From Imaging Guidance to Dosimetric and Staging Superiority: Impact of ⁶⁸Ga-PSMA PET/CT and mpMRI on Target Delineation, Organs at Risk Sparing, and Biochemical Response in Prostate Cancer Radiotherapy—A Single-Center Experience

Abstract

Purpose: To evaluate inter-modality agreement and differences in detection rates between ⁶⁸Ga-PSMA PET/CT and mpMRI for primary staging in prostate cancer, and to assess their impact on radiotherapy (RT) target-volume personalization, organ-at-risk (OAR) sparing, and post-RT prostate-specific antigen (PSA) dynamics. Methods: We retrospectively analyzed 98 patients with biopsy-proven prostate adenocarcinoma who underwent pretreatment ⁶⁸Ga-PSMA PET/CT and definitive intensity-modulated or volumetric-modulated arc RT; baseline mpMRI was available in 75 patients. Treatment plans were classified into six categories according to fractionation (conventional vs. moderate hypofractionation), pelvic nodal coverage, and focal boost to dominant intraprostatic lesions (DILs). Inter-modality agreement was assessed using Cohen’s kappa. DVH parameters for target volumes and OARs, including physical dose, EQD2, and BED, were compared between fractionation regimens. Post-RT PSA levels were evaluated at 1 week–1 month, 3–6 months, 12 months, and 24 months. Results: For primary T-stage (n = 70), overall agreement between mpMRI and ⁶⁸Ga-PSMA PET/CT was 32.9% (23/70), with slight agreement (Cohen’s κ = 0.19; 95% CI, 0.04–0.34; p = 0.008). Among 47 patients classified as having organ-confined disease (T2a–T2c) on mpMRI, 7 (14.9%) were upstaged to T3–T4 by ⁶⁸Ga-PSMA PET/CT. For pelvic nodal disease (n = 74), ⁶⁸Ga-PSMA PET/CT detected N1 disease more frequently than mpMRI (25.7% [19/74] vs. 10.8% [8/74]; p = 0.012, McNemar test), with fair agreement (κ = 0.26; 95% CI, 0.05–0.47; overall percent agreement, 77.0%). For seminal vesicle invasion (n = 73), agreement was moderate (κ = 0.47; 95% CI, 0.24–0.70; p < 0.001; overall percent agreement, 82.2%). Compared with conventional RT, moderately hypofractionated RT provided significantly lower absolute OAR exposure: mean rectal dose (21.57 ± 3.58 vs. 28.46 ± 6.55 Gy; p < 0.001), mean rectal BED₃ (39.54 ± 6.56 vs. 47.67 ± 10.96 Gy₃; p < 0.001), rectal V50Gy (11.32 ± 3.65% vs. 14.39 ± 5.17%; p < 0.001), mean bladder dose (24.17 ± 8.63 vs. 31.02 ± 8.84 Gy; p < 0.001), and maximum femoral-head dose (35.12 ± 7.46 vs. 47.20 ± 7.99 Gy; p < 0.001). Target coverage, when normalized to the prescription dose, was comparable between fractionation groups. The median PSA decreased from 0.96 ng/mL before RT to 0.18 ng/mL at 1 week–1 month, 0.06 ng/mL at 3–6 months, and 0.04 ng/mL at 12 months. In exploratory subgroup analyses, baseline clinicopathological variables, mpMRI PI-RADS category, and ⁶⁸Ga-PSMA PET/CT parameters were not significantly associated with the magnitude of post-RT PSA decline (all p > 0.05). Conclusions: In this retrospective cohort, ⁶⁸Ga-PSMA PET/CT provided complementary staging information to mpMRI, particularly by identifying additional pelvic nodal disease and selected local upstaging, thereby supporting individualized target-volume delineation. Moderately hypofractionated image-guided RT achieved comparable relative target coverage with lower absolute OAR doses. Early and sustained PSA suppression likely reflects the combined effect of precise RT and risk-adapted androgen deprivation therapy. Owing to the retrospective design, absence of histopathological validation, and limited long-term follow-up, these findings require confirmation in larger prospective cohorts.

1. Introduction

Prostate cancer remains one of the most frequently diagnosed malignancies and a leading cause of cancer-related mortality among men worldwide [1]. External beam radiotherapy (RT) is established as a standard curative modality for localized and regionally advanced prostate cancer, yielding long-term outcomes comparable to radical prostatectomy [2]. The primary goal of contemporary radiation oncology is to achieve optimal tumor control within target volumes while maintaining steep dose gradients to minimize radiation exposure and dosimetric burden to surrounding organs at risk (OARs), including the rectum, bladder, and femoral heads [3,4].

The evolution of advanced delivery techniques, such as intensity modulated radiation therapy (IMRT) and volumetric modulated arc therapy (VMAT), has enabled high precision in dose conformity and OAR sparing [5]. However, the therapeutic efficacy of these techniques relies heavily on accurate pre-treatment target volume delineation and rigorous anatomical and biological staging. Limitations inherent to traditional anatomical imaging in detecting subclinical microscopic extension and occult nodal metastases carry a risk of geographical miss or unnecessary irradiation of normal tissue [5,6,7].

Multiparametric magnetic resonance imaging (mpMRI) is widely regarded as the gold standard for intraprostatic lesion visualization and local T-staging, particularly for identifying extracapsular extension and seminal vesicle involvement [8,9]. Nevertheless, mpMRI exhibits limited sensitivity for detecting pelvic lymph node metastases due to its reliance on size-based morphologic criteria [10]. Conversely, gallium-68 labeled prostate-specific membrane antigen positron emission tomography/computed tomography (⁶⁸Ga-PSMA PET/CT) targets molecular PSMA expression, enabling the identification of subcentimeter, morphologically occult nodal disease with high specificity [11]. Integrating mpMRI and ⁶⁸Ga-PSMA PET/CT via image fusion combines anatomical detail with molecular imaging information, providing a robust platform for personalized RT target volume definition [1,5,11].

Another critical aspect of modern RT strategy involves the selection of optimal fractionation schedules and focal dose escalation [7,8]. Because prostate carcinoma exhibits a uniquely low α/β ratio (approximately 1.5 Gy), moderately hypofractionated RT regimens offer radiobiological advantages by enhancing the therapeutic ratio while reducing integral OAR doses [12,13]. Furthermore, applying a boost to dominant intraprostatic lesions (DILs) delineated by dual-modality fusion represents an effective approach to escalate target dose without exceeding acceptable OAR toxicity thresholds [5,14].

The objective of this study was to retrospectively evaluate a cohort of prostate cancer patients staged with both ⁶⁸Ga-PSMA PET/CT and mpMRI who underwent definitive IMRT/VMAT. We aimed to compare the diagnostic staging performance (T and N staging concordance and upstaging) of both modalities, assess OAR dosimetric sparing across six distinct RT target field classifications (incorporating conventional vs. hypofractionated regimens and focal boost applications), and investigate whether personalized biological RT planning achieves favorable OAR dose-volume profiles while delivering uniform post-treatment PSA response kinetics.

2. Materials and Methods

2.1. Patient Population and Study Design

In this retrospective study, 98 patients with biopsy-proven prostate adenocarcinoma who presented to the Department of Radiation Oncology, Faculty of Medicine, Recep Tayyip Erdogan University between January 2022 and December 2024 were evaluated. Inclusion criteria required baseline primary staging with ⁶⁸Ga-PSMA PET/CT and/or multiparametric magnetic resonance imaging (mpMRI) according to institutional protocols, followed by definitive primary RT.

The present cohort substantially overlaps with our previously published study [5]. Initially, 98 patients treated between January 2022 and December 2024 who underwent pretreatment ⁶⁸Ga-PSMA PET/CT imaging formed the cohort of the present study. Subsequently, in our previously published study [5], which was designed to evaluate the clinical impact of ⁶⁸Ga-PSMA PET/CT fusion-based radiotherapy planning in a broader population, 95 of these 98 patients (96.9%) were included, and the enrollment period was extended until May 2025. Therefore, 40 additional patients treated after the present study period were included in [5]. These 40 newly added patients represented 29.6% of the total 135-patient cohort in [5], demonstrating that [5] was a substantially expanded and separately analyzed cohort. Importantly, the two studies address distinct scientific questions and employ different analytical frameworks. The previous study compared ⁶⁸Ga-PSMA PET/CT fusion-based radiotherapy planning with non-PET-based planning regarding target delineation, dosimetry, and early PSA response. In contrast, the present study evaluates the concordance between mpMRI and ⁶⁸Ga-PSMA PET/CT for primary T-stage, pelvic nodal status, and seminal vesicle invasion, alongside dosimetric outcomes based on fractionation, pelvic nodal coverage, focal boost, and longitudinal post-radiotherapy PSA kinetics. No outcome analysis or result is duplicated between the two manuscripts.

Demographic, clinical, radiological, nuclear medicine, and dosimetric data of all eligible patients were extracted from the hospital information management system and radiation treatment planning systems. Patients were stratified according to NCCN Guidelines (Risk Stratification Categories) into Low-Risk, Intermediate-Risk (Favorable/Unfavorable), and High-Risk/Very High-Risk groups based on clinical T-category, initial diagnostic PSA level, and ISUP Grade Group.

Per institutional protocol aligned with NCCN guidelines for definitive RT, androgen deprivation therapy (ADT) was administered as follows:

• Low-Risk (n = 11, 11.2%): Received definitive RT alone without ADT.
• Intermediate-Risk (n = 53, 54.1%): Received definitive RT combined with short-to-intermediate-course ADT (1 year total duration; neoadjuvant, concurrent, and adjuvant phases).
• High-Risk/Very High-Risk (n = 34, 34.7%): Received definitive RT combined with long-term ADT (2 to 3 years total duration; neoadjuvant, concurrent, and adjuvant phases).

Overall, ADT was administered predominantly as LHRH agonists—primarily leuprolide acetate and occasionally goserelin acetate. To prevent the testosterone flare effect associated with LHRH agonist initiation, concurrent oral bicalutamide (50 mg/day) was administered for 1 month at the start of therapy. Consequently, the marked reduction from Initial Diagnostic PSA (median: 11.39 ng/mL) to Pre-RT Baseline PSA (median: 0.96 ng/mL) is consistent with PSA suppression during neoadjuvant ADT.

Regarding distant metastatic status, ⁶⁸Ga-PSMA PET/CT identified oligometastatic bone disease (M1b, defined as ≤5 non-visceral bone lesions) in 5 patients (5.1%). In accordance with contemporary NCCN and EAU guidelines advocating local prostate-directed radiotherapy for low-volume/oligometastatic prostate cancer to improve local control and systemic outcomes, these patients received definitive prostate radiotherapy alongside long-term ADT. All 5 M1b patients were retained in the overall cohort and included in the primary staging, dosimetric, and longitudinal PSA analyses.

2.2. Imaging Protocols and Staging

• mpMRI Acquisition and Evaluation: Pre-treatment pelvic mpMRI scans (3.0 Tesla MRI scanner GE Discovery MR750w, GE Healthcare, Waukesha, WI, USA) included T2-weighted, diffusion-weighted (DWI/ADC), and dynamic contrast-enhanced (DCE) sequences. Images were systematically analyzed according to Prostate Imaging Reporting and Data System (PI-RADS v2.1) criteria [15] to determine the primary intraprostatic lesion location, extracapsular extension (ECE), seminal vesicle invasion (SVI), and regional pelvic lymph node status. Among the 75 patients with available mpMRI scans, definitive T-staging concordance was evaluable in n = 70 cases. In 5 patients, mpMRI showed no focal suspicion of prostatic malignancy (reported as normal/PI-RADS 1–2 or unidentifiable intraprostatic lesion), whereas ⁶⁸Ga-PSMA PET/CT successfully localized the primary lesion and assigned a definitive PSMA T-stage. These non-localized mpMRI cases were excluded from local T-stage cross-tabulation to maintain strict paired category alignment.
• ⁶⁸Ga-PSMA PET/CT Evaluation: Whole-body ⁶⁸Ga-PSMA PET/CT scans (Biograph mCT, Siemens Healthcare GmbH, Erlangen, Germany) were evaluated at the molecular level to determine primary tumor focus, regional lymph node involvement, and distant metastases. Pathological foci exhibiting PSMA avidity were anatomically categorized, and maximum standardized uptake values (SUV_max) were quantitatively recorded.

2.3. Target Volume Delineation, Dosimetry, and Radiotherapy Planning

All treatment plans were computed on the Eclipse Treatment Planning System (Eclipse™ v13.6; Varian Medical Systems, Palo Alto, CA, USA) using the Anisotropic Analytical Algorithm (AAA). Treatments were delivered on a Varian linear accelerator (Trilogy, Varian Medical Systems, Palo Alto, CA, USA) using daily kilovoltage image-guided RT (kV-IGRT). Planning CT scans were acquired on a single dedicated CT simulator (Aquilion™ LB, Toshiba Medical Systems, Tokyo, Japan). To eliminate inter-observer variability, all target volumes and OARs were delineated by a single senior radiation oncologist. Multimodality rigid image fusion (⁶⁸Ga-PSMA PET/CT, mpMRI, and planning CT) was executed directly on the Eclipse TPS.

Target volume definitions strictly adhered to the NRG Oncology/RTOG consensus guidelines for definitive prostate cancer RT [5,16,17,18].

• Gross Tumor Volume (GTV): Defined as pathologically avid foci corresponding to dominant intraprostatic lesions (DILs) and/or metastatic pelvic lymph nodes identified on ⁶⁸Ga-PSMA PET/CT and/or mpMRI.
• Primary Target Volume: To avoid dosimetric confounding between elective pelvic nodal basins and the primary tumor site, the target volume evaluated in primary target dosimetric analyses refers specifically to the primary clinical target volume (CTV_primary). In cases where pelvic nodal basins were not irradiated, or when evaluating the primary prostatic component, this target encompassed the prostate gland alone or the prostate combined with the proximal or entire seminal vesicles according to NCCN risk criteria, strictly excluding elective pelvic lymph node CTVs.
• Clinical Target Volume (CTV): Encompassed the entire prostate, the proximal 1 cm of the seminal vesicles or the entire seminal vesicles, depending on risk category based on NCCN risk stratifications [1], and elective pelvic nodal basins (common, external, internal iliac, presacral, and obturator) when pelvic nodal irradiation was indicated.
• Planning Target Volume (PTV): Constructed by adding anisotropic margins of 3–5 mm (3 mm posteriorly toward the rectum) to the CTV to account for organ motion and setup uncertainties (5 mm for pelvic nodes) [19].
• Focal Boost Application: In patients with identifiable localized foci, a focal boost was targeted specifically to the PET/MRI-defined DIL while strictly respecting OAR constraints [14,17,20].

