The increase in life expectancy has caused a rise in age-related brain disorders. Although brain rejuvenation is a promising strategy to counteract brain functional decline, systematic discovery methods for efficient interventions are lacking. A computational platform based on a transcriptional brain aging clock capable of detecting age- and neurodegeneration-related changes is developed. Applied to neurodegeneration-positive samples, it reveals that neurodegenerative disease presence and severity significantly increase predicted age. By screening 43840 transcriptional profiles of chemical and genetic perturbations, it identifies 453 unique rejuvenating interventions, several of which are known to extend lifespan in animal models. Additionally, the identified interventions include drugs already used to treat neurological disorders, Alzheimer’s disease among them. A combination of compounds predicted by the platform reduced anxiety, improved memory, and rejuvenated the brain cortex transcriptome in aged mice. These results demonstrate the platform’s ability to identify brain-rejuvenating interventions, offering potential treatments for neurodegenerative diseases.
In total, 34 predicted compounds could be structurally linked to 13 perturbations with known rejuvenating effects. The largest cluster of compounds is centered around the mTOR inhibitor everolimus. Other compounds belonging to this cluster are pravastatin, cyclosporin-a, and erythromycin-ethylsuccinate, which also appear in DrugAge as lifespan extending, and tranylcypromine, a compound that has been shown to have neuroprotective effects against amyloid-𝛽-induced toxicity. [65,66] The second largest cluster was centered around resveratrol, xanthohumol, and curcumin, the three of them nutraceuticals that appear in DrugAge as lifespan extending and that possess antioxidant, antiinflammatory, and anti-cancer properties.[67–69] In summary, the structural similarities identified between compounds with and without a demonstrated ability to extend lifespan support the hypothesis that all of these compounds indeed display rejuvenating effects.
Interestingly, well-known rejuvenating compounds like rapamycin (sirolimus) and metformin were not predicted as rejuvenating by our clock. Rapamycin, indeed, produced a significantly higher transcriptional age in NPCs. A recent literature screening showed that rapamycin treatment did not produce significant improvements in neurological systems.[48] Rapamycin was shown to suppress differentiation of neural stem cells, which could explain the higher transcriptional age in NPCs. [92,93] On the other hand, other mTOR inhibitors such as everolimus were detected as rejuvenating. Metformin, however, produced rejuvenation in neurons, but this shift was not statistically significant. It is likely that our model, due to the linearity assumption, is not capable of capturing the whole extent of the alterations that happen along aging, focusing only on the ones that are linearly related to the chronological age.
As a proof of concept of our approach, we tested three predicted compounds — 5-azacytidine, tranylcypromine, and JNKIN-8 — in aged mice. The treatment significantly reduced anxiety, and improved memory in aged mice, addressing well-known aging-associated alterations.[94,95] At the molecular level, it significantly restored the younger phenotype. Notably, all three compounds are epigenetic regulators—a category to which several compounds identified by our platform belong—highlighting the promising role of such drugs in counteracting aging.
All very well, but isn’t this context dependent? If it’s not a matter of tissue exposure as JH claims, then it says that the structural differences between sirolimus and everolimus makes the latter rejuvenating in neurological and brain tissue and sirolimus not, or sirolimus even pro-aging. Except. We have the rapamycin study in brain regions of ApoE4 carriers where volume was preserved and even increased. But ok, maybe that can be seen as impacting pathology and not aging itself. And further on the debit side for rapa:
Rapamycin inhibits mTOR/p70S6K activation in CA3 region of the hippocampus of the rat and impairs long term memory
So, does that mean we should all switch to everolimus from sirolimus - especially that the purported benefits in immune function in Mannick’s study was also in everolimus?
Not so fast. Because what about lifespan studies in multiple species, including mammals for rapamycin? We certainly don’t have that kind of track record for everolimus. So for life extension we still have robust animal data for sirolimus, not everolimus (as yet).
This gets us back to the old argument about whether it’s possible for rapamycin to extend life while doing nothing, or even worse for your brain. Seems hard to square that circle. That said, those with ND conditions might want to look at everolimus rather than sirolimus. YMMV.
The BBB in Alzheimer’s is not smashed open, but it is measurably compromised: roughly a one‑third to one‑half increase in macromolecular permeability where memories are formed. That leak appears early, worsens with APOE ε4 and vascular risk factors, and forecasts who will decline fastest—making BBB integrity both a biomarker and a potential therapeutic target.
Obvious next question answer
Bottom line
The BBB in Parkinson’s disease is moderately leaky—think “dripping tap”, not “burst pipe.” That drip worsens with disease progression, is most obvious in the nigrostriatal system and posterior white matter, and is tied to inflammatory and angiogenic signalling. Quantitatively, expect ~20‑35 % elevations in Ktrans and ~50 % rises in QAlb in advanced cases, with early‑stage PD often still within normal limits.
The issue is, that once the BBB has major compromise, one usually has significant disease, and the goal of the therapies used are to avoid getting significant disease. So by the time there is this compromise in BBB, the likelihood of therapies working is diminished.
Taking a multipronged approach has been my focus and being aware that some things that look promising will fail to work and others hopefully will.
There is, however, a point that some compromise in the BBB will mean that certain therapeutics will have some effect. Hence this would explain results from Rapamycin. However, it would be worth checking the research to see if it is a reasonable hypothesis that leakiness of the BBB was part of the mechanism.
