Histology-Based Radiomics for [F]FDG PET Identifies Tissue Heterogeneity in Pancreatic Cancer.

Radiomics features can reveal hidden patterns in a tumor but usually lack an underlying biologic rationale. In this work, we aimed to investigate whether there is a correlation between radiomics features extracted from [F]FDG PET images and histologic expression patterns of a glycolytic marker, monocarboxylate transporter-4 (MCT4), in pancreatic cancer. A cohort of pancreatic ductal adenocarcinoma patients ( = 29) for whom both tumor cross sections and [F]FDG PET/CT scans were available was used to develop an [F]FDG PET radiomics signature. By using immunohistochemistry for MCT4, we computed density maps of MCT4 expression and extracted pathomics features. Cluster analysis identified 2 subgroups with distinct MCT4 expression patterns. From corresponding [F]FDG PET scans, radiomics features that associate with the predefined MCT4 subgroups were identified. Complex heat map visualization showed that the MCT4-high/heterogeneous subgroup was correlating with a higher MCT4 expression level and local variation. This pattern linked to a specific [F]FDG PET signature, characterized by a higher SUV and SUV and second-order radiomics features, correlating with local variation. This MCT4-based [F]FDG PET signature of 7 radiomics features demonstrated prognostic value in an independent cohort of pancreatic cancer patients ( = 71) and identified patients with worse survival. Our cross-modal pipeline allows the development of PET scan signatures based on immunohistochemical analysis of markers of a particular biologic feature, here demonstrated on pancreatic cancer using intratumoral MCT4 expression levels to select [F]FDG PET radiomics features. This study demonstrated the potential of radiomics scores to noninvasively capture intratumoral marker heterogeneity and identify a subset of pancreatic ductal adenocarcinoma patients with a poor prognosis.