Designed concave octahedron heterostructures decode distinct metabolic patterns of epithelial ovarian tumors.

Epithelial ovarian cancer (EOC) is a polyfactorial process associated with alterations in metabolic pathways. High-performance screening tool for EOC is in high demand to improve prognostic outcome but still missing. Here, we develop a concave octahedron Mn 2 O 3 /(Co, Mn)(Co, Mn) 2 O 4 (MO/CMO) composite with a heterojunction, rough surface, hollow interior, and sharp corners to record metabolic patterns of ovarian tumors with laser desorption/ionization mass spectrometer (LDI MS). The MO/CMO composites with multiple physical effects induce enhanced light absorption, preferred charge transfer, increased photothermal conversion and selective trapping of small molecules. The MO/CMO shows ∼2-5-fold signal enhancement compared to mono- or dual-enhancement counterparts, and ∼10-48-fold compared to the commercialized products. Subsequently, serum metabolic fingerprints of ovarian tumors are revealed by MO/CMO assisted LDI-MS, achieving high reproducibility of direct serum detection without treatment. Furthermore, machine learning of the metabolic fingerprints distinguishes malignant ovarian tumors from benign controls with the area under the curve value of 0.987. Finally, we screen seven metabolites associated with the progression of ovarian tumors as potential biomarkers. Our approach guides the future depiction of the state-of-the-art matrix for intensive MS detection and accelerates the growth of nanomaterials-based platforms toward precision diagnosis scenarios. This article is protected by copyright. All rights reserved.