Ultra-High Definition PET/CT Imaging of Head and Neck Cancer Enabled by Next Generation Digital PET Technology

Objectives To assess the clinical feasibility and potential clinical benefit of next generation digital PET detector technology for ultra-high definition imaging in head and neck cancer oncology staging and compare to existing conventional photomultipler tube detector PET/CT technology.

Methods A next-generation, solid-state, digital photon counting, high temporal time of flight resolution PET/CT system (dPET) (Vereos) is being used to prospectively image patients with head and neck cancer and compare its imaging characteristics, by intra-individual comparison, with standard of care conventional photomultiplier tube detector time-of-flight PET/CT (cPET) (Gemini). At the time of submission, 8 head and neck cancer patients agreed to participate in the ongoing comparison study. Standard cPET imaging was performed using a target 481 MBq FDG dose 75 min p.i. with investigational digital PET imaging being performed either at about ~55 min p.i. or after the SOC ~95 min p.i. All other imaging acquisition aspects were kept identical. Digital PET images were reconstructed using Time-of-Flight with different voxel volumes 4x4x4 mm3 (standard definition), 2x2x2 mm3 (high definition), and 1x1x1 mm3 (ultra-high definition). Readers evaluated the image sets in regard to overall quality, lesion detection, confidence of characterization and quality of background by blinded individual datasets and then subsequently matched exam review.

Results Overall image quality was rated highest for the ultra-high definition reconstruction of the dPET. When compared with conventional PET/CT, digital PET imaging of head and neck cancer patients produced substantially improved image quality for the visual and quantitative assessment of head and neck primary as well as metastatic lesions. While SUV analysis of larger lesions remained comparable, smaller sub-centimeter lesions were better delineated and demonstrated higher SUV values on digital PET imaging. Overall, dPET enabled the visualization of lesions which appeared more sharply delineated with improved detail regarding heterogeneity, all leading to higher diagnostic confidence of the readers especially in smaller (≤15mm) lesions.

Conclusions FDG PET imaging in head and neck cancer is overall of great clinical value for staging. This study presents the initial demonstration of ultra-high definition dPET imaging which was found to markedly improve overall image quality, lesion detectability (especially for lesions ≤ 15mm) and diagnostic confidence of the readers. The next generation dPET technology appears to help us further advance the quality and precision of FDG PET for diagnostic and therapy planning decisions in head and neck oncology patients.