Comparison of 68 Ga-PSMA ligand PET/CT versus conventional cross-sectional imaging for target volume delineation for metastasis-directed radiotherapy for metachronous lymph node metastases from prostate cancer.

PURPOSE: To assess the differences in the target volume (TV) delineation of metachronous lymph node metastases between 68  Ga-PSMA ligand PET/CT and conventional imaging in a comparative retrospective contouring study.

PATIENTS AND METHODS: Twenty-five patients with biochemical prostate cancer recurrence after primary prostatectomy underwent 68  Ga-PSMA ligand PET/CT in addition to conventional imaging techniques such as CT and/or MR imaging for restaging. All patients were diagnosed with at least one lymph node metastasis. TVs were manually delineated in two different ways: (a) based on conventional imaging (CT/MRI) and (b) based on conventional imaging (CT/MRI) plus 68  Ga-PSMA ligand PET/CT. The size of TVs, overlap rates, and subjective assessment of the difficulty of TV delineation reported by the radiation oncologist (easy/moderate/difficult) were compared.

RESULTS: With the additional information from PSMA ligand PET, 47 lymph node metastases were identified and included in the gross tumor volume (GTV). The median clinical target volume (CTV) of non-PET-based TV delineation was statistically larger than the CTV based on PET imaging (134.8 ml [range 6.9-565.2] versus 44.9 ml [range 4.9-481.3; p = 0.001]). The CTV based on CT/MRI enclosed only 81.3% (39/48) of PET-positive lymph nodes. The CT/MRI-based CTV did not enclose all PET-positive lymph nodes in 24% (6/25) of patients. In 12% (3/25) of patients, all PET-positive lymph nodes were outside of the CT/MRI-based CTV. The median overlap rates (TVPET /TVCT/MRI  × 100) were 45.7% (range 0-96.9) for the GTV and 71.7% (range 9.8-98.2) for the CTV. The assessment of difficulty of contouring revealed that contouring with the additional imaging information of the PET was categorized as easy/moderate in 92% (23/25) and as difficult in 8% (2/25) of the cases, whereas contouring based on CT/MRI without PET was categorized as difficult in 56% (14/25) and as easy/moderate in 44% of the cases (11/25; p = 0.003).

CONCLUSION: 68  Ga-PSMA ligand PET/CT is superior to conventional cross-sectional imaging for the delineation of lymph node metastases from prostate cancer. PET-based TV delineation allows for smaller target volumes and should be considered the standard for irradiation of metachronous lymph node metastases in recurrent prostate cancer. Conventional imaging is not sufficiently sensitive for radio-oncological treatment concepts in oligometastatic prostate cancer.