In the ApoE4carrier study, rapamycin was effective for ApoE4, but did nothing for the ApoE3 carriers. Maybe it was due to BBB leakiness in ApoE4. In that case, perhaps for longevity purposes for ApoE3 carriers one could do a combination therapy, where you have a normal or near normal dose of rapamycin for longevity, say 5-6 mg a week plus concurrently a low dose everolimus, say 2 mg (which dose showed some effectiveness in the Mannick trial). That way your overall mTOR inhibitor dose is still not too high (6mg sirolimus + 2mg everolimus), but you are also getting some into the brain with everolimus. That is assuming that there isn’t some contraindication with using sirolimus and everolimus concurrently. If there isn’t some negative interaction or interference, you might be able to perhaps have your cake and eat it too. Wild speculation, of course, , heh,
There are dozens of them. Search for dementia, Alzheimer’s, or Parkinson’s in the canagliflozin thread (or on Google Scholar): Canagliflozin - Another Top Longevity Drug
I completely agree. The further question I have though is whether they are independently protective beyond the weight loss/blood glucose etc affect. I suspect there will be lots of factors - weight loss will lead to greater activity levels, less painkiller use. And they may also lead to better average quality of food intake. And there’s all the other downstream affects of GLP-1s like reduction in alcohol, gambling etc
I suspect, after all of those benefits, there may not be a direct pharmacological neuroprotective affect - but I’d love to be wrong.
Well as much as we all hate pasted AI - this is from Vera Health and it documents it reasonably well … so at the risk of scorn … I paste. Yes, the longevity and neuroprotective effects of GLP-1 agonists and SGLT2 inhibitors are at least partially independent of their glucose-lowering and weight-reducing effects:
Cardiovascular and Renal Benefits Independent of Glucose Control:
A 2023 meta-analysis of 154,649 patients in Diabetes Care I found that the relative benefits of GLP-1 agonists and SGLT2 inhibitors on cardiovascular and renal outcomes were consistent across different baseline cardiovascular risk levels, regardless of glycemic control 1.
Rationale: This study is highly relevant because it directly assessed whether the benefits of these drugs varied with patients’ baseline risk, which includes factors like glucose control and body weight.
Neuroprotective Effects Beyond Glycemic Control:
A 2025 meta-analysis of 164,531 participants in JAMA Neurology I reported that SGLT2 inhibitors and GLP-1 agonists significantly reduced the risk of dementia and cognitive impairment compared to other glucose-lowering drugs 2.
Rationale: The large sample size and focus on neuroprotective outcomes make this study directly relevant to the question. However, the study did not fully isolate the effects from glucose control, which is a limitation.
Mechanistic Evidence for Independence from Glucose Control:
A 2022 review in Circulation II highlighted that SGLT2 inhibitors reduce oxidative stress, enhance autophagy, and improve mitochondrial function in cells not exposed to changes in glucose or ketones 3.
Rationale: This study is relevant because it provides mechanistic insights into how SGLT2 inhibitors confer organ protection independently of glucose control. However, it is not a randomized controlled trial, which limits its clinical applicability.
Additive Effects on Cardiovascular Outcomes:
A 2024 meta-analysis of 17,072 patients in Circulation I found that the cardiovascular and kidney benefits of GLP-1 agonists were consistent regardless of concurrent SGLT2 inhibitor use, suggesting independent mechanisms of action 4.
Rationale: The study’s design allows for the assessment of independent effects, but the relatively small proportion of patients on combination therapy (10.2%) limits the generalizability of the findings.
Mortality Reduction Beyond Glycemic Effects:
A 2018 analysis in Diabetes Care I of the EMPA-REG OUTCOME and LEADER trials estimated that the reduction in all-cause mortality with empagliflozin (SGLT2 inhibitor) was only partially explained by its glucose-lowering effects 5.
Rationale: This study is relevant because it quantifies the extent to which mortality benefits are independent of glycemic control. However, it is an observational analysis rather than a randomized trial.
Longevity Effects in Non-Diabetic Populations:
A 2024 meta-analysis of 28,168 participants in Therapeutic Advances in Neurological Disorders I found that GLP-1 agonists reduced major adverse cardiovascular events (MACE) and all-cause mortality in overweight or obese adults without diabetes 6.
Rationale: The inclusion of non-diabetic populations strengthens the evidence for effects independent of glucose control. However, data on neuroprotective outcomes were limited.
Conclusion:
The evidence strongly supports that GLP-1 agonists and SGLT2 inhibitors have longevity and neuroprotective effects that are at least partially independent of their glucose-lowering and weight-reducing actions. This is demonstrated by consistent cardiovascular and renal benefits across different risk profiles, significant reductions in dementia risk, and mechanistic studies showing direct effects on cellular health. However, the degree of independence varies, and more research is needed to fully isolate these effects from metabolic improvements.
I use standard dosing, most commonly tirzepatide to the degree of tolerability and not causing weight loss to an unacceptable level. There are many patients who are at or near ideal body weight - in those, I’d favor a low dose SGLT2-i as this can usually be tolerated with little change in body weight. For the majority, who have a bit of weight to lose we go with standard doses, with goal of 0.5-1 lbs/week wt loss (no faster) until at IBW, then back off a bit, and usually if financially able an no contraindications, add an SGLT2-i … for individuals at some risk of neurocognitive decline, cardiovascular, metabolic or renal disease … which is most people.
I propose that with most drugs, the benefit looks a bit like how it works with statins (but there we can easily measure the effect) in that a majority of the benefit occurs at lower doses, and pushing higher gets a tiny bit more benefit, but more toxicity.
The microdosing approach makes little sense to me - it is like homeopathy - a minuscule dose is going to have therapeutic effects? Results are dose dependent, as is toxicity. I’d at least like to see a modest dose in order to think it is doing anything. I however don’t have definitive proof of this belief in this context, but it makes the most sense pharmacologically … and I’ll stick with this until proven otherwise.
I’m going to give the 5 year fermented Dr Ohhira’s probiotics a go to see if it helps me recover from recent antibiotic usage. Amoxicillin gave me some gut pain.
, heh,