Value of 3D-presumed circular area ratio in MRI assessment of benign prostatic hyperplasia Authors: Jianli Yang, Wenwei Shi, Hao Xie, Yuan Zhang, Jiaying Li, Gang Guo, Mengxuan Yuan, Zhenyu Zou, Xuetao Mu Published: 2026 ABSTRACT AbstractObjectivesTo quantitatively assess morphological alterations in benign prostatic hyperplasia (BPH) using MRI-based three-dimensional presumed circular area ratio (3D-PCAR), and to explore its correlation with International Prostate Symptom Score (IPSS).Materials & methodsA total of 335 patients with pathologically confirmed BPH who underwent preoperative prostate MRI and completed IPSS questionnaire were retrospectively analyzed. MRI-derived volume parameters included total prostate volume (TPV), transition zone volume (TZV), and transition zone index (TZI, TZI = TZV / TPV); morphological parameters included intravesical prostatic protrusion (IPP), PCAR (axial, coronal, and sagittal), and peripheral zone thickness (PZ-thickness). Correlations between MRI-derived parameters and total IPSS, IPSS storage sub-score (IPSS-ss) and IPSS voiding sub-score (IPSS-vs) were analyzed using Pearson correlation, one-way analysis of variance (ANOVA) and multivariable regression.ResultsAll MRI parameters showed significant correlations with IPSS scores (all p < 0.05). Positive correlations with IPSS were observed for IPP, TPV, TZV, TZI, and PCAR (r = 0.18–0.59), whereas PZ-thickness showed negative correlations (r = − 0.18 to − 0.24). Among the evaluated MRI-derived parameters, PCAR measurements in all three planes demonstrated moderate-to-strong correlations with IPSS (axial r = 0.485, 0.474, 0.405; coronal r = 0.592, 0.553, 0.524; sagittal r = 0.545, 0.508, 0.484, all p < 0.001) and remained significant predictors of IPSS in multivariable analyses (p < 0.05).Conclusion3D-PCAR reflects prostate sphericity and correlates with symptom severity in BPH, demonstrating potential as a novel non-invasive imaging biomarker for evaluating intraprostatic pressure and urinary symptoms.Supplementary InformationThe online version contains supplementary material available at https://doi.org/10.1007/s00345-026-06676-8.Keywords: Magnetic resonance imaging, Presumed circular area ratio, Benign prostatic hyperplasia, Prostate morphology Introduction Benign prostatic hyperplasia (BPH) is a prevalent condition among middle-aged and elderly male populations, frequently accompanied by lower urinary tract symptoms (LUTS) [1]. Objective and accurate assessment of disease severity is clinically essential. Current evaluation modalities primarily include the International Prostate Symptom Score (IPSS), serum prostate-specific antigen (PSA) testing, and imaging examinations. While the IPSS effectively reflects clinical symptom severity and serves as a treatment guidance indicator [2], it cannot directly evaluate prostate volume or morphological characteristics. In contrast, imaging provides direct visualization of prostatic architectural features and plays an increasingly pivotal role in BPH assessment [3]. Transrectal ultrasound (TRUS) remains the cornerstone for BPH evaluation due to its cost-effectiveness and procedural accessibility. However, TRUS has notable limitations, including compromised accuracy in volume measurement and operator-dependent variability in examination results. MRI has been established as a precise diagnostic modality capable of accurately measuring both total prostate volume and zonal anatomical volumes [4–6]. Accumulated evidence indicates that the transitional zone index (TZI), calculated as the ratio of transitional zone volume to total prostate volume, serves as a significant volumetric parameter strongly associated with BPH severity, and exhibits predictive value for LUTS and urodynamic parameters [4]. Recently, Taychert et al. further demonstrated that MRI-derived TZI independently predicts urodynamic bladder outlet obstruction, supporting the translation of MRI morphometry toward harder clinical endpoints [7]. Nevertheless, the urinary symptoms of BPH extend beyond volumetric factors. Emerging research has revealed a significant correlation between the degree of lower urinary tract obstruction and prostatic morphological characteristics [8]. Notably, the presumed circle area ratio (PCAR) has been identified as a notable morphological parameter linked to LUTS: as transitional zone enlargement leads to elevated intraprostatic pressure, the gland undergoes progressive spherical deformation [9]. This morphological change is accompanied by compression of the peripheral zone, resulting in its progressive thinning—a phenomenon corroborated by histopathological-MRI correlative studies [10], which have provided tissue-level evidence for peripheral zone compression and glandular atrophy in larger prostates. To date, however, PCAR has primarily been used only in the horizontal axial plane to assess the extent to which prostate shape approaches roundness, limiting its ability to comprehensively represent the three-dimensional (3D) morphology of the prostate. In parallel, Ji et al. recently proposed prostate spherical volume ratio (PSVR) as a novel 3D metric derived from full prostate segmentation [11], which demonstrated significant correlation with LUTS and showed additional predictive value for BPH clinical progression, highlighting the growing interest in 3D morphological assessment. In this study, we propose the use of MRI-based three-plane (axial, coronal, sagittal) PCAR measurements, referred to as 3D-PCAR, to comprehensively evaluate prostate morphology. We further aim to investigate the correlation between 3D-PCAR