For inverse planning optimization, treatment field design strictly maintained institutional OAR dose-volume constraints, including the rectum (V_65Gy < 25%, V_50Gy < 50%), urinary bladder (V_65Gy < 35%), bilateral femoral heads (D_max < 50 Gy), and penile bulb (D_mean < 40 Gy).

2.4. Radiotherapy Planning and Target Volume Classification

Radiotherapy plans were generated using IMRT or VMAT, with either conventional fractionation (2.0 Gy/fraction) or moderate hypofractionation (2.5 Gy/fraction). Plans were retrospectively classified into six treatment categories according to the fractionation schedule, the inclusion of pelvic lymph node irradiation, and the use of a focal boost. Pelvic irradiation was delivered specifically to patients with lymph node involvement detected on ⁶⁸Ga-PSMA PET/CT or mpMRI. A focal boost was delivered sequentially (rather than as a simultaneous integrated boost [SIB]) to visible DILs in selected patients.

• Field 1 (Hypo 70-P): Hypofractionated RT (Prostate ± Seminal Vesicles Only): Target volume encompassed the prostate ± seminal vesicles; a total dose of 70 Gy in 28 fractions (2.5 Gy/fraction) was delivered using moderate hypofractionation.
• Field 2 (DIL boost-Hypo 72.5-P): Hypofractionated RT with Intraprostatic Boost (Prostate ± Seminal Vesicles Only). While 70 Gy was delivered to the prostate ± seminal vesicles (2.5 Gy/fraction), a focal boost delivered a total dose of 72.5 Gy in 29 fractions (2.5 Gy/fraction) to the DIL identified on ⁶⁸Ga-PSMA PET/CT and mpMRI fusion.
• Field 3 (Conv 76-78-P): Conventional RT (Prostate ± Seminal Vesicles Only): Conventional fractionation schedule targeting the prostate ± seminal vesicles without pelvic lymph node irradiation, delivering a total dose of 76–78 Gy in 38–39 fractions (2.0 Gy/fraction).
• Field 4 (Conv 76-78 P-LN): Conventional RT with Pelvic Nodal Irradiation: Whole-pelvis conventional RT protocol encompassing elective pelvic lymph nodes in addition to the prostate and seminal vesicles, delivering 76–78 Gy in 38–39 fractions (2.0 Gy/fraction).
• Field 5 (DIL boost-Conv 80-P): Conventional RT with Intraprostatic Boost (Prostate ± Seminal Vesicles Only): Prostate ± seminal vesicles received 78 Gy in 39 fractions (2.0 Gy/fraction), while a focal boost reached a total dose of 80 Gy in 40 fractions (2.0 Gy/fraction) to the PET/MR-defined DIL.
• Field 6 (DIL boost Conv 80 P-LN): Conventional RT with Nodal Irradiation and Intraprostatic Boost: The largest target volume category, incorporating whole-pelvis, seminal vesicle, and prostate irradiation (78 Gy in 39 fractions) with a focal boost to the DIL reaching a total dose of 80 Gy in 40 fractions.

2.5. Imaging Evaluation and Inter-Modality Agreement

Diagnostic findings were evaluated by comparing anatomical stage derived from multiparametric MRI (mpMRI) with molecular findings from ⁶⁸Ga-PSMA PET/CT. In the absence of histopathological validation (pelvic lymph node dissection or prostatectomy specimen), differences in nodal and organ confined stage identification were categorized as inter-modality discordance and higher detection rate, rather than true diagnostic sensitivity or accuracy. Inter-modality agreement for local T-staging and nodal involvement was assessed using Cohen’s kappa (κ) statistics.

2.6. Dosimetric and Organs at Risk Analysis

Dose-volume histogram (DVH) metrics for target volumes (Prostate/CTV_primary/PTV) and OARs—including the rectum, urinary bladder, bilateral femoral heads, penile bulb, and external body contour—were extracted from the radiation treatment planning system (TPS) for each patient. The evaluated dosimetric parameters comprised:

• CTV_primary: Percentage minimum (% Min), maximum (% Max), and mean (% Mean) relative doses.
• Rectum: Percentage Min, Max, and Mean relative doses, along with the percentage of organ volume receiving an absolute dose of at least 65 Gy (V_65Gy) and 50 Gy (V_50Gy).
• Urinary Bladder, Femoral Heads, Penile Bulb, and Body: Percentage Min, Max, and Mean relative dose values.

2.7. Follow-Up and Biochemical Outcome Definitions

Serum prostate-specific antigen (PSA) levels were recorded during routine post-treatment outpatient follow-ups at pre-specified intervals: 1 week–1 month (first control), 3–6 months, 12 months (1 year), and 24 months (2 years). To evaluate longitudinal treatment efficacy and biochemical dynamics, interval changes (ΔPSA), absolute PSA reductions, and percentage changes were computed relative to the preceding follow-up interval and baseline levels.

Biochemical failure was defined according to the ASTRO-RTOG Phoenix consensus criterion as a serum PSA rise of ≥2.0 ng/mL above the post-RT nadir PSA. Biochemical failure was assessed for patients with available long-term follow-up (n = 20 at 24 months).

2.8. Statistical Analysis

Statistical analyses were performed using SPSS software (Version 26.0; IBM Corp., Armonk, NY, USA). Continuous variables were evaluated for normality using the Kolmogorov–Smirnov and Shapiro–Wilk tests. Descriptive statistics were presented as counts and percentages (%) for categorical variables. For continuous variables, data were reported as mean ± standard deviation or median with interquartile range (IQR, 25th–75th percentiles) and minimum–maximum range, depending on distribution.

Due to significant right-skewness and high variance observed in serum PSA changes (ΔPSA), non-parametric methods were prioritized for statistical inference across Section 3.2.6, Section 3.2.7, Section 3.2.8 and Section 3.2.9. Differences between two independent groups were evaluated using the Mann–Whitney U test, whereas comparisons across three or more groups were analyzed using the Kruskal–Wallis test. All tests were two-tailed, and a p-value <0.05 was considered statistically significant.

Non-parametric continuous and ordinal variables across multiple subgroups (up to six fields) were compared using the Kruskal–Wallis test. To identify specific inter-group differences while controlling for Type I error, pairwise post-hoc comparisons were conducted using Dunn’s test with Bonferroni correction for multiple testing. Tied ranks were assigned average rank values during non-parametric computations. Missing data were handled using complete-case analysis for each specific paired comparison without imputation. Inter-modality diagnostic concordance between mpMRI and ⁶⁸Ga-PSMA PET/CT was evaluated using Cohen’s kappa (κ) coefficient and reported alongside asymptotic 95% confidence intervals (95% CI) and overall percent agreement (OPA).

2.9. Ethics Approval and Institutional Permissions

This study was approved by the Non-Interventional Clinical Research Ethics Committee of Recep Tayyip Erdogan University (Date: 29 April 2025; Reference No: E-40465587-050.01.04-1506; Decision No: 2025/181). Institutional permission was granted by the Scientific Research Application Review Commission of the Rize Provincial Health Directorate, Republic of Turkiye Ministry of Health (Date: 27 January 2025; Reference No: E-64960800-799-266637787). The study was conducted in strict accordance with the ethical principles of the Declaration of Helsinki and relevant patient privacy regulations.

Given the retrospective design of the study and the use of fully anonymized historical data, the need for study-specific informed consent was waived by the Ethics Committee. However, written informed consent regarding the routine clinical use and scientific research publication of de-identified medical data had been routinely obtained from all adult patients prior to treatment initiation and maintained in their institutional medical records. All patient data were fully anonymized prior to extraction and statistical analysis to ensure strict confidentiality, and no identifiable personal information or clinical images are included in this manuscript.

3. Results

3.1. Diagnostic and Clinical Characteristics

3.1.1. Demographic, Clinical, and Pathological Characteristics

Demographic, clinical, pathological, and imaging characteristics of the 98 included patients are summarized in Table 1.

Table 1. Demographic, Pathological, Clinical, and Imaging Characteristics of Patients (N = 98).

Parameters	Median (Min–Max)	Mean ± SD	n (%)
Demographic and Clinical Characteristics
Age (years)	70 (47–84)	69.03 ± 7.53	-
Follow-up duration (months)	16.0 (1–109)	19.58 ± 15.83	-
Initial Diagnostic PSA (ng/mL)	11.39 (0.04–99.0)	18.88 ± 19.32	-
Stratification by initial diagnostic PSA (ng/mL)
<10	-	-	41 (41.8%)
10–20	-	-	29 (29.6%)
>20	-	-	28 (28.6%)
Pre- and Post-RT Serum PSA Kinetics (ng/mL)
Pre-RT Baseline PSA (Post-Neoadjuvant ADT) PSA (n = 97)	0.96 (0.00–27.82)	2.38 ± 3.77	-
Post-RT 1 week–1 month (n = 96)	0.18 (0.01–4.90)	0.53 ± 0.91	-
Post-RT 3–6 months (n = 97)	0.06 (0.00–3.74)	0.22 ± 0.54	-
Post-RT 1 year (n = 68)	0.04 (0.01–4.07)	0.22 ± 0.63	-
Post-RT 2 years (n = 20)	0.05 (0.01–4.36)	0.51 ± 1.16	-
Pathological Features
Gleason Score Sum (GS)
6 (3 + 3)	-	-	16 (16.3%)
7 (3 + 4/4 + 3)	-	-	59 (60.2%)
8 (4 + 4)	-	-	17 (17.3%)
9 (4 + 5/5 + 4)	-	-	6 (6.1%)
Perineural Invasion (PNI)
Present	-	-	41 (41.8%)
Absent	-	-	55 (56.1%)
Unknown	-	-	2 (2.0%)
Necrosis
Present	-	-	3 (3.1%)
Absent	-	-	85 (86.7%)
Unknown	-	-	10 (10.2%)
Imaging Findings
mpMRI PI-RADS Category (n = 75)
PI-RADS 0 †	-	-	6 (8.0%)
PI-RADS 1	-	-	0 (0%)
PI-RADS 2	-	-	3 (4.0%)
PI-RADS 3	-	-	6 (8.0%)
PI-RADS 4	-	-	19 (25.33%)
PI-RADS 5	-	-	41 (54.67%)
mpMRI Nodal Involvement (n = 75)
Positive (N1)	-	-	8 (10.7%)
Negative (N0)	-	-	67 (89.3%)
mpMRI Seminal Vesicle Invasion (n = 75)
Positive	-	-	15 (20.0%)
Negative	-	-	60 (80.0%)
68Ga-PSMA PET/CT Primary Prostatic Avidity (N = 98)
Present	-	-	92 (93.9%)
Prostate SUVmax (n = 92)	6.45 (2.20–84.30)	9.93 ± 10.92	-
Prostate SUVmax ≤ 10	-	-	65 (70.65%)
Prostate SUVmax > 10	-	-	27 (29.35%)
68Ga-PSMA PET/CT Nodal Involvement (N = 98)
Positive (N1)	-	-	23 (23.5%)
Negative (N0)	-	-	75 (76.5%)
Nodal SUVmax (n = 23)	4.1 (1.7–88.8)	11.3 ± 18.8	-
68Ga-PSMA PET/CT Metastatic Stage (M-Stage), n (%)
M0 (Non-metastatic)			93 (94.9%)
M1b (Oligometastatic bone disease)			5 (5.1%)
68Ga-PSMA PET/CT Seminal Vesicle Invasion (n = 73)
Positive	-	-	16 (21.9%)
Negative	-	-	57 (78.1%)

Table Footnotes: Values are expressed as median (minimum–maximum), mean ± standard deviation (SD), or count (n) and percentage (%). Abbreviations: PSA: Prostate-Specific Antigen; RT: Radiotherapy; PNI: Perineural Invasion; mpMRI: Multiparametric Magnetic Resonance Imaging; PI-RADS: Prostate Imaging Reporting and Data System; PSMA: Prostate-specific membrane antigen; PET: Positron Emission Tomography; SUV_max: Maximum Standardized Uptake Value; Min: Minimum; Max: Maximum; SD: Standard Deviation. † Note: PI-RADS category 0 represents an internal institutional designation for mpMRI scans with no visible focal prostatic lesion (completely normal prostatic parenchyma), rather than a standard PI-RADS v2.1 score. PI-RADS category 1 was retained in the classification table for completeness, though no focal lesions corresponding to PI-RADS 1 were identified in the cohort (n = 0).

The median age at diagnosis was 70 years (range: 47–84 years). Overall follow-up duration was calculated from the date of initial histological diagnosis to the last clinical contact, yielding a median follow-up of 16.0 months (range: 1–109 months). This extended upper range (up to 109 months) reflects patients with long-standing primary diagnoses who underwent initial active surveillance or prolonged neoadjuvant ADT prior to definitive RT. Post-radiotherapy clinical follow-up ranged specifically from 1 month to 24 months (1–2 years) across the cohort (data cut-off date: December 2025). Among the total cohort (N = 98), long-term 24-month post-RT PSA data were available for a maturing subset of 20 patients (n = 20), as patients treated in late 2023 and 2024 had not yet reached the 2-year post-treatment landmark at the time of analysis. The initial diagnostic median serum PSA level was 11.39 ng/mL (range: 0.04–99.0 ng/mL; mean ± SD: 18.88 ± 19.32 ng/mL). Stratification by initial diagnostic PSA levels revealed three distinct sub-groups comprising <10 ng/mL in 41.8% (n = 41), 10–20 ng/mL in 29.6% (n = 29), and >20 ng/mL in 28.6% (n = 28) of the patient cohort, respectively (Table 1). Following neoadjuvant ADT, the median Pre-RT baseline PSA level precipitously declined to 0.96 ng/mL (range: 0.00–27.82 ng/mL; mean ± SD: 2.38 ± 3.77 ng/mL).

