Cited by 54Genetics, Aging, and Longevity in Model OrganismsTelomeres, Telomerase, and SenescenceStatistical Methods in Clinical Trials

The DOI matches its Crossref record; 23 of the 23 references it cites were found in a registry or an index; 1 of the 11 sources its claims cite were checked for support.

Endpoints for geroscience clinical trials: health outcomes, biomarkers, and biologic age

Open access

Abstract

Treatments that target fundamental processes of aging are expected to delay several aging-related conditions simultaneously. Testing the efficacy of these treatments for potential anti-aging benefits will require clinical trials with endpoints that reflect the potential benefits of slowing processes of aging. There are several potential types of endpoints to capture the benefits of slowing a process of aging, and a consensus is needed to standardize and compare the results of these trials and to guide the analysis of observational data to support trial planning. Using biomarkers instead of clinical outcomes would substantially reduce the size and the duration of clinical trials. This requires validation of surrogate markers showing that treatment induced change in the marker reliably predicts the magnitude of change in the clinical outcome. The surrogate marker must also reflect the biological mechanism for the effect of treatment on the clinical outcome. “Biological age” is a superficially attractive marker for such trials. However, it is essential to establish that treatment induced change in biological age reliably predict the magnitude of benefits in the clinical outcome. Reaching consensus on clinical outcomes for geroscience trials and then validating potential surrogate biomarkers requires time, effort, and coordination that will be worthwhile to develop surrogate outcomes that can be trusted to efficiently test the value of many anti-aging treatments under development.

Overview

AI snapshot

Written from the paper’s full text.

TL;DR

Geroscience trials need agreed clinical endpoints and validated surrogate markers to test treatments that may delay multiple age-related conditions. Biomarkers and “biological age” could greatly reduce trial size and duration, but they are not valid surrogates unless treatment-induced changes reliably predict health-outcome changes through the same causal pathway.

Key findings

  • A valid surrogate requires that treatment, marker change, and health outcome lie on the same causal pathway; predictive or prognostic biomarkers are not automatically surrogates.
  • Hip bone mineral density (BMD) is cited as a validated surrogate: a meta-analysis found R² = 0.73 between change in hip BMD and reduction in hip fracture risk, enabling future osteoporosis trials of a few hundred participants over 2 years instead of trials typically involving 5,000–10,000 women with 3–5 years of follow-up.
  • Failed surrogate examples include intensive glucose lowering in ACCORD, which reduced HbA1c but increased cardiovascular disease events; MK-0773, which increased lean/appendicular mass but did not improve leg press strength or gait speed; and amyloid-reducing dementia trials that failed to improve cognition or dementia risk.
  • “Biological age,” especially epigenetic age, is attractive but not yet established as a surrogate; without understanding its mechanisms, it may be insensitive to interventions targeting one aging process and could produce misleading trial results.

Methods

  • Narrative review/perspective, not an empirical study; it discusses endpoint selection for geroscience clinical trials and the pathway from response biomarkers to predictive/prognostic biomarkers to surrogate markers.
  • It considers potential clinical endpoints including total mortality, disability-free survival, specific diseases, advancing multimorbidity, frailty/deficit indices, and frailty phenotype, summarized in Table 1.
  • It uses prior trial and meta-analysis examples, plus conceptual figures, to illustrate valid and failed surrogate markers and to argue for consensus, standardization, and centralized data sharing.

Limitations

  • The paper states there is neither consensus nor a process to generate consensus on acceptable geroscience trial endpoints, and that what morbidities should be included is unclear.
  • It states that biological mechanisms linking epigenetic clocks or other biological-age measures to health outcomes are uncertain, and that biological age is unlikely to be a valid surrogate without such mechanistic understanding.
  • Evident from the text: this is a conceptual review without primary data, systematic search, or formal analysis; its recommendations depend on future coordination, standardized measurements, and trial data sharing that are not yet established.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.1007/s11357-022-00671-8

Cites

23 of 23 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Verification

Claims the paper makes

1 of 11 cited sources checked · 0 supported · 0 flagged · accuracy not yet measured

  • Interpretationreasoning

    Current work to identify biomarkers is not aligned with the goal of accelerating the evaluation of promising interventions.

    “Thus, current work to identify biomarkers is not aligned with the goal of accelerating the evaluation of promising interventions.”
    Quote in the paperSubstantial claim
    • Rests on an assumption: There is neither consensus nor a process to generate consensus on what constitutes an acceptable endpoint for a geroscience-based prevention trial.
  • Premisecitation

    Hip bone mineral density (BMD) is an example of a valid surrogate marker.

    “Bone density (BMD) of the hip is an example of a valid surrogate marker.”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [13] Black DM, et al. Diabetes Endocrinol. 2020;8(8):672–82. https://doi.org/10.1016/S2213-8587(20)30159-5. source
  • Interpretationreasoning

    Using validated surrogate biomarkers instead of clinical outcomes would substantially reduce the size, duration, and cost of clinical trials.

