Prospective evaluation of PSMA-PET imaging for preoperative lymph node staging in prostate cancer

2014 
20 Objectives To prospectively evaluate the value of a new 68Ga-labelled PSMA-ligand (Glu-NH-CO-NH-Lys-(Ahx)-[(68)Ga(HBED-CC)]) targeting the prostate-specific membrane antigen (PSMA) for preoperative lymph node (LN) staging. Methods PSMA-PET imaging was performed in 37 intermediate and high risk PCa patients. After injection of 131±19 MBq 68Ga-PSMA 5 patients underwent contrast-enhanced PET/CT and 32 fully-diagnostic PET/MR. All patients had radical prostatectomy and extended pelvic LN dissection. All LN stations were evaluated on PET using a five-point-scale. Depending on the uptake of 68Ga-PSMA in the primary tumor results were given separately for group A (moderate/high uptake) and group B (no/faint uptake). Results Histopathology revealed LN metastases in 10 of 37 (27.0%) of patients and 19 of 214 (8.9%) stations on field-based analysis. In PET-imaging the primary tumor showed moderate/intense uptake in 33/37 patients with 4/37 only presenting with no/faint uptake. For group A on patient-based analysis, sensitivity and specificity for PSMA-PET were 75.0% (6/8) and 96.0% (24/25), respectively (AUC 0.848). On field-based analysis, these numbers were 64.7% (11/17) and 98.2% (168/71), respectively (AUC 0.813). Of note, histopathology revealed only micrometastases in the two false-negative patients. In group B two patients showed LN metastases in each one LN field which were not detected by PET-imaging. Conclusions For LN staging in PCa 68Ga-PSMA offers a high specificity both on patient and field based analysis. Our preliminary data show a moderate to high sensitivity for detecting LN metastases in patients with PET-positive primary tumors. However, our results also indicate that care must be taken in 68Ga-PSMA negative primary tumors as LN metastases appear to be PET-negative.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    0
    References
    10
    Citations
    NaN
    KQI
    []