A sensor/autophagy biomarker panel as a surrogate endpoint for human trials

Q: What changed since this question was written?

The field has explicitly separated three things this question merges: a target-engagement marker, a biomarker of ageing, and a validated surrogate endpoint (Moqri, Cell 2023, PMID 37657418). The 2025 Longevity Biotechnology Association recommendations for trial biomarkers list clocks, omics and functional measures, and do not include a pathway-activity panel (Herzog, npj Aging 2025, PMID 41436473). The current answer to shortening geroscience trials is hierarchical clinical composites with biomarkers placed at the lowest tier (Abdellatif, Nature Aging 2026, PMID 42342910), alongside organ-specific proteomic clocks (Goeminne, Cell Metab 2024, PMID 39488213). On this panel's own components: there is no human blood assay for SAMTOR or GATOR state in this corpus, and the earliest measurement of physiological autophagic flux in a large human trial that this Atlas holds reached significance only post hoc (Bensalem, J Physiol 2025, PMID 40345145, n=121, six months). A second intervention-responsive human readout has since reported: in a placebo-controlled experimental-medicine study in older adults, low-dose rapamycin significantly reduced p21, a marker of DNA-damage-induced senescence, in immune cells (Kell, Aging Cell 2026, PMID 41524558). That is a pharmacodynamic marker moving under controlled conditions, and nothing links it to clinical benefit.

Q: What is still open?

Split in two, because the halves have different answers. As a pharmacodynamic readout for choosing dose and schedule, an mTORC1 against mTORC2 substrate panel is tractable and trials are running. As a surrogate endpoint it is not: surrogacy requires showing the marker mediates clinical benefit across intervention classes, and no ageing biomarker of any kind — pathway panel, epigenetic clock or organ clock — has met that bar. The consensus papers say so themselves.

Q: What's the testable hypothesis?

Two different claims that were being treated as one. The easier one: blood tests can show the drug is hitting the pathway, which helps pick a dose. The hard one, still unmet: a blood test that stands in for living longer or healthier, well enough that a regulator would accept it instead of waiting.

Q: How could this be tested?

Back-test the panel across the animal longevity studies in the corpus (does it separate lifespan-extending from null interventions?), then validate prospectively against EVERLAST's daily-vs-weekly arms. The mTORC1-vs-mTORC2 substrate-phosphorylation arm still requires assay development, and the experiment has to name it: state the biological variable each readout measures, validate complex specificity independently rather than assuming it, and fix sample processing, dynamic-range calibration and prospective clinical validation before any panel is read as a pharmacodynamic marker. AIMTOR (BOU2020) can inform that development but does not supply it. It is a genetically encoded reporter that must be expressed in cells rather than applied to an unmodified blood sample; in the primary paper rapamycin and the pan-mTOR inhibitor Torin1 lowered its signal to a similar degree, so it does not resolve mTORC1 from mTORC2; and the authors state they cannot rule out contributions from mTORC2 or other PIKK-family kinases. Because the reporter also integrates endogenous phosphatase activity alongside kinase activity, no equal-AIMTOR-average result may be described as proof of equal integrated kinase catalysis - a constraint that applies to the matched-average proposals in F1-F3 as well.

Human-endpoint gap · editorial confidence 70% · evidence stands at: Pharmacodynamics in humans · Atlas ID H5

Editorial confidence is the curator's judgement of how likely the hypothesis is to hold, given the studies in the Atlas. It is set by hand, not computed from data, and it is not a calibrated probability.

The gap

Among the human trials collected in this Atlas, none has shown a clinical or functional healthspan benefit: trials aimed at such outcomes either missed their main goal or only checked safety. The goals that were met are immune-response measures, such as a better flu-vaccine response in older adults, not proof of healthier or longer lives. The wins that do exist are for specific diseases (like certain cancers), not for ageing itself. Testing ageing directly takes decades, so researchers want a faster stand-in measurement. A tool that tracks mTOR activity live inside cells does exist, but it has to be built into the cells first, and it cannot tell the two mTOR complexes apart – so the blood test this question asks for still has to be developed and validated.