    “Using biomarkers instead of clinical outcomes would substantially reduce the size and the duration of clinical trials.”
    • Also stated in: § Introduction · hedged
      “The availability of validated surrogate biomarkers would obviate the need conduct large, lengthy, and costly trials to establish treatment efficacy.”
    Quote in the paperNot a substantial claim
  • Premisenone given

    Meaningful health endpoints for geroscience trials would be ascertainable by interview and simple tests.

    “In general, meaningful health endpoints for geroscience trials would be ascertainable by interview, for example, about activities of daily living for disability, and by questions combined with simple tests, such as gait speed and change in body weight, for frailty.”
    Quote in the paperSubstantial claim
  • Premisereasoning

    To be a valid surrogate, the biological effect of the treatment, the change in the marker, and change in the health outcome must all line up in the same causal pathway.

    “Thus, the biological effect of the treatment, the change in the marker, and change in the health outcome must all line up in the same causal pathway (Fig. 1).”
    Quote in the paperSubstantial claim
  • It may state two separate claims as one. 8
  • Probably not a substantial claim. 17
  • It may state two separate claims as one.

    Premisecitation

    In a randomized clinical trial, MK-0773 substantially increased lean body and appendicular mass but had no effect on leg press strength, gait speed, or other health endpoints.

    “Similarly, a promising treatment for sarcopenia, MK-0773, substantially increased lean body and appendicular mass, but in a randomized clinical trial, the treatment had no effect on leg press strength, gait speed, or other health endpoints [15].”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites [15] Papanicolaou DA, Ather SN, Zhu H, Zhou Y, Lutkiewicz J, Scott BB, Chandler J. A phase IIA randomized, placebo-controlled clinical trial to study the efficacy and safety of the selective androgen receptor modulator (SARM), MK-0773 in female participants with sarcopenia. J Nutr Health Aging. 2013;17(6):533–43. https://doi.org/10.1007/s12603-013-0335-x. source

      The Results (Efficacy) report significant increases in total and appendicular lean body mass, but no statistically significant advantage over placebo for leg-press strength, gait speed, or several physical-function measures at six months. The broader statement that MK-0773 had no effect on other health endpoints overstates the paper: the Discussion reports increases in serum transaminases, hemoglobin, hematocrit, and systolic blood pressure. (LLM, conf=0.86)

      Experimental — accuracy not yet measured
  • It may state two separate claims as one.

    Premisereasoning

    Composite endpoints that include multiple effects such as multimorbidity or deficit accumulation have higher event rates and require proportionately smaller sample sizes.

    “Endpoints that include a composite of multiple effects, such as multi-morbidity or accumulation of physical and cognitive deficits (deficit accumulation indices), have the advantages of higher rates requiring proportionately smaller sample sizes.”
    Quote in the paperSubstantial claim
  • It may state two separate claims as one.

    Interpretationreasoning

    Sample sizes for incident health events such as Alzheimer's disease are generally very large and require long follow-up.

    “Sample sizes for incident health events, such as a diagnosis of Alzheimer’s disease, are generally very large and require long follow-up because these age-related diseases are uncommon.”
    Quote in the paperSubstantial claim
    • Rests on an assumption: Age-related diseases are uncommon.
  • Probably not a substantial claim. The paper’s conclusions may not depend on it.

    Premisenone given

    Prognostic biomarkers pertain to individuals who already have the health condition; the marker or change in that marker predicts change in the severity or level of the condition.

    “This type of marker pertains to individuals who already have the health condition; the marker or change in that marker predicts change in the severity or level of the condition.”
    Quote in the paperNot a substantial claim
  • Probably not a substantial claim. The paper may give no reason why it holds.

    Premisenone given

    A valid surrogate marker must reflect the biological mechanism for the effect of treatment on the clinical outcome.

    “The surrogate marker must also reflect the biological mechanism for the effect of treatment on the clinical outcome.”
    Quote in the paperNot a substantial claim
  • Probably not a substantial claim. The paper may give no reason why it holds.

    Premisenone given

    Stronger associations with age indicate that an endpoint is more likely to be responsive to interventions that slow aging.

    “Each is age-related, and it is reasonable to posit that stronger associations with age indicate that the endpoint is more likely to be responsive to interventions that slow aging.”
    Quote in the paperNot a substantial claim
  • Probably not a substantial claim. The paper’s conclusions may not depend on it. The paper may give no reason why it holds.

    Interpretationreasoning

    Predictive markers may have clinical benefits by estimating a patient's risk of the condition and, therefore, the magnitude of benefit from the treatment.

