Abstract
218
Objectives: Next-generation digital photon counting PET detector (dPET) technology is an innovative leap over existing conventional photomultiplier tube-based PET (cPET) technologies and readily enables higher definition whole-body PET imaging with smaller voxel volumes. The objective of this study is to assess the feasibility and clinical potential of higher definition PET/CT imaging with cPET/CT technology in patients with head and neck cancer.
Methods: Ten newly diagnosed head and neck cancer patients agreed to participate in an ongoing intra-individual matched pair comparison study of 18F-FDG PET using cPET/CT (Gemini ToF 64, Philips) and dPET/CT (pre-commercial release Vereos, Philips) imaging systems. Standard cPET/CT was performed using a target dose of 481 MBq FDG at ~75 min p.i. and investigational dPET/CT was performed at ~55 min p.i. or ~95 min p.i. All cPET and dPET images were reconstructed with ToF using a standard definition (SD) voxel volume of 4x4x4 mm3 and a high definition (HD) volume of 2x2x2 mm3. Images were compared blindly and then subsequently matched review for overall image quality, background quality, lesion detectability, and diagnostic confidence by a three reader panel.
Results: All 40 cPET and dPET data sets were rated as evaluable. The HD cPET and HD dPET reconstructions were consistently rated best in terms of overall image quality for the evaluation of primary tumors and metastatic lesions with no appreciable effect on the background FDG activity when compared with respective SD reconstructions. Both HD cPET and dPET data sets demonstrated improved lesion detectability with sharper delineation of malignant/metastatic lesions, likely due to reduced partial volume, to better appreciate lesion heterogeneity. A total of 58 lesions were identified with 52 lesions noted on both SD cPET and dPET and an additional 6 subcentimeter lesions also noted on the HD cPET and dPET data sets. Quantitative lesion assessment also demonstrated that larger lesions had comparable SUVmax values on HD reconstructions whereas smaller lesions demonstrated higher SUVmax values on HD imaging. For lesions < 15mm, average SUVmax increased by 32% for HD cPET and 26% for HD dPET whereas average SUVmax only increased by 7% for HD cPET and 8% for HD dPET in lesions > 15 mm.
Conclusion: FDG PET imaging has proven clinical utility in head and neck oncology, and is routinely utilized for the detection of occult malignancy, staging and assessment of therapeutic response. Higher definition PET/CT imaging represents another step towards improved disease staging and therapeutic planning by improving lesion detectability (especially in smaller lesions), lesion characterization (especially in larger heterogeneous lesions), and diagnostic confidence for the reader. Although next-generation dPET/CT systems will routinely enable higher definition whole-body PET imaging, this study demonstrates that substantial improvements are even feasible and achievable with existing photomultiplier-tube-based cPET/CT systems. Further development and optimization of higher definition PET image reconstruction approaches will enable new precision nuclear medicine practices for cancer patients and, in particular, strengthen FDG PET utilization as a more effective imaging biomarker in head and neck oncology. Research Support: ODSA TECH 13-060, ODSA TECH 10-012, and ODSA TECH 09-028