HealthAge was founded in 2023 and is a large-scale, multi-center collaborative project involving the Gérontopôle at Toulouse University Hospital in conjunction with the University of Toulouse and the Institut National de la Santé et de la Recherche Médicale (Inserm) [1]. The main objective of HealthAge is to extend healthspan as human life expectancy has dramatically increased, while healthspan has not followed suit, therefore negatively impacting the quality of life of older adults [2]. In HealthAge, this ambitious goal is tackled by connecting function at the public health level with processes at a biological level centered on the concept of intrinsic capacity (IC) [3].

In 2015, the World Health Organization (WHO) published the first World Report on Ageing and Health [3] in order to implement a new public-health strategy on aging. In this report, the WHO defined healthy aging as “the process of developing and maintaining the functional ability that enables people to be and to do what they have reason to value.” Functional ability is determined by the IC of an individual, which is the composite of all the physical and mental capacities plus interacting environmental influences [2,3,4]. To operationalize IC in a clinical context, the WHO have identified six connected core components reflecting multiple physiological functions relevant to aging: locomotion, cognition, vitality, vision, audition, and psychological well-being. Evidence demonstrates that deficits in IC strongly predict adverse health outcomes including falls, care dependency, disability, and mortality [6]. IC and its six domains can be assessed clinically according to the recommendations in the WHO Integrated Care for Older People (ICOPE) handbook (5 step process) [7]. This can be achieved through the use of an application (ICOPE Monitor that was developed and trialed at the Gérontopôle, Toulouse University, France), a web platform or through a telephone call with a trained research nurse [8, 9]. ICOPE represents an innovative, patient-centered system to manage and optimize health and well-being. Note this review is founded on the original ICOPE program for human IC assessment; however, ICOPE 2 (4 step process) has recently been released by the WHO, which represents a refined version with expanded assessments and clearer pathways for its integration into primary care [10].

As healthcare systems face the growing burden of aging populations, ICOPE provides a structured framework that combines preventive care, clinical management and social support services. This holistic approach aims to enhance health outcomes through long-term follow-up, reducing hospitalizations, long-term care dependency and overall healthcare costs. Implementing ICOPE in primary care will generate valuable insights into aging trajectories, enabling researchers to identify patterns, assess intervention efficacy and refine strategies for extending healthspan. This new shift in the health paradigm will in turn advance the field of geroscience through the promotion of healthy aging. Recently, a large-scale feasibility study performed at the Gérontopôle, Toulouse University Hospital (France) has provided evidence to support the implementation of ICOPE on a large scale in clinical practice [8, 11]. Through the use of the ICOPE program, the WHO aims to reduce the number of dependent people by 15 million worldwide by 2025 [12].

The main purpose of this narrative review is to describe the unique HealthAge initiative encompassing the human, outbred SWISS mice and African Turquoise killifish (TK) cohorts and describe how IC is measured longitudinally in each species. The animal models with their accelerated aging profiles in conjunction with human phenotyping will expedite discovery and validation of novel biomarkers and drug targets that protect against IC declines.

The INSPIRE translational (INSPIRE-T) human cohort is a 10-year, observational, repeated measures study, designed to capture aging trajectories based on the WHO ICOPE (original version) program and the assessment of IC. The INSPIRE-T study was approved by the French Ethics Committee located in Rennes (CPP Ouest V) and the study was registered at http://clinicaltrials.gov (NCT04224038). Recruitment to the study began in October 2019 and all participants gave signed informed consent with the target number of subjects (n = 1000) being recruited by December 2021. The open cohort (ongoing recruitment) consists of 1109 participants with data available at baseline (61.9% female) aged from 20 to 102 years old (mean age ± SD, 62.4 ± 19.0 years) recruited from the Toulouse area (France) exhibiting varying levels of functional status. Sample size calculations were not performed due to the exploratory nature of HealthAge. We considered an approach based on the potential to obtain parameter estimates with sufficient precision informed by 1000 human subjects, which represents the maximum number of people that can be monitored with the available funding. The same rationale was used to derive the numbers of mice (n = 1576) and TK (n = 300). Participants in the INSPIRE-T cohort are stratified into 10-year age groups with oversampling of participants ≥ 70 years old to compensate for increased drop-out in this age bracket and to enable the better investigation of frailty and age-related disease. Recruitment to the study remains permanently open to maintain a minimum pool of 1000 participants. The rationale, methodology, and objectives of the INSPIRE-T study are comprehensively reviewed elsewhere [12, 13].

