Semaglutide 3-Month Pre-Treatment Optimizes Sofwave Efficacy, Boosting Collagen Density by 41% in Virtual Trial

The growing use of GLP-1 receptor agonists (GLP-1RAs) for weight loss often leads to concerns about premature facial aging and skin laxity. Non-invasive skin tightening procedures like Sofwave rely on fibroblast mechanotransduction to stimulate collagen production, a process known to be impaired in obesity and potentially influenced by metabolic changes from GLP-1RAs. The optimal timing for combining these interventions to maximize aesthetic outcomes and mitigate adverse effects remains a critical, unaddressed clinical gap.

Researchers conducted a novel fully computational, mechanistically constrained virtual randomized trial. A cohort of 20,000 digital twins (all with BMI >30 kg/m2) was randomized into five arms. All arms received identical Sofwave schedules, but semaglutide start times varied: concurrent, 1 month, 3 months, or 6 months pre-treatment, versus a Sofwave-only control. The multiscale model integrated a finite element model for ultrasound-induced tissue stress with an agent-based model for fibroblast dynamics, ECM remodelling, and semaglutide pharmacodynamics. The primary endpoint measured was the time-integrated cycling fibroblast burden (AUC).

Fibroblast response to mechanical stress was significantly modulated by insulin resistance (IR). Higher IR increased the fibroblast activation threshold from 36.8 kPa to 47.9 kPa and reduced the maximum cycling probability from 0.34 to 0.21. Semaglutide timing produced a non-monotonic effect on treatment efficacy. > The 3-month pre-treatment arm yielded the greatest net benefit, showing a +0.092 increase in cycling AUC and a +41% increase in collagen density. This benefit was attributed to a balance between direct fibroblast priming (+0.118 AUC) and indirect adipose support loss (-0.026 AUC). Prolonged 6-month pre-treatment proved detrimental in high-IR phenotypes, with a 90th percentile ΔAUC of -0.006. Additionally, the 3-month pre-treatment arm resulted in a more isotropic ECM, with an anisotropy index of 0.26 compared to 0.41 in the control group.