Biomarker
Immune Age Signals
Source Summary
16 reviewed sources across 6 biomarker evidence groups.
- IMM-AGE2
- iAge / inflammatory age2
- Validation / methods / frameworks4
- Other immune-clock / model families5
- FDA / regulatory context2
- NCBI / terminology1
Biomarker Snapshot
Umbrella category spanning distinct model families.
Not one settled measurement or diagnostic test.
Inputs differ by model; no model is assumed to use every modality.
Compared as distinct approaches, not interchangeable scores.
Includes longitudinal, cross-sectional, and single-cell model development.
No individual clinical use is inferred.
Model development and association evidence are stronger than utility evidence.
External validation, transportability, and longitudinal responsiveness remain uneven.
Clinical decision usefulness is not supported by this source set.
Framework discussion does not validate a surrogate or clinical endpoint.
Qualification must be assessed for an exact biomarker and context of use.
Commercial availability would not itself establish validation.
Original studies describe models; reviews and regulator pages define limits.
Confidence applies to the cautious scope shown here.
Executive Summary
This umbrella record covers distinct immune-aging model families rather than one settled measurement. IMM-AGE, iAge, IMMAX, sc-ImmuAging, and related immune-clock approaches use different data, cohorts, and model tasks. The reviewed sources support research-stage model development, cohort association, methods, and trial-readiness context. They do not establish clinical decision usefulness, consumer-test readiness, or interchangeability.
Why It Matters
Age-related immune-system changes are relevant to aging research, and model families attempt to summarize different cellular, inflammatory, transcriptomic, and multimodal patterns. The field is useful for separating model development and association from external validation, endpoint readiness, clinical utility, and regulatory qualification. That separation matters because the phrase immune age can sound more clinically certain than the reviewed evidence supports.
Model Family Map
| Model / family | Data modality | Development context | Reported research use | Evidence layer | Validation posture | Limitation | Source |
|---|---|---|---|---|---|---|---|
| IMM-AGE | Longitudinal high-dimensional immune profiling | 135 healthy adults monitored over nine years | Research metric of immune-system change; mortality association reported in Framingham | Model development and external association | Association reported; clinical utility not established | Specialized assays and computation; not a diagnostic or individual treatment tool | Alpert et al. 2019, PMID 30842675 |
| iAge / inflammatory age | Blood immunome and systemic inflammation patterns | 1,001 individuals aged 8-96; deep-learning model | Research associations with multimorbidity, immunosenescence, frailty, cardiovascular aging, and centenarian status | Model development, cohort association, mechanistic follow-up | Reported associations; individual clinical utility not established | Assay and cohort context may not generalize; association is not diagnosis or causality | Sayed et al. 2021, PMID 34888528 |
| IMMAX | Flow-cytometry blood-cell subpopulations | Age-specific centiles in 1,605 individuals | Method and reference-interval work approximating immune-age patterns | Method development and reference centiles | Confirmatory longitudinal analysis remains needed | Biomarker selection affects results; IMMAX is not identical to IMM-AGE | Brode et al. 2023, PMID 37685992 |
| sc-ImmuAging / single-cell approaches | Single-cell RNA sequencing | Cell-type-specific clocks from 1,081 healthy individuals aged 18-97 | Research application to infection and vaccination transcriptome contexts | Single-cell model development and research application | Technically specialized; clinical decision usefulness not established | Not directly comparable with flow-cytometry, inflammatory, or multimodal scores | Li et al. 2025, PMID 40044970 |
Measurement and Data-Modality Map
| Modality | Typical research context | What it may capture | Major limitation |
|---|---|---|---|
| Cellular composition signals | Flow cytometry and immune-cell subset analysis | Age-related differences in immune-cell proportions | Composition is platform-, gating-, and population-dependent. |
| Immune-cell phenotypes | High-dimensional longitudinal immune profiling | Coordinated cellular-state patterns over time | Specialized assays and computation limit transportability. |
