506 BO22 – Unveiling the athletic future: GPS & multi-omics power to foresee elite female football injuries

Background In the pursuit of injury prevention in elite female football, accurate predictive models are invaluable tools for coaches and athletes. Addressing a gap in existing research, we aim to create a robust injury risk prediction model by integrating player characteristics, workload (external load), genetic predisposition and omic data (internal load) within a longitudinal framework. Design Retrospective (2019–2020, 2020–2021, and 2021–22 seasons) and prospective (2021–23) longitudinal data for model building and validation, respectively. Setting Women's elite soccer players. Participants FC Barcelona women's team (n=24). Assessment of risk factors Anthropometric information, genomic data (108 single nucleotide polymorphisms – SNPs) and urinary concentration of 69 metabolites measured every 3 months. Main outcome Non-contact injury (muscle, tendon, ligament and cartilage). Results Association between injury risk and player load (p=0.045) as well as decelerations (p=0.017). Six SNPs located at DCN, ADAMTS5, FBN2, ESRRB, and FBN2 genes exhibited associations with non-contact injuries after adjusting for player load (p<0.001). We devised a genetic score (Figure 1A) revealing substantial differences in injury risk probability (Figure 1B). Further, three metabolites, including b-alanine, displayed statistically significant associations with non-contact injuries (Figure 1C). Our predictive model, sequentially incorporating player load, decelerations, the genetic score, and one metabolite, showcased enhanced performance. Notably, the C-index in the validation set increased from 0.60 with player load alone to 0.68 with the inclusion of metabolites, and further to 0.74 with genetics (Figure 1D). Conclusion This ground-breaking study underscores the substantial enhancement of injury risk prediction achievable through omics technologies. Coaches can now implement tailored interventions rooted in athletes' genetic predispositions. The incorporation of metabolomic data allows for ongoing internal workload monitoring, enabling dynamic modification of injury risk in response to the current state of women.players. This research represents a pivotal stride toward safeguarding the well-being and performance of elite female football players.