On pathological evaluation, the majority of cases were categorized as Gleason Score (GS) 7 (60.2%, n = 59) and GS 8 (17.3%, n = 17). Perineural invasion (PNI) was documented in 41.8% (n = 41) of patients, whereas necrosis was identified in 3.1% (n = 3) of pathological specimens.

3.1.2. Response to Radiotherapy and Longitudinal PSA Kinetics

A significant and sustained biochemical response was observed in post-RT serum PSA levels. The median pre-RT PSA value was 0.96 ng/mL (range: 0.00–27.82), which precipitously declined to 0.18 ng/mL (range: 0.01–4.90) at 1 week–1 month post-RT, 0.06 ng/mL (range: 0.00–3.74) at 3–6 months, and reached a nadir-like level of 0.04 ng/mL (range: 0.01–4.07) at 1 year (Table 1).

3.1.3. Inter-Modality Discordance and Nodal Detection Rates

In the paired analysis cohort (n = 74), ⁶⁸Ga-PSMA PET/CT demonstrated a higher overall nodal detection rate compared to mpMRI (25.7% [19/74] vs. 10.8% [8/74], respectively). However, analysis of nodal discordance revealed a bidirectional pattern: while ⁶⁸Ga-PSMA PET/CT detected avid nodes in 14 patients who were node-negative on mpMRI, 37.5% (3/8) of nodes classified as suspicious on mpMRI were PSMA-negative. For local T-stage evaluation (particularly extraprostatic extension, ECE), agreement between modalities was poor. Rather than proving clinical understaging by mpMRI, this discordance highlights the distinct imaging mechanics of soft-tissue anatomical resolution (mpMRI) versus molecular tracer uptake (⁶⁸Ga-PSMA PET/CT).

3.1.4. General Findings of ⁶⁸Ga-PSMA PET/CT and Multiparametric MRI (mpMRI)

Pretreatment ⁶⁸Ga-PSMA PET/CT imaging was performed in all study participants (N = 98), while baseline mpMRI data were available for 76.5% (n = 75) of the cohort.

• mpMRI Findings (n = 75): Among patients evaluated with mpMRI (n = 75), focal prostatic lesions corresponding to standard PI-RADS categories 2 to 5 were identified in 92.0% (n = 69) of cases, with PI-RADS 5 being the most predominant (54.67%, n = 41). In six patients (8.0%), no visible focal lesion was identified and these were categorized as PI-RADS 0, while no patients presented with PI-RADS 1 lesions (0.0%, n = 0).
• ⁶⁸Ga-PSMA PET/CT Findings: Among 92 patients exhibiting primary prostatic PSMA avidity, the mean prostate SUVmax was 9.93 ± 10.92 (median: 6.45, range: 2.20–84.30), with 29.35% of the total cohort (n = 27) presenting with a high uptake profile of prostate SUVmax > 10. In 23 patients with detected pelvic lymph node involvement, the mean nodal SUV_max was 11.35 ± 18.83 (median: 4.10, range: 1.70–88.80).

3.1.5. Local Tumor Staging (T Stage) Comparison: mpMRI vs. ⁶⁸Ga-PSMA PET/CT

Data from 70 patients who completed local T-staging using both mpMRI and ⁶⁸Ga-PSMA PET/CT are presented in a cross-tabulation format in Table 2. In this cohort (spanning T0–T4 stages), mpMRI classified 24 patients (34.3%) as T2a, whereas ⁶⁸Ga-PSMA PET/CT identified 17 patients (24.3%) as T2a. However, complete concordance between the two modalities for stage T2a was limited to 9 patients (12.9%). Similarly, 4 patients (5.7%) were concordantly confirmed as T2b by both modalities.

Table 2. Cross-Tabulation of T-Stages Between mpMRI and ⁶⁸Ga-PSMA PET/CT (n = 70).

PSMA PET T-Clinical Stage	mpMRI T2a	mpMRI T2b	mpMRI T2c	mpMRI T3a	mpMRI T3b	mpMRI T4	Total n (%)
T0 †	2	0	2	0	0	0	4 (5.7%)
T2a	9 *	2	1	3	2	0	17 (24.3%)
T2b	6	4 *	2	1	1	0	14 (20.0%)
T2c	4	3	4 *	1	2	0	14 (20.0%)
T3a	1	1	2	1 *	1	1	7 (10.0%)
T3b	2	1	1	1	2 *	2	9 (12.9%)
T4	0	0	0	0	2	3 *	5 (7.1%)
Total (n)	24	11	12	7	10	6	70 (100%)

Table Footnotes: Bold cells marked with an asterisk (*) indicate exact concordance between mpMRI and ⁶⁸Ga-PSMA PET/CT T-stage classifications. Overall percent agreement (OPA) was 32.9% (23/70). Inter-modality agreement was slight (Cohen’s κ = 0.19; 95% CI, 0.04–0.34; p = 0.008). † Category T0 denotes cases in which no localized primary tumor was identified on the respective imaging modality. On ⁶⁸Ga-PSMA PET/CT, T0 represents scans without pathological primary prostatic uptake (n = 6). On mpMRI, non-localized or negative findings without a visible focal lesion were categorized using the institutional designation PI-RADS 0 (n = 6). Abbreviations: mpMRI, multiparametric magnetic resonance imaging; PSMA, prostate-specific membrane antigen; PET, positron emission tomography; CT, computed tomography; OPA, overall percent agreement; CI, confidence interval; PI-RADS, Prostate Imaging Reporting and Data System.

The overall percent agreement (OPA) between the two staging modalities for primary T-stage (n = 70) was 32.9% (23/70 patients), yielding slight inter-modality agreement (Cohen’s κ = 0.19, 95% CI: 0.04–0.34, p = 0.008). Among 47 patients classified with organ-confined disease (T2a–T2c) on mpMRI, 14.9% (n = 7/47; comprising 3 T2a, 2 T2b, and 2 T2c cases) were upstaged by ⁶⁸Ga-PSMA PET/CT to locally advanced disease (n = 4 T3a and n = 3 T3b). Furthermore, ⁶⁸Ga-PSMA PET/CT confirmed T4 organ invasion in 50.0% (n = 3) of the 6 patients classified as T4 on mpMRI, and in 20.0% (n = 2) of the 10 patients classified as T3b on mpMRI. All 5 patients identified with T4 disease on ⁶⁸Ga-PSMA PET/CT were consistently in the locally advanced group (T3b or T4) on mpMRI.

For pelvic nodal status (n = 74), ⁶⁸Ga-PSMA PET/CT detected N1 disease in 25.7% (19/74) compared to 10.8% (8/74) by mpMRI (p = 0.012, McNemar test), with fair overall concordance (Cohen’s κ = 0.26, 95% CI: 0.05–0.47, OPA = 77.0%). Finally, for seminal vesicle invasion (n = 73), ⁶⁸Ga-PSMA PET/CT and mpMRI demonstrated moderate diagnostic concordance (Cohen’s κ = 0.47, 95% CI: 0.24–0.70, p < 0.001, OPA = 82.2%).

3.1.6. Pelvic Lymph Node Involvement (N-Stage) Comparison: mpMRI vs. ⁶⁸Ga-PSMA PET/CT

Concordance between mpMRI and ⁶⁸Ga-PSMA PET/CT for detecting pelvic lymph node metastases (N-stage) was evaluated in 74 patients with available dual-imaging data (Table 3). Based on concordant negative and positive classifications, both modalities agreed on node-negative (N0) status in 52 patients and node-positive (N1) status in 5 patients, yielding an Overall Percent Agreement (OPA) of 77.0% (57/74).

Table 3. Comparison of Pelvic Lymph Node Involvement Between mpMRI and ⁶⁸Ga-PSMA PET/CT (n = 74).

68Ga-PSMA PET/CT LN Stage	mpMRI LN Negative, n (%)	mpMRI LN Positive, n (%)	Total, n (%)	p-Value	Cohen’s Kappa (κ)
68Ga-PSMA PET/CT LN Negative	52 (78.8%)	3 (37.5%)	55 (74.3%)	0.023	0.26
68Ga-PSMA PET/CT LN Positive	14 (21.2%)	5 (62.5%)	19 (25.7%)	-	(Fair Agreement)
Total	66 (100%)	8 (100%)	74 (100%)	-	-

Table Footnotes: Overall Percent Agreement (OPA) = 77.0% (57/74). Diagnostic agreement: Cohen’s $\mathsf{\kappa }=0.26$ (95% CI: 0.05–0.47, p = 0.023, Fair Agreement). Detection rate comparison: ⁶⁸Ga-PSMA PET/CT detected significantly more N1 cases than mpMRI (25.7% [19/74] vs. 10.8% [8/74], p = 0.012, McNemar test). Analysis restricted to n = 74 patients with complete paired dual-modality nodal staging data. Abbreviations: LN: Lymph Node; mpMRI: Multiparametric Magnetic Resonance Imaging; PSMA: Prostate-specific membrane antigen; PET: Positron Emission Tomography.

Inter-modality agreement analysis for pelvic nodal staging demonstrated fair concordance (Cohen’s κ = 0.26, 95% CI: 0.05–0.47, p = 0.023). However, ⁶⁸Ga-PSMA PET/CT demonstrated a significantly higher nodal detection rate compared to mpMRI (25.7% [19/74] vs. 10.8% [8/74], p = 0.012 via McNemar test for paired binary data).

Importantly, among the 66 patients classified as pelvic lymph node-negative (N0) on mpMRI, ⁶⁸Ga-PSMA PET/CT identified avid nodal metastases (N1) in 14 cases (21.2%), leading to nodal upstaging. Conversely, out of 8 patients reported as node-positive (N1) on mpMRI, 3 (37.5%) showed no corresponding PSMA avidity on ⁶⁸Ga-PSMA PET/CT.

3.1.7. Seminal Vesicle Invasion (SV/T3b) Comparison: mpMRI vs. ⁶⁸Ga-PSMA PET/CT

Concordance between mpMRI and ⁶⁸Ga-PSMA PET/CT for detecting seminal vesicle invasion (SVI) was evaluated in 73 patients with complete dual-modality data (Table 4). Both imaging modalities demonstrated concordant negative findings in 51 patients and concordant positive findings in 9 patients, yielding an Overall Percent Agreement (OPA) of 82.2% (60/73).

Table 4. Comparison of Seminal Vesicle Invasion Between mpMRI and ⁶⁸Ga-PSMA PET/CT (n = 73).

68Ga-PSMA PET/CT SVI Status	mpMRI SVI Positive, n (%)	mpMRI SVI Negative, n (%)	Total, n (%)
Positive	9 (60.0%)	7 (12.1%)	16 (21.9%)
Negative	6 (40.0%)	51 (87.9%)	57 (78.1%)
Total	15 (100%)	58 (100%)	73 (100%)

Table Footnotes: Overall Percent Agreement (OPA) = 82.2% (60/73). Statistical Analysis: Cohen’s $\mathsf{\kappa }=0.47$ (Moderate agreement, p < 0.001). Abbreviations: SVI: Seminal Vesicle Invasion; mpMRI: Multiparametric Magnetic Resonance Imaging; PSMA: Prostate-specific membrane antigen; PET: Positron Emission Tomography.

Statistical analysis of inter-modality agreement yielded a Cohen’s Kappa coefficient of κ = 0.47 (p < 0.001), indicating moderate and statistically significant concordance between mpMRI and ⁶⁸Ga-PSMA PET/CT for detecting SVI.

Among the 58 patients reported as negative for SVI on mpMRI, 7 (12.1%) were found to have intense focal PSMA avidity in the seminal vesicles on ⁶⁸Ga-PSMA PET/CT, resulting in T3b upstaging. Conversely, out of 15 patients classified as positive for SVI on mpMRI, 6 (40.0%) demonstrated no corresponding tracer uptake on ⁶⁸Ga-PSMA PET/CT.

3.2. Dosimetric Analyses and Organs at Risk Comparisons

3.2.1. Target Volume and Organs at Risk Dosimetric Parameters

When analyzing the overall dosimetric parameters across the entire cohort (N = 98), the median mean dose to the CTV_primary was 100.9% (range: 100.2–101.2%) of the prescribed dose. Regarding OAR dosimetry, the median mean doses were 32.9% (range: 27.7–38.4%) for the rectum and 37.1% (range: 27.9–46.9%) for the urinary bladder. Evaluation of high-dose rectal volumes demonstrated a median V_65Gy of 5.2% (range: 3.9–7.1%) and a median V_50Gy of 11.8% (range: 9.9–15.0%) (Table 5).

Table 5. Cohort-Wide Target Volume and Organs at Risk DVH Metrics (N = 98).

Organ/Dosimetric Parameter	Median (Cohort Min–Max) (Range Across Patients) *
${\mathrm{CTV}}_{\mathrm{primary}}$Dosimetry
% Min Dose	96.7 (95.0–97.7)
% Max Dose	104.8 (103.8–105.9)
% Mean Dose	100.9 (100.2–101.2)
Rectal Dosimetry
% Min Dose	3.0 (2.1–4.4)
% Max Dose	103.5 (102.6–104.6)
% Mean Dose	32.9 (27.7–38.4)
$V_{65\mathit{Gy}}$ (%)	5.2 (3.9–7.1)
$V_{50\mathit{Gy}}$ (%)	11.8 (9.9–15.0)
Urinary Bladder Dosimetry
% Min Dose	3.7 (2.1–11.0)
% Max Dose	103.7 (102.5–105.5)
% Mean Dose	37.1 (27.9–46.9)
Femoral Head Dosimetry
% Min Dose	1.2 (0.9–2.9)
% Max Dose	54.8 (48.4–61.6)
% Mean Dose	21.4 (17.3–26.2)
Penile Bulb Dosimetry
% Mean Dose	34.3 (25.1–44.9)
Body Dosimetry
% Max Dose	105.8 (104.5–106.8)

Table Footnotes: Values are presented as median (minimum-maximum). Percentage (%) dose values are normalized relative to the prescribed target dose. Abbreviations: Min: Minimum; Max: Maximum; Mean: Mean Dose; V_65Gy: Volume receiving at least 65 Gy; V_50Gy: Volume receiving at least 50 Gy. * Values represent the cohort Median and the [Minimum–Maximum] range of that specific single DVH metric measured across all N = 98 patients. Percentage (%) dose values are normalized to the prescribed target dose. (Note: The range in brackets reflects how a single parameter, such as CTV_primary % Mean Dose, varies from the lowest to the highest patient value across the population, whereas Table 6, Table 7 and Table 8 compare distinct metrics like Minimum, Maximum, and Mean doses across subgroups).

