At Ending Age-Related Diseases 2021, Greg Fahy discussed TRIIM-X, the ongoing continuation of his original TRIIM trial, along with some surprising findings from it.
Script
I’m just going to try to update everybody on the TRIIM-X clinical trial. I will explain where that came from, just to make sure everybody has the same background in a second. I just wanted to introduce the talk by pointing out what my original inspiration was that drove what you’re going to see today.
That was a paper published in 1986 showing that it’s possible to regrow the thymus. As you can see on the upper panel on the left, if you intervene in the aging of the thymus at 16 months by implanting pituitary adenoma cells into an old rat, you can restore the proliferative capacity of both B and T cells at 16 months, compared to about an 80% or a 90% decline in those functions at the same age without intervention. You can intervene later in life as well, and partly regrow the thymus, although the results are not quite as good.
We think that thymus regeneration is an important aging intervention. It’s a strategic one, and it has potentially early clinical significance. As I think everybody in this audience knows, the thymus is the master gland of the immune system. It’s located in the chest cavity beneath the breastbone, and it withers away starting around the time of puberty, particularly until there’s very little functional mass left as you get to be even 25 to 50 years of age.
Unfortunately, this process, called thymic involution, depletes the functionality of T cells, and T cells are needed to protect you from infectious disease and cancer. Infectious disease and cancer, of course, are major sources of age-related morbidity and mortality as you can see in the examples to the right, showing your risk of dying from the flu, skyrocketing as you go above the age of 60, which corresponds to this crash in immune system functionality in the left figure. Also, the probability of developing cancer goes up with age, including prostate cancer, and the immune system may be able to protect us from that problem.
These difficulties seem to coincide with the collapse of the immune system, and so we think that thymus regrowth, regeneration, is imperative and could have major effects on human longevity. In that experiment, growth hormone was produced by these pituitary adenoma cells that were implanted into the mice. We’ve chosen to use growth hormone as well, for a number of reasons, but one of the reasons is that growth hormone is a known quantity, and drug repurposing has a higher probability of making it through the FDA approval process than an unknown chemical entity.
Growth hormone was approved for adult use in 1996. It has a good safety record. It has some side effects, but they’re mild and reversible, and it does many other positive things as Joao Pedro was mentioning yesterday: improving wound healing, various aspects of brain function, cartilage health, and as depicted in the figures, these profound regenerative features that involve regrowing not just the thymus but most of the other tissues in the body as well. Whereas, as you can see, in the left figure, lean body mass and organ mass tend to decline with age except for fat, growth hormone actually reverses those trends, and this may have implications for the production of senescence cells in fat.
Also, as Joao Pedro mentioned yesterday, he thought that growth hormone has benefits in the short run that might not be good in the long run. The beauty of our approach is that you don’t need to use growth hormone in the long run because T cells live at least 12 years. If you were to use growth hormone for a one-year course once every 12 years or so, that would be relatively safe and avoid any long-term effects.
We also chose growth hormone because it’s the only pathway with IGF-1 shown to reverse thymic involution in many species: rats, dogs, cats, and human immunodeficiency victims. As you can see, there are many other choices available to us, but most of these things are very exotic or undesirable compared to growth hormone, so we felt the growth hormone was the best choice.
Joao Pedro also mentioned a standard objection to growth hormone among biogerontologists, which is that it seems to be associated with accelerated aging rather than retarded aging. I think that this observation may be less significant than it seems at first glance, because the longevity mutants that are produced by knocking out growth hormone and IGF-1 signaling seem to, in part, be due to the alteration of the anatomy of the brain early in life.
That leads to a prevention of brain inflammation in growth hormone knockout animals, such as the Ames dwarf mouse, and you can see on the left upper figure that the control mouse at 18 months has an inflamed brain. These knockouts don’t, and the inflammation is associated with damage: you can see the inflammation going up in the middle there. When the inflammation is prevented, you reverse a lot of aging.
