Optimizing acquisition times for total-body positron-emission tomography/computed tomography with half-dose 18F-fluorodeoxyglucose in oncology patients

Introduction: To explore the boundary of acquisition time while maintaining acceptable quality of PET images and further identify a suitable acquisition protocol for half-dose total-body PET oncological imaging.

Methods: To search the acceptable threshold of acquisition time, an exploration dataset (October 2019 to December 2019) of 46 oncology patients (34/12 male/female, age 61.5 ± 8.8 years) who underwent total-body PET/CT with weight-based ¹⁸F-FDG injection (1.88 ± 0.09 MBq/kg) in our department was first retrospectively studied. The acquisition time for all patients was 15 min and the acquired images were reconstructed and further split into 15-, 8-, 5-, 3-, 2, 1-min duration groups, which were abbreviated as G15, G8, G5, G3, G2, and G1. A qualitative assessment on the overall image quality was performed using a 5-point scale (5, excellent; 1, non-diagnostic). Quantitative analysis parameters of PET image quality included the SUVmax, SUVmean and standard deviation (SD) of the liver; the SUVmax, SUVmean and SD of the blood pool; and the SUVmax, SUVpeak and tumour-to-background ratio (TBR) of the lesions. The lesion detection rate was calculated to quantify lesion detectability by referring to the number of lesions identified in G15. Subsequently, the filtered protocols from the exploration dataset were further validated in 147 oncology patients (December 2019 to June 2021). All PET raw data were reconstructed and further divided into 5 groups, referred as Gs, G8, G5, G3, and G2, respectively. A 5-point scale and the lesion detection rate were used to evaluate the overall image quality and lesion detectability.

Results: In the exploration dataset, the subjective image scores for G1, G2, G3, G5, and G8 were 2.0 ± 0.2, 2.8 ± 0.3, 3.1 ± 0.2, 3.9 ± 0.3, and 4.1 ± 0.2, respectively. Two cases in G1 was rated as 1 point, while patients in the remaining groups scored at least 2. The scores of G5 and G8 were significantly higher than those of the other three groups (all p＜0.05), and no significant difference was observed between G5 and G8 (p=0.888). For quantitative analysis, the groups with a longer acquisition time showed lower SUV values in the liver and mediastinal blood pool. The SUVmax of the liver and blood pool were not statistically different between G8 and G15 (all p＞0.05). G15 showed significantly lower values of lesion SUVmax, SUVpeak, and TBR than those in the remaining groups (all p≤0.039), while no significant difference among the remaining groups was found (all p≥0.078). G15 served as the reference, and all lesions (100%) could be identified by all groups except for G1 (85.3%) and G2 (97.1%). In the validation dataset, the subjective score was 2.9 ± 0.2 for G2, 3.0 ± 0.0 for G3, 3.5 ± 0.4 for G5, 4.0 ± 0.2 for G8, and 4.5 ± 0.4 for Gs. The scores of G2 and G3 were at about the same level or slightly lower than 3 points. And significant difference of the scores was observed between any two groups (all p＜0.05), except for G2 and G3 (p=1.0). A total of 204 lesions were analyzed in the lesion detection rate, and the detection rates were 100% and 99.0% in G8 and G5 and significantly reduced to 94.1-89.7% in G3 and G2 (all p＜0.001).

Conclusions: A 2-min acquisition protocol, though somewhat lowering quality, provided acceptance performance in certain groups and specific medical situations. Protocols with acquisition times ≥ 5 min by half-dose total-body PET/CT could provide comparable lesion detectability as regular protocols, showing better compatibility and feasibility with clinical practice.

• Download figure
• Open in new tab
• Download powerpoint