🧬 Can muscle hypertrophy be improved by building the vasculature first? A new preprint explores a clever sequential gene-therapy strategy: VEGF-A165 plasmid → AAV9-follistatin (FST344) to couple angiogenesis with muscle growth. In C57BL/6 mice, the combination produced: 🔹 Greater muscle hypertrophy — R/L muscle mass ratio 1.176 vs 1.113 with AAV-FST alone; combination p=0.004, Cohen’s d=2.04. 🔹 ~2× higher CD31⁺ vessel density, while simultaneously producing the largest muscle fibres. 🔹 ~2.1× higher muscle FST mRNA than AAV-FST alone at Day 115. 🔹 A 2.6× higher muscle-to-liver transgene expression ratio, suggesting that angiogenic preconditioning may improve local AAV targeting/transduction. The biology is compelling: anabolism without adequate capillarization may produce bigger muscle, but not necessarily better muscle. VEGF preconditioning could help coordinate fibre growth with vascular adaptation—a potentially important concept for sarcopenia, cachexia, and neuromuscular disease. Important caveat: this is a small mouse study and muscle strength/function was not measured, so whether the added mass translates into functional benefit remains unknown. 📄 Sequential VEGF-A165 plasmid and AAV-follistatin gene therapy enhances muscle hypertrophy and capillarisation in C57BL/6 mice 🔗 DOI: 10.64898/2026.08.26.747237 #Sarcopenia #Muscle #GeneTherapy #Follistatin #VEGF #AAV #Angiogenesis #Cachexia #Aging #RegenerativeMedicine

Fibroblasts aren't just bystanders in liver cancer—they actively shape whether immunotherapy succeeds. A new PNAS study introduces a quantitatively calibrated spatial quantitative systems pharmacology (spQSP) model that integrates mechanistic modeling with spatial omics to predict immunotherapy response in hepatocellular carcinoma (HCC). Key findings: 🔹 The authors expanded an HCC spQSP framework by incorporating cancer-associated fibroblasts (CAFs) and calibrated the model using Bayesian inference (pyABC) with pretreatment CODEX spatial proteomics. 🔹 The calibrated model accurately reproduced outcomes from nivolumab + cabozantinib clinical trials, including objective response rates and key tumor microenvironment characteristics. 🔹 Spatial neighborhood analysis (FunCN) revealed that: • ↑ CD8–tumor and CD4–tumor proximity predicts better survival. • ↑ Fibroblast clustering, Treg clustering, and CD8–macrophage interactions associate with worse outcomes. 🔹 Virtual tumors resistant to immunotherapy consistently exhibited: • dense fibroblast networks • increased ECM deposition • elevated TGF-β • physical exclusion of T cells from tumor regions. 🔹 Validation with independent IMC and Visium datasets confirmed that fibroblast-rich stromal regions coincide with poor T-cell infiltration, supporting a model where CAFs function as physical immune barriers. Perhaps the most exciting aspect is methodological rather than biological. Instead of treating spatial transcriptomics as a static descriptive technology, this work uses static spatial omics to infer dynamic cellular behaviors (such as cell motility) through Approximate Bayesian Computation, enabling mechanistic simulations of future tumor evolution. This moves spatial biology closer to true digital twin oncology. The study also reinforces an emerging therapeutic concept: Targeting fibroblast-mediated immune exclusion—not simply increasing T-cell abundance—may substantially improve checkpoint blockade efficacy in HCC. A compelling demonstration of how spatial omics, mechanistic modeling, and virtual clinical trials can converge to enable biomarker discovery and precision immunotherapy. #HCC #Immunotherapy #SpatialBiology #SystemsBiology #QSP #DigitalTwin #CancerResearch #ComputationalBiology