3.2.2. Dosimetric Variations According to CTV_primary

No statistically significant differences were observed among CTV subgroups (25.4–67.6 cm³, 67.6–130.5 cm³, and >130.5 cm³) regarding target coverage parameters (minimum, maximum, and mean target doses, p > 0.05). However, progressive enlargement of the CTV was significantly associated with elevated maximum femoral dose (p = 0.009), mean femoral dose (p = 0.004), and maximum body dose (p = 0.032) (Table 6).

Table 6. Comparison of Dosimetric Parameters According to CTV_primary Volume Subgroups.

% Dose	CTVprimaryVolume 25.4–67.6 cm3	CTVprimaryVolume 67.6–130.5 cm3	CTVprimaryVolume >130.5 cm3	p-Value
CTVprimary
Min	95.19 ± 5.15; 96.85 (71.6–99.0)	94.65 ± 6.10; 96.60 (64.1–98.7)	94.32 ± 5.01; 95.70 (85.8–99.1)	0.467
Max	104.28 ± 3.49; 104.55 (81.5–108.7)	104.37 ± 3.66; 104.80 (83.3–108.2)	106.40 ± 1.59; 106.30 (104.4–108.8)	0.359
Mean	100.28 ± 3.32; 100.85 (77.7–103.0)	100.34 ± 3.61; 101.10 (78.9–105.5)	101.48 ± 0.98; 100.90 (100.7–102.9)	0.323
Rectum
Min	3.16 ± 1.52; 2.60 (1.0–7.3)	3.40 ± 1.58; 3.10 (1.0–9.8)	4.42 ± 2.20; 3.50 (2.7–8.2)	0.371
Max	103.66 ± 1.42; 103.65 (99.6–106.8)	103.31 ± 1.55; 103.40 (98.9–106.8)	105.34 ± 1.52; 105.10 (103.1–106.9)	0.318
Mean	32.00 ± 7.41; 29.60 (19.0–47.8)	34.43 ± 7.29; 33.50 (20.6–54.0)	40.42 ± 6.08; 39.20 (32.8–49.4)	0.089
V65Gy	5.53 ± 2.88; 5.05 (2.0–20.6)	6.30 ± 4.06; 5.10 (1.2–21.7)	9.96 ± 3.76; 8.80 (5.6–15.3)	0.634
V50Gy	12.15 ± 4.09; 11.10 (4.4–28.9)	13.20 ± 5.09; 12.20 (6.9–29.2)	19.24 ± 4.04; 19.10 (14.1–24.9)	0.380
Urinary Bladder
Min	5.96 ± 5.41; 3.60 (0.7–18.5)	7.81 ± 8.11; 3.80 (1.0–35.5)	8.46 ± 5.55; 8.60 (2.4–15.4)	0.279
Max	103.64 ± 1.90; 103.50 (99.7–108.0)	104.26 ± 2.31; 103.80 (100.1–108.5)	105.78 ± 2.33; 106.00 (102.2–108.1)	0.149
Mean	34.68 ± 11.55; 36.50 (11.6–55.8)	38.78 ± 12.55; 36.70 (16.2–68.8)	44.64 ± 5.17; 42.30 (40.1–50.9)	0.186
Femoral Heads
Min	2.46 ± 3.34; 1.20 (0.0–20.6)	1.74 ± 1.25; 1.12 (0.1–3.60)	1.82 ± 1.28; 1.90 (0.1–3.6)	0.780
Max	51.88 ± 10.71; 52.60 (33.0–72.7)	57.13 ± 10.95; 57.60 (16.0–77.7)	70.84 ± 8.95; 70.80 (58.1–82.3)	0.009
Mean	20.15 ± 5.49; 20.40 (7.8–35.4)	23.84 ± 7.24; 23.10 (1.4–41.0)	28.60 ± 7.02; 30.00 (16.9–34.9)	0.004
Penile Bulb
Mean	35.54 ± 12.42; 34.20 (13.7–64.6)	35.94 ± 14.81; 34.50 (14.4–68.9)	32.42 ± 17.37; 34.30 (12.5–53.1)	0.985
Body
Max	105.37 ± 1.40; 105.15 (103.2–108.0)	105.99 ± 1.44; 106.20 (101.7–108.5)	106.94 ± 1.89; 106.90 (104.4–109.6)	0.032

Table Footnote: Values are presented as Mean ± Standard Deviation; Median (Minimum–Maximum). Target Volume Definitions: CTV_primary represents the primary Clinical Target Volume, encompassing the prostate gland ± proximal or entire seminal vesicles according to NCCN risk stratification categories, strictly excluding elective pelvic lymph node basins. CTV_primary volume subgroup brackets are defined as: 25.4–67.6 cm³, 67.6–130.5 cm³, and >130.5 cm³. Dosimetric Parameters: Dose values for target volumes and organs at risk (OARs) are reported as the percentage of the prescribed target dose (% Dose) received by the respective structure. V_65Gy and V_50Gy denote the percentage of rectal volume receiving absolute physical doses of at least 65 Gy and 50 Gy, respectively.

Table 6. Comparison of Dosimetric Parameters According to CTV_primary Volume Subgroups.

% Dose	CTVprimaryVolume 25.4–67.6 cm3	CTVprimaryVolume 67.6–130.5 cm3	CTVprimaryVolume >130.5 cm3	p-Value
CTVprimary
Min	95.19 ± 5.15; 96.85 (71.6–99.0)	94.65 ± 6.10; 96.60 (64.1–98.7)	94.32 ± 5.01; 95.70 (85.8–99.1)	0.467
Max	104.28 ± 3.49; 104.55 (81.5–108.7)	104.37 ± 3.66; 104.80 (83.3–108.2)	106.40 ± 1.59; 106.30 (104.4–108.8)	0.359
Mean	100.28 ± 3.32; 100.85 (77.7–103.0)	100.34 ± 3.61; 101.10 (78.9–105.5)	101.48 ± 0.98; 100.90 (100.7–102.9)	0.323
Rectum
Min	3.16 ± 1.52; 2.60 (1.0–7.3)	3.40 ± 1.58; 3.10 (1.0–9.8)	4.42 ± 2.20; 3.50 (2.7–8.2)	0.371
Max	103.66 ± 1.42; 103.65 (99.6–106.8)	103.31 ± 1.55; 103.40 (98.9–106.8)	105.34 ± 1.52; 105.10 (103.1–106.9)	0.318
Mean	32.00 ± 7.41; 29.60 (19.0–47.8)	34.43 ± 7.29; 33.50 (20.6–54.0)	40.42 ± 6.08; 39.20 (32.8–49.4)	0.089
V65Gy	5.53 ± 2.88; 5.05 (2.0–20.6)	6.30 ± 4.06; 5.10 (1.2–21.7)	9.96 ± 3.76; 8.80 (5.6–15.3)	0.634
V50Gy	12.15 ± 4.09; 11.10 (4.4–28.9)	13.20 ± 5.09; 12.20 (6.9–29.2)	19.24 ± 4.04; 19.10 (14.1–24.9)	0.380
Urinary Bladder
Min	5.96 ± 5.41; 3.60 (0.7–18.5)	7.81 ± 8.11; 3.80 (1.0–35.5)	8.46 ± 5.55; 8.60 (2.4–15.4)	0.279
Max	103.64 ± 1.90; 103.50 (99.7–108.0)	104.26 ± 2.31; 103.80 (100.1–108.5)	105.78 ± 2.33; 106.00 (102.2–108.1)	0.149
Mean	34.68 ± 11.55; 36.50 (11.6–55.8)	38.78 ± 12.55; 36.70 (16.2–68.8)	44.64 ± 5.17; 42.30 (40.1–50.9)	0.186
Femoral Heads
Min	2.46 ± 3.34; 1.20 (0.0–20.6)	1.74 ± 1.25; 1.12 (0.1–3.60)	1.82 ± 1.28; 1.90 (0.1–3.6)	0.780
Max	51.88 ± 10.71; 52.60 (33.0–72.7)	57.13 ± 10.95; 57.60 (16.0–77.7)	70.84 ± 8.95; 70.80 (58.1–82.3)	0.009
Mean	20.15 ± 5.49; 20.40 (7.8–35.4)	23.84 ± 7.24; 23.10 (1.4–41.0)	28.60 ± 7.02; 30.00 (16.9–34.9)	0.004
Penile Bulb
Mean	35.54 ± 12.42; 34.20 (13.7–64.6)	35.94 ± 14.81; 34.50 (14.4–68.9)	32.42 ± 17.37; 34.30 (12.5–53.1)	0.985
Body
Max	105.37 ± 1.40; 105.15 (103.2–108.0)	105.99 ± 1.44; 106.20 (101.7–108.5)	106.94 ± 1.89; 106.90 (104.4–109.6)	0.032

Table Footnote: Values are presented as Mean ± Standard Deviation; Median (Minimum–Maximum). Target Volume Definitions: CTV_primary represents the primary Clinical Target Volume, encompassing the prostate gland ± proximal or entire seminal vesicles according to NCCN risk stratification categories, strictly excluding elective pelvic lymph node basins. CTV_primary volume subgroup brackets are defined as: 25.4–67.6 cm³, 67.6–130.5 cm³, and >130.5 cm³. Dosimetric Parameters: Dose values for target volumes and organs at risk (OARs) are reported as the percentage of the prescribed target dose (% Dose) received by the respective structure. V_65Gy and V_50Gy denote the percentage of rectal volume receiving absolute physical doses of at least 65 Gy and 50 Gy, respectively.

3.2.3. Dosimetric Impact of Focal Boost Delivery

Comparative analysis between patients who received a boost (n = 42) and those who did not (n = 56) demonstrated that boost application significantly improved target dose homogeneity while sparing OARs. In the boost group, CTV_primary Min (93.9 ± 6.0% vs. 95.7 ± 5.0%; p < 0.001), CTV_primary Max (103.6 ± 3.4% vs. 105.1 ± 3.5%; p < 0.001), and CTV_primary Mean (99.9 ± 3.5% vs. 100.8 ± 3.2%; p < 0.001) doses were significantly lower, reflecting a more homogeneous target dose distribution.

Furthermore, boost delivery resulted in a significant reduction in maximum OAR doses. Rectum Max (103.0 ± 1.4% vs. 104.1 ± 1.5%; p = 0.001), Bladder Max (102.9 ± 2.0% vs. 104.9 ± 1.8%; p < 0.001), and Body Max (104.8 ± 1.4% vs. 106.4 ± 1.2%; p < 0.001) doses were significantly higher in the non-boost group. No statistically significant differences were observed between the two groups regarding minimum and mean OAR doses, as well as volumetric parameters (V₆₅, V₅₀, Femoral, and Penile Bulb doses) (p > 0.05) (Table 7).

Table 7. Dosimetric Comparison According to Boost Delivery Status.

Organ/% Dose	Boost Yes (n = 42)	Boost No (n = 56)	p-Value
CTVprimary
Min	93.9 ± 6.0; 95.4 (64.1–98.2)	95.7 ± 5.0; 97.4 (71.6–99.1)	<0.001
Max	103.6 ± 3.4; 104.0 (83.3–106.8)	105.1 ± 3.5; 105.3 (81.5–108.8)	<0.001
Mean	99.9 ± 3.5; 100.3 (78.9–105.5)	100.8 ± 3.2; 101.1 (77.7–103.0)	<0.001
Rectum
Min	3.1 ± 1.4; 2.6 (1.0–6.8)	3.5 ± 1.7; 3.3 (1.0–9.8)	0.318
Max	103.0 ± 1.4; 103.2 (98.9–105.4)	104.1 ± 1.5; 103.9 (101.4–106.9)	0.001
Mean	32.5 ± 8.0; 30.3 (20.6–51.4)	34.2 ± 7.1; 33.9 (19.0–54.0)	0.173
V65Gy	5.8 ± 2.5; 5.5 (2.1–16.5)	6.3 ± 4.2; 5.0 (1.2–21.7)	0.606
V50Gy	12.5 ± 4.1; 11.1 (6.2–26.0)	13.3 ± 5.2; 12.2 (4.4–29.2)	0.459
Urinary Bladder
Min	6.9 ± 6.9; 3.7 (1.1–35.5)	6.9 ± 6.6; 3.8 (0.7–23.5)	0.960
Max	102.9 ± 2.0; 102.5 (99.7–107.9)	104.9 ± 1.8; 104.9 (101.2–108.5)	<0.001
Mean	36.2 ± 11.5; 36.3 (13.4–68.8)	37.5 ± 12.4; 39.4 (11.6–61.3)	0.506
Femoral Heads
Min	2.7 ± 4.3; 1.1 (0.0–26.2)	5.5 ± 15.0; 1.2 (0.1–105.0)	0.819
Max	54.9 ± 11.1; 53.0 (34.7–74.3)	55.1 ± 12.0; 55.9 (16.0–82.3)	0.652
Mean	20.9 ± 6.3; 20.7 (1.4–36.9)	23.1 ± 6.9; 22.7 (7.8–41.0)	0.122
Penile Bulb
Mean	35.6 ± 11.9; 34.3 (14.4–60.5)	35.5 ± 14.9; 35.2 (12.5–68.9)	0.960
Body
Max	104.8 ± 1.4; 104.7 (101.7–107.9)	106.4 ± 1.2; 106.5 (103.2–109.6)	<0.001

Table Footnotes: Values are presented as Mean ± Standard Deviation; Median (Minimum–Maximum). Percentage (%) dose values are normalized relative to the prescribed target dose. Abbreviations: Min: Minimum Dose; Max: Maximum Dose; Mean: Mean Dose; V_65Gy: Volume receiving at least 65 Gy; V_50Gy: Volume receiving at least 50 Gy. Statistically significant p-values are highlighted in bold.

