Effect of data-driven gating on the quantification of prostate cancer metastases in 68Ga-PSMA- and ¹⁸F-DCFPyL-PSMA PET/CT.

Abstract

Background: PSMA PET/CT using ¹⁸F‑DCFPyL and ⁶⁸Ga‑labelled tracers is central to prostate cancer staging, but ⁶⁸Ga‑PSMA may show reduced sensitivity and quantitative accuracy, partly due to respiratory motion and lower spatial resolution. Device-based motion‑correction methods have limitations. Data‑driven gating (DDG) offers a hardware‑free alternative, though its impact on PSMA PET, especially ⁶⁸Ga‑PSMA has not been systematically evaluated. To evaluate the quantitative and qualitative performance of the DDG motion-correction method in ⁶⁸Ga-PSMA and ¹⁸F-DCFPyL-PSMA PET/CT, focusing on metastatic lymph node, lung, and bone lesions, and to assess its robustness across tracers.

Methods: This retrospective study included patients with biopsy‑proven prostate cancer who underwent ⁶⁸Ga‑PSMA or ¹⁸F‑DCFPyL PSMA PET/CT for primary staging or biochemical recurrence and had metastatic disease. Motion correction was applied using the OncoFreeze AI™ data‑driven gating (OFAI‑DDG) method. Metastatic lesions were segmented semi‑automatically, and SUV metrics, image noise, and contrast‑to‑noise ratio were quantified by anatomical site in both OFAI-DDG and static images. Qualitative image quality was independently scored by two experienced readers. Statistical analyses included non‑parametric tests, Bland-Altman agreement, regression analyses, tracer comparison, and inter‑reader reliability assessment.

Results: After exclusions, 23 ⁶⁸Ga‑PSMA and 40 ¹⁸F‑PSMA patients were analyzed, yielding 133 and 192 metastatic lesions, respectively. OFAI‑DDG reconstruction significantly increased SUV_mean and SUV_max compared with static imaging for both tracers across bone, lymph node, and lung lesions (p < 0.0001), with the largest gains in lung lesions. CNR improved for most lesion types, and image noise was significantly reduced. Qualitative analysis showed better image quality, sharpness, and noise with OFAI‑DDG, with high inter‑reader agreement. OFAI‑DDG performance was robust across tracers and largely independent of lesion location.

Conclusions: This study demonstrates that OFAI‑DDG significantly improves PSMA PET/CT image quality and quantitative accuracy compared with standard static imaging. The improvements were observed for both ⁶⁸Ga‑PSMA and ¹⁸F‑PSMA tracers, indicating tracer‑independent robustness. The OFAI‑DDG simplified the workflow and reduced time used for patient set-up, while enhancing lesion visualization and quantitative reliability without altering lesion detection, supporting its clinical value for metastatic prostate cancer imaging.