Potential of FGF2 and GIPR as immunological biomarkers and therapeutic targets in recurrent pregnancy loss

Recurrent pregnancy loss (RPL) is a complex obstetric complication whose immunological mechanisms remain incompletely understood. This study aimed to identify immune-related biomarkers for RPL by integrating the GSE165004 dataset with immune-related genes from the MSigDB database. A total of 234 immune-related differentially expressed genes (IDEGs) were identified, which were enriched in pathways such as natural killer cell-mediated cytotoxicity, JAK-STAT signaling, and chemokine signaling. Using machine learning algorithms (LASSO, SVM-RFE, RF), four hub genes—FGF2, TINAGL1, CKLF, and GIPR—were screened. Among them, FGF2 and GIPR were significantly downregulated in the RPL group ( P < 0.05). Diagnostic efficacy was evaluated using ROC curves, and a nomogram was constructed. Immune cell infiltration analysis via CIBERSORT revealed that FGF2 showed a negative correlation with monocyte infiltration, whereas GIPR was positively correlated with resting dendritic cells ( P < 0.05), suggesting their involvement in modulating the immune microenvironment. Upstream regulator analysis using NetworkAnalyst predicted that FGF2 may be regulated by miRNAs such as miR-15a-5p, while GIPR was associated with transcription factors SP1/SP3. Validation using single-cell transcriptomics and preliminary clinical samples (RT-qPCR/WB) confirmed the downregulation of FGF2 and GIPR in RPL ( P < 0.05). In conclusion, our findings suggest that altered FGF2 and GIPR expression may be associated with immune dysregulation in RPL and warrant further functional investigation. These findings provide preliminary insights for potential immunotherapeutic strategies in RPL, although larger-scale validation and functional studies are needed.