Introduction

Positron emission tomography / computed tomography (PET/CT) scan with 11C-choline or 68Ga-prostate-specific-membrane-antigen (PSMA) ligand is used to diagnose nodal recurrence of prostate cancer. However, no previous study directly compared these two tracers in men receiving salvage lymph node dissection (SLND).

Methods

The study included 266 patients affected by a single nodal recurrence and treated with SLND at eight tertiary referral centers. All patients were diagnosed with PET/CT scan using either 11C-choline (n=196; 74%) or 68Ga-PSMA ligand (n=70; 26%). The study outcomes were the following: (i) rate of negative histologic report at final pathology; (ii) concordance between site of positive imaging and location of positive nodes (iii) biochemical response (BR) defined as PSA ≤0.2 ng/ml at one month after SLND.

Results

PSA level at SLND was lower in patients diagnosed with PSMA-ligand (2.2 vs. 1.5 ng/ml, p=0.011). The single positive spot at pre-operative imaging was pelvic and retroperitoneal in 178 (91%) and 18 (9%) patients diagnosed with 11C-choline, versus 62 (89%) and 8 (11%) patients diagnosed with PSMA ligand (p=0.6). In the 11C-choline group, 41 out of 178 (23%) patients with a single pelvic spot received also retroperitoneal SLND, compared to 21 out of 62 (34%) patients in the PSMA group. In the 11C-choline group 12 out of 18 (67%) patients with a single retroperitoneal spot received also pelvic SLND, compared to 8 out of 8 (100%) patients in the PSMA group. The median number of lymph nodes removed was not significantly different (18 vs. 16, p=0.3). Overall, 50 (26%) patients diagnosed with 11C-choline and 19 (27%) patients diagnosed with PSMA had negative histologic report at final pathology (p=0.7), whereas 92 (47%) and 20 (30%) patients had ≥2 positive nodes, respectively (p=0.013). The concordance between site of positive imaging and location of positive nodes was not significantly different among groups (72% vs. 69%, p=0.3). 0verall, 93 (47%) and 23 (32%) patients had positive nodes outside the positive sites of PET/CT scan (p=0.013). The BR rate was significantly higher in the PSMA group (33% vs. 43%, p=0.016).

Conclusions

In patients with a single positive spot at PET/CT scan, the rate of negative histology is not significantly different when 11C-choline or PSMA are used. Moreover, both ligands significantly underestimate the true nodal tumor burden. However, using PSMA ligand tracer, similar diagnostic results may be achieved at lower PSA level.

Funding

none