Underestimation of positron emission tomography/computerized tomography in assessing tumor burden in prostate cancer nodal recurrence: Head-to-head comparison of 68Ga-PSMA and 11C-choline in a large, multi-institutional series of extended salvage lymph node dissections

Purpose:We compared the use of 11C-choline and 68Ga-prostate specific membrane antigen in men undergoing salvage lymph node dissection for nodal recurrent prostate cancer.Materials and Methods:The study included 641 patients who experienced prostate specific antigen rise and nodal recurrence after radical prostatectomy and underwent salvage lymph node dissection. Lymph node recurrence was documented by positron emission tomography/computerized tomography using 11C-choline (407, 63%) or 68Ga-PSMA ligand (234, 37%). The outcome was underestimation of tumor burden (difference between number of positive nodes on final pathology and number of positive spots at positron emission tomography/computerized tomography). Multivariable analysis tested the association between positron emission tomography/computerized tomography tracer (11C-choline vs 68Ga-PSMA) and tumor burden underestimation.Results:Overall the extent of tumor burden underestimation was significantly higher in the 11C-choline group compared to the 68Ga-PSMA group (p <0.0001), which was confirmed on multivariable analysis (p=0.028). Repeating these analyses according to prostate specific antigen, tumor burden underestimation was lower with 68Ga-PSMA only when prostate specific antigen was 1.5 ng/ml or less. Conversely, the underestimation of the 2 tracers became similar when prostate specific antigen was greater than 1.5 ng/ml. Furthermore, we evaluated the risk of underestimation by number of positive spots on positron emission tomography/computerized tomography. The higher the number of positive spots the higher the underestimation of tumor burden regardless of the tracer used (p=0.2).Conclusions:Positron emission tomography/computerized tomography significantly underestimates the burden of prostate cancer recurrence, regardless of the tracer used. 68Ga-PSMA was associated with a lower rate of underestimation in patients with a prostate specific antigen below 1.5 ng/ml and a limited nodal tumor load. In all other men there was no benefit from 68Ga-PSMA over 11C-choline in assessing the extent of nodal recurrence.