Planning a Single Person Trial of Senolytic Drug Candidates

This post should be considered as part of an ongoing and yet to be concluded process of thinking out loud on the topic of self-experimentation with senolytic drug candidates. These are compounds that to some degree selectively destroy senescent cells in animal studies. Some have been shown to have positive effects in animal studies of various sorts in the years prior to the present wave of interest in senescent cell clearance, and some of those effects might be plausibly linked to removal of senescent cells. Some were tested as cancer therapeutics, or analgesics, or for other uses. Some have serious and harmful side effects, as is the case for most prospective chemotherapeutics. They are intended to destroy cells, and they are nowhere near as discriminating as we'd all like them to be. Nonetheless, all of these drug candidates are to varying degrees available for purchase, and thus available for self-experimentation.

Now, self-experimentation has a long and storied history in the scientific community. Many noted researchers at some point obtained the first human data from their own bodies, and that seems to me the most ethical of approaches: the researcher assumes the risks. Setting aside for a moment the question of risk, the point to take away from this history is that there is absolutely no point in doing this unless you measure and publish what you did and what happened. Guessing at outcomes or using drug candidates merely in the hope that effects will carry over from studies in mice helps no-one. The same goes for picking easily measured outcomes just because they are easy to measure. The objective here is to learn something and transmit that learning, which is possible even in an environment of single person tests without controls, provided we are seeking effects that are both large and reliable, and provided we go about this is a sensible manner. In this context, self-experimentation can help to point the way for those with the resources to run more rigorous experiments capable of better quantifying effect size, optimal dosage, and the like.

Obtain a Cooperative Physician

The first step is to ensure that you have a physician who understands what you are intending to do and achieve, and is willing to order up the required tests. You will need an interface and guide to the local medical establishment, especially for the more expensive scanning and testing. This usually isn't all that hard to obtain, since you'll be paying.

Obtain a Cooperative Laboratory Company

You will need a company to act as an interface with suppliers, as many of them will not accept orders for senolytics from individuals. In this age of drug prohibition, it also smooths the way for biochemical deliveries across national borders for them to be between laboratory companies. You will also need a company with laboratory resources, or that can act as an interface to laboratory services for some of the work you might want carried out. The ideal situation here is to work with someone within the community, via your connections, as it would otherwise require some legwork to find a company willing to work with you.

Determine the Health Metrics to be Assessed

The ideal set of data desired at the end of a short test of senolytics includes (a) the degree to which senescent cells were removed, and (b) the degree to which relevant measures of aging were reversed. The reality is that both require some compromises given the current state of medical testing. After some reading around and thinking on what would likely be affected by cellular senescence, given what is presently known, I settled on the following tests for consideration. One important item is that the normal values obtained from healthy individuals for a given test must vary to a large enough degree across the age range of 30 to 60 to make it useful to run the test if you are something other than very old. This is definitely not true for as many of the available tests as you might think would be the case.

Firstly, there is standard bloodwork and urinalysis. This is actually not all that likely show anything interesting if comparing before and after measures, especially in people who are not in their 60s or later, but it is cheap and a useful demonstration to show that nothing terrible took place. Further, some of the measures in bloodwork are needed for other parts of the testing. In particular, it is possible to see indications of tumor lysis syndrome resulting from senescent cell destruction. If there is a characteristic change in such measures immediately following use of a senolytic drug, it is an indication that something is happening, which is useful evidence.

When looking at liver function, none of the values obtained from normal bloodwork are particularly helpful. The numbers for normal function don't vary enough with age, and do vary a fair amount with circumstances and lifestyle choices. However, hepatobiliary scintigraphy results do change characteristically with age. This is a nuclear medicine procedure involving use of a radioactive tracer, so expect to pay accordingly.

For kidney function, the desired measure is glomular filtration rate. Now there are numerous ways of obtaining this result. There is the direct and expensive nuclear medicine approach with tracers, but also estimated approaches using data obtained from standard bloodwork. There are a number of resources that explain the differences in some detail, such as a PDF from the National Kidney Foundation. The estimated approaches suffer from various degrees of inaccuracy for the levels one would expect to find in a healthy individual, sad to say. The MDRD Study equation method should not be used, but the alternative CKD-EPI equation seems worth trying.

Given the evidence for a relationship between cellular senescence and calcification of blood vessels, calcium scans and scoring at first seem interesting. This is especially the case since it is apparently very hard to reduce a calcium score; it is something achieved only gradually over years, and with great attention to lifestyle changes. Calcium scans are just a standard CT scan followed by semi-automated analysis, producing an Agatston score or lesion-specific calcium score. Unfortunately, even later in life a large percentage of people score zero - as many as half or more in the late 40s and early 50s, for example. There is an online calculator from one of the research groups involved in this work if you are interested in exploring the numbers. All of this makes calcium scoring nowhere near as helpful as it might be, given the cost of a CT scan. It is probably only worth trying for people in their 70s and later, or who already have a score to hand and know it is non-zero.