| Cytokine and inflammatory signals | Blood immunome and inflammaging models | Patterns of systemic inflammatory signaling | Inflammation is context-sensitive and does not equal overall immune function. |
| Transcriptomic signals | Bulk or cell-type-specific expression models | Age-associated gene-expression patterns | Batch effects, tissue context, and model recalibration can change output. |
| Single-cell approaches | scRNA-seq immune-cell clocks | Cell-type-specific aging patterns and heterogeneity | Complex workflows are not directly comparable with routine blood panels. |
| Multimodal and composite scores | Integrated immune-aging models and atlases | Combined cellular, molecular, or inflammatory patterns | Different inputs and tasks produce non-equivalent composite scores. |
Evidence-Stage Map
| Evidence layer | Reviewed posture | What the evidence supports | What remains missing |
|---|---|---|---|
| Model development | Present; model-specific | Published derivation of IMM-AGE, iAge, IMMAX, and sc-ImmuAging approaches | A single standardized immune-age construct is not established. |
| Internal validation | Partial; model-specific | Within-study performance and model checks reported by individual papers | Methods and validation designs are not harmonized across model families. |
| External validation | Limited | Selected external associations, including the reported IMM-AGE Framingham analysis | Broad cross-platform, cross-population, and independent replication remains uneven. |
| Cohort association | Present; model-specific | Reported associations with aging-related phenotypes in source cohorts | Association does not establish diagnosis, causality, or individual utility. |
| Replication | Limited | Multiple approaches point to measurable immune-aging heterogeneity | Direct replication using equivalent assays, tasks, and endpoints is not universal. |
| Longitudinal evidence | Partial | IMM-AGE includes longitudinal development; some model applications examine change | Within-person stability, responsiveness, and clinically meaningful change are not established broadly. |
| Trial-endpoint discussion | Present as framework | Current sources define evaluation criteria and trial-readiness gaps | Framework discussion does not validate a specific exploratory, pharmacodynamic, or surrogate endpoint. |
| Clinical utility | Not established | The reviewed literature identifies translation requirements and gaps | Evidence that using a score improves clinical decisions or outcomes. |
| Regulatory qualification | Not established in this review; model-specific | FDA sources explain qualification and context-of-use requirements | An exact qualified immune-age model and context of use must be demonstrated, not assumed. |
Validation and Generalizability
Cohort dependence
Model outputs may reflect derivation-cohort characteristics. Association in one cohort does not establish usefulness in another.
Assay and platform dependence
Flow cytometry, cytokine assays, transcriptomics, and single-cell platforms introduce distinct technical dependencies.
Population and age-range dependence
Population composition and covered age ranges can affect calibration and interpretation.
Recalibration
Methods reviews identify recalibration, batch effects, and endpoint mismatch as recurring concerns.
External and longitudinal validation
Independent replication, within-person stability, responsiveness, and longitudinal meaning remain uneven across families.
Clinical transfer
Reproducible model performance does not by itself show that the measure improves clinical decisions or outcomes.
Trial-Readiness and Endpoint Status
Research signal
Published models can support research analysis within their documented assay and cohort contexts.
Exploratory endpoint
Use would require a defined model, assay, analysis plan, and interpretation.
Pharmacodynamic marker
Responsiveness to an intervention and the meaning of change must be demonstrated for the exact model.
Trial-stratification candidate
Predictive value and transportability would need model-specific prospective evidence.
Surrogate endpoint
No blanket claim that an immune-age score substitutes for a clinical outcome is supported.
Validated clinical endpoint
The reviewed sources do not establish these models as validated clinical endpoints.
Regulatory-qualified biomarker
FDA qualification is an exact biomarker and context-of-use process, not validation of a field label.
