Introduction: Human ageing is a dynamic process depending on intrinsic and extrinsic factors and its evolution is a continuum of transitions, involving multifaceted processes at multiple levels. It is recognized that frailty and sarcopenia are shared by the major age-related diseases, thus contributing to elderly morbidity and mortality. They are major health issues in elderly populations, given their high prevalence and association with several adverse outcomes. Due to their complex phenotypes and underlying pathophysiology, the need for robust and multidimensional biomarkers is now essential to move towards a more personalized care and prevention. Methods: The NU-AGE project [1] regroups 1250 free-living elderly people (65–79 y.o., men and women), free of major diseases, recruited within 5 European centres. Twenty percent of the subjects were pre-frail as defined by the criteria proposed by Fried et al.[2]. Six hundred twenty five volunteers were randomly assigned to an intervention group (1-year Mediterranean diet). A sub-cohort consisting in first, 120 subjects, half pre-frail randomly selected from the Italian and Polish centres, and secondly, 92 subjects shifting their frailty status were included for untargeted serum metabolomics at T0 (recruitment) and T1 (after diet intervention). Results: Metabolomics enables to discriminate sub-phenotypes of pre-frailty both at the gender level and depending on the pre-frailty progression and reversibility. Additionally, early and/or predictive markers of pre-frailty were identified in both populations. Conclusion: These results open the door, through multivariate strategies, to a possibility of monitoring the disease progression over time and/or in response to interventions at a very early stage