Abstract
3076
Introduction: PSMA-PET/CT has shown higher accuracy for detecting prostate cancer (PCa) metastases compared to Fluciclovine-PET/CT. However, reader confidence interpreting the PET/CT scans with the different radiotracers remains unknown. In May 2020, a certainty scale was implemented at our institution to standardize communication of the degree of diagnostic certainty in radiology reports (Fig.1.). Our aim was to compare the degree of diagnostic certainty to describe metastases in patients with biochemical recurrence (BCR) of PCa on PSMA-PET/CT versus Fluciclovine-PET/CT.
Methods: A radiologist retrospectively reviewed all consecutive Fluciclovine-PET/CT reports from Aug-Dec 2020 and PSMA-PET/CT reports from Aug-Dec 2021 to extract the degree of diagnostic certainty based on the presence of the specific terms within in the impression section. Negative reports and those using non-recommended terms were excluded. History of prostatectomy, radiotherapy and PSA levels were obtained from medical records. The degree of diagnostic certainty to describe nodal, osseous, and other metastases was compared between Fluciclovine and PSMA PET/CT reports using Mann-Whitney U test among all patients, per treatment strategy (prostatectomy and radiotherapy), and PSA level. PSA levels were also compared between PSMA and Fluciclovine PET/CTs using Mann-Whitney U test. A p value < 0.05 was considered statistically significant.
Results: A total of 62% (57/107) of Fluciclovine and 63% (122/193) of PSMA PET/CT reports described positive findings for PCa using the recommended terminology (Fig. 2.). PSA levels were significantly lower in patients who underwent PET/CT with PSMA (median= 1.26 ng/mL, ranging from <0.02 to 503.5) compared to Fluciclovine (median= 2.82 ng/mL, ranging from <0.02 to 482.5, p=0.02). PSMA and Fluciclovine-PET/CT reports described osseous metastases in 38% (46/ 122) and 44% (25/57), nodal metastases in 52% (63/122) and 54% (31/57), and other metastases (e.g. solid organ and peritoneal) in 7% (9/122) and 5% (3/57) of cases, respectively. Osseous metastases were described with significantly higher certainty qualifiers on PSMA-PET/CT reports (41% [19/46] with absolute certainty, 24% [11/46] very high probability, 26% [12/46] high probability, and 9% [4/46] intermediate probability) compared to Fluciclovine PET/CT reports (respectively, 23% [6/26], 27% [7/26], 19% [5/26], and 31% [8/26]; U=736, p=0.044) (Fig. 3.). After prostatectomy, nodal metastases were described with significantly higher confidence on PSMA-PET/CT reports (respectively, 10% [4/41], 34% [14/41], 37% [15/41], and 20% [8/41]) compared to Fluciclovine PET-CT reports (respectively, 9% [1/11], 9% [1/11], 9% [1/11], and 73% [8/11]; U=334.5, p=0.011) (Fig. 4). Fluciclovine-PET/CT reports had no greater diagnostic certainty than PSMA-PET/CT reports to describe metastases in any category (among all patients, after prostatectomy or after radiation). There was no significant difference in diagnostic confidence between PSMA-PET/CT and Fluciclovine-PET/CT to describe other metastases.
Conclusions: Despite their lower PSA levels, patients undergoing PSMA-PET/CTs had osseous metastases described with higher diagnostic certainty compared to those who had Fluciclovine PET/CT. After prostatectomy, nodal metastases were described with higher certainty on PSMA than Fluciclovine PET/CT reports. Usage of PSMA rather than Fluciclovine for PET/CT may result in increased reader confidence to diagnose metastases in patients with biochemically recurrent PCa.