Tests in lung tissue suggest that removal of senescent cells can somewhat reverse loss of tissue elasticity. So it seems worth looking at measures of skin elasticity. These can be obtained using cutometer or ballistometer commercial devices, with a number of papers commenting on reliability of the results. You might have to find a plastic surgeon or one of those dubious anti-aging clinics, however, rather than a standard dermatology practice. Possibly more useful is the indirect measure of blood vessel elasticity via pulse wave velocity, which is an easy test to arrange, and which does have a significant degree of change over the middle years of life. The question there, as with all matters cardiovascular, is the degree to which normal readings change because of primary (including the effects of senescent cells) versus secondary (weight gain and lack of exercise) causes of aging. The testing that is being accomplished here is as much of the relevance of the tests as it is of the effects of senolytic therapies. For that and other reasons, you can't just pick one test.

Another cardiovascular measure with a useful profile of changes over time is heart rate variabilility. Measurement here is also easily arranged. Of note, the Palo Alto Prize founders chose heart rate variability as their measure of aging for the interventions produced by competing teams.

Biomarkers of aging based on DNA methylation are well on the way towards becoming a practical possibility these days, though there is as yet no one consensus approach that everyone agrees upon. Nonetheless, Osiris Green is offering a DNA methylation biomarker of aging implementation at an affordable price. This is cheap enough to put into contention, even though there is as much validation of the test needed as validation of senolytics.

If you can stretch to custom lab work, then it is worth looking into the existing cellular senescence tests, or the skin sample test noted this morning, both of which require a biopsy. In the former case there are kits and the tests are well established, at least in the research community, with a question mark on how the biopsy process will interact with the role of cellular senescence in wound healing to make the results unhelpful. In the latter case, the paper provides enough details for someone to repeat the protocol, but it is anyone's guess as to how useful it will be in practice. This is another case where calibrating the test is as much the goal as calibrating the effects of a senolytic.

Pick the Senolytic Drug Candidates

Right at the start, let us throw out dasanitib, navitoclax, and similar items targeting the Bcl-2 family. They are comparatively indiscriminate chemotherapeutics, and almost everything else that the research community has identified as a potential senolytic drug is better, judging from the animal data: either more discriminating, less harmful, or both. Of the remaining compounds, it makes sense to try a combination, as some studies have suggested synergies exist between drug candidates, or that different senolytics work on different types of senescent cell. Also, the academic and corporate studies will not at the outset tend to run trials for drug combinations. It is better to raise the odds of finding interesting new data.

The compounds that seem worth looking into fall into two categories. The first are easily obtained supplement-like items that are comparatively cheap, taken orally, and well characterized for safety. In this category are fisetin and quercetin, though there is some debate over whether or not the latter is in fact senolytic. The second are more recently identified senolytics that are less easily obtained and used, in some cases with little to no human data on safety and usage, but that seem promising given recent research. Here, I'd include piperlongumine and FOXO4-DRI. In each case, you would want to read around on what is known of the pharmacology, the studies that used the compound, current thinking on how it works, and make a call on whether or not you are willing to take the risk of trying it. This will certainly involve digging through research papers, and will certainly be an individual choice. Don't blindly follow anyone's recommendations: choose for yourself.

Establish Dosage and Schedule

Figuring the likely dose for a human study involves reading the existing literature on animal studies to find the most relevant dose used there, usually expressed in mg/kg, and then multiply accordingly. You will quickly find that for most senolytics there is no easy way to come to a recommended dose, and you'll be forced to use your best judgement. For example, piperlongumine has so far only been studied in cell cultures for its effects on senescent cells. Looking at the literature, it was tested as an analgesic at levels of 1-250 mg/kg, for cancer suppression at 2.5-5 mg/kg, for sensitizing cancer to other treatments at 1 mg/kg, and for more direct cancer ablation at 2.5 mg/kg. In some cases these were single doses followed by an assessment, in others they continued for weeks.

Similarly for fisetin, there are no published animal studies for effects on senescent cells. For other purposes in past years, however, you'll find data on the pharmacokinetics for doses of 10-250 mg/kg, another study providing 10-45 mg/kg, twice a day for weeks, and yet another for cancer suppression at 5 mg/kg twice over a period of a few weeks.