In brief, data collection in the INSPIRE-T human cohort consists of information on the six domains of IC assessed using the ICOPE program. Participants had IC (ICOPE Step 1) monitored every 4 months during the first year of the study then every 6 months for the remainder of the study. Participants undergo an additional thorough clinical assessment (ICOPE Step 2) and blood sampling if IC declines are detected, the latter enabling the investigation of blood-borne mediators expressed at the same time that IC declines are apparent. Demographic measures, information on health status, cognition (Mini Mental State Examination, MMSE [14]), Fried’s frailty phenotype [15], functional ability [16, 17], oral health [18], lifestyle (including physical activity, sleep statistics, diet [19, 20], smoking status and alcohol consumption), as well as participant-reported outcomes (cognition [21], mobility, fatigue, social isolation [22]) are also collected to characterize participants and provide supplementary information on IC domains. Participants will be evaluated once a year for all clinical assessments throughout the 10-year study. Table 1 (column 1) provides a full list of clinical tests that are performed in the INSPIRE-T human cohort in order to capture IC.

To further probe the biology underlying IC, all participants underwent dual-energy X-ray absorptiometry (DXA) measurements for body composition at baseline and year 2 (with subsequent 5 yearly follow-ups planned). Whole body and brain magnetic resonance imaging (MRI) was also performed at study baseline in a subset of participants (n = 113). Furthermore, maximum oxygen consumption (VO2 max) as well as isokinetic muscle strength (Cybex) was measured between baseline and 24 months in a subset of participants (n = 279 and n = 286 respectively).

A biobank was established from biospecimens including blood, urine, saliva, and dental plaque, which were collected at study baseline, then annually as detailed elsewhere [12]. Nasopharyngeal swabs, stools, hair follicles, skin surface samples, and skin biopsies were also collected optionally at baseline (stool and hair follicle samples will be subsequently collected every 2 years throughout the study).

The INSPIRE mouse cohort consists of 1576 outbred SWISS mice (Janvier Labs, France) and is described in detail elsewhere [23, 24]. Outbred SWISS mice were chosen as a model system as they retain substantial genetic variability in comparison to the more traditional inbred laboratory mouse strains and thereby provide a more suitable murine model to reflect the diversity of human genetics. The colony consists of both male and female mice enabling inter-sex differences to be explored throughout the course of aging. The mice were housed 4 per cage in Digital Ventilated Cages (DVC®) (Tecniplast, Italy) in a temperature-controlled environment (21 °C ± 2 °C) with a 12-h light–dark cycle and had unrestricted access to food and water. This caging system enables 24 h/7 d data collection on locomotion, body weight, as well as food and water consumption.

The mice were divided into four groups to better model human habits: control animals, high fat/high sucrose diet fed (HF/HSD), “wheel” (to enable voluntary physical activity) and HF/HSD diet + wheel. This experimental paradigm mimics “healthy” versus “unhealthy” human lifestyles focusing on nutrition and exercise as the lifestyle variables. Subgroups of mice from these four groups were culled at 6, 12, 18, and 24 months, and these mice were designated the “cross-sectional cohort” (n = 1456) [23, 24]. These time points correspond to an approximate age range of 30 – 80 years old in humans. The remaining 120 mice (60 females and 60 males) were designated the “longitudinal cohort” and were left to live a natural lifespan solely to determine strain longevity under our particular housing conditions (without intervention or assessment). Of note, the mice in the longitudinal cohort were under control feeding conditions.

