Integrated Ultrasound Localization Microscopy and Multi-Omics for Mechanism-Driven Stratification of Azoospermia

Accurate differentiation between non-obstructive azoospermia (NOA) and obstructive azoospermia (OA) is crucial for therapeutic decision-making, yet current diagnostic approaches predominantly depend on invasive testicular biopsy. Herein, we establish a mechanism-driven stratification framework integrating ultrasound localization microscopy (ULM) with multi-omics profiling. Using murine models representing diverse etiologies, high-frame-rate ULM following intravenous microbubble administration (acquiring over 120,000 compounded frames at 1000 Hz) overcomes the spatial-resolution limit of conventional ultrasound to achieve in vivo quantitative mapping of the testicular microvasculature. Super-resolution maps unveil distinct etiology-specific microvascular phenotypes: NOA models showed fragmented and hypoperfused vascular networks with reduced larger intratesticular vessel diameter (BU:166 ± 9.9μm, p < 0.01; Pnldc1:175± 4.9μm, p < 0.05), whereas OA displayed increased vessel density (7.944 ± 0.52; p < 0.01), mean vessel diameter (43.99 ± 1.168 μm; p < 0.01), and mean flow velocity (8.333 ± 0.523 mm/s; p<0.001). Transcriptomic and metabolomic analyses linked these imaging phenotypes to angiogenesis, extracellular matrix remodeling, lipid peroxidation, and energy-metabolism rewiring. In a small preliminary human cohort, clinical ULM showed concordant differences in vascular morphology and mean vessel diameter between NOA and OA. These findings support ULM as a candidate contrast-enhanced imaging adjunct for mechanism-informed azoospermia stratification, requiring further validation in larger prospective cohorts.