Table 7. Dosimetric Comparison According to Boost Delivery Status.

Organ/% Dose	Boost Yes (n = 42)	Boost No (n = 56)	p-Value
CTVprimary
Min	93.9 ± 6.0; 95.4 (64.1–98.2)	95.7 ± 5.0; 97.4 (71.6–99.1)	<0.001
Max	103.6 ± 3.4; 104.0 (83.3–106.8)	105.1 ± 3.5; 105.3 (81.5–108.8)	<0.001
Mean	99.9 ± 3.5; 100.3 (78.9–105.5)	100.8 ± 3.2; 101.1 (77.7–103.0)	<0.001
Rectum
Min	3.1 ± 1.4; 2.6 (1.0–6.8)	3.5 ± 1.7; 3.3 (1.0–9.8)	0.318
Max	103.0 ± 1.4; 103.2 (98.9–105.4)	104.1 ± 1.5; 103.9 (101.4–106.9)	0.001
Mean	32.5 ± 8.0; 30.3 (20.6–51.4)	34.2 ± 7.1; 33.9 (19.0–54.0)	0.173
V65Gy	5.8 ± 2.5; 5.5 (2.1–16.5)	6.3 ± 4.2; 5.0 (1.2–21.7)	0.606
V50Gy	12.5 ± 4.1; 11.1 (6.2–26.0)	13.3 ± 5.2; 12.2 (4.4–29.2)	0.459
Urinary Bladder
Min	6.9 ± 6.9; 3.7 (1.1–35.5)	6.9 ± 6.6; 3.8 (0.7–23.5)	0.960
Max	102.9 ± 2.0; 102.5 (99.7–107.9)	104.9 ± 1.8; 104.9 (101.2–108.5)	<0.001
Mean	36.2 ± 11.5; 36.3 (13.4–68.8)	37.5 ± 12.4; 39.4 (11.6–61.3)	0.506
Femoral Heads
Min	2.7 ± 4.3; 1.1 (0.0–26.2)	5.5 ± 15.0; 1.2 (0.1–105.0)	0.819
Max	54.9 ± 11.1; 53.0 (34.7–74.3)	55.1 ± 12.0; 55.9 (16.0–82.3)	0.652
Mean	20.9 ± 6.3; 20.7 (1.4–36.9)	23.1 ± 6.9; 22.7 (7.8–41.0)	0.122
Penile Bulb
Mean	35.6 ± 11.9; 34.3 (14.4–60.5)	35.5 ± 14.9; 35.2 (12.5–68.9)	0.960
Body
Max	104.8 ± 1.4; 104.7 (101.7–107.9)	106.4 ± 1.2; 106.5 (103.2–109.6)	<0.001

Table Footnotes: Values are presented as Mean ± Standard Deviation; Median (Minimum–Maximum). Percentage (%) dose values are normalized relative to the prescribed target dose. Abbreviations: Min: Minimum Dose; Max: Maximum Dose; Mean: Mean Dose; V_65Gy: Volume receiving at least 65 Gy; V_50Gy: Volume receiving at least 50 Gy. Statistically significant p-values are highlighted in bold.

3.2.4. Dosimetric Performance Across Fractionation Regimens

Comparison between conventional RT (n = 52; delivering 76–80 Gy in 38–40 fractions, prescription BED_1.5 range: 177.3–186.7 Gy) and hypofractionated RT (n = 46; delivering 70–72.5 Gy in 28–29 fractions, prescription BED_1.5 range: 186.7–193.3 Gy) plans demonstrated significantly higher absolute CTV_primary doses in the conventional arm for Min (73.89 ± 5.63 vs. 68.22 ± 1.88 Gy; p < 0.001), Max (81.92 ± 3.65 vs. 74.40 ± 1.20 Gy; p < 0.001), and Mean dose (78.68 ± 3.65 vs. 71.57 ± 1.28 Gy; p < 0.001). This difference primarily reflected the higher total prescription dose in the conventional group; when doses were normalized to the prescription dose, target coverage parameters did not differ significantly (Table 8).

Hypofractionated RT provided significantly lower absolute OAR exposure. Rectum Min (1.86 ± 0.76 vs. 3.10 ± 1.36 Gy; p < 0.001), Max (73.72 ± 1.22 vs. 81.37 ± 1.64 Gy; p < 0.001), Mean dose (21.57 ± 3.58 vs. 28.46 ± 6.55 Gy; p < 0.001), Mean EQD2 (23.72 ± 3.94 vs. 28.60 ± 6.57 Gy; p < 0.001), Mean BED₃ (39.54 ± 6.56 vs. 47.67 ± 10.96 Gy; p < 0.001), V65 (5.25 ± 3.85% vs. 6.81 ± 3.15%; p < 0.001), and V50 (11.32 ± 3.65% vs. 14.39 ± 5.17%; p < 0.001) were all significantly reduced in the hypofractionated arm.

Similarly, Bladder Min (2.92 ± 2.50 vs. 7.32 ± 6.18 Gy; p < 0.001), Max (74.28 ± 1.69 vs. 81.41 ± 1.46 Gy; p < 0.001), Mean dose (24.17 ± 8.63 vs. 31.02 ± 8.84 Gy; p < 0.001), Mean EQD2 (26.58 ± 9.49 vs. 31.17 ± 8.88 Gy; p = 0.015), and Mean BED₃ (44.31 ± 15.82 vs. 51.96 ± 14.80 Gy; p = 0.015) were significantly lower with hypofractionation. Femur Max (35.12 ± 7.46 vs. 47.20 ± 7.99 Gy; p < 0.001), Femur Mean (14.06 ± 4.27 vs. 19.02 ± 5.22 Gy; p < 0.001), and Penile Bulb Mean dose (24.11 ± 8.30 vs. 29.11 ± 11.81 Gy; p = 0.032) were also significantly reduced. Conversely, Body Max was significantly higher in the conventional arm (82.90 ± 1.24 vs. 75.34 ± 1.17 Gy; p < 0.001), consistent with the higher prescription dose (Table 8).

Table 8. Dosimetric Comparison According to Fractionation Regimen.

Organ/Parameter	Conventional RT (n = 52)	Hypofractionated RT (n = 46)	p Value
Prescription BED1.5 (Gy)	177.3–186.7 Gy (76 Gy/38 fr: 177.3; 78 Gy/39 fr: 182.0; 80 Gy/40 fr: 186.7)	186.7–193.3 Gy (70 Gy/28 fr: 186.7; 72.5 Gy/29 fr: 193.3)
CTVprimary doses (Gy)
Min (Gy)	73.89 ± 5.63; 75.97 (51.28–78.56)	68.22 ± 1.88; 68.42 (60.97–71.12)	<0.001
Max (Gy)	81.92 ± 3.65; 82.60 (63.57–85.44)	74.40 ± 1.20; 74.54 (72.66–77.00)	<0.001
Mean (Gy)	78.68 ± 3.65; 79.64 (60.61–81.76)	71.57 ± 1.28; 70.94 (70.14–76.49)	<0.001
Rectum
Min (Gy)	3.10 ± 1.36; 2.89 (0.91–7.64)	1.86 ± 0.76; 1.78 (0.70–3.57)	<0.001
Max (Gy)	81.37 ± 1.64; 81.30 (77.14–84.32)	73.72 ± 1.22; 73.78 (71.33–75.83)	<0.001
Mean (Gy)	28.46 ± 6.55; 28.55 (14.44–42.12)	21.57 ± 3.58; 20.93 (14.94–29.22)	<0.001
Mean EQD2 (Gy)	28.60 ± 6.57; 28.55 (14.44–42.12)	23.72 ± 3.94; 23.02 (16.43–32.14)	<0.001
Mean BED3 (Gy)	47.67 ± 10.96; 47.58 (24.07–70.20)	39.54 ± 6.56; 38.37 (27.38–53.57)	<0.001
V65 (%)	6.81 ± 3.15; 5.95 (2.00–16.50)	5.25 ± 3.85; 4.75 (1.20–21.70)	<0.001
V50 (%)	14.39 ± 5.17; 13.65 (4.40–29.20)	11.32 ± 3.65; 10.75 (6.20–28.90)	<0.001
Bladder
Min (Gy)	7.32 ± 6.18; 5.08 (0.76–28.40)	2.92 ± 2.50; 2.17 (0.49–12.32)	<0.001
Max (Gy)	81.41 ± 1.46; 81.59 (77.90–85.60)	74.28 ± 1.69; 74.22 (70.84–78.23)	<0.001
Mean dose (Gy)	31.02 ± 8.84; 31.64 (10.64–55.04)	24.17 ± 8.63; 22.92 (8.12–42.14)	<0.001
Mean EQD2 (Gy)	31.17 ± 8.88; 31.81 (10.64–55.60)	26.58 ± 9.49; 25.22 (8.93–46.35)	0.015
Mean BED3 (Gy)	51.96 ± 14.80; 53.01 (17.73–92.67)	44.31 ± 15.82; 42.03 (14.89–77.26)	0.015
Femoral heads
Min (Gy)	2.85 ± 4.82; 0.97 (0.00–22.00)	3.51 ± 11.02; 0.83 (0.21–73.50)	0.440
Max (Gy)	47.20 ± 7.99; 46.57 (25.08–64.19)	35.12 ± 7.46; 36.18 (11.20–50.89)	<0.001
Mean (Gy)	19.02 ± 5.22; 18.16 (8.89–31.98)	14.06 ± 4.27; 14.57 (1.01–26.75)	<0.001
Penile bulb
Mean (Gy)	29.11 ± 11.81; 27.88 (9.50–52.36)	24.11 ± 8.30; 24.18 (10.15–45.22)	0.032
Body
Max (Gy)	82.90 ± 1.24; 82.90 (80.10–85.60)	75.34 ± 1.17; 75.08 (72.87–78.23)	<0.001

Table Footnote: Values are presented as mean ± standard deviation; median (minimum–maximum). Radiobiological Modeling: Equivalent dose in 2 Gy fractions (EQD2) and biologically effective dose (BED) for the rectum and urinary bladder were calculated using $\mathsf{\alpha }/\mathsf{\beta }=3\mathrm{Gy}$; target volume prescription BED was calculated using $\mathsf{\alpha }/\mathsf{\beta }=1.5\mathrm{Gy}$ for prostate adenocarcinoma. Prescription BED_1.5 was calculated using the linear-quadratic model: $\mathrm{BED}=\mathrm{D}\left(1+{\displaystyle \frac{\mathrm{d}}{\mathsf{\alpha }/\mathsf{\beta }}}\right)$, where D is the total prescription dose, d is the dose per fraction, and $\mathsf{\alpha }/\mathsf{\beta }=1.5\mathrm{Gy}$. Calculated prescription BED_1.5 ranged from 177.3 to 186.7 Gy in the conventional arm (76–80 Gy in 2.0 Gy/fraction) and from 186.7 to 193.3 Gy in the hypofractionated arm (70–72.5 Gy in 2.5 Gy/fraction).

Table 8. Dosimetric Comparison According to Fractionation Regimen.

Organ/Parameter	Conventional RT (n = 52)	Hypofractionated RT (n = 46)	p Value
Prescription BED1.5 (Gy)	177.3–186.7 Gy (76 Gy/38 fr: 177.3; 78 Gy/39 fr: 182.0; 80 Gy/40 fr: 186.7)	186.7–193.3 Gy (70 Gy/28 fr: 186.7; 72.5 Gy/29 fr: 193.3)
CTVprimary doses (Gy)
Min (Gy)	73.89 ± 5.63; 75.97 (51.28–78.56)	68.22 ± 1.88; 68.42 (60.97–71.12)	<0.001
Max (Gy)	81.92 ± 3.65; 82.60 (63.57–85.44)	74.40 ± 1.20; 74.54 (72.66–77.00)	<0.001
Mean (Gy)	78.68 ± 3.65; 79.64 (60.61–81.76)	71.57 ± 1.28; 70.94 (70.14–76.49)	<0.001
Rectum
Min (Gy)	3.10 ± 1.36; 2.89 (0.91–7.64)	1.86 ± 0.76; 1.78 (0.70–3.57)	<0.001
Max (Gy)	81.37 ± 1.64; 81.30 (77.14–84.32)	73.72 ± 1.22; 73.78 (71.33–75.83)	<0.001
Mean (Gy)	28.46 ± 6.55; 28.55 (14.44–42.12)	21.57 ± 3.58; 20.93 (14.94–29.22)	<0.001
Mean EQD2 (Gy)	28.60 ± 6.57; 28.55 (14.44–42.12)	23.72 ± 3.94; 23.02 (16.43–32.14)	<0.001
Mean BED3 (Gy)	47.67 ± 10.96; 47.58 (24.07–70.20)	39.54 ± 6.56; 38.37 (27.38–53.57)	<0.001
V65 (%)	6.81 ± 3.15; 5.95 (2.00–16.50)	5.25 ± 3.85; 4.75 (1.20–21.70)	<0.001
V50 (%)	14.39 ± 5.17; 13.65 (4.40–29.20)	11.32 ± 3.65; 10.75 (6.20–28.90)	<0.001
Bladder
Min (Gy)	7.32 ± 6.18; 5.08 (0.76–28.40)	2.92 ± 2.50; 2.17 (0.49–12.32)	<0.001
Max (Gy)	81.41 ± 1.46; 81.59 (77.90–85.60)	74.28 ± 1.69; 74.22 (70.84–78.23)	<0.001
Mean dose (Gy)	31.02 ± 8.84; 31.64 (10.64–55.04)	24.17 ± 8.63; 22.92 (8.12–42.14)	<0.001
Mean EQD2 (Gy)	31.17 ± 8.88; 31.81 (10.64–55.60)	26.58 ± 9.49; 25.22 (8.93–46.35)	0.015
Mean BED3 (Gy)	51.96 ± 14.80; 53.01 (17.73–92.67)	44.31 ± 15.82; 42.03 (14.89–77.26)	0.015
Femoral heads
Min (Gy)	2.85 ± 4.82; 0.97 (0.00–22.00)	3.51 ± 11.02; 0.83 (0.21–73.50)	0.440
Max (Gy)	47.20 ± 7.99; 46.57 (25.08–64.19)	35.12 ± 7.46; 36.18 (11.20–50.89)	<0.001
Mean (Gy)	19.02 ± 5.22; 18.16 (8.89–31.98)	14.06 ± 4.27; 14.57 (1.01–26.75)	<0.001
Penile bulb
Mean (Gy)	29.11 ± 11.81; 27.88 (9.50–52.36)	24.11 ± 8.30; 24.18 (10.15–45.22)	0.032
Body
Max (Gy)	82.90 ± 1.24; 82.90 (80.10–85.60)	75.34 ± 1.17; 75.08 (72.87–78.23)	<0.001

