Baseline versus 2-week change in suPAR: distinct predictive and early-response roles in rTMS for adolescent depression

Abstract Objective Adolescent depression is a prevalent condition with limited effective treatment options. The inflammatory hypothesis of depression has gained increasing attention, yet evidence in adolescent populations remains scarce. Soluble urokinase-type plasminogen activator receptor (suPAR), a stable chronic inflammatory marker, exhibits dysregulation in major depressive disorder (MDD). However, the relationships between the level of suPAR and the effects of rTMS treatment remain unclear. This study aimed to investigate the effects of rTMS combined with sertraline on clinical symptoms and inflammatory markers (suPAR, CRP, IL-6) in adolescents with depression. Methods This study has been registered with the China Clinical Trial Center (ChiCTR2500112243). A total of 120 adolescents with depression were enrolled to receive rTMS plus sertraline ( n = 84) or sham stimulation plus sertraline ( n = 36). Demographic data and childhood trauma (CTQ) were collected. The severity of depressive symptoms was assessed using HAMD. Pain catastrophizing and pain vigilance were evaluated with the PCS and the PVAQ respectively. Levels of serum inflammatory factors, including suPAR, CRP and IL-6 were assessed. Additionally, all assessment and sampling procedures were performed at baseline, 2 weeks, and 4 weeks. Results After four weeks of treatment, the rTMS group showed greater improvement in PCS scores compared to the sham group. For HAMD-24, a significant between-group difference was observed at Week 4 ( P = 0.02), but the overall time×group interaction did not reach statistical significance ( P = 0.160). However, no significant difference was observed in PVAQ scores. Furthermore, the 2-week change in suPAR levels was significantly correlated with the improvement of depressive symptoms in the rTMS group ( r = -0.290, P = 0.008). The response rate was higher in the rTMS group than in the sham group (64.29% vs. 44.44%, P = 0.04). Baseline suPAR level showed a significant negative predictive value for treatment response (OR = 0.609, P = 0.03). A prediction model combining clinical variables and suPAR moderately discriminated responders from non-responders (AUC = 0.786). Conclusion suPAR shows preliminary promise as a potential predictive biomarker for pre-treatment patient selection, while 2-week changes in suPAR may serve as an on-treatment response indicator.