For quercetin, one can look at the original study identifying it as senolytic to see that the researchers used a single dose of 50 mg/kg. For FOXO4-DRI, there is a very little data beyond the one recent study announcing its effects and another equally recent focused on cancer. Both are paywalled and unfortunately the details of the dosage are not in the main body of the original paper, but rather in the supplemental materials that I've yet to obtain. Still, it is there for consideration when I get to it.

Bear in mind that you are certainly going to want to try a very tiny dose at the outset, and then work your way up to the final dose. This precaution is only sensible and is done for a variety of reasons. In some cases these senolytic compounds are poorly or not at all tested in humans. Secondly, how certain are you that the suppliers did everything absolutely correctly, and that the testing of their compounds worked as desired? Further, if trying combinations yet to be tested in any published paper, there is always the possibility of unforeseen interactions. Lastly, if things actually work well and you started out with a high load of senescent cells, you do have to worry about the possibility of tumor lysis syndrome due to too many cells dying at once. All of these are very good reasons to ease into the desired dosage over time.

There is very definitely a spectrum of safety in the compounds I've mentioned here, from quercetin (sold in stores, manufactured by many supplement companies, in existence for years) through to FOXO4-DRI (comparatively new, barely manufactured at all, must be custom ordered, with no published human data, and only a couple of papers for animal studies). When you pick your poison, do so in full knowledge of the level of risk you undertake.

Figure out the Logistics

Quercetin and fisetin are things you put in bottles on a shelf and can leave there for months. You take the pills orally. That is all pretty easy. Piperlongumine requires freezer storage, and possibly powdering or compounding to be taken orally. FOXO4-DRI is a short lifespan protein, must be keep in freezer storage, then reconstituted and given via injection: intraperitoneal injection in mice, but most likely intravenous injection would be the most desirable option in humans. If you are familiar at all with how diabetics manage their insulin supplies, the situation is very similar.

Management of injection logistics is something that you want a lab company and probably a physician to help with, rather than embark upon it alone. In this context it is very much worth noting that, given the drug war nonsense that has gripped the world these past few decades, you want to be careful as to how you go about obtaining needles for any compounds that must be injected. This is another good reason to arrange everything in conjunction with a friendly lab company and physician.

Determine Suppliers and Order Products

Finding suppliers for the chosen senolytics varies considerably in difficulty. For quercetin, you walk across the street to pick up a few bottles from the nearest supplement store, and by going with a trusted brand can probably feel good about skipping the step of validating that the contents are what they say they are. Or you may be able to find an existing review of the supplier's products online. Fisetin can still be ordered in bottles, but here the number and quality of suppliers is more of an unknown, so the need to test the product comes into play.

For piperlongumine, you will be ordering from a chemical supplier and paying a considerable amount - hundreds of dollars for a single dose, going by the levels used in animal studies. For FOXO4-DRI, it is likely that the best course, given the very small number of suppliers, is to have it synthesized as a custom batch by a company that specializes in protein synthesis. This is expensive, and is where you will need the lab company. In both cases, suppliers will be reluctant to supply anyone they think is going to use it for human testing outside the formal trial system or a research institution.

Test the Products

You will also need the friendly lab company for the task of determining how to validate the quality of products when they arrive from the suppliers. Validation of quality is not a completely straightforward process, and may require digging up specialist services, which is better done through a company already in that ecosystem than to try it yourself. It is a matter of great importance to establish that you are getting what you pay for, both to avoid wasting the time and resources spent on the exercise of self-experimentation, as well as for reasons of personal safety. Even with the best of intentions, compounds that are not mass manufactured can have bad batches.

Run the Experiment

The first step is to run all the desired tests to obtain a set of initial baseline values. For many of these, such as standard bloodwork, it makes sense to run them twice, perhaps a few weeks apart, since numbers tend to vary with circumstances. Then follow the dosage schedule. Then run two more sets of tests, one a few days after the end of dosage, and one a month later. Precisely because many of the measures can vary with lifestyle, it is important to be consistent in your diet, exercise, and so forth across this period of time.

Then, once done, wrap it all up by publishing the data online for the community to look over.

Considering the Easy versus the Not So Easy Options

It should be apparent from reading the above notes and the linked materials that the choice of candidate senolytics and assays makes a big difference to the amount of work required to run a useful exercise in self-experimentation. It also makes a big difference to the level of personal risk undertaken. I picked the senolytics discussed in this post in part to make this point. To cut down to the easiest and safest level of self-experimentation, it would be possible to try only fisetin and quercetin and largely avoid the need for laboratory services, just relying upon a friendly physician to order bloodwork, cardiovascular, and other established tests. One could also be fairly confident that the risk of adverse effects in that scenario is lower than it is in the others. Sadly these are also the more dubious senolytic candidates; there is no such thing as a free lunch, it seems.


