Abstract
<title>Abstract</title> <p>Background. Accurate pelvic lymph node (N1) staging governs treatment selection in intermediate-to-high-risk prostate cancer, yet conventional imaging is insensitive to small nodal deposits. Prostate-specific membrane antigen positron emission tomography/computed tomography (PSMA PET/CT) is increasingly used for primary staging, but its per-patient accuracy against a histopathologic reference standard—and the factors that modulate it—remain incompletely defined. Methods. Following PRISMA-DTA 2020 guidance and a prospectively registered protocol, we searched five bibliographic databases from inception to June 8, 2026, with registry, grey-literature, and citation searches. Eligible studies evaluated PSMA PET/CT for primary per-patient N1 staging in treatment-naïve intermediate-to-high-risk disease against histopathologic extended pelvic lymph node dissection (ePLND). We fitted bivariate random-effects models, derived hierarchical and Moses–Shapiro–Littenberg summary ROC (SROC) curves, and performed prespecified subgroup analyses, meta-regression, 14 sensitivity analyses, leave-one-out and cumulative meta-analyses, and publication-bias assessment. Results. Thirty studies (5,927 patients; nodal prevalence 25.5%) were analyzed. Pooled sensitivity was 56.9% (95% CI 48.3–65.0%; I² = 86.9%) and specificity 93.2% (92.1–94.2%; I² = 26.0%), with a diagnostic odds ratio of 19.03 (12.82–28.23), positive and negative likelihood ratios of 7.44 and 0.50, and SROC area under the curve 0.90 (Q* = 0.895). The sensitivity prediction interval was wide (20.2–87.2%) but that for specificity narrow (89.8–95.6%). 68Ga-PSMA-11 achieved the highest sensitivity (65.1%) and 18F-DCFPyL the lowest (34.0%), while specificity was uniform across tracers. Partial verification inflated apparent sensitivity (81.7% vs 49.4% with complete verification). Meta-regression identified median PSA (p = 0.017), partial verification (p = 0.007), and prior conventional imaging (p = 0.006) as independent predictors of sensitivity. Estimates were robust across all 14 sensitivity analyses (sensitivity 43.1–65.1%; specificity 92.9–94.0%); significant small-study effects were present (Deeks’ p = 0.0009). Conclusions. PSMA PET/CT couples high, reproducible specificity with moderate, heterogeneous sensitivity for pelvic N1 staging. A positive scan reliably confirms nodal disease and supports treatment intensification, whereas a negative scan cannot exclude micrometastatic involvement and should be integrated with validated nomograms and selective ePLND rather than used in isolation to guide surgical decision-making.</p>