Abstract
1131
Introduction: Oxygen-15 (t½ = 124s) has been used for nuclear imaging experiments since the beginning of the field. With the advent of simultaneous hybrid PET/MR technology, [15O]Water has seen a resurgence, particularly for validating novel MR sequences. Indeed, for quantitative blood flow studies it remains the gold standard method. For patient safety, contemporary [15O]Water production methods must adhere to current GMP. Given the short half-life, this presented a non-trivial challenge for our facility when establishing production protocols. Herein we present our production validation protocol along with 1 year of GMP data from production for PET/MR imaging sessions.
Methods: 15O was produced via the d,n reaction using a 1% O2/N2 target on an 8.4 MeV GE PETTrace. After a 6 min, 40 μA bombardment, the [15O]O2 gas was released and passed through an ascarite/charcoal column. While mixing with H2 over a 400oC Pt wire, [15O]water formed and this vapour was bubbled into mixing vial with phosphate buffered saline over 2 min. The solution was then transferred via positive pressure through a 0.22 μm filter into a final patient vial. For production validation (PV), six batches of [15O]water were prepared: (1) sterility, (2) quality control testing, and (3 - 6) patient doses. PV was completed in quadruplicate. To ensure the quality was maintained, the final dose during PV were also tested. Radiochemical identity and purity were assessed using a gas chromatograph with serial thermal conductivity and radiation detection. Half-life measurement was used to determine radionuclidic identity and gamma spectroscopy was used for purity. Sterility and pyrogenicity were assessed by filter integrity testing and endotoxin measurement, respectively. Retrospective testing confirmed sterility. Breakthrough of Pt was measured via inductively coupled plasma mass spectroscopy.
Results: During initial production validation, we found the back pressure created by passing the vapour through the sterile filter caused inadequate mixing of H2 and [15O]O2 over the heated platinum wire. This resulted in poor radiochemical yields and large radioactive gas releases. By adding an initial mixing vial, consistent yields of 10 - 12 GBq (270 - 325 mCi) were obtained 4 min after bombardment. Twenty four [15O]Water batches were tested for quality assurance from Sep 2018 - Jan 2020. All batches produced met release specification for human injection as outlined in the table below. No adverse events were reported from the 7 control and 7 dementia patients injected with our product. Imaging results from our [15O]Water were consistent with previous reports. Preliminary imaging results will be presented (Fig 1).
Conclusions: The multiple batch protocol proved to be a safe and effective means for producing [15O]Water for human injection. Furthermore, this protocol could be readily adapted by any facility attempting to produce [15O]Water for clinical studies. Compared with previous attempts at our site, the protocol outlined here was more consistent and reliable, improved production workflow, and led to more available radioactivity for participant injection and QC testing. More activity was particularly important given our PET/MR was 6 floors above and across the hospital from the cyclotron. Acknowledgements: This work was funded through grants from the Canadian Institutes of Health Research, the Alzheimer’s Drug Discovery Foundation, the Canadian Foundation of Innovation, St. Joseph’s Health Care London, and the Lawson Health Research Institute. We thank all team members of the Lawson Cyclotron and PET/MR facility for their contributions to this work.