Technical framing: QUANTIFIED human-endpoint desert: every completed human trial in this corpus with a clinical or functional aging/healthspan PRIMARY endpoint missed it or was safety-only. The exception is immune-response surrogates, which were met: MAN2014 (influenza-vaccine response, about 20%) and the MAN2018 phase 2a (self-reported infections) - which is why H7 reports positive human data while this card reports none on hard outcomes. PEARL (MOE2025): visceral fat NOT met (secondary bone density OR 0.24, 95% CI 0.06-0.93, p=0.04). RTB101 phase 3 (MAN2021): symptomatic-RTI primary NOT met. Resveratrol (POU2013): fully negative. CALERIE (ROM2016): the paper held here is the SAFETY analysis of the two-year randomised caloric-restriction trial in 218 non-obese adults - CR was safe and well tolerated overall, but bone density fell significantly more than in controls, so 'no significant differences' would be wrong. Scope note: CALERIE is a caloric-restriction trial, not an mTOR-inhibitor trial; it belongs here as a geroscience comparator, not as direct evidence on an mTOR intervention. The human 'wins' are disease indications (BOLERO-2 / BAS2012: PFS roughly doubled with exemestane, overall survival not significantly improved; RADIANT-3 / YAO2011: median PFS 11.0 vs 4.6 months, HR 0.35, p<0.001) or surrogate biomarkers (CHU2019: topical rapamycin epidermal p16 reduced, p=0.008). Not 'few trials' but none with a hard aging endpoint. On the assay side, a live-cell readout exists for mTORC1-associated signalling -- AIMTOR (BOU2020), a genetically encoded BRET biosensor characterised in cell lines and in primary myoblasts from a single human muscle biopsy, which reads mTOR activity live with subcellular resolution (lysosome/cytosol/nucleus/mitochondria). Its scope is narrower than this panel needs. The authors report that it responds mainly to mTORC1, that rapamycin and the pan-mTOR inhibitor Torin1 lowered the signal to a similar degree, and that contributions from mTORC2 or other PIKK-family kinases cannot be ruled out; the signal also integrates endogenous phosphatase activity alongside mTOR kinase activity. So it is a working assay for mTORC1-associated signalling in cells engineered to carry it, not a validated blood panel that distinguishes mTORC1 from mTORC2, and it does not remove the assay-development step from this question.

What changed since this question was written

The field has explicitly separated three things this question merges: a target-engagement marker, a biomarker of ageing, and a validated surrogate endpoint (Moqri, Cell 2023, PMID 37657418). The 2025 Longevity Biotechnology Association recommendations for trial biomarkers list clocks, omics and functional measures, and do not include a pathway-activity panel (Herzog, npj Aging 2025, PMID 41436473). The current answer to shortening geroscience trials is hierarchical clinical composites with biomarkers placed at the lowest tier (Abdellatif, Nature Aging 2026, PMID 42342910), alongside organ-specific proteomic clocks (Goeminne, Cell Metab 2024, PMID 39488213). On this panel's own components: there is no human blood assay for SAMTOR or GATOR state in this corpus, and the earliest measurement of physiological autophagic flux in a large human trial that this Atlas holds reached significance only post hoc (Bensalem, J Physiol 2025, PMID 40345145, n=121, six months). A second intervention-responsive human readout has since reported: in a placebo-controlled experimental-medicine study in older adults, low-dose rapamycin significantly reduced p21, a marker of DNA-damage-induced senescence, in immune cells (Kell, Aging Cell 2026, PMID 41524558). That is a pharmacodynamic marker moving under controlled conditions, and nothing links it to clinical benefit.

What is still open

Split in two, because the halves have different answers. As a pharmacodynamic readout for choosing dose and schedule, an mTORC1 against mTORC2 substrate panel is tractable and trials are running. As a surrogate endpoint it is not: surrogacy requires showing the marker mediates clinical benefit across intervention classes, and no ageing biomarker of any kind — pathway panel, epigenetic clock or organ clock — has met that bar. The consensus papers say so themselves.

The hypothesis

Two different claims that were being treated as one. The easier one: blood tests can show the drug is hitting the pathway, which helps pick a dose. The hard one, still unmet: a blood test that stands in for living longer or healthier, well enough that a regulator would accept it instead of waiting.

Technical framing: Two hypotheses. Tractable: a blood mTORC1/mTORC2 substrate-phosphorylation panel plus an autophagic-flux assay can select dose and schedule for mTOR-directed geroprotectors in humans. Still negative: that any ageing biomarker satisfies formal surrogate-endpoint criteria sufficient for regulatory acceptance.