    “Predictive markers may have clinical benefits by estimating a patient’s risk of the condition and, therefore, the magnitude of benefit from the treatment.”
    Quote in the paperNot a substantial claim
  • Probably not a substantial claim. The paper may give no reason why it holds.

    Premisenone given

    Information linking outcomes to fundamental biological processes of human aging will typically come from model systems where the impact of perturbing aging biology can be rigorously evaluated.

    “This information will typically come from model systems where the impact of perturbing aging biology can be rigorously evaluated.”
    Quote in the paperNot a substantial claim
  • Probably not a substantial claim.

    Premisereasoning

    To establish that a biomarker is a surrogate, treatment-induced change in the biomarker must predict change in the health outcome.

    “What matters to establish that a biomarker is a surrogate is that treatment induced change in the biomarker predicts change in the health outcome.”
    Quote in the paperSubstantial claim
  • It may state two separate claims as one.

    Interpretationreasoning

    A treatment may have opposing effects on components of a composite index, which may reduce the sensitivity of the index and the size of the treatment effect.

    “Sometimes, a treatment may have opposing effects on components of an index. In this case, a composite endpoint may reduce the sensitivity of the index and the size of the treatment effect.”
    Quote in the paperSubstantial claim
  • Probably not a substantial claim. The paper may give no reason why it holds.

    Premisenone given

    Predictive markers are an essential link in the pathway from treatment to change in a health condition.

    “Predictive markers are an essential link in the pathway from treatment to change in a health condition (Fig. 1).”
    Quote in the paperNot a substantial claim
  • Probably not a substantial claim. It may be too vague to check against evidence. The paper may give no reason why it holds.

    Premisenone given

    Outcomes for geroscience trials must be linked, in part, to fundamental biological processes of human aging.

    “Outcomes must be linked, in part, to fundamental biological processes of human aging.”
    Quote in the paperNot a substantial claim
  • Probably not a substantial claim.

    Premisenone given

    Establishing that a biomarker is a response biomarker establishes the first part of the pathway for a biomarker to be a potential surrogate.

    “Establishing that a biomarker is a response biomarker establishes the first part of the pathway for a biomarker to be a potential surrogate (Fig. 1).”
    Quote in the paperSubstantial claim
  • Probably not a substantial claim. The paper’s conclusions may not depend on it. It may state two separate claims as one.

    Premisenone given

    Incident diseases such as cardiovascular disease are feasible to collect but may require validation in medical records.

    “Incident diseases, such as cardiovascular disease are feasible to collect but may require validation in medical records.”
    Quote in the paperNot a substantial claim
  • Probably not a substantial claim. It may be too vague to check against evidence. The paper may give no reason why it holds.

    Premisenone given

    Ideally, a marker used to include participants is in the biological pathway from treatment effect to health outcome.

    “Ideally, a marker used to include participants is in the biological pathway from treatment effect to health outcome.”
    Quote in the paperNot a substantial claim
  • It may state two separate claims as one.

    Premisenone given

    Predictive and prognostic biomarkers are useful in trial design by estimating a potential participant's risk, allowing inclusion of participants at high risk of the outcome.

    “Predictive and prognostic biomarkers are useful in the design of trials by estimating a potential participant’s risk. This allows the inclusion of participants at high risk of the outcome.”
    Quote in the paperNot a substantial claim
  • Probably not a substantial claim. The paper may give no reason why it holds.

    Premisenone given

    Human observational research can link putative aging biomarkers to patterns of human disease.

    “Human observational research can link putative aging biomarkers to patterns of human disease.”
    Quote in the paperNot a substantial claim
  • Probably not a substantial claim.

    Premisenone given

    It is often assumed that showing that a biomarker, or change in that marker, predicts the health endpoint establishes that biomarker as a "surrogate."

    “It is often assumed that showing that a biomarker, or change in that marker, predicts the health endpoint establishes that biomarker as a “surrogate.””
    Quote in the paperSubstantial claim
  • Probably not a substantial claim.

    Premisereasoning

    Validating a surrogate marker requires showing that treatment-induced change in the marker reliably predicts the magnitude of change in the clinical outcome.

    “This requires validation of surrogate markers showing that treatment induced change in the marker reliably predicts the magnitude of change in the clinical outcome.”
    • Also stated in: § Predictive and prognostic biomarkers · plain
      “To be a valid surrogate, treatment-induced change in the marker must reliably predict a change health endpoint (Fig. 1).”
    Quote in the paperSubstantial claim
  • Probably not a substantial claim. It may be too vague to check against evidence. The paper may give no reason why it holds.

    Premisenone given

    Randomized clinical trials will be required to establish with rigor whether targeting aging biology is effective and safe.

    “Ultimately, randomized clinical trials will be required to establish with rigor whether the targeting aging biology is effective and safe.”
    Quote in the paperNot a substantial claim
  • It may state two separate claims as one.