Consumer Claim-Risk Watch
| Claim pattern | Why it is risky | Safer evidence-based framing |
|---|---|---|
| Your immune age is X | Treats heterogeneous model outputs as one settled measurement. | A specified research model produced a score in a defined assay and cohort context. |
| Diagnosis or disease-risk certainty | Cohort association does not establish diagnosis or individual prognosis. | Some studies report model-specific associations that require independent validation and careful interpretation. |
| Validated consumer test | Commercial access or a numeric result does not establish clinical validation. | Consumer-test readiness is not established by the reviewed source set. |
| Treatment or supplement recommendation | Model studies do not establish treatment selection, dosing, or personalized guidance. | Use the evidence only for research-stage model and validation analysis. |
| Biological-age reversal | A score change is not proof of reversal, benefit, or improved outcomes. | Any observed score change remains model- and context-specific. |
| Clinically actionable | Clinical decision usefulness has not been demonstrated. | Clinical utility remains an evidence gap. |
| Scores are interchangeable | Models differ in inputs, tasks, cohorts, and endpoints. | Compare model families descriptively without equating their numeric outputs. |
| FDA validated or qualified | Qualification applies to an exact biomarker and context of use. | Describe FDA qualification as a context-specific process unless an exact qualified use is documented. |
What This Can Support
Research-stage immune-aging context
Organize published model-development and immune-aging biology evidence.
Model-family comparison
Compare inputs, tasks, cohorts, and limitations without treating scores as equivalent.
Cohort and methods analysis
Map how study design and measurement platforms shape model interpretation.
Validation-gap mapping
Track external validation, transportability, reproducibility, and longitudinal gaps.
Trial-readiness discussion
Evaluate what model-specific evidence would be needed for exploratory trial use.
Qualification and context-of-use analysis
Apply FDA terminology and process boundaries without assuming qualification.
Consumer claim-risk review
Identify where public language outruns research-stage evidence.
What This Does Not Prove
No diagnosis or individual prediction
These sources do not establish an immune-age diagnosis, biological-age certainty, individual disease prediction, or clinical decision utility.
No consumer or intervention guidance
They do not establish a validated consumer test or support treatment, dosing, supplement, anti-aging, or personalized recommendations.
No outcome benefit from score change
Changing a score is not proof of improved lifespan, healthspan, clinical outcomes, or intervention benefit.
No blanket endpoint or regulatory status
No surrogate-endpoint, validated clinical-endpoint, or FDA qualification claim is supported without an exact model and context of use.
No model interchangeability
IMM-AGE, iAge, IMMAX, sc-ImmuAging, inflammatory-age models, and other immune clocks should not be treated as equivalent.
Timeline / Milestones
-
IMM-AGE publication
Alpert et al., Nature Medicine
Longitudinal model development and a reported external association; not clinical utility.
-
iAge publication
Sayed et al., Nature Aging
Inflammatory aging model and cohort associations; not diagnosis or individual actionability.
-
IMMAX method study
Brode et al., IJMS
Flow-cytometry centile approach; confirmatory longitudinal work remains needed.
-
Validation and translation frameworks
Moqri et al.; Herzog et al.
General frameworks map comparability, generalizability, and clinical-translation requirements.
-
Single-cell clocks and immune atlas work
Li et al.; Chander et al.
Expands model and biology context without establishing a consumer or clinical test.
-
Trial-readiness and methods frameworks
Cipriano et al.; Yu et al.
Defines evaluation criteria and heterogeneity gaps; not validation of one model.
-
LongevityNext source review
Source-reviewed editorial record
Cautious public scope retained; clinical utility and consumer readiness remain not established.
Source Posture
Original model publications
Support model descriptions, derivation contexts, and source-reported findings.
Cohort and association evidence
Supports reported associations within study contexts, not diagnosis, causality, or clinical actionability.
Validation and methods reviews
Map heterogeneity, comparability, recalibration, generalizability, and translation gaps.
FDA regulatory context
Explains qualification and context-of-use requirements; it does not approve the models in this record.
NCBI terminology context
Supports biomarker and endpoint terminology rather than immune-age model validation.