How it could be tested

Back-test the panel across the animal longevity studies in the corpus (does it separate lifespan-extending from null interventions?), then validate prospectively against EVERLAST's daily-vs-weekly arms. The mTORC1-vs-mTORC2 substrate-phosphorylation arm still requires assay development, and the experiment has to name it: state the biological variable each readout measures, validate complex specificity independently rather than assuming it, and fix sample processing, dynamic-range calibration and prospective clinical validation before any panel is read as a pharmacodynamic marker. AIMTOR (BOU2020) can inform that development but does not supply it. It is a genetically encoded reporter that must be expressed in cells rather than applied to an unmodified blood sample; in the primary paper rapamycin and the pan-mTOR inhibitor Torin1 lowered its signal to a similar degree, so it does not resolve mTORC1 from mTORC2; and the authors state they cannot rule out contributions from mTORC2 or other PIKK-family kinases. Because the reporter also integrates endogenous phosphatase activity alongside kinase activity, no equal-AIMTOR-average result may be described as proof of equal integrated kinase catalysis - a constraint that applies to the matched-average proposals in F1-F3 as well.

Related studies

Year	Evidence	Study
2014	H	mTOR inhibition improves immune function in the elderly MAN2014 Older volunteers took the rapalog everolimus (RAD001) before a flu shot, and their antibody response to the vaccine rose by about 20% at doses that were fairly well tolerated. The share of T cells carrying the exhaustion marker PD-1 also fell. The PEARL authors summarise the doses as 0.5 mg daily or 5 mg weekly for 6 weeks. This is the human result behind the weekly low-dose idea. Boundary: a stronger vaccine response is a lab marker, not fewer infections, and the later RTB101 trial (MAN2021) shows how that gap can matter.
2018	H	TORC1 inhibition enhances immune function and reduces infections in the elderly MAN2018 A phase 2a RCT in 264 older adults: 6 weeks of a low-dose combination of BEZ235 (a catalytic PI3K/mTOR inhibitor) and everolimus (RAD001) was associated with fewer reported infections over the following year, plus higher antiviral gene expression and a better flu-vaccine response. The later phase 3 trial of RTB101 (BEZ235) alone (MAN2021) did not confirm a clinical benefit.
2025	H	Influence of rapamycin on safety and healthspan metrics after one year: PEARL trial results MOE2025 The authors describe PEARL as the longest trial of rapamycin for healthy ageing so far: 48 weeks, once a week, ages 50 to 85, 114 people finished (40 on 5 mg, 35 on 10 mg, 39 on placebo). The question the trial was built to answer was visceral fat on a DXA scan, and that did not change. Most of what gets quoted comes from secondary results. Lean mass rose in women on 10 mg, and these women also reported less pain, but pain came from a questionnaire the authors did not list as an endpoint, and only eight women were in the 10 mg group, hence the very wide interval (odds ratio 28, 95% CI 2.4 to 324). Side effects were similar to placebo, with more stomach complaints on rapamycin. Bone density was the one signal pointing the wrong way: in the 5 mg group the odds of an improvement in bone mineral density were lower than on placebo (odds ratio 0.24, 95% CI 0.06 to 0.93, p = 0.04), but 10 mg showed no change (odds ratio 1.00) and the comparisons between groups were not significant. Boundary: the capsules were compounded, and the authors found they reached about a third of the blood level of the commercial drug, so 10 mg here is closer to 3 mg of generic sirolimus. Dosing was self-reported, and the group was about two thirds men and health-conscious.
2021	H	Targeting the biology of ageing with mTOR inhibitors to improve immune function in older adults: phase 2b and phase 3 randomised trials MAN2021 This paper reports two trials of RTB101, a drug that blocks mTOR, taken once a day for 16 weeks by adults over 65. In the phase 2b trial (652 people), fewer people on 10 mg had a lab-confirmed respiratory infection than on placebo (19% against 28%), but that result comes from a pooled analysis the authors did not correct for multiple testing, and the corrected analysis did not reach significance. The phase 3 trial (1,024 people) used a different main endpoint, set with the FDA: symptoms with or without a lab test. It found no difference (26% against 25%, p = 0.65), although blood tests still showed more antiviral genes switched on in both trials. In phase 2b, twice-daily dosing and a combination with everolimus did not reach significance, which the authors read as a hint that on-off inhibition may beat constant inhibition. Those numbers are "data not shown" in the paper. Boundary: RTB101 is not rapamycin or everolimus, the phase 3 group was healthier (no COPD, no smokers), and that trial was too small to test the lab-confirmed endpoint. The only everolimus arm (RTB101 plus everolimus 0.1 mg daily) was a combination and did not reach significance, which is why this paper is listed against the everolimus and immune function link on the pathway map. It cannot show what everolimus alone does.
2013	H	High-dose resveratrol supplementation in obese men: an investigator-initiated, randomized, placebo-controlled clinical trial of substrate metabolism, insulin sensitivity, and body composition POU2013 4 weeks of high-dose resveratrol had no effect on insulin sensitivity, blood pressure, or body composition - directly contradicting the promising rodent data that drove the original hype.
