Jialiang Deng, Guoxiong Ma, Weijin Chen, Songhao Wen, Nanhui Chen
[177Lu]Lu-PSMA radioligand therapy (LuPSMA RLT) has improved outcomes for patients with metastatic castration-resistant prostate cancer (mCRPC), yet clinically meaningful response heterogeneity persists even among patients selected on the basis of PSMA PET positivity. A central limitation of patient-level PSMA eligibility is that it establishes target availability in at least part of the tumor burden, but it does not show whether all metabolically active lesions have sufficient PSMA expression, therapeutic retention, and delivered radiation dose. In this focused narrative review, we examine FDG/PSMA discordance as an imaging-suspected lesion-level phenotype of suspected target insufficiency rather than as a generic biomarker. We propose a provisional target-sufficiency framework that separates five research-testable states: concordant target-available disease, PSMA-positive target-density risk, focal suspected target-insufficient discordance, widespread suspected target-insufficient discordance, and delivery-insufficient target-positive disease. We synthesize dual-tracer eligibility studies, post-trial analyses of VISION- and TheraP-like criteria, FDG metabolic burden analyses, PSMA heterogeneity studies, post-therapy SPECT evidence, and LuPSMA dosimetry studies. Across predominantly retrospective cohorts, FDG-avid PSMA-low or PSMA-negative disease has been associated with adverse outcomes; however, this phenotype should not be conflated with high whole-body FDG metabolic tumor volume, low global PSMA uptake, measured absorbed-dose failure, or a proven resistance mechanism.