Feasibility demonstration of ultra-high definition Na-¹⁸F digital PET/CT imaging and quantitative assessment of osteoblastic tumor burden in patients with metastatic genitourinary cancers

Objectives: For patients undergoing routine oncologic and non-oncologic bone imaging, consistent qualitative and accurate quantitative assessment of sodium fluoride-18 (¹⁸F)-avid osteoblastic lesions is essential. Fluoride-18 is a PET radionuclide which forms chemical complexes at sites of increased bone turnover such as malignant/metastatic osteoblastic lesions. A recent technology innovation has led to the replacement of conventional photomultiplier tubes in the PET gantry with next-generation, solid-state, digital photon counting PET detectors (dPET). This new dPET detector technology has the potential to improve ¹⁸F PET bone imaging using higher definition reconstruction (i.e., decreasing voxel volumes). The objective of this study is to assess the feasibility and potential clinical benefit of dPET/CT technology in quantifying ¹⁸F-avid osteoblastic tumor burden using standard and higher definition image reconstructions.

Methods: In an ongoing clinical trial, 9 patients with metastatic genitourinary cancers underwent same-day sequential FDG and ¹⁸F PET/CT utilizing a pre-commercial release dPET/CT system (Vereos, Philips). Whole-body dPET/CT imaging was first performed using a target 185 MBq FDG dose 60 min p.i.. Immediately following the FDG dPET imaging, a target dose of 130 MBq Na¹⁸F was administered and repeat whole-body dPET/CT imaging was performed after another 60 min p.i.. 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). A reader panel assessed both FDG and ¹⁸F image data sets to identify only ¹⁸F-avid (and not FDG-avid) osteoblastic tumor lesions. In addition, threshold based semi-automated isocontouring of these ¹⁸F-avid lesions was performed using a MIM Vista workstation and the following metrics were extracted: highest SUVmax (hSUVmax), average of SUVmax value, fluoride volume (FV) and total lesion fluoride (TLF).

Results: All 54 data sets were evaluable. UHD dPET imaging for both FDG and ¹⁸F were rated best for image quality. Among the patients, there was a variable amount of osteoblastic tumor burden. Of the 9 patients, 5 had ¹⁸F-avid osteoblastic lesions with no associated FDG uptake. Of these, 4 had 8 discrete lesions whereas 1 had disseminated osteoblastic metastatic disease. On average, hSUVmax increased by 13% when lesions were evaluated using high definition reconstructions and 27% for ultra-high definition reconstructions. Similarly, lesion FV increased by 11% for high definition and 28% for ultra-high definition reconstructions. In addition, there were slight increases in the average SUVmax and TLF for lesions at higher definition reconstructions.

Conclusion: This study presents the initial demonstration of ultra-high definition dPET imaging for ¹⁸F-avid osteoblastic lesions. This dPET technology enables higher definition reconstructions and appears to advance image quality and quantitative precision of ¹⁸F PET bone imaging by reducing partial volume effects. As a consequence, the use of fixed threshold-based quantitative methodologies with higher definition PET reconstruction will likely contribute to true increases in both SUV and volume metrics when compared with standard definition PET reconstruction. Research Support: NCI R01CA195513, ODSA TECH 13-060, TECH 10-012, and TECH 09-028