Following finalisation of the mechanical design of structural parts of the PET–RF insert and the PET detector mounting system, the RF coil prototype was developed and tested, and safety measurements were performed. Electronics prototypes of the PET component were developed and tested (coincidence testing), and performance testing of prototype PET detector electronics inside the MRI environment was performed. The auxiliary systems were assembled, including the detector stacks, cooling system, power supply and associated cables and tubes. With the operational and control software prepared, the insert was introduced into the clinic for testing of its PET and MRI capabilities. In parallel, novel functional MR acquisition methods (MRI fingerprinting) were established and published. Using the PET component of the device, image reconstruction was gauged with a point source and phantoms. Successful data acquisition and image quality demonstrate the success of the integration of the scanner hardware and placement of the detector blocks. The head-to-head comparison of phantom measurements with HYPMED confirmed superior PET sensitivity compared to conventional whole-body PET-MR systems. The HYPMED device achieved a resolution of <1.5 mm – unattainable with conventional whole-body PET-MRI systems. Refinement will further improve PET resolution down to ~1 mm.
To test the diagnostic performance of conventional whole-body PET-MRI, 376 patients have been included, who received [18F]FGD, [18F]FMISO or [18F]choline as tracers. Histopathological specimen were shipped to enable pathological analyses. Whole-body PET-MRI data were analysed to assess diagnostic accuracy in terms of discrimination of benign and malignant lesions as well as characterization of tumour biology. All data (imaging and respective stage & histopathology information on cancers) have been stored in an anonymized database accessible to the HYPMED consortium for future comparison with the clinical performance of the HYPMED device.
The documentation and tests necessary for regulatory approval of the HYMPED device were investigated, concluding requirements for a feasibility study are very similar to those for approval for fundamental research with human subjects. Although the path for regulatory approval of the HYPMED device under the new MDR is uncertain, guidance was produced to outline the relevant norms and regulatory standards in risk management, electrical and biological safety, software and other areas, as well as design specifications and documentation of design processes. A design failure mode and effect analysis was done to identify and mitigate risks from operation, and to ensure radiation safety for patients, a specific absorption rate study conducted by NORAS, PHI and UKA was also described. The identified list of verification tests would effect a comprehensively verified system.
A tissue biobank consisting of 281 pre-invasive and invasive breast cancers and 30 normal samples was set up. For tissue biomarker development, cases were examined by conventional immunohistochemistry and by the Opal system to study relations between different cellular components and their interplay with tumour cells for correlation with whole-body PET-MR and HYPMED imaging findings.
Dissemination activities continued, and an online presence was maintained through the project website and social media. HYPMED was promoted at major congress events mostly via online formats due to COVID-19, but also onsite at ECR 2022. A dedicated HYPMED workshop held in March 2022 drew an audience of more than 270 attendees.
