A novel partial volume correction method for accurate quantification of [18F] flortaucipir in the hippocampus

Emma E Wolters; Sandeep S V Golla; Tessa Timmers; Rik Ossenkoppele; Chris W J van der Weijden; Philip Scheltens; Lothar Schwarte; Robert C Schuit; Albert D Windhorst; Frederik Barkhof; Maqsood Yaqub; Adriaan A Lammertsma; Ronald Boellaard; Bart N M van Berckel

doi:10.1186/s13550-018-0432-2

A novel partial volume correction method for accurate quantification of [¹⁸F] flortaucipir in the hippocampus

EJNMMI Res. 2018 Aug 15;8(1):79. doi: 10.1186/s13550-018-0432-2.

Authors

Emma E Wolters^{1

2}, Sandeep S V Golla³, Tessa Timmers^{1

2}, Rik Ossenkoppele^{1

2

4}, Chris W J van der Weijden¹, Philip Scheltens², Lothar Schwarte⁵, Robert C Schuit¹, Albert D Windhorst¹, Frederik Barkhof^{1

6}, Maqsood Yaqub¹, Adriaan A Lammertsma¹, Ronald Boellaard¹, Bart N M van Berckel^{1

2}

Affiliations

¹ Departments of Radiology & Nuclear Medicine, VU University Medical Center, PO Box 7057, 1007MB, Amsterdam, The Netherlands.
² Neurology & Alzheimer Center, VU University Medical Center, Amsterdam, The Netherlands.
³ Departments of Radiology & Nuclear Medicine, VU University Medical Center, PO Box 7057, 1007MB, Amsterdam, The Netherlands. s.golla@vumc.nl.
⁴ Lund University, Clinical Memory Research Unit, Lund, Sweden.
⁵ Anaesthesiology, Amsterdam Neuroscience, VU University Medical Center, Amsterdam, The Netherlands.
⁶ Institutes of Neurology & Healthcare Engineering, UCL, London, UK.

Abstract

Background: Off-target binding in the choroid plexus (CP) may cause spill-in of the tau PET tracer [¹⁸F] flortaucipir into the adjacent hippocampus region. The impact of this spill-in on hippocampal uptake was assessed using a novel partial volume correction method (PVC).

Methods: PVC was performed on 20 [¹⁸F] flortaucipir dynamic PET scans (10 probable AD and 10 controls). Volumes of interest (VOIs) were defined for both hippocampus and CP. The correlation between hippocampal and CP distribution volume (V_T), with and without PVC, was determined. Both anatomically defined and eroded VOIs were used.

Results: For controls, the correlation between hippocampal and CP V_T was significantly reduced after using PVC along with an eroded VOI (r² = 0.59, slope = 0.80 versus r² = 0.15, slope = 0.15; difference: p < 0.05). The same was true for AD patients (p < 0.05).

Conclusion: PVC together with an optimized hippocampal VOI resulted in effective reduction of CP spill-in and improved accuracy of hippocampal V_T.