Table Footnote: Values are presented as mean ± standard deviation; median (minimum–maximum). Radiobiological Modeling: Equivalent dose in 2 Gy fractions (EQD2) and biologically effective dose (BED) for the rectum and urinary bladder were calculated using $\mathsf{\alpha }/\mathsf{\beta }=3\mathrm{Gy}$; target volume prescription BED was calculated using $\mathsf{\alpha }/\mathsf{\beta }=1.5\mathrm{Gy}$ for prostate adenocarcinoma. Prescription BED_1.5 was calculated using the linear-quadratic model: $\mathrm{BED}=\mathrm{D}\left(1+{\displaystyle \frac{\mathrm{d}}{\mathsf{\alpha }/\mathsf{\beta }}}\right)$, where D is the total prescription dose, d is the dose per fraction, and $\mathsf{\alpha }/\mathsf{\beta }=1.5\mathrm{Gy}$. Calculated prescription BED_1.5 ranged from 177.3 to 186.7 Gy in the conventional arm (76–80 Gy in 2.0 Gy/fraction) and from 186.7 to 193.3 Gy in the hypofractionated arm (70–72.5 Gy in 2.5 Gy/fraction).

Dosimetric Parameters & Statistical Analysis: V_65Gy and V_50Gy represent the percentage of rectal volume receiving absolute physical doses of at least 65 Gy and 50 Gy, respectively, and are expressed as percentage volumes rather than physical doses. Inter-group comparisons between fractionation regimens were performed using the Mann–Whitney U test.

3.2.5. Dose Distributions According to Treatment Target Volumes

Comparison among the six RT target-volume protocols demonstrated significant differences in absolute CTV_primary coverage parameters, including Min (p < 0.001), Max (p < 0.001), and Mean dose (p < 0.001), largely reflecting the different prescription levels across protocols (Table 9).

Table 9. Dosimetric Comparison Across Radiotherapy Target-Volume Regimens.

Parameter	Field 1 (n = 30)	Field 2 (n = 16)	Field 3 (n = 16)	Field 4 (n = 10)	Field 5 (n = 18)	Field 6 (n = 8)	p Value
Prescription dose/fractionation	70 Gy/28 fr	72.5 Gy/29 fr	78 Gy/39 fr	78 Gy/39 fr	80 Gy/40 fr	80 Gy/40 fr
Prescription BED1.5 (Gy)	186.7	193.3	182.0	182.0	186.7	186.7
CTVprimary
Min (Gy)	67.78 ± 1.51; 68.18 (61.88–69.09)	68.90 ± 2.20; 68.98 (60.97–71.12)	71.71 ± 6.40; 74.14 (55.86–77.22)	74.84 ± 3.45; 75.97 (64.27–77.06)	73.85 ± 7.03; 76.88 (51.28–78.32)	76.07 ± 2.34; 76.00 (72.00–78.56)	<0.001
Max (Gy)	73.74 ± 0.80; 73.64 (72.66–75.74)	75.44 ± 0.97; 75.44 (72.79–77.00)	80.27 ± 4.95; 81.65 (63.57–83.46)	82.15 ± 1.14; 81.90 (80.81–84.79)	82.20 ± 4.13; 82.88 (66.64–84.88)	83.63 ± 0.97; 83.84 (82.00–85.44)	<0.001
Mean (Gy)	70.74 ± 0.28; 70.77 (70.14–71.19)	72.87 ± 1.15; 72.79 (70.98–76.49)	76.67 ± 4.80; 77.30 (60.61–79.79)	79.04 ± 0.81; 78.94 (77.30–80.34)	79.28 ± 4.20; 80.16 (63.12–81.12)	80.22 ± 0.92; 80.24 (78.96–81.76)	<0.001
Rectum
Min (Gy)	1.93 ± 0.75; 1.82 (0.70–3.57)	1.76 ± 0.77; 1.70 (0.72–3.55)	2.51 ± 1.02; 2.62 (0.91–4.13)	4.10 ± 1.68; 3.63 (2.18–7.64)	2.47 ± 0.96; 2.32 (1.36–4.64)	3.90 ± 0.80; 3.84 (2.72–5.44)	<0.001
Max (Gy)	72.93 ± 0.80; 72.77 (71.33–74.76)	74.94 ± 0.56; 74.93 (74.02–75.83)	80.02 ± 1.43; 80.38 (77.14–81.82)	81.11 ± 1.34; 81.04 (79.25–83.30)	82.08 ± 1.24; 82.16 (79.12–83.92)	82.46 ± 1.60; 82.72 (79.68–84.32)	<0.001
Mean (Gy)	21.59 ± 3.36; 20.93 (16.31–28.14)	21.53 ± 3.99; 20.92 (14.94–29.22)	25.79 ± 5.89; 25.99 (14.44–38.92)	32.78 ± 4.63; 31.94 (26.36–42.12)	25.03 ± 6.35; 24.00 (18.40–41.12)	33.34 ± 3.80; 34.24 (26.80–38.24)	<0.001
Mean EQD2 (Gy)	23.75 ± 3.70; 23.02 (17.94–30.95)	23.68 ± 4.39; 23.01 (16.43–32.14)	25.79 ± 5.89; 25.99 (14.44–38.92)	32.78 ± 4.63; 31.94 (26.36–42.12)	25.28 ± 6.41; 24.25 (18.59–41.54)	33.68 ± 3.84; 34.59 (27.07–38.63)	<0.001
Mean BED3 (Gy)	39.58 ± 6.17; 38.37 (29.90–51.59)	39.47 ± 7.32; 38.35 (27.38–53.57)	42.98 ± 9.82; 43.32 (24.07–64.87)	54.63 ± 7.72; 53.23 (43.94–70.20)	42.14 ± 10.69; 40.41 (30.98–69.24)	56.14 ± 6.40; 57.65 (45.12–64.39)	<0.001
V65 (%)	5.51 ± 4.86; 4.05 (1.20–21.70)	4.84 ± 1.19; 5.20 (2.90–6.60)	7.35 ± 3.32; 6.85 (2.00–15.40)	6.77 ± 3.88; 5.25 (2.90–15.30)	6.67 ± 3.13; 5.70 (3.70–16.50)	6.30 ± 2.04; 5.70 (4.20–9.50)	0.011
V50 (%)	11.57 ± 4.36; 10.70 (6.20–28.90)	10.93 ± 2.20; 10.90 (7.30–14.50)	14.70 ± 5.79; 14.60 (4.40–29.20)	15.52 ± 6.08; 14.25 (7.70–26.80)	13.29 ± 4.44; 12.60 (8.40–26.00)	14.50 ± 4.57; 13.50 (9.40–22.20)	0.030
Urinary bladder
Min (Gy)	3.13 ± 2.47; 2.59 (0.49–12.32)	2.61 ± 2.59; 1.99 (0.80–10.44)	4.22 ± 4.28; 2.09 (0.76–14.27)	11.96 ± 5.47; 13.03 (1.17–18.33)	4.74 ± 6.51; 2.88 (1.04–28.40)	10.82 ± 3.31; 11.68 (3.28–14.00)	<0.001
Max (Gy)	73.61 ± 1.33; 73.53 (70.84–75.95)	75.33 ± 1.69; 75.18 (72.94–78.23)	80.67 ± 1.35; 80.68 (77.90–82.37)	81.04 ± 1.29; 80.96 (78.94–83.15)	82.11 ± 1.50; 81.92 (79.76–85.60)	81.75 ± 1.25; 81.84 (80.16–84.00)	<0.001
Mean (Gy)	25.16 ± 8.94; 25.69 (8.12–42.14)	22.62 ± 8.13; 21.21 (9.71–37.63)	27.23 ± 8.70; 28.92 (10.64–39.16)	36.42 ± 6.96; 37.21 (18.25–47.81)	28.21 ± 9.56; 26.32 (17.12–55.04)	35.01 ± 4.73; 37.76 (28.96–40.00)	<0.001
Mean EQD2 (Gy)	27.68 ± 9.83; 28.26 (8.93–46.35)	24.88 ± 8.95; 23.33 (10.69–41.39)	27.23 ± 8.70; 28.92 (10.64–39.16)	36.42 ± 6.96; 37.21 (18.25–47.81)	28.50 ± 9.66; 26.59 (17.30–55.60)	35.37 ± 4.78; 38.15 (29.26–40.41)	0.003
Mean BED3 (Gy)	46.14 ± 16.39; 47.10 (14.89–77.26)	41.46 ± 14.91; 38.88 (17.81–68.98)	45.38 ± 14.50; 48.19 (17.73–65.26)	60.69 ± 11.60; 62.01 (30.42–79.69)	47.50 ± 16.10; 44.32 (28.83–92.67)	58.95 ± 7.96; 63.58 (48.76–67.35)	0.003
Femoral heads
Min (Gy)	3.87 ± 13.77; 0.70 (0.21–73.50)	2.94 ± 4.46; 1.23 (0.29–19.00)	3.11 ± 5.64; 1.08 (0.08–22.00)	4.93 ± 6.74; 1.68 (0.62–20.98)	2.08 ± 3.31; 0.80 (0.00–13.36)	1.12 ± 0.70; 1.00 (0.24–2.24)	0.117
Max (Gy)	35.72 ± 8.25; 37.73 (11.20–50.89)	34.20 ± 6.15; 34.76 (25.16–48.21)	41.59 ± 7.50; 41.92 (25.08–53.74)	52.67 ± 6.54; 51.91 (43.68–64.19)	45.87 ± 6.85; 45.04 (35.04–59.12)	51.15 ± 6.46; 50.64 (42.00–59.44)	<0.001
Mean (Gy)	14.13 ± 3.46; 14.70 (5.46–19.67)	13.95 ± 5.39; 13.67 (1.01–26.75)	17.75 ± 5.92; 16.61 (8.89–31.98)	23.09 ± 4.41; 22.97 (16.93–31.20)	17.51 ± 4.57; 17.12 (12.00–27.52)	18.42 ± 3.94; 20.00 (12.08–22.72)	<0.001
Penile bulb
Mean (Gy)	23.60 ± 8.77; 23.87 (10.15–45.22)	24.90 ± 7.69; 24.61 (10.44–43.72)	28.14 ± 14.77; 26.21 (9.50–52.36)	31.48 ± 11.46; 33.81 (10.69–51.48)	29.03 ± 11.60; 27.44 (13.28–48.40)	27.62 ± 8.49; 26.72 (13.92–38.96)	0.290
Body contour
Max (Gy)	74.72 ± 0.66; 74.66 (72.87–75.95)	76.31 ± 1.13; 76.02 (74.97–78.23)	81.99 ± 0.90; 81.99 (80.10–83.30)	82.22 ± 0.93; 82.09 (80.50–83.93)	83.68 ± 1.09; 83.52 (81.36–85.60)	83.73 ± 0.80; 83.84 (82.80–85.44)	<0.001

Table Footnote: Values are presented as mean ± standard deviation; median (minimum–maximum). Field Descriptions: Field 1: hypofractionated RT, 70 Gy/28 fr; Field 2: DIL boost hypofractionated RT, 72.5 Gy/29 fr; Field 3: conventional RT, 76–78 Gy/38–39 fr; Field 4: conventional RT with pelvic lymph nodes, 76–78 Gy/38–39 fr; Field 5: DIL boost conventional RT, 80 Gy/40 fr; Field 6: DIL boost conventional RT with pelvic lymph nodes, 80 Gy/40 fr.

Protocols including elective pelvic lymph-node irradiation showed the greatest OAR burden. Rectal Mean dose was highest in Field 6 (33.34 ± 3.80 Gy) and Field 4 (32.78 ± 4.63 Gy), corresponding to Mean BED₃ values of 56.14 ± 6.40 Gy and 54.63 ± 7.72 Gy, respectively. In contrast, prostate/seminal-vesicle hypofractionated protocols yielded lower rectal exposure, particularly Field 2 (21.53 ± 3.99 Gy; BED₃ 39.47 ± 7.32 Gy) and Field 1 (21.59 ± 3.36 Gy; BED₃ 39.58 ± 6.17 Gy) (p < 0.001). Rectum Min, Max, V65, and V50 also differed significantly across protocols (p < 0.001, p < 0.001, p = 0.011, and p = 0.030, respectively).

Bladder Min and Mean doses were significantly elevated in pelvic nodal cohorts, with the highest values in Field 4 (Min 11.96 ± 5.47 Gy; Mean 36.42 ± 6.96 Gy) and Field 6 (Min 10.82 ± 3.31 Gy; Mean 35.01 ± 4.73 Gy); Bladder Mean EQD2 and BED₃ also differed significantly across protocols (both p = 0.003). Femoral-head exposure increased with larger treatment volumes: Femur Max was highest in Field 4 (52.67 ± 6.54 Gy) compared with Field 2 (34.20 ± 6.15 Gy), and Femur Mean was 23.09 ± 4.41 Gy in Field 4 versus 13.95 ± 5.39 Gy in Field 2 (both p < 0.001). Penile Bulb Mean dose did not differ significantly among protocols (p = 0.290), whereas Body Max was highest in Field 6 (83.73 ± 0.80 Gy) and Field 5 (83.68 ± 1.09 Gy; p < 0.001), consistent with the 80-Gy prescription doses (Table 9).