Unfortunately, all of this happens, it’s all programmed to happen early in life. You can reverse the Ames phenotype by giving growth hormone before puberty, but you cannot reverse it when you give the growth hormone after puberty. This mechanism of prevention by growth hormone has nothing to do with adults. It’s an artificial model, so I don’t think we should pay too much attention to that aspect. Another final thing about growth hormone and its safety is that it’s concerning that growth hormone produces IGF-1, which is the growth factor might have some relationship to cancer.
As you can see, if you actually give growth hormone to old, or adult animals, or humans, it does not shorten lifespan, it does not increase the incidence of cancer. Growth hormone does have one side effect that raises insulin, but this can be blocked with other age-opposing agents, such as DHEA and metformin, and you know pretty much about these agents. DHEA has a number of therapeutic effects against aging and so does metformin, in addition to having an insulin-lowering effect in both of these cases.
For all of these reasons and more, we started the TRIIM trial in 2015, that was completed in 2017 and was based at Stanford. You see cohort one in the middle and cohort two on the right. As you probably know, this involves a combination of growth hormone, DHEA, and metformin and involved nine men up between the ages of 51 and 65. We treated them with this cocktail four times a week in a very patient-specific way, and we published results in 2019.
As I think you know, and we’ll just breeze through this, we did see the regeneration of the thymus that we were expecting to see, we saw evidence for increased production of new T cells of various flavors, CD4 and CD8. We saw also a reduction in apparent risk factors for cancer, such as reduction in PD1, an improvement in PSA risk factor scores and an improvement in the lymphocyte-to-monocyte ratio as determined by CyTOF
We also saw a reduction in the inflammatory marker CRP, and we were able to more or less prevent the insulin increase that’s normally caused by growth hormone, although it was not perfect. Things that we did not expect to see were hair darkening, improved kidney function, and CyTOF evidence for possibly reduction in monocytes. The most significant thing, perhaps, that we saw, though, was unexpected, and that was a reversal of epigenetic aging.
Just to explain how we calculate that: everyone, if you have your epigenetic age measured by any so-called epigenetic aging clock, there will be an offset, there will be a difference between your epigenetic age and your chronological age. That’s that epigenetic age minus age at time zero. We call that the offset to your chronological age, or, as Steve Horvath calls it, the epigenetic aging acceleration, and this will vary from person to person.
It makes it pretty messy to look at data with people who have a variable baseline, so we subtract it off the baseline offset in all cases. We look at how this difference between epigenetic age and age changes over time, rather than what the original difference was and how that changed over time. You can see in panels A to D here what you probably already know, which is that we saw a reduction in epigenetic age using all four of the epigenetic aging clocks that were applied by Steve Horvath at UCLA. In each case, we reached a strong statistical significance, even though we only had nine people on the trial.
We summarized the overall concept of the study by relating changes in epigenetic age that we observed in TRIIM to what you would expect of normal aging and what you’d have in order to reach what is called Longevity Escape Velocity: essentially, being able to reverse aging by one year for every year that goes by, therefore escaping from aging. We did better than that. At the end of the trial, on average, people were one and a half years younger than they were before they started the trial, or two and a half years younger than they would have been if they had not entered the trial.
Those are the background results that you’ve seen before; here’s another result from the TRIIM trial that came about recently: Steve Horvath went back and applied another aging clock, a fifth aging clock, to the original TRIIM data recently. This clock is called the plasma PhenoAge clock, and it has a number of advantages.
Number one, you don’t have to wait until the end of the trial to analyze frozen cells; you can actually calculate this clock based on factors that you can measure quite cheaply from blood samples. All you need is basically a complete blood count, a high-sensitivity CRP score, and a comprehensive metabolic panel, and you basically have everything that you need in order to calculate this clock.
