TO THE EDITOR: 11C-(+)-4-propyl-9-hydroxynaphthoxazine ((+)-PHNO) is a new PET ligand developed by our group. Binding assays show that (+)-PHNO displays high affinity and selectivity for the D2 receptor (1). Recently, it has been noted that 11C-(+)-PHNO has a preferential affinity and selectivity in vivo for the D3 receptors (2). Because 11C-(+)-PHNO is an agonist radiotracer for D2 and D3, it is likely to produce pharmacologic effects, in contrast to antagonist radiotracers. We reviewed all 11C-(+)-PHNO consecutive scans obtained in our PET center for side effects. Mass injected (μg), subjects' weight (kg), and dose (μg/kg) were included in the analysis. Side effects were recorded on the basis of the subjects' self-report either during or right after finalization of the scan. A physician was available at all times to confirm and treat any possible side effects. Side effects were coded as 0 (no effect), 1 (nausea), or 2 (vomiting), based on our early experience with 11C-(+)-PHNO (3). Odds ratios (ORs) were calculated using logistic regression analyses to investigate the relationship between dose, mass, and effects.
The number of reviewed 11C-(+)-PHNO scans totalled 486. Injected mass ranged from 0.85 to 5.56 μg, with a mean of 2.30 μg (SE, 0.024 μg). Injected doses ranged from 0.01 to 0.08 μg/kg, with a mean of 0.03 μg/kg (SE, 0.0004 μg/kg). No effect was present in 84.6% of the scans reviewed; nausea was present in 14.3%, and vomiting in 1.1%. Symptoms arose 3–5 min after the injection and subsided within 7–12 min in all cases. In none of the cases was any medical action required.
In a logistic regression model including all subjects, nausea was significantly predicted by dose (Wald = 21.70, P < 0.001, OR 1.99) and mass (Wald = 16.319, P < 0.001, OR = 2.826), and vomiting was significantly predicted by dose (Wald = 7.31, P < 0.007, OR = 2.66) but not by mass injected (Wald = 0.694, P = 0.405, OR = 1.810). When only drug-free volunteers were analyzed (n = 209), no effect was present in 79.8% of the cases, nausea was present in 18.7%, and vomiting in 1.5%. In a logistic regression model including only drug-free volunteers, nausea was significantly predicted by dose (Wald = 6.98, P < 0.008, OR 1.54) and mass (Wald = 11.981, P = 0.001, OR = 2.843). Vomiting was predicted at a trend level by dose (Wald = 3.33, P < 0.06, OR 2) but not by mass (Wald = 0.105, P = 0.746, OR = 1.303). When only antipsychotic-treated participants were analyzed (n = 66), no effect was present in 97% of the cases, nausea was present in 3%, and no vomiting was present in any. In a logistic regression model including only these subjects, nausea was significantly predicted neither by dose (Wald = 2.25, P < 0.13) nor by mass (Wald = 0.000, P = 0.99). In all cases, when an injected dose of 0.029 μg/kg or less was selected, there was no relationship between dose and nausea.
The side effects reported in this study are consistent with the expected agonism at the D2- and D3-receptor (4–7).
We conclude that doses of 11C-(+)-PHNO of 0.029 μg/kg or less are highly unlikely to produce any side effects in humans and that 11C-(+)-PHNO is a safe agonist radiotracer for PET in human studies of health and disease.
Acknowledgments
We acknowledge the scientists and technicians from the PET center who carried out the 11C-(+)-PHNO scans. This work is supported in part by the CIHR and the OMHF New Investigator Awards.
Footnotes
-
COPYRIGHT © 2010 by the Society of Nuclear Medicine, Inc.