18FDG-PET/CT in the assessment of lymph node involvement in patients with invasive penile carcinoma

Objectives The extent of lymph node involvement (LNI) is the most relevant prognostic factor in patients with penile cancer. This prospective study was conducted to analyze the diagnostic accuracy of 18FDG-PET/CT in the assessment of LNI in patients with invasive penile carcinoma.

Methods Thirty two patients (mean age: 74 y) with invasive penile carcinoma were staged by 18FDG-PET/CT and conventional CT scan before surgery. According to TNM 2009 classification, 22 patients were clinically node-negative (cN0) and 10 patients clinically node-node positive (cN+: 6 cN1 and 4 cN2-3). PET/CT were performed on a Discovery LS hybrid PET scanner (GE), 60 min after injection of 5-7 MBq of 18FDG and blindly evaluated by experienced nuclear medicine physicians. A FDG uptake exceeding background activity was defined as pathological. All of the patients were operated and histopathology (pN) served as gold standard for the statistical analyzes.

Results In the 22 cN0 patients (44 inguinal area = IA), LNI was finally pathologically assessed in 5 IA (4 patients), Sensitivity, specificity, predictive positive value, negative predictive value and accuracy per IA of PET were 60, 87, 37, 94 and 84% respectively, compared to 80, 79, 33, 97 and 79% for conventional CT scan (p=ns). PET was falsely positive in 5 IA, in 4 patients with inflammatory primitive lesions. In 2 IA (2 patients), micro-metastases (≤9 mm) were missed by PET. In the 10 cN+ patients, PET correctly evaluated LNI extent. Especially, PET was negative in the 2 pN0 patients and correctly upstaged 3 patients cN1 to pN2.

Conclusions 18FDG-PET/CT scanning appears promising for assessing LNI extent in cN+ patients with invasive penile carcinoma, with potential impact on therapeutic strategy. In cN0 patients, PET helps to detect LNI, but because of its low sensitivity to detect micrometastases, surgical staging oriented or not by sentinel node detection remains necessary