The clock was originally presented by Levine et al in the original PhenoAge paper, but there have been a couple of refinements since then to make the accuracy of the calculation greater. There’s actually a longevity calculator that Johnny Adam, that the LAGRG group, has made available, and I can make that available to anyone who’s interested in that as well. The basic result is that the plasma PhenoAge, which you can measure in real time, which is a huge advantage, added on to its closest cousin, which is the PhenoAge clock, which is shown in the left from the originally published TRIIM data, we can see the curves are remarkably similar, the p-values are remarkably similar. This provides us a potentially important new tool to track how we’re doing in real time rather than having to wait until the end of the trial.
We were pretty excited about the results of the trial, but there’s so much more to learn, and we have these outstanding questions that we’d love to see answered. What happens if the treatment lasts longer than 12 months? As you can see, in the left panel, the results change more rapidly between month 9 and month 12 than they change between month 0 and month 9. We would love to know how to extrapolate that forward, there’s all kinds of ways you could do it, you could do it with a linear extrapolation; turns out a quadratic fit gives you an R value of 1.0.
That may have no meaning, but that’s the best curve fit that we can find. According to that curve fit, if we had treated people for just another three months, we might have doubled the anti-aging effect of the treatment: gone back in time not just by two and a half years but five years, at least according to the average clock result, but we don’t know. We may be able to develop more information about that by extending the treatment in TRIIM-X, our follow-up trial.
Another thing we’d like to know is how long this effect lasts. The different clocks gave different results with respect to that; the GrimAge clock, which is the best predictor of longevity, didn’t show any diminution of the aging offset improvement that we saw with the treatment, but what would happen greater than 6 months after the end of the trial? This is the original data, mostly just going out 18 months or 6 months after the end of the trial.
The other question that we’d like to know is, if we re-treat somebody, can we add on to the previous epigenetic aging reversal that we already established? We’re beginning to get data on the second and third question, and we hope to eventually get data on the first question, and I will share some of the information we have on these second few questions in this talk.
For these reasons, and many others, including that we didn’t have women in the original trial, et cetera, we launched an extension of the TRIIM-X trial. It’s being extended by including women, a broader age range, men and women up to the age of 80 and down to the age of 40 this time. We launched this on November 23 of last year, so we’re roughly at the 9-month point for many people. We’re basing the trial not at Stanford anymore, because it’s more convenient to operate at our own local area where our headquarters is, at the BioLabs at the Lundquist Institute.
You can see our first cohort as of November 23, standing with the BioLabs building in the background, including our first woman, the first woman ever to be treated with this approach.
Very, very briefly, the Lundquist venue is part of a larger medical research complex that involves Harvard UCLA Medical Center and the Lundquist institute itself. Within the Lundquist Institute, you have our building, which is the medical research laboratory, which is in easy walking distance to three other buildings, where we can do other tests that are related to longevity and see if we can correlate the results of those tests with the results of our own endpoints, particularly GrimAge and other aging clock results.
The enrollment statistics are pretty much as you see here. We began enrollment, as I said, in November; since then, we’ve enrolled a total of about 23 people. Of those people, five of them are women. We had two dropouts because they did not want to follow the study guidelines, so we had to remove them from the trial.
Not shown here are two control volunteers who were gracious enough to join us in June. They’re untreated controls, so they’re receiving no treatment. The way the trial works, for various logistic reasons, is that we usually have dose updates on Saturday and Sunday of every week, and then any changes in dosing are usually started on a Sunday. We dose also on Tuesday, Thursday, and Friday of the week. This enables us to have esoteric assays run which otherwise might happen over the weekend.
What are some of the early results of the TRIIM-X trials? Some of the volunteers have told me some interesting things like “I can feel myself getting younger,” or “people ask me how come I look so young, what have I been doing lately,” or “I lost five pounds without any effort, and before the trial, I couldn’t lose any weight,” or “I feel more energetic,” or “I just feel great.” Those are good, encouraging signs.