    Interpretationnone given

    Trials with biomarker endpoints generally may require a much smaller number of participants, shorter follow-up times, and lower cost than trials with clinical or health outcomes.

    “Surrogate markers are very attractive because trials with biomarker endpoints generally may require a much small number of participants, shorter follow-up times, and, therefore, much lower cost than trials with clinical or health outcomes (Table 1).”
    Quote in the paperSubstantial claim
  • It may state two separate claims as one.

    Interpretationreasoning

    Change in lean mass does not increase muscle strength and function, perhaps because "lean mass" is a poor measure of skeletal muscle mass.

    “This suggests that change in lean mass does not increase muscle strength and function perhaps because “lean mass” is a poor measure of skeletal muscle mass.”
    Quote in the paperSubstantial claim
  • Probably not a substantial claim.

    Premisecitation

    The difference between biological and calendar age, called "age acceleration," predicted a variety of health outcomes.

    “Ideally, the difference between biological and calendar age, called “age acceleration,” predicted a variety of health outcomes. [e.g., 7]”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [7] Lu AT, et al. DNA methylation GrimAge strongly predicts lifespan and healthspan. Aging (Albany NY). 2019;11(2):303–27. https://doi.org/10.18632/aging.101684. source
  • Probably not a substantial claim.

    Interpretationreasoning

    A key issue for the use of epigenetic age or other measures of biological age is understanding the mechanism for its association with calendar age and with health outcomes.

    “Therefore, a key issue for the use of epigenetic age—or other measures of biological age—is understanding the mechanism for its association with calendar age and with health outcomes.”
    Quote in the paperSubstantial claim
24 of 24
  • The paper relies on this without support.

    There is neither consensus nor a process to generate consensus on what constitutes an acceptable endpoint for a geroscience-based prevention trial.

    “However, there is neither a consensus nor a process to generate consensus on what would constitute an acceptable endpoint for a geroscience-based prevention trial.”
  • The paper relies on this without support.

    Animals are poor models of many human diseases, and genetically manipulated models for one disease may not be relevant for other diseases.

    “However, animals are poor models of many human diseases, and genetically manipulated models for one disease may not be relevant for other diseases.”
  • The paper relies on this without support.

    The US Food and Drug Administration requires that drugs being evaluated for potential approval must affect how a patient feels, functions, or survives.

    “From a regulatory perspective, the US Food and Drug Administration requires that drugs being evaluated for potential approval must affect how a patient feels, functions, or survives.”
  • The paper relies on this without support.

    The sample size and duration of trials is generally the most critical determinant of feasibility.

    “The sample size and duration of trials is generally the most critical determinant of feasibility.”
  • The paper relies on this without support.

    Age-related diseases are uncommon.

    “Sample sizes for incident health events, such as a diagnosis of Alzheimer’s disease, are generally very large and require long follow-up because these age-related diseases are uncommon.”
  • The paper relies on this without support.

    Some diseases have a long prodromal period and the benefits of even a successful treatment might not be evident for a number of years.

    “Some diseases have a long prodromal period and the benefits of even a successful treatment might not be evident for a number of years.”
  • The paper relies on this without support.

    There is an intense thirst for treatments to slow aging.

    “There is an intense thirst for treatments to slow aging.”
7 of 7

Citations

Cites

Flagged first, then by citations

Cited by

Sorted by citations

Metadata

Authors
Cummings, S. R., & Kritchevsky, S. B.
Affiliations
  • University of California, San Francisco (US)
  • Wake Forest University Health Sciences (US)
Corresponding author
Cummings, Steven R.
Venue
GeroScience
Published
2022-10-19
Volume
44
Issue
6
Pages
2925-2931
ISSN
  • 2152-4041 (ISSN-L)
  • 2509-2715
  • 2509-2723
OpenAlex
W4306772674
Cited by
54
References
23
Citations by year
2026: 25 · 2025: 12 · 2024: 9 · 2023: 7 · 2022: 1
FWCI
2.36
Type
journal article
Language
en
Open access
Locations
  • GeroScience (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • eScholarship (California Digital Library) (submittedVersion, cc-by)
Funders
National Institute on Aging (AG059416, AG061456, AG021332, AG058556)
Topics
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Aging › Genetics, Aging, and Longevity in Model Organisms
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Cell Biology › Telomeres, Telomerase, and Senescence
  • Physical Sciences › Mathematics › Statistics and Probability › Statistical Methods in Clinical Trials
Keywords
clinical trial endpoints, surrogate markers, biological age, health outcomes, anti-aging therapy, aging, biomarker validation
MeSH terms
Geroscience, Biological Products, Biomarkers, Outcome Assessment, Health Care

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

A language model drew the claims from the paper on 2026-11-10, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).

A second model labelled each claim on 2026-11-10 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).

Cited sources are checked with a language model’s help (gpt-6-sol).

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review