Signaling pathways alteration involved in head and neck cancer can be identified through textural features analysis in 18F-FDG PET images: a prospective study

Objectives Numerous signaling pathways alterations with impact on patients’ outcome have been well described in head and neck cancer (HNC). Patients with HNC undergo FDG PET/CT scan at diagnosis. A complete characterization of tumors properties can be achieved through the calculation of features including metabolically active tumor volume (MATV), SUVs and heterogeneity of the intratumoral FDG uptake. It has been hypothesized that some of these features may be linked to the underlying tumor physiology. The objective of this study was to recover signaling pathways alterations from FDG PET images.

Methods 52 patients were prospectively recruited from two French centers. All patients underwent an FDG PET scan and biopsies of both healthy and tumor tissues that went through a transcriptomic analysis performed in four spates on 4×44k chips from Agilent. Tumors were characterized in PET images with 10 parameters including heterogeneity quantification through textural features. A module network algorithm followed by functional annotation was used to link PET parameters to signaling pathways alterations.

Results Some intratumor FDG heterogeneity quantitative features were found to discriminate differentially expressed genes between tumor and healthy tissue (fold-change >2, p<0.01) into co-regulated 30 groups (p<0.05). Functional annotations of these genes groups were able to associate them with pathways relative to cell proliferation or these involved with apoptosis.

Conclusions The characterization of FDG tracer uptake heterogeneity in PET tumor images may provide quantitative information about tumor underlying physiological processes and could allow identifying specific altered pathways.