Efficient measurement of total tumor microvascularity ex vivo using a mathematical model to optimize volume subsampling

Abstract

We introduce immunofluorescence and automated image processing protocols for serial tumor sections to objectively and efficiently quantify tumor microvasculature following antivascular therapy. To determine the trade-off between tumor subsampling and throughput versus microvessel quantification accuracy, we provide a mathematical model that accounts for tumor-specific vascular heterogeneity. This mathematical model can be applied broadly to define tumor volume samplings needed to reach statistical significance, depending on the biomarker in question and the number of subjects. Here, we demonstrate these concepts for tumor microvessel density and total microvascularity (TMV) quantification in whole pancreatic ductal adenocarcinoma tumors ex vivo. The results suggest that TMV is a more sensitive biomarker for detecting reductions in tumor vasculature following antivascular treatment. TMV imaging is a broadly accessible technique that offers robust assessment of antivascular therapies, and it offers promise as a tool for developing high-throughput assays to quantify treatment-induced microvascular alterations for therapeutic screening and development.