Radiobiological Modeling: Equivalent dose in 2 Gy fractions (EQD2) and biologically effective dose (BED) for the rectum and urinary bladder were calculated using α/β = 3 Gy; target volume prescription BED was calculated using α/β = 1.5 Gy for prostate adenocarcinoma. Prescription BED_1.5 was calculated using the linear-quadratic model: $\mathrm{BED}=D\left(1+{\displaystyle \frac{d}{\alpha /\beta }}\right)$, where D is the total prescription dose, d is the dose per fraction, and α/β = 1.5 Gy. Calculated prescription BED_1.5 values were 186.7 Gy for Field 1, 193.3 Gy for Field 2, 182.0 Gy for Fields 3 and 4, and 186.7 Gy for Fields 5 and 6.

Dosimetric Parameters & Statistical Analysis: V_65Gy and V_50Gy represent rectal percentage volumes receiving absolute physical doses of at least 65 Gy and 50 Gy, respectively. Inter-group dosimetric comparisons across the six treatment field categories were performed using the Kruskal–Wallis test.

3.2.6. Impact of Clinical, Biological, Dosimetric, and Imaging Parameters on Post-Treatment PSA Kinetics

To evaluate longitudinal biochemical response following RT, post-treatment delta PSA (ΔPSA) values were analyzed across sequential follow-up milestones: first post-RT control, 3–6 months, 1 year, and 2 years. Due to the marked right-skewness and high variance of serum PSA values, non-parametric statistical methods were employed, and median values alongside interquartile ranges (IQRs) are prioritized for comparative interpretation.

3.2.7. Fractionation Schedule, Focal Boost, and Target Volume Configurations

Neither the RT fractionation schedule (Conventional vs. Hypofractionated) nor the integration of a focal dose boost exerted a statistically significant influence on post-treatment PSA decay dynamics at any follow-up interval (p > 0.05). Specifically, for conventional vs. hypofractionated schedules, the mean ΔPSA reductions showed no significant differences at the first control (p = 0.679), 3–6 months (−0.24 ± 0.37 ng/mL vs. −0.40 ± 0.84 ng/mL, p = 0.715), 1 year (p = 0.485), or 2 years (p = 0.739). Likewise, focal boost integration did not significantly accelerate early PSA clearance compared to non-boosted regimens (first control: p = 0.590; 3–6 months: −0.39 ± 0.83 ng/mL vs. −0.25 ± 0.43 ng/mL, p = 0.949; 1 year: p = 0.890; 2 years: p = 0.146) (Table 10).

Table 10. Post-Treatment PSA Dynamics (ΔPSA, ng/mL) Stratified by Radiotherapy Regimen and Dosimetric Characteristics.

Variables	First Control ΔPSA	3–6 Month ΔPSA	1-Year ΔPSA	2-Year ΔPSA
Fractionation Schedule
Conventional (n = 52)	−1.64 ± 2.61; −0.55	−0.24 ± 0.37; −0.08	−0.04 ± 0.20; 0.00	0.18 ± 1.25; 0.00
Hypofractionated (n = 46)	−2.17 ± 4.62; −0.48	−0.40 ± 0.84; −0.06	−0.03 ± 0.11; −0.01	0.34 ± 0.89; 0.01
p-value	0.679	0.715	0.485	0.739
Focal Boost
Present (n = 42)	−2.22 ± 4.67; −0.48	−0.39 ± 0.83; −0.08	−0.05 ± 0.19; 0.00	0.02 ± 0.97; −0.01
Absent (n = 56)	−1.62 ± 2.63; −0.64	−0.25 ± 0.43; −0.08	−0.03 ± 0.14; −0.01	0.41 ± 1.27; 0.00
p-value	0.590	0.949	0.890	0.146
Target Field Regimens
Field 1 (n = 30)	−1.50 ± 2.56; −0.58	−0.30 ± 0.53; −0.06	−0.04 ± 0.12; −0.01	0.00 ± 0.02; 0.00
Field 2 (n = 16)	−3.11 ± 6.58; −0.45	−0.55 ± 1.20; −0.04	−0.02 ± 0.09; −0.03	0.68 ± 1.28; 0.13
Field 3 (n = 16)	−1.20 ± 2.36; −0.37	−0.13 ± 0.17; −0.08	−0.01 ± 0.02; −0.01	0.01 ± 0.03; 0.00
Field 4 (n = 10)	−2.59 ± 3.22; −2.04	−0.32 ± 0.39; −0.19	0.00 ± 0.27; −0.02	1.47 ± 2.38; 0.16
Field 5 (n = 18)	−1.87 ± 3.10; −0.48	−0.28 ± 0.51; −0.06	−0.13 ± 0.26; −0.01	−0.05 ± 0.07; −0.02
Field 6 (n = 8)	−1.18 ± 0.91; −1.23	−0.31 ± 0.32; −0.27	0.05 ± 0.14; 0.00	−0.86 ± 1.20; −0.86
p-value	0.716	0.918	0.659	0.125
Anatomical Target Volume
Pelvic Lymph Nodes	−1.75 ± 2.39; −1.59	−0.30 ± 0.33; −0.27	0.02 ± 0.20; 0.00	0.54 ± 2.19; 0.03
Prostate+Seminal Vesicles	−1.56 ± 2.35; −0.71	−0.19 ± 0.40; −0.08	−0.04 ± 0.16; −0.01	−0.04 ± 0.09; 0.00
Prostate Only	−2.13 ± 4.62; −0.41	−0.39 ± 0.82; −0.06	−0.06 ± 0.14; −0.02	0.38 ± 0.87; 0.01
p-value	0.824	0.541	0.398	0.308
CTVprimary Volume (cm3)
25.4–67.6	−1.95 ± 4.40; −0.44	−0.19 ± 0.36; −0.05	−0.04 ± 0.12; −0.01	0.61 ± 1.36; 0.00
67.6–130.5	−1.87 ± 2.77; −0.91	−0.49 ± 0.87; −0.16	−0.06 ± 0.19; −0.02	−0.29 ± 0.63; −0.02
>130.5	−1.08 ± 1.64; −0.11	−0.10 ± 0.11; −0.10	0.14 ± 0.27; 0.01	0.02 ± 0.08; 0.02
p-value	0.547	0.014	0.040	0.121

Table Footnotes: Values are presented as Median (IQR); Mean ± Standard Deviation. ΔPSA values represent absolute changes in serum prostate-specific antigen (ng/mL) relative to the preceding follow-up interval (negative values indicate a decrease in PSA concentration). Abbreviations: ΔPSA: Delta Prostate-Specific Antigen; LN: Lymph Nodes; SV: Seminal Vesicles; P: Prostate. Statistically significant p-values (p < 0.05) are presented in bold. Non-parametric tests (Mann–Whitney U and Kruskal-Wallis) were applied due to the right-skewed distribution and high variance of ΔPSA dynamics. Medians and IQRs are prioritized for group comparisons and text reporting.

Furthermore, variations in target volume definitions—evaluated across six distinct field configurations (Fields 1–6; all p ≥ 0.125) and anatomical extent of coverage (Pelvic Lymph Nodes vs. Seminal Vesicles vs. Prostate-only; all p ≥ 0.308)—yielded no statistically significant differences in longitudinal PSA clearance patterns (Table 10).

3.2.8. Baseline Clinical Target Volume

In contrast, baseline Clinical Target Volume (CTV) emerged as the sole significant determinant of intermediate-term post-RT PSA dynamics. Statistically significant differences in median ΔPSA were observed across CTV tertiles (25.4–67.6 cm³, 67.6–130.5 cm³, and >130.5 cm³) at the 3–6-month (p = 0.014, Kruskal–Wallis test) and 1-year (p = 0.040, Kruskal–Wallis test) follow-up intervals (Table 10):

3–6 Months (p = 0.014): Patients in the intermediate volume group (67.6–130.5 cm³) achieved the most pronounced early median serum PSA reduction [median: −0.16 ng/mL], outperforming both the smaller volume group (25.4–67.6 cm³) [median: −0.05 ng/mL] and the severe prostatomegaly group (>130.5 cm³) [median: −0.10 ng/mL] (Table 10).

1 Year (p = 0.040): While patients with small (25.4–67.6 cm³) and intermediate (67.6–130.5 cm³) CTV maintained a continued median reduction in serum PSA levels [median: −0.01 ng/mL and −0.02 ng/mL, respectively], those with severe prostatomegaly (>130.5 cm³) exhibited blunted clearance dynamics, characterized by a median ΔPSA demonstrating stabilization or slight positive drift [median: 0.01 ng/mL] (Table 10).

3.2.9. Post-RT PSA Dynamics, Exploratory Sub-Group Analyses, and Biochemical Failure Rates

Baseline histopathologic variables (Gleason score, perineural invasion, clinical T stage, necrosis, seminal vesicle involvement) and imaging metrics (⁶⁸Ga-PSMA PET/CT SUV_max, mpMRI PI-RADS category, PET T/N/M stage) demonstrated no statistically significant association with the magnitude of post-treatment PSA decline at any follow-up interval in this exploratory analysis (Table 11, all p > 0.05). Furthermore, the single statistically significant association observed between baseline CTV and intermediate-term PSA decline (3–6 months p = 0.014, 1 year p = 0.040) must be interpreted as an exploratory finding within the context of multiple unadjusted statistical testing. While these observations reflect consistent early biochemical response across risk categories in a cohort receiving combined RT and ADT, they do not establish formal statistical equivalence or true risk neutralization due to sample size constraints in specific subgroups.

Table 11. p-value Summary of Post-Treatment PSA Dynamics Across Subgroups.

Parameter/Variable	Subgroups	First Control ΔPSA (p-Value)	3–6 Month ΔPSA (p-Value)	1-Year ΔPSA (p-Value)	2-Year ΔPSA (p-Value)
Gleason Score (GS 1.scor)	3/4	0.246	0.754	0.484	0.540
Gleason Score (GS 2.scor)	3/4/5	0.275	0.851	0.895	0.459
Total Gleason Score (GS-T)	6/7/8/9	0.392	0.774	0.781	0.496
Perineural Invasion (PNI)	Absence/Presence	0.318	0.118	0.332	0.393
Clinical T Stage	1C/2A/2B/2C/3A	0.129	0.479	0.751	0.136
Necrosis	Absence (n = 85)/Presence (n = 3)	0.576	0.790	0.884	NA
Seminal Vesicle Invasion	Absence/Presence	0.978	0.532	0.905	0.208
Prostate SUVmaxGroup	1–4.9/5–9.9/10–19.9/≥ 20	0.926	0.718	0.769	0.757
mpMRI PI-RADS Category T	Non-localized (0)/2/3/4/5	0.847	0.315	0.171	0.934
68Ga-PSMA PET/CT T Stage	Non-avid (0)/2a/2b/2c/3a/3b/4	0.081	0.921	0.754	0.504
68Ga-PSMA PET/CT T Stage (3-tier)	0/2a–c/3a–4	0.663	0.424	0.309	0.174
68Ga-PSMA PET/CT Nodal Status (N)	Absence/Presence	0.357	0.123	0.976	0.866
68Ga-PSMA PET/CT Metastasis (M)	Absence/Presence	0.201	0.287	0.078	0.100
68Ga-PSMA PET/CT Focal Lesion Count	0/1/2/3/4/Diffuse	0.947	0.620	0.800	0.871

Table Footnotes: ΔPSA values represent absolute changes in serum prostate-specific antigen (ng/mL) relative to the preceding follow-up interval. Abbreviations: ΔPSA: Delta Prostate-Specific Antigen; mpMRI: Multiparametric Magnetic Resonance Imaging; PI-RADS: Prostate Imaging Reporting and Data System; ⁶⁸Ga-PSMA PET/CT: Prostate-specific membrane antigen Positron Emission Tomography/Computed Tomography; SUV: Standardized Uptake Value; NA: Not Applicable.

While median serum PSA reached a nadir of 0.04 ng/mL at 12 months, positive mean ΔPSA values were observed at the 24-month follow-up across several sub-cohorts, including conventional fractionation (+0.18 ng/mL) moderate hypofractionation (+0.34 ng/mL), Field 2 (+0.68 ± 1.28 ng/mL), Field 4 (+1.47 ± 2.38 ng/mL), and the pelvic nodal irradiation group (+0.54 ng/mL).

Applying the Phoenix definition (nadir + 2 ng/mL), a total of 2 out of 20 patients (10.0%) with complete 24-month follow-up experienced true biochemical failure (both cases belonged to the high-risk, nodal-positive Field 4 cohort receiving pelvic nodal irradiation). The remaining mean serum PSA elevation observed across the cohort was attributable to benign serum PSA rebound following the planned ADT withdrawal between 18 and 24 months, which may partly reflect physiological PSA recovery after its cessation.

Similarly, metabolic and anatomical imaging parameters derived from pre-treatment ⁶⁸Ga-PSMA PET/CT and mpMRI demonstrated no statistically significant impact on longitudinal PSA kinetics:

• Prostate SUV_max Groups (1–4.9/5–9.9/10–19.9/>20): p ≥ 0.718 across all intervals.
• mpMRI PI-RADS Scores (0/2/3/4/5): p ≥ 0.171 across all intervals.
• ⁶⁸Ga-PSMA PET/CT Staging: Overall T stage (p ≥ 0.081), condensed 3-tier T stage (p ≥ 0.174), nodal involvement (N stage; p ≥ 0.123), distant metastasis (M stage; p ≥ 0.078), and intraprostatic focal lesion count (p ≥ 0.620) showed no significant associations with post-RT PSA decay (Table 11).