I’ve seen some side effects, however, as you might expect. The most serious ones, from a subjective point of view, at least, are carpal tunnel syndrome, we actually had to take one man and one woman off treatment for a couple of weeks to allow this to reverse. It does seem to have reversed, largely, so those people are back on treatment, or shortly will be. That side effect is a known side effect of growth hormone, and it does happen in a minority of people.
Other people, related to this, have had stiffness in their fingers, but we asked them to try using a squeeze ball to exercise the area, and that seems to have helped a bit. Unfortunately, metformin comes in 500-milligram increments, and when we gave 500 milligrams more to one volunteer, he vomited. We just built that up again, more gradually; we have no current problem with that.
One thing that I’m looking at now is that, for reasons that escape me, some of our volunteers seem to be showing a decrease in hematocrit and hemoglobin. I don’t understand that; I don’t have complete data, so I can’t really comment on that further. I don’t really know much about that, but it’s something that I’ve noticed.
There are other things that we can look at in blood, that we looked at in blood in TRIIM, to see whether we can reproduce some of those effects. One of those things was C-reactive protein, we saw a reduction in that in TRIIM, and we’re seeing in many people a reduction of that in TRIIM-X as well. Here, we’re seeing it this effect in both men and women.
You’ll also see that the curve is bumpy on occasion. This has to do with the COVID-19 pandemic. Turns out that when you get vaccinated for COVID-19, it raises your high-sensitivity CRP. After a while, this comes right back down again, and you come back down to basically where you were before, which in all of these examples is way below where you started from.
This is a slightly different format, it’s the same kind of data, everything to the left of the vertical line is before treatment; everything to the right of it is after treatment. We saw an improvement in estimated GFR in TRIIM. We’re seeing the same thing in TRIIM-X in several volunteers. I’ve just given you three examples here, which happened to be men. We’ve probably seen it in women, but I’d have to check on that again.
Here’s another example of something that looks like we’re reproducing from the original TRIIM study: we saw a reduction in PSA and an increase in percent free PSA, both of which are good for the prospect of prostate cancer. We’re seeing similar trends in TRIIM-X, although the trends seem to be slower to kick in. Occasionally, we’ll see a bump in people’s results like you see in the big data points near the bottom. It may be hard to appreciate, but in that case, the PSA at the end of that curve is still lower than it was at the beginning.
There are some things that we’re noticing in TRIIM-X that we did not notice in TRIIM. One of those things has to do with blood lipids. Usually, we would expect triglycerides to go up with treatment, at least transiently, because growth hormone tends to mobilize fat, and I have a feeling that that may result in triglyceride spikes as the fat is taken out of storage depots and put into the bloodstream. But, people who are particularly high in triglycerides are sometimes seeing a reduction in their triglyceride level, and we’re going to be following that much more closely to see how consistent that is.
Here’s an interesting trend we did not notice in TRIIM. I have to go back and re-examine the data. This is LDL cholesterol. You can see examples of a couple guys who showed a very steady downward trend in their LDL cholesterol, and they were still above the upper limit. That’s considered normal, but at the bottom you see two guys or two gals who had their LDL cholesterol taken from above normal to below the upper limit of normal, so that was actually therapeutic effect, potentially.
Then we started seeing some new things that were really not anticipated. I’m not totally sure what to make of this, but blood urea nitrogen is normally a marker of kidney function, but even people whose kidneys did not seem to be improving yet showed a reduction in BUN in several cases, and we’re not sure what this means, but I speculate that we’re taking nitrogen out of the bloodstream, and we’re putting it into into the creation of new lean body mass. That would be an anabolic effect that maybe we can track just using the simple blood test. We’re pretty interested in that.
Another completely anticipated change that we’ve seen in several people, particularly people whose lung function is questionable because they have high carbon dioxide levels in their bloodstream, is the reduction in carbon dioxide levels. I’ve given examples here from two men and two women of varying ages, but we’re seeing this in men and women up to the age of 77 to 78. That’s pretty exciting to us. In some cases, there’s a little bit of backsliding after the initial reduction in carbon dioxide, but we hope that this will straighten out. Again, these are all preliminary results.
