Daily Measures
An example of one approach to the daily cardiovascular measures is as follows, adding extra measures as a way to demonstrate the level of consistency in the tools:
- Put on the Polar H10; this is involved enough to increase heart rate a little for a short period of time, so get it out of the way first.
- Sit down in a comfortable position and relax for a few minutes.
- Measure blood pressure and pulse on the left arm using the Omron 10.
- Measure blood pressure and pulse on the right arm using the Omron 10.
- Measure pulse wave velocity on the left index fingertip over a 30 second period using the iHeart system.
- Measure pulse wave velocity on the right index fingertip over a 30 second period using the iHeart system.
- Measure heart rate variability for a ten minute period using the Polar 10 and Selfloops.
Consistency is Very Important
Over the course of an experiment, from first measurement to last measurement, it is important to maintain a consistent weight, diet, and level of exercise. Sizable changes in lifestyle can produce results that may well prevent the detection of any outcome resulting from a first generation senolytic pharamaceutical using the simple tests outlined here. Further, when taking any measurement, be consistent in time of day, distance in time from last exercise or meal, and position of the body. Experimentation with measurement devices will quickly demonstrate just how great an impact these line items can have.
The costs given here are rounded up for the sake of convenience, and in some cases are blurred median values standing in for the range of observed prices in the wild.
- Business mailbox, such as from UPS: $250 / year
- Baseline tests from WellnessFX: $220 / test
- MyDNAage kits: $310 / kit
- Osiris Green sample kits: $70 / kit
- Omron 10 blood pressure monitor: $80
- Polar H10 heart monitor: $100
- iHeart monitor: $210
- American Weigh Gemini-20 microscale: $90
- Miscellenous equipment: spatulas, labels, vials, pill capsules, etc: $60
- 2 x 2g orders of dasatinib via Alibaba: $300
- 2 x 5g orders of venetoclax via Alibaba: $1300
- 2 x 2g orders of alvespimycin via Alibaba: $400
- Store-purchased quercetin capsules: $10
- Shipping and LC-MS analysis of samples: $120 / sample
One might expect the process of discovery, reading around the topic, ordering materials, and validating the pharmaceuticals to take a couple of months. Once all of the decisions are made and the materials are in hand, pick a start date. The schedule for the self-experiment is as follows:
- Day 1-10: Once or twice a day, take measures for blood pressure, pulse wave velocity, and heart rate variability.
- Day 10: Bloodwork and DNA methylation test.
- Day 11-15: Test a 1/10 dose of the senolytic compounds used, one by one, and abandon the effort if issues are experienced.
- Day 16: Start to carry out the program of treatment.
- Day 31-40: Repeat the blood pressure, pulse wave velocity, and heart rate variability measures.
- Day 40: Repeat the bloodwork and DNA methylation test.
The exact timing is not really important, but it is a good idea to allow enough time following the end of the dosage for things to settle down. In animal studies, senolytic effects occurred fairly rapidly, as did the benefits, but allowing a few weeks of time in a human self-experiment still sounds like a good idea. Certainly it costs nothing to take that step.
If you run a self-experiment and keep the results to yourself, then you helped only yourself. The true benefit of rational, considered self-experimentation only begins to emerge when many members of community share their data, to an extent that can help to inform formal trials and direction of research and development. There are numerous communities of people whose members self-experiment with various compounds and interventions, with varying degrees of rigor. One can be found at the LongeCity forums, for example, and that is a fair place to post the details and results of a personal trial with senolytics. Equally if you run your own website or blog, why not there?
When publishing, include all of the measured data, the compounds and doses taken, duration of treatment, and age, weight, and gender. Fuzzing age to a less distinct five year range (e.g. late 40s, early 50s) is fine. If you wish to publish anonymously, it should be fairly safe to do so, as none of that data can be traced back to you without access to the bloodwork provider. None of the usual suspects will be interested in going that far. Negative results are just as important as positive results. For example, given the measures proposed in this post it is entirely plausible that positive changes as a result of present senolytic treatments in a basically healthy late 40s or early 50s individual will be too small to identify - they will be within the same range as random noise and measurement error. Data that confirms this expectation is still important and useful for the community, as it will help to steer future, better efforts.
Given all of the cautions above, why not wait? Waiting can be a very sensible strategy. The state of senolytic therapies is progressing rapidly. New and less chemotherapeutic senolytics are emerging, such as FOXO4-DRI. At some point in the next few years, reliable direct tests for senescence will arrive on the scene, allowing a much better view of whether or not these treatments are actually achieving the claimed results. That said, it doesn't hurt to plan, and it doesn't hurt to tinker with some of the component parts of a plan. That is how we can determine whether or not it is worthwhile to experiment now versus waiting to experiment later with better tools.