2016	H	Safety of two-year caloric restriction in non-obese healthy individuals ROM2016 Two years of sustained 25% caloric restriction in healthy non-obese adults was safe overall, though it lowered bone density more than controls.
2019	H	Topical rapamycin reduces markers of senescence and aging in human skin: an exploratory, prospective, randomized trial CHU2019 A small human trial testing whether rapamycin can slow aging in a tissue you can actually see and biopsy - skin. Topical rapamycin significantly lowered the senescence marker p16 and raised collagen VII, with visible improvement in skin appearance. Early, small-scale evidence (17 of 36 completed) that topical rapamycin can reduce a senescence marker in human skin.
2012	H	Everolimus in postmenopausal hormone-receptor-positive advanced breast cancer (BOLERO-2) BAS2012 A phase 3 RCT (n=724) proving mTOR matters in a common cancer. Resistance to hormone therapy in breast cancer is associated with mTOR activation. Adding everolimus more than doubled progression-free survival (10.6 vs 4.1 months by central review) - leading to FDA approval. Main toxicity was stomatitis.
2011	H	Everolimus for advanced pancreatic neuroendocrine tumors (RADIANT-3) YAO2011 A phase 3 RCT (n=410) that made everolimus a standard treatment for pancreatic neuroendocrine tumors. It more than doubled progression-free survival (11.0 vs 4.6 months, a 65% reduction in risk of progression/death) with mostly mild side effects. Another FDA-approved indication built on blocking mTOR.
2024	S	Targeting ageing with rapamycin and its derivatives in humans: a systematic review LEE2024 Described by its authors as the first systematic review of rapamycin/rapalogs in humans for aging. Screened 18,400 articles, included 19 studies. Found improvements in immune, cardiovascular, and skin (integumentary) parameters; NO significant effect on endocrine, muscular, or neurological systems. No serious adverse events in healthy people, but more infections and raised cholesterol/triglycerides in people with age-related disease. This is the highest-tier human-evidence summary in the whole Atlas - it aggregates many individual human studies rather than reporting one.
	RT	Everolimus Aging Study (EVERLAST): Clinical Evaluation of mTORC1 Inhibition for Geroprotection NCT05835999 Phase 2 trial (NCT05835999, EVERLAST) testing this Atlas's open dosing hypothesis directly: whether daily low-dose (0.5 mg/day) or weekly (5 mg/week) everolimus for 24 weeks can improve physiological and molecular hallmarks of aging in insulin-resistant or prediabetic adults aged 55-80, without the metabolic penalty seen with continuous higher-dose rapalog use. Double-blind and placebo-controlled with four arms (everolimus daily, everolimus weekly, and a matching placebo for each), n=106, plus a separate untreated 18-35 cohort for comparison. Primary endpoint is peripheral insulin sensitivity by dual-tracer OGTT; mTORC1 and mTORC2 signalling are secondary endpoints in muscle and blood cells. STATUS: the registry records the trial as COMPLETED, with primary completion 2026-06-30 and completion 2026-07-22. No results are posted yet, so there is nothing to read into the outcome either way.
2020	M	AIMTOR, a BRET biosensor for live imaging, reveals subcellular mTOR signaling and dysfunctions BOU2020 AIMTOR is a genetically encoded BRET biosensor that reads out mTOR activity live in single cells and in specific subcellular compartments (cytosol, lysosome surface, nucleus, near mitochondria).
2013	M	mTORC1 phosphorylation sites encode their sensitivity to starvation and rapamycin KAN2013 Each mTORC1 phosphorylation site has its own intrinsic 'substrate quality'. Good substrates stay phosphorylated when mTORC1 is partly inhibited by rapamycin or starvation, poor ones drop out first, and changing a few residues around a site shifts its sensitivity in cells. One level of mTORC1 activity can therefore produce different output profiles.
2026	H	Reduced dietary protein intake does not alter autophagy in human blood: A randomized crossover study in healthy adults SIN2026 In a rigorous randomized crossover trial, 63 healthy adults completed both a normal-protein (20% energy) and a reduced-protein (10% energy) diet for 4 weeks each. Despite preclinical evidence that dietary protein restriction should activate autophagy via mTORC1 inhibition, directly measured autophagic flux in blood immune cells did not change between the two diets. This is one of the few human studies to test the protein-restriction-to-autophagy link with a dynamic, flux-based (not just steady-state marker) readout, and the negative result suggests that moderate protein reduction alone, without an accompanying caloric deficit, is not sufficient to engage autophagy in humans the way it does in some animal models.
2026	H	Autophagy Flux Is Remodeled Sex- and Cell Type-Specifically During Human Aging, and Is Linked to Reduced Physical Function in Older Adults MOR2026 Autophagy flux in humans is remodeled with age in a sex- and cell-type-specific way, uncoupled from autophagy-gene transcription; in adults over 70, higher PBMC autophagy flux associates with reduced physical function, and 12 weeks of mild exercise lowers PBMC flux while improving function.

