Prospective comparison of simultaneous [Ga]Ga-PSMA-11 PET/MR versus PET/CT in patients with biochemically recurrent prostate cancer.

OBJECTIVES

PSMA-PET has become the PET technique of choice to localise the site of biochemically recurrent prostate cancer (PCa). With hybrid PET/MRI, the advantages of MRI are added to molecular characteristic of PET. The aim of this study was to investigate the incremental value of PET/MR versus PET/CT in patients with biochemically recurrent PCa by head-to-head comparison.

METHODS

Thirty-four patients with biochemically recurrent PCa were prospectively included. They underwent [Ga]Ga-PSMA-11 PET/CT, followed by simultaneous PET/MR. All PET (PET, PET), CT and MR images were evaluated for number of lesions and location. The number of lesions at specific sites was compared using Wilcoxon-sign-rank test. For PET, the maximum and mean standardised uptake values (SUVs) were calculated for each lesion compared using a two-sided paired t test.

RESULTS

PET and PET scans were positive in 19 and 20 patients, detecting 73 and 79 lesions respectively. All lesions detected on PET were also detected on PET. CT and MRI only were positive in 14 and 17 patients, detecting 38 and 50 lesions, respectively, which was significantly lower than PET and PET respectively. Combined interpretation showed more lesions on PET/MR than on PET/CT (88 vs 81). No significant difference in detection of presence of local recurrence nor distant metastases was found. SUV and SUV values were significantly higher on PET than on PET in local recurrence and lymph node metastases.

CONCLUSIONS

[Ga]Ga-PSMA-11 PET/MR was able to detect biochemically recurrent PCa at least as accurately as PET/CT for local recurrence, lymph node metastasis and distant metastasis.

KEY POINTS

• PSMA PET/MRI detects the location of biochemical recurrence at least as accurately as PET/CT. • Substitution of PET/CT by PET/MRI adds sensitivity in PSMA lesion detection also in the setting of distant recurrence due to both the MR and TOF PET components.