These findings suggest that personalized RT planning guided by ⁶⁸Ga-PSMA PET/CT and mpMRI image fusion achieves comparable and robust post-treatment biochemical clearance regardless of baseline adverse histopathologic or metabolic risk profiles. No statistically significant subgroup differences in early ΔPSA were detected; these exploratory findings do not establish equivalence across risk profiles.

4. Discussion

The integration of advanced molecular imaging techniques, specifically ⁶⁸Ga-PSMA PET/CT and multiparametric magnetic resonance imaging (mpMRI), into RT planning represents a major paradigm shift in the management of localized and regionally advanced prostate cancer [1,21,22]. In the present study, we evaluated the impact of dual-modality image-guided RT planning on staging accuracy, target volume definition, OAR sparing, and biochemical response patterns. Our findings demonstrate that ⁶⁸Ga-PSMA PET/CT significantly alters staging and field design, while hypofractionated regimens combined with DIL boost effectively optimize OAR dosimetric sparing across dose, EQD2, and BED metrics.

Accurate anatomical and biological staging is critical in radiation oncology to avoid both geographical miss (leading to local/regional recurrence) and unnecessary irradiation of healthy surrounding structures [19]. Although mpMRI is widely recognized as the gold standard for local tumor visualization and primary T-staging, its sensitivity in detecting pelvic lymph node involvement remains limited [23].

In our cohort, the agreement between mpMRI and ⁶⁸Ga-PSMA PET/CT for primary T-staging was surprisingly low (overall concordance of 32.9%, Cohen’s Kappa κ ≈ 0.19). Notably, 14.9% of patients categorized as having organ-confined disease (T2a–T2c) on mpMRI were upstaged by ⁶⁸Ga-PSMA PET/CT due to the detection of extracapsular extension, seminal vesicle invasion, or adjacent organ involvement. Moderate agreement was observed for seminal vesicle involvement (κ = 0.47, p < 0.001).

The most pronounced discrepancy emerged in pelvic nodal (N) staging. While conventional mpMRI identified nodal metastases in only 10.8% of cases, ⁶⁸Ga-PSMA PET/CT detected nodal positivity in 25.7% of the patient population. Crucially, among 66 patients classified as N0 by mpMRI, 14 (21.2%) were found to harbor PSMA-avid pelvic lymph node metastases. These findings align closely with the landmark proPSMA trial, which demonstrated a 27% absolute greater accuracy for ⁶⁸Ga-PSMA PET/CT compared to conventional imaging in identifying pelvic nodal and distant metastatic disease [10]. From a clinical perspective, this high diagnostic accuracy directly informed target volume delineation in our study, ensuring the appropriate inclusion of elective pelvic nodal volumes (Field 4 and Field 6) and allowing high-precision dose escalation to involved nodes.

A central observation of this study is the marked discordance between ⁶⁸Ga-PSMA PET/CT and mpMRI in nodal detection and local T-stage categorization. While ⁶⁸Ga-PSMA PET/CT yielded a substantially higher nodal detection rate (25.7% vs. 10.8%, n = 74) we explicitly note that in the absence of surgical histopathological confirmation, a higher detection rate cannot be equated with superior diagnostic sensitivity or accuracy.

Indeed, nodal discordance was observed in both directions. The presence of mpMRI-suspicious nodes that lacked PSMA avidity (3/8 nodes, 37.5%) points to potential limitations of molecular imaging, including low-level PSMA expression, small lesion size below the spatial resolution of PET, or true non-malignant/inflammatory lymphadenopathy. Conversely, PSMA-avid nodes missed by mpMRI may represent false-positive tracer uptake in non-prostatic benign tissue.

Regarding local T-staging, the fair inter-modality agreement (κ = 0.19, p < 0.05) for overall local T-stage classification should not be interpreted as definitive mpMRI understaging. ⁶⁸Ga-PSMA PET/CT possesses inferior spatial resolution compared to high-field mpMRI for local anatomical boundary definition, making mpMRI the established modality for local T-staging and extraprostatic extension detection. Therefore, ⁶⁸Ga-PSMA PET/CT should be viewed as a complementary tool for target volume optimization rather than a standalone replacement for anatomical T-staging.

4.1. Dosimetric Performance and Organs at Risk Sparing Dynamics

The implementation of IMRT and VMAT allows for steep dose gradients, delivering ablative doses to target volumes while shielding critical OARs [3]. Our study comprehensively analyzed how fractionation schemes, target volumes, and focal boost to DILs affect normal tissue dosimetry.

• Impact of Fractionation Schemes: Prostate cancer is characterized by a low α/β ratio (approximately 1.5 Gy), making it inherently sensitive to higher dose-per-fraction schedules. Moderately hypofractionated RT (Fields 1 and 2) demonstrated superior normal-tissue sparing compared with conventionally fractionated schedules across comparable prostate-and-seminal-vesicle target volumes (Fields 3 and 5). Specifically, hypofractionated plans yielded significantly lower mean rectal doses (21.57 ± 3.58 Gy vs. 28.46 ± 6.55 Gy, p < 0.001), reduced rectal V50 (11.32 ± 3.65% vs. 14.39 ± 5.17%, p < 0.001), lower mean bladder doses (24.17 ± 8.63 Gy vs. 31.02 ± 8.84 Gy, p < 0.001), and decreased maximum femoral-head doses (35.12 ± 7.46 Gy vs. 47.20 ± 7.99 Gy, p < 0.001) (Table 8). These results reinforce the radiobiological rationale that moderate hypofractionation optimizes the therapeutic ratio by reducing the absolute dose to organs at risk without compromising relative target coverage [13].
• Role of Focal Boost to Dominant Intraprostatic Lesions (DILs): Escalating the radiation dose to DIL represents a promising strategy to prevent local failure, a concept validated in the phase III FLAME trial using mpMRI [17] and further refined here using mpMRI and ⁶⁸Ga-PSMA PET/CT fusion. Interestingly, in our cohort, patients receiving a focal boost exhibited significantly lower maximum doses (D_max) to the rectum (103.0% vs. 104.1%, p = 0.001), bladder (102.9% vs. 104.9%, p < 0.001), and overall body (104.8% vs. 106.4%, p < 0.001) compared to non-boost plans. This dosimetric advantage demonstrates that biological imaging-guided focal dose escalation does not compromise OAR safety; rather, modern inverse optimization algorithms achieve superior dose conformity around target volumes, effectively reducing hot spots within adjacent normal tissues.
• Influence of Clinical Target Volume (CTV): Large CTVs (>130.5 cm³) did not adversely impact target coverage but were significantly correlated with increased maximum femoral dose (p = 0.009), mean femoral dose (p = 0.004), and maximum body dose (p = 0.032). This observation reflects the anatomical displacement and broader beam angles required to encompass larger target geometries, leading to an increased low-to-intermediate dose bath in peripheral musculoskeletal structures [24].

When evaluating dosimetric outcomes across different fractionation schemes, the influence of target volume geometry—specifically elective pelvic nodal irradiation in conventional fields versus prostate-only fields—warrants nuanced interpretation. Geometrically, expanded nodal fields inherently increase the volumetric exposure of adjacent OARs such as the rectum and bladder. However, in modern inverse-planned RT, stringent OAR dose-volume constraints are strictly enforced irrespective of field extent. The favorable OAR sparing observed in hypofractionated regimens reflects both the steep dose fall-off achieved through PET/MRI-guided planning and the specific target geometry. While prostate-only subset comparisons confirm the isolated advantage of hypofractionation, the overall data demonstrate that advanced image guidance maintains acceptable OAR dose profiles even when target volumes are expanded to encompass pelvic lymph nodes.

4.2. Longitudinal PSA Dynamics, Sub-Group Exploratory Analyses, and 2-Year PSA Rebounds

In our cohort, serum PSA levels declined rapidly following definitive RT and ADT, reaching a median of 0.04 ng/mL at 12 months. In exploratory bivariate analyses, baseline risk parameters—including high Gleason scores (≥7), advanced T-stage, and ⁶⁸Ga-PSMA PET/CT nodal positivity—were not significantly associated with differences in the magnitude of initial post-RT PSA decline.

However, we explicitly acknowledge that failure to reject the null hypothesis (p > 0.05) in these sub-group comparisons does not constitute statistical proof of clinical equivalence or “risk neutralization.” Crucially, an honest appraisal of the 24-month longitudinal data reveals positive mean ΔPSA values across several high-risk subsets, most prominently in Field 4 (+1.47 ± 2.38 ng/mL) and pelvic nodal-positive patients (+0.54 ng/mL).

Rather than reflecting uniform long-term disease control, these 2-year PSA increments stem from two distinct physiological phenomena: first, the expected serum PSA rebound/recovery following ADT withdrawal after completion of 18–24 months of total androgen suppression; and second, true early biochemical failure (nadir + 2 ng/mL), which occurred in n = 2 patients (10.0%) in the pelvic nodal irradiation cohort. These findings underscore that adverse baseline features, particularly nodal involvement, retain their unfavorable prognostic significance over time despite high-precision image-guided RT, highlighting the necessity of long-term biochemical surveillance. Furthermore, single significant associations—such as that observed with baseline CTV—should be viewed as exploratory findings generated within the context of multiple testing, requiring validation in larger, appropriately powered prospective cohorts.

4.3. Study Limitations and Future Directions

Several methodological aspects and limitations of this study should be noted. First, its retrospective, single-center design may limit broader generalizability. Second, while the 16-month median follow-up is well-suited for capturing diagnostic staging performance and acute response kinetics, extended multi-year monitoring is necessary to establish long-term oncologic outcomes such as late toxicity and biochemical relapse-free survival.

Third, regarding post-treatment PSA kinetics, 88.8% of the cohort received risk-stratified ADT per NCCN guidelines. Because ADT administration and its planned duration (1 year vs. 2–3 years) were intrinsically collinear with NCCN clinical risk categories, ADT could not be modeled as an independent statistical covariate. Consequently, all post-RT PSA kinetics were evaluated relative to the Pre-RT Baseline PSA (median: 0.96 ng/mL) measured immediately prior to the first RT fraction to account for prior neoadjuvant suppression. Therefore, our findings reflect the combined local cytoreductive efficacy of high-precision RT superimposed on systemic therapy, and the isolated independent effect of RT versus ADT cannot be disentangled.

Fourth, the sub-group analyses evaluating longitudinal PSA dynamics were exploratory and descriptive in design. Because no formal adjustment for multiple statistical testing was applied and certain sub-groups contained small sample sizes (such as patients with histological necrosis, n = 3, or those with available 2-year follow-up, n = 20), these comparisons lacked adequate statistical power to detect small effect sizes or rule out minor non-equivalence. Consequently, non-significant p-values must not be interpreted as definitive evidence of treatment equivalence across risk strata.

Fifth, a major limitation of this study is the lack of a histopathological reference standard (such as radical prostatectomy or pelvic lymph node dissection specimens) to validate imaging findings. Because all patients received definitive, non-surgical RT, we were unable to determine true sensitivity, specificity, positive predictive value, or negative predictive value. Consequently, nodal and T-stage discrepancies reflect inter-modality detection differences rather than histopathologically proven diagnostic accuracy.

Sixth, clinical acute and late gastrointestinal (GI) and genitourinary (GU) toxicity outcomes (e.g., CTCAE or RTOG criteria) were not systematically reported in this dosimetric and biochemical study. While hypofractionated regimens and focal boost delivery demonstrated significant dose, EQD2, and BED reductions in OARs (rectum and bladder), these dosimetric advantages represent surrogates for normal tissue sparing rather than proven clinical toxicity reduction. Prospective trials incorporating patient-reported outcome measures (PROMs) and standardized clinical toxicity grading are required to translate these dosimetric benefits into definitive clinical toxicity endpoints.

Seventh, while the initial cohort comprised 98 patients evaluated across early post-treatment milestones, long-term 24-month follow-up data were available for a subset of 20 patients (n = 20). This reduction in patient numbers at 24 months represents natural longitudinal cohort maturation (with patients treated in late 2023 and 2024 yet to reach the 2-year post-treatment mark) rather than selective loss to follow-up. Consequently, findings at 24 months—including post-ADT withdrawal serum PSA dynamics—should be interpreted as exploratory secondary endpoints requiring validation as the full cohort matures.

Eighth, multimodality image registration relied on rigid fusion algorithms. Potential soft-tissue deformation between diagnostic scans and planning CT represents an inherent limitation of rigid fusion.

5. Conclusions

This study demonstrates the clinical value of dual-modality ⁶⁸Ga-PSMA PET/CT and mpMRI in primary staging accuracy, target volume definition, normal tissue sparing, and post-RT PSA kinetics in definitive prostate cancer RT planning. In the paired analysis cohort, ⁶⁸Ga-PSMA PET/CT exhibited a significantly higher pelvic lymph node detection rate compared to mpMRI (25.7% vs. 10.8%), directly optimizing target volume delineation and potentially reducing the risk of regional target miss.

From a dosimetric perspective, when evaluated across equivalent target volumes, moderately hypofractionated RT regimens achieved favorable OAR sparing (rectum and bladder) while maintaining robust target coverage compared to conventional fractionation. Furthermore, focal dose escalation to DILs achieved high dose conformity while maintaining dosimetric acceptability within defined OAR tolerance limits.

Clinically, image-guided RT combined with risk-stratified ADT achieved robust early serum PSA suppression across the cohort (median post-RT PSA of 0.04 ng/mL at 12 months). While exploratory analyses showed consistent early biochemical decline across diverse baseline histopathologic and imaging risk profiles, these descriptive findings require validation in larger cohorts to assess long-term comparative outcomes and oncologic efficacy.

Nevertheless, given the retrospective, single-center design, the absence of histopathological validation, and the potential confounding influence of androgen deprivation therapy on PSA kinetics, these findings should be interpreted as hypothesis-generating rather than definitive. The observed imaging discordances reflect inter-modality detection differences rather than confirmed diagnostic accuracy, and the reported PSA suppression likely reflects the combined effect of definitive radiotherapy and risk-adapted androgen deprivation therapy. Further prospective, multicenter studies with histopathological correlation, standardized toxicity assessment, and longer follow-up are required to validate these results.