Correlation Between Clinical factors and PSMA Expression in Prostate Cancer: Variations Along the Disease Continuum

Introduction: Prostate-specific membrane antigen (PSMA) is recognized for its overexpression in prostate cancer cells, offering a target for PSMA PET imaging. Despite the widespread use of various clinical factors as biomarkers for prostate cancer, the nuanced relationship between these factors and PSMA expression throughout the disease continuum remains inadequately understood. This study aims to investigate the correlation between PSMA expression and key clinical parameters across the three phases of prostate cancer: initial workup, Biochemical Recurrence (BCR), and Castration-Resistant Prostate Cancer (CRPC) groups.

Methods: A retrospective study comprising 119 prostate cancer patients who underwent PSMA-targeted imaging was conducted. The patients were categorized into three groups based on their disease status: initial workup (n=35), BCR (n=22), and CRPC (n=62) groups. PSMA expression was assessed through Metabolic Tumor Volume (MTV), Total Lesion PSMA-expression (TLP), SUVmean, and the highest SUVmax measured in [68Ga] Ga-PSMA-11 PET. Clinical parameters included serum PSA levels, Androgen-Deprivation Therapy (ADT) status, PSA doubling time, PSA nadir, and initial treatment (radiation or prostatectomy). Spearman correlation coefficients were calculated to assess the relationship between PSA levels and PSMA PET/CT parameters in each group. Subgroup analyses were performed based on ADT status and initial treatment. Point-biserial correlation analysis was performed to evaluate the correlation between the status of ADT and PSMA expression in initial workup group.

Results: In the initial workup group, PSA exhibited a significant positive correlation with MTV (ρ=0.512, p=0.0017), TLG (ρ=0.517, p=0.0015), and the highest SUVmax (ρ=0.483, p=0.0033). The subgroup analysis revealed higher correlation coefficients in patients on ADT compared to those not on ADT (0.802 vs 0.525 for MTV, and 0.776 vs 0.445 for TLP), suggesting potential ADT effects on PSA levels and PSMA expression. In the BCR group, PSA demonstrated significant correlations with MTV (ρ=0.445, p=0.0379), TLG (ρ=0.656, p=0.0009), SUVmean (ρ =0.577, p=0.0050), and the highest SUVmax (ρ=0.627, p=0.0018). The CRPC group displayed significant correlations of PSA with MTV (ρ=0.751, p<0.0001) and TLG (0.716, p<0.0001), with subgroup analyses indicating distinct correlations (MTV : ρ=0.803 for ADT vs ρ=0.668 for prostatectomy) (TLG : ρ=0.842 for ADT vs ρ=0.475 for prostatectom) based on treatment modalities.

Conclusions: Our study revealed a consistent positive correlation between PSA levels and MTV/TLP across the disease continuum, particularly pronounced in the CRPC group. The subgroup analysis suggested potential effects of ADT on PSA levels and PSMA expression, emphasizing the need for further investigation into the interplay between hormonal therapies and PSMA-related imaging parameters. These findings underscore the complex relationship between clinical parameters and PSMA expression, providing valuable insights for advancing personalized diagnosis, and treatment approaches in prostate cancer.