Revision history

ShowHide

2026-07-09: AIMTOR (BOU2020) biosensor added as a candidate readout.
2026-08-30: The basis previously cited RADIANT-3 (YAO2011) as 'PFS 5% vs 2%'. Those were the grade 3/4 hyperglycaemia rates from the adverse-event table, not a progression-free-survival result.
2026-09-04: Trial acronyms paired with Atlas study codes inline.
2026-09-23: The hypothesis was split in two (it was previously one).
2026-09-29: 'Every trial missed' limited to clinical and functional healthspan endpoints; the immune-response surrogates that were met (MAN2014, MAN2018) are now named.
2026-10-03: KAN2013 added - shows in human and mouse cell lines that individual mTORC1 phosphorylation sites differ in intrinsic substrate quality and therefore in their sensitivity to rapamycin and starvation, which constrains how a substrate-phosphorylation panel can be read as a pharmacodynamic marker (cell lines, not human blood).
2026-10-03: SIN2026 added - a randomised crossover trial in 63 healthy young adults found no change in whole-blood LC3B-II autophagic flux after 4 weeks of reduced (10%) vs average (20%) protein intake, showing a human flux assay is feasible but did not respond to moderate protein restriction (human, PBMCs only; mTORC1 not measured).
2026-10-03: MAN2014 and MAN2018 added to Supporting_Studies. They were named in the basis on 2026-09-29 as the two met endpoints but never linked, so the card cited evidence the reader could not open (validate_claims.py R10).
2026-10-05: MOR2026 added (daily check, unreviewed) - a healthy-aging human cohort study found PBMC autophagic flux is measurable, cell type- and sex-specific, associates with reduced physical function after age 70, and decreases over 12 weeks of mild exercise in parallel with improved function, further strengthening the case that an autophagic-flux readout is a feasible, intervention-responsive human assay (human, observational + pilot intervention, not a validated surrogate endpoint).
2026-10-07: KEL2026 (PMID 41524558) added - p21 in immune cells as a human PD readout that responded to low-dose rapamycin under placebo control. Strengthens the pharmacodynamic half of the split; the surrogate half is unchanged (revision of all 17 cards against the literature).
2026-10-10: Audit N04. The card treated AIMTOR (BOU2020) as a ready-made readout for the mTORC1-versus-mTORC2 arm of the panel, so that arm 'can now build directly on AIMTOR rather than needing de novo assay development'. Checked against the primary full text: AIMTOR is a genetically encoded reporter requiring expression in the cell, it responds mainly to mTORC1, rapamycin and Torin1 lower its signal to a similar degree so it does not resolve the two complexes, the authors cannot exclude mTORC2 or other PIKK kinases, and the signal integrates phosphatase as well as kinase activity. The human material was primary myoblasts from one muscle biopsy, not blood. The assay-development step is back in the proposed experiment, with specificity, sample processing, dynamic range and prospective validation named, and the phosphatase constraint on matched-average readings is now stated.

← All open questions · Frontier questions

Was this page useful?