Does PSA Nadir + 2 ng/mL Always Indicate Biochemical Recurrence? A PSA Kinetics-Based Evaluation Following Carbon Ion Radiotherapy for Localized High-Risk Prostate Cancer

<b>Background:</b> Biochemical recurrence after radiotherapy for prostate cancer is commonly defined by the Phoenix criteria (prostate-specific antigen [PSA] nadir + 2 ng/mL). However, some patients experience PSA elevation without clinical recurrence, which is known as PSA bounce. This study aimed to evaluate PSA kinetics after scanning-method carbon ion radiotherapy (CIRT) in patients with high-risk prostate cancer (HR-PCa) and to assess the clinical validity of the Phoenix criteria. <b>Methods:</b> We retrospectively analyzed 171 patients with HR-PCa who underwent CIRT and 2 years of androgen deprivation therapy. Patients were classified into three groups based on post-treatment PSA kinetics: non-recurrence, pseudo-recurrence (PR; PSA > 2 ng/mL followed by spontaneous decline without salvage therapy), and recurrence (R; PSA > 2 ng/mL with salvage therapy). PSA bounce was defined as a transient PSA increase > 0.4 ng/mL followed by spontaneous decline. Kaplan-Meier and receiver operating characteristic (ROC) analyses were used to evaluate biochemical relapse-free survival and determine the optimal PSA cutoff. <b>Results:</b> Among 171 patients, 18 (10.5%) met the Phoenix criteria (R+PR), of whom 6 (33.3%) experienced spontaneous PSA decline. The 5-year biochemical relapse-free survival rate was 90.0%. PSA bounce occurred in 33.9%. ROC analysis identified an optimal PSA cutoff of 1.91 ng/mL (area under the curve: 0.985), whereas the positive predictive value at the 2 ng/mL cutoff was as low as 61.1% due to the influence of PSA bounce. <b>Conclusions:</b> After CIRT, a PSA rise of >2 ng/mL does not always indicate HR-PCa recurrence and should be interpreted with caution to avoid overtreatment.