Detailed Sources
View detailed source table
IMM-AGE
| Source | Model / family | Type | Supports | Limitation | Link |
|---|---|---|---|---|---|
| Alpert et al. 2019 - PubMed | IMM-AGE | Primary study | Longitudinal model development and reported Framingham mortality association | Research metric; not diagnostic or clinical-utility validation | PubMed PMID 30842675 |
| Alpert et al. 2019 - Nature Medicine | IMM-AGE | Publisher article | Article record for the IMM-AGE study | Same study, not independent replication | Nature Medicine |
iAge / inflammatory age
| Source | Model / family | Type | Supports | Limitation | Link |
|---|---|---|---|---|---|
| Sayed et al. 2021 - PubMed | iAge | Primary study | Inflammatory aging model, cohort associations, and CXCL9 interpretation | Does not establish individual clinical utility or treatment relevance | PubMed PMID 34888528 |
| Sayed et al. 2021 - Nature Aging | iAge | Publisher article | Article record for the iAge study | Same study, not independent replication | Nature Aging |
Validation / methods / frameworks
| Source | Model / family | Type | Supports | Limitation | Link |
|---|---|---|---|---|---|
| Cipriano et al. 2026 - PubMed | Immune-aging biomarkers | Trial-readiness framework | Evaluation criteria and remaining gaps for trial contexts | Framework; not validation of a specific endpoint | PubMed PMID 42399672 |
| Cipriano et al. 2026 - Nature Medicine | Immune-aging biomarkers | Publisher article | Article record for the trial-readiness framework | Same framework, not a separate validation study | Nature Medicine |
| Moqri et al. 2024 - PubMed | Aging biomarkers | Validation framework | Comparability, generalizability, and validation requirements | General aging-biomarker framework, not immune-specific | PubMed PMID 38355974 |
| Herzog et al. 2024 - PubMed | Aging biomarkers | Translation framework | Clinical-translation barriers and individual-level validation needs | General framework, not immune-specific | PubMed PMID 39285015 |
Other immune-clock / model families
| Source | Model / family | Type | Supports | Limitation | Link |
|---|---|---|---|---|---|
| Yu et al. 2026 - PubMed | Immune aging clocks | Methods review | Tasks, modalities, models, recalibration, and common limitations | Review-level evidence; does not validate clinical use | PubMed PMID 41827856 |
| Yu et al. 2026 - PMC | Immune aging clocks | Full-text methods review | Accessible methods and heterogeneity context | Same review, not independent validation | PMC12984438 |
| Li et al. 2025 - PubMed | sc-ImmuAging | Primary study | Single-cell model development and research applications | Specialized research tool; not consumer or diagnostic validation | PubMed PMID 40044970 |
| Chander et al. 2025 - PubMed | Human Immune Health Atlas | Primary resource study | Multidimensional, non-linear immune-aging biology context | Resource study, not an immune-age clinical tool | PubMed PMID 41162704 |
| Brode et al. 2023 - PubMed | IMMAX | Method study | Flow-cytometry index and age-specific centile approach | Confirmatory longitudinal analysis remains needed | PubMed PMID 37685992 |
FDA / regulatory context
| Source | Model / family | Type | Supports | Limitation | Link |
|---|---|---|---|---|---|
| FDA - About Biomarkers and Qualification | Context of use | Regulator guidance | Biomarker and qualification terminology and context-specific interpretation | Not immune-age-specific and not evidence of model qualification | FDA qualification overview |
| FDA - Status of Qualification Submissions | Qualification status process | Regulator status resource | Public pathway for checking qualification projects and statuses | Status must be checked for an exact biomarker and context of use | FDA submission status |
NCBI / terminology
| Source | Model / family | Type | Supports | Limitation | Link |
|---|---|---|---|---|---|
| FDA-NIH BEST glossary | Biomarker terminology | Government glossary | Biomarker, endpoint, and context-of-use terminology | Not immune-age-specific and not model validation | NCBI Bookshelf NBK326791 |
Confidence / Methodology
Editorial status
Editorial review complete; source freshness checked
Evidence separation
This review keeps model families and evidence layers separate. Reported cohort associations are not treated as clinical utility. Trial-readiness discussion is not treated as endpoint validation. FDA qualification is described only as a model- and context-of-use-specific process, not as blanket qualification of immune-age signals.
Last reviewed
2026-07-09