ロボット支援根治的前立腺摘除術における膀胱鏡下ICGガイド下センチネルリンパ節郭清：前向き研究

To evaluate the diagnostic accuracy, procedural feasibility, and clinical outcomes of a novel cystoscopic-guided indocyanine green (ICG) injection technique followed by fluorescence-guided sentinel lymph node dissection (SLND) during robotic-assisted radical prostatectomy (RaRP), hereafter referred to as CysICG-SLND.

We prospectively enrolled patients with clinically localized prostate cancer (cT1-cT3, cN0) scheduled for RaRP between November 2019 and May 2020. Each patient received cystoscopic-guided intraprostatic injections of ICG, with injection volumes tailored according to prostate size. SLND was performed using the da Vinci Xi Surgical System equipped with near-infrared imaging and followed by a routine extended pelvic lymph node dissection (ePLND). Diagnostic accuracy, procedural complications, and medium-term oncological outcomes were analyzed.

A total of 34 patients were included (risk category: low = 2, intermediate = 11, high = 21). Cystoscopic ICG injection was technically successful in all cases. The overall fluorescence detection rate was 97.1%, with excellent per-patient sensitivity (100%) and negative predictive value (NPV) (100%) across all patients. However, per-node diagnostic performance was moderate (sensitivity 66.7%, specificity 47.8%). Notably, in the high-risk group, per-node performance was modest, with sensitivity, specificity, positive predictive value (PPV), and NPV of 66.7%, 55.2%, 20.5%, and 90.7%, respectively. No metastatic nodes were identified in low-risk and intermediate-risk groups, demonstrating a reliable 100% NPV in these subpopulations. High-risk patients had significantly higher biochemical recurrence (BCR) rates compared to lower-risk groups (p = 0.017). There were no allergic reactions or postoperative complications attributable to ICG.

CysICG-SLND is a feasible, accurate approach for nodal staging in prostate cancer, particularly beneficial in low- to intermediate-risk groups, achieving excellent sensitivity and NPV. Although diagnostic performance is limited in high-risk patients, high per-patient accuracy still supports clinical utility. Larger prospective studies are warranted to validate these results and further elucidate the technique's clinical role.