Key components of IC were assessed in the cross-sectional mouse cohort at 6, 12, 18, and 24 months as an approximation to ICOPE vitality, locomotion, cognition, and psychological well-being screening. Vision and audition were not assessed in the mouse cohort at this time. In a future planned mouse aging study, vision and audition will be monitored and evaluated using a novel digitalized system that is currently under development. Thus, the four aforementioned IC domains were assessed through a non-invasive “IC score,” an adapted version of the previously described Valencia Frailty Score for mice [25]. This score includes as measures of vitality: % body weight change (over 1 month), grip strength, and motor coordination assessed using the tight-rope test. Locomotion was also measured through the assessment of spontaneous walking speed using the open-field test and running speed (and time) using the incremental treadmill test [24]. To better mirror human ICOPE IC screening, cognition (working memory) was evaluated using the Y-Maze spontaneous alternation test [26], and as a measure of psychological well-being, the open-field test was performed to assess anxiety [27]. The same open-field test as for locomotion was used where anxiety-like behaviors such as reluctance to explore the center of the arena and increased thigmotaxis (wall following) were observed in an unfamiliar open space.

Cardiorespiratory function has been suggested to indirectly influence the vitality domain of IC through metabolic changes [5], thus cardiac systolic and diastolic functions were examined by echocardiography and Doppler echocardiography at 6, 12, 18, and 24 months in the mice. Urinary function was also measured at the same time points using a spontaneous void spot assay due to the association of urinary incontinence with social hindrance and loneliness in humans [23, 24]. Social isolation and loneliness have in turn both been associated with worse cognition and Alzheimer’s disease in older adults [28]. Indeed, ICOPE 2 now includes clinical assessment and support for people with urinary incontinence [10]. Furthermore, complete blood counts were conducted in the mice at 6, 12, 18, and 24 months to assess hematological factors including red blood cells, reticulocytes, and platelets as well as myeloid and lymphoid cell populations [24]. At the time of endpoint euthanasia of the mice subgroups, blood, feces, urine, and organs were collected and stored for future biomarker studies.

The African Turquoise Killifish (TK) (Nothobranchius furzeri) are short-lived vertebrates with a median lifespan of 3–8 months, which therefore lend themselves to aging studies [29]. The rapid aging of TK shares several characteristics with human aging including telomere shortening, cancer, immuno-compromisation, reduced locomotion and cognitive decline [29]. The INSPIRE TK cohort (GRZ strain) consists of a replenished, ongoing cohort of about 300 TK with an approximate 50:50 ratio of males to females, thus enabling inter-sex differences in aging to be explored as with the SWISS mice cohort. The TK were raised at 28 °C in a central filtration recirculating system with a 14-h light/10-h dark cycle under standard laboratory conditions. The fish were housed individually, with one fish per 3 L of system water. The influences of lifestyle factors (diet and exercise) on IC were not assessed as a function of aging as with the mice cohort. All TK were left to live a natural lifespan, and in the main cohort, noninvasive IC measures of vitality, locomotion, and cognition were taken over time.

To provide measures of vitality, in a sub-population of TK (n = 30), growth (weight and length) was measured every 2 weeks throughout lifespan. Furthermore, metabolic blood parameters including glucose and lactate levels were measured in young TK (2 months) versus old TK (5–6 months). Vitality was also determined by measuring the reaction time for fish to reach food at the water surface, and food intake was assessed using fluorescent artemias (shrimp) through the subsequent detection of the fluorescent signal in the gut and feces.