In terms of the epigenetic aging aspects, or in this case, the plasma aging aspect, we’re beginning to piece together some information about what happens after treatment stops and whether we can regain a positive effect once treatment is resumed. We have two examples. I’ve only really worked up one of them so far.
This is a gentleman who was in TRIIM originally and that has now signed up and is part of TRIIM-X as well, so we have his complete dataset. Using this new on-the-fly assay, the plasma PhenoAge rather than an epigenetic aging clock, which at least gives us a preliminary concept of what might be going on. At the start of the TRIIM trial, this gentleman actually had a plasma PhenoAge that was 3.7 years higher than his chronological age, at the end of TRIIM, that switched out, so his plasma PhenoAge was 4.4 years younger than his chronological age.
Basically, nine months after he completed the trial, he had lost another 3.3 years based on this plasma PhenoAge clock. So the the treatment effect actually, it didn’t wear off, and it wasn’t horizontal, it actually got better for a while after the trial ended. There was some regression of that benefit at the time that we started the TRIIM-X trial. At that time, he had gone from being 7.7 years below his chronological age to only .9 years below his chronological age.
On the other hand, you have to consider where he would have been if he hadn’t been treated in the first place; he would have been more like 62 or 63 years of age, biologically, and compared to that, there was still a gain of 4.6 years, and that’s 5.3 years later after this beginning of the trial. Of course, now he’s begun the TRIIM-X trial. Before we go to that, you can see that compared to where he was at the beginning of TRIIM, he had shown no net aging from 5.3 years before.
Effectively, we’ve been able to hold aging at bay if you will, at least based on this questionable plasma PhenoAge clock, but it is just something interesting to note. Now that he’s on TRIIM-X, we’ve seen him go back another 4.2 years below his chronological age or, compared to where he would have been if he’d never been in TRIIM in the first place, a 7.9-year advantage, so that’s pretty cool. We’ll have to see how other people pan out along those lines.
Speaking about other people, other people in the trial are beginning to show the same reduction in plasma PhenoAge, including two women and two men of varying ages from fairly young to fairly, moderately old, including one lady who was 77 years of age here. Again, these results are very preliminary, but we’re encouraged by that.
I have to caution you about one thing, and that is, here’s the rest of the story about the plasma PhenoAge. I showed you before that if you compare just the data out to 12 months against the ancestor of the plasma PhenoAge, namely the DNA methylation PhenoAge clock, you get almost identical results.
If you look at what happens 6 months later, 18 months, the plasma PhenoAge clock continues to go down, whereas the DNA methylation clock regresses a little bit, you get less of a net aging reversal effect. If you compare the PhenoAge clock to the totality of all of the other clocks that we measured and averaged in TRIIM, they more agree with the DNA methylation version of PhenoAge than with the plasma PhenoAge.
We don’t really know what this means. All I can say is that it’s interesting and informative, and certainly this clock is telling us something, it’s just telling us something a little different than what the other clocks are telling us. We’ll have to unravel what that may mean, in time.
Unfortunately, after all of these encouraging results, I have to present some unhappy news to the audience: we did have one very likely case, basically a confirmed case of prostate cancer in one man who was in our treatment group. However, you have to understand that we had two cases of prostate that were caught before treatment could even begin. So far, we don’t have any statistical proof that there’s actually an increased risk due to treatment, and we have to keep in mind the fact that we’re trying to battle prostate cancer, other forms of cancer, and immune senescence, as it affects infectious disease.
As you can see, the risk of getting prostate cancer increases with age, and we’re pushing our volunteer population out to the age of 80. The very thing that we’re trying to escape becomes more probable as we expand the trial in this direction. We’re fighting against the enemy, but we hope to win that battle. So far, we don’t have proof that there’s a problem with the treatment per se. Life is not yet risk free, unfortunately. Maybe some of you can help us cure cancer. In the meantime, be careful, everybody out there.