Speaking of FOXO4-DRI (which is patented) there are some limitations to patent law (at least here in Germany) which could be used as a loophole to use it legally without any license. These are:
- Patents are limited to 20 years (not very useful).
- Patents don't apply to scientific and private use. Scientist is not a protected title - everybody can claim to be one.
- Patents apply within state borders only. This is true for almost any law and regulation. You can do a lot of stuff offshore.
- Clinical trials are excempt from patents, which is quite interesting. So everybody could start one.
- There's also an exception for prescription drugs that are custom made in a pharmacy.

The 20 year rule is notable in the regard that the clock is ticking already. That's 19 years left to find an investor, do clinical trials, create a company and a market, pay off your investor and make some money yourself. Every year lost is missing at the end.

Posted by: Matthias F at May 22nd, 2017 5:16 PM

@Reason Thanks for this comprehensive guideline.

My original intent was to give Quercetin a try, in fact, I already ordered it. However, what I learned about Quercetin here over the course of the last few days, I got a caution.

I probably do a 7 days fasting instead, just drinking tea. I have done that before but without having any clue what senescent cells and senolytics are and not for that reason. Now with a better understanding, I will ask my doctor for a fresh blood test before I start and one after I'm done fasting.

Also, I might submit the results of my blood tests to

- even if they just try to predict your gender and chronologic age for now - to see if their model can recognize any difference.

Posted by: bardu at May 22nd, 2017 9:53 PM

You may not want to brush off blood work and liver measurements. I know you're looking for benefits, but it's also possible that the drugs will cause some toxicity. While they may not change with age, liver enzymes are a good readout of toxicity and are important for clinical testing of new drugs.

Posted by: mike at May 23rd, 2017 3:25 AM

The simplest and most affordable experiment would be trying fisetin, doesn't it? Finding the right dosage for a human seems to bump the price. But I would try this.

Posted by: Incongruous at May 24th, 2017 12:13 PM

I reckon getting a lab to produce a peptide such as foxo4 dri would be both difficult and expensive, which is why the home brew approach suggested by Matthius on the other blog post is attractive.

Posted by: Jim at May 24th, 2017 9:43 PM

Simple? Sort of, aside from figuring out (read: wildly guessing) a dose from a purely in vitro study of what is likely an extensively-metabolized molecule. Affordable: probably, depending on where your blind dart-chuck lands.

But sensible, no. All the other molecules here have been shown to have senolytic effects in vivo; we have nothing but Petri dish results for fisetin's senolytic effect as of yet. And as noted above, previous results in vitro from the same group with a closely-related compound (quercetin) have proven challenging for two independent groups. It may be significant that the Mayo group did not in vivo results with quercetin alone, but only in combination with dasatinib; the reported results could therefore be attributable to dasatinib alone, or quercetin might enhance dasatinib's effects but have no benefit as a monotherapy.

Please don't say "We-e-ell, it can't hurt, and it might help: it's natural" ;) .

Posted by: Michael at May 24th, 2017 10:16 PM


It's still possible to ask a lab for an offer. Some of them have price lists online. Prices range from 60$-80$ per amino acid for 100mg and 160$-200$/AA for 1000mg. The sequence (including targeting) has 46 AA, so that should be 3000$-4000$/100mg and 7000$-9000$/1000mg. Once you can split that between multiple people it's not such a big deal.

The big deal are the risks involved. Reason's plan has some good ideas already and a proper risk assessment might also be a good idea:
- Make a list of what might go wrong. Write it down. Ask an expert on it.
- Determine the likeliness and severity of each point in the list.
- Create a backup plan. How can the risk be avoided or detected early. What to do if it happens nonetheless.

I've seen people doing risky stuff. If you ever went e.g. skydiving - there are those that prepare properly and those that don't do it for a long time.

Posted by: Matthias F at May 25th, 2017 7:36 AM

Getting together a group of like-minded self-experimenters would perhaps motivate careful monitoring of effects/consequences and allow the group to figure out a way to obtain some of these hart-to-get therapies at a cheaper cost (per individual, anyway). I'm a graduate student in NYC with plenty of time on my hands, and would be very happy to get together with others in the vicinity to plan and organize something toward this effect.

Posted by: gheme at May 26th, 2017 11:55 AM

Post a comment; thoughtful, considered opinions are valued. Comments incorporating ad hominem attacks, advertising, and other forms of inappropriate behavior are likely to be deleted.

Note that there is a comment feed for those who like to keep up with conversations.