Locomotion both spontaneous and forced (using a swim tunnel) as well as other behaviors including aggressiveness and temerity were investigated using a 24 h/7 d infra-red video tracking system. Anxiety was evaluated using the Novel tank diving assay where an increase in over-time vertical swimming from the initial position at the bottom of the tank is interpreted as a reduction in anxiety [30]. Collectively, anxiety, aggressiveness, and temerity provide measures of fish psychological well-being. At present, recorded swimming patterns are under analysis using computer-assisted methods in collaboration with the Computing Research Institute in Toulouse (Institut de Recherche en Informatique de Toulouse). Further pertaining to the locomotion domain of IC, the cross-sectional area of the fish trunk was measured to monitor muscle loss with age as a model of sarcopenia in the fish.

Cognition in young TK (2 months) versus old TK (5–6 months) was assessed using X and T maze set-ups using a Pavlovian conditioning model to provoke swimming [31]. Mitochondrial morphology was also examined in skeletal muscle of young versus old TK (2 months versus 5–6 months) using electron microscopy providing a measure of age-related metabolic changes, which have been implicated in the vitality domain of IC [5]. To date, vision and audition have not been assessed in the TK. However, vision will be assessed using an optokinetic test chamber [32], and audition will be investigated by studying the auditory-evoked startle response and hearing sensitivity to acoustic startle stimuli in future experiments [33]. Additionally, histo-morphological analysis of the retina will be performed in both the larvae and adult fish to examine structural changes in the eye occurring with age. Complete blood counts will also be performed in young versus old TK from samples isolated from the head kidney (playing a similar role to the bone marrow in humans), blood, and gills to mirror the hematological assessments performed in the mice [24]. In contrast to the INSPIRE mouse cohort, organs and blood samples were not routinely collected and stored due to the lack of biological matter yielded per TK for subsequent laboratory analysis. Instead, biomarker studies will be conducted on tissue from whole TK.

Both targeted and untargeted (discovery-based) techniques will be employed to identify novel biomarkers of IC declines per se as well as biomarkers of declines in its constituent domains using the biological specimens collected from all three species. In line with the expertise available in the HealthAge research community, the following bio-processes will be examined: organ function, cognition, immune-function/inflammation, metabolism and mitochondrial biology. Specific targets to be investigated include the following: ATPase inhibitory factor 1 (IF1), T-cell subset composition, fat distribution, and apelin. The reasons for this are that these pleiotropic mediators play a role in physiological processes that are altered in disease and are thereby intimately linked to IC. ATPase inhibitory factor 1 (IF1) is an endogenous inhibitor of mitochondrial ATP synthase (of the electron transport chain) and plays a pathological role in disease states such as cancer, cardiovascular disease, and diabetes [34]. Altered T-cell subset distribution contributes to immuno-senescence leading to increased susceptibility to a multitude of age-associated diseases [35] and increasing evidence suggests that fat distribution in different depots affects overall health and particularly metabolic health [36]. Lastly, apelin, a putative biomarker of sarcopenia [37], is of interest because it is involved in many biological processes including cardiovascular homeostasis, neuroprotection, immunity, oxidative stress and metabolic disorders [38, 39]. The multiple functions of apelin are attributed to the wide expression pattern of its receptor, APJ [38, 39].

In terms of the untargeted biomarker discovery research arm, multi-omics including transcriptomic, metabolomic, lipidomic, proteomic as well as whole genome epigenetic studies will be performed on the collected biospecimens. Multi-omics will be used to identify the most abundantly changing targets to produce ranked lists in the three species for comparative analysis. We will subsequently use linear mixed effect models to investigate whether single biomarkers associate with IC declines over time. Receiver Operating Characteristic (ROC) curve and Area Under Curve (AUC) analyses will also be used to assess biomarker performance. Machine learning algorithms will be employed to analyze the ability of multiple biomarkers to predict IC declines where interactions and more complex non-linear relationships might be at play. The tissues/organs obtained from the animals provide a valuable resource for biomarker identification as such parallel tissues are not available in humans in HealthAge. Nevertheless, putative biomarkers of IC declines will be cross-validated where possible in humans. Furthermore, the causal role of biomarkers in loss of IC will be explored where mechanistically relevant using the animal models in order to identify potential gero-therapeutic targets for the prevention of IC declines in humans.