and IPSS, to explore its potential clinical value as a non-invasive imaging biomarker for assessing BPH morphology and symptom severity. Materials and methods Study population and process This multicenter retrospective cohort study was conducted from January 2021 to December 2024 with institutional review board approval (approval number: KY2025-001, waiver of informed consent obtained). Patients meeting the following criteria were enrolled: (1) urological treatment at participating institutions with biopsy-confirmed BPH diagnosis, and (2) availability of complete preoperative prostate MRI and IPSS questionnaire data. Exclusion criteria comprised: (1) MRI contraindications; (2) concurrent conditions potentially confounding LUTS assessment (urethral strictures, bladder calculi, prostate cancer, pelvic trauma history, or neurogenic bladder); (3) prior prostate-directed therapies (5α-reductase inhibitors, surgical interventions including TURP/prostatectomy, radiation therapy, or ablation); and (4) > 3-month intervals between MRI, IPSS assessment, and pathological confirmation. After applying the inclusion and exclusion criteria, 335 eligible patients were stratified into three symptom severity groups based on IPSS scores: Group 1 (mild symptoms, IPSS 0–7), Group 2 (moderate symptoms, IPSS 8–19), and Group 3 (severe symptoms, IPSS 20–35). The study population flow chart is illustrated in Fig. 1. The study population flow chart MR acquisition and postprocessing All MR examinations were performed using a 3.0-T MRI scanner (Siemens, Germany, Trio TIM) with an 8-channel phased-array coil or 5.0-T a prototype whole-body MR scanner (uMR Jupiter, United Imaging Healthcare, Shanghai, China) with a 24-channel body coil. An endorectal coil was not employed in this study to avoid deformation of the prostate gland, which could interfere the accuracy of morphological parameters, such as the TZI (due to peripheral zone compression) and PCAR (due to gland deformation). The imaging coverage comprehensively encompassed the anatomical regions of both the prostate and seminal vesicles, with all subjects positioned in the standard supine position. The scanning protocol comprised axial T1-weighted imaging (T1WI) and multiplanar (axial, coronal, and sagittal) T2-weighted imaging (T2WI). Detailed acquisition parameters for each sequence are provided in Table 1. Notably, Post-hoc subgroup analysis showed no significant differences in any key parameters (PCAR-axial, PCAR-coronal, PCAR-sagittal, TPV, TZV, and TZI) between the 3.0T and 5.0T subgroups (all p > 0.05). Table 1Clinical Multi-parametric magnetic resonance imaging (mp-MRI) sequence parameters used for this study3.0 T5.0 TT1WIT2WIT1WIT2WISequenceFast spin echoFast spin echoFast spin echoFast spin echoImaging planeAxialAxial, coronal, sagittalAxialAxial, coronal, sagittalEcho time (ms)1191, 96, 989.88202, 180, 190Repetition time (ms)7003650, 3760, 34708864441, 4648, 3490Flip angle (°)150150, 140, 1409090, 90, 90Field of view (mm)300 × 300300 × 300, 350 × 350, 250 × 250300 × 300300 × 300, 220 × 220, 300 × 300Matrix (frequency× phase)256 × 256256 × 256, 256 × 256, 256 × 256288 × 202400 × 300, 384 × 288, 384 × 288Slice thickness (mm)44, 4, 42.51.5, 3, 2.5Slice gap (mm)002.751.65, 3.3, 2.75Fat saturationNoNo, Yes, YesNoNo, No, NoAcquisition time (min)01:3201:36, 01:54, 02:1300:3901:11, 00:39, 00:53 Clinical Multi-parametric magnetic resonance imaging (mp-MRI) sequence parameters used for this study Parameter measurements Measurement and processing of prostate volumetric parameters (TPV, TZV and TZI) First, the original DICOM files were converted to NIfTI format using MRIcron. Subsequently, the whole prostate and transition zone were manually segmented independently layer by layer using ITK-SNAP (Version 3.8) by two radiologists with 10 and 5 years of experience in prostate MRI interpretation, respectively. Segmentations were saved as binary mask images. Based on these masks, TPV and TZV were calculated. The intra-operator consistency coefficient (ICC) was 0.85, indicating good consistency in the segmentation process. The TZI was calculated according to the formula (TZI = TZV/TPV). Measurement of prostate morphological parameters Measurement of the intravesical prostatic protrusion (IPP): The measurement was performed on sagittal T2WI by determining the distance from the internal urethral orifice to the terminal point of hyperplastic prostatic tissue protruding into the bladder. Measurement of the peripheral zone (PZ)-thickness: PZ-thickness was defined as the maximum distance between the inner and outer boundaries of the peripheral zone, measured along a line drawn from the center of the transitional zone to the outer margin of the peripheral zone. Note that, the examiner first identified the maximal cross-sectional area on axial T2WI. Bilateral PZ-thickness measurements (left and right) were then obtained, and the mean value of these three measurements (left, right, and their average) was recorded as the definitive PZ-thickness [12]. To ensure measurement accuracy, two radiologists (see Volume Parameter Measurement section for methodology) independently measured PZ-thickness, and the mean of their measurements was used for analysis. Measurement of the PCAR: The ratio of the maximum cross-sectional area of the prostate to the area of a hypothetical circle with equal circumference was measured separately on axial, sagittal, and coronal T2WI sequences (Fig. 2). A 64-year-old male patient with a total International Prostate Symptom Score (IPSS) of 8, IPSS storage subscore (IPSS-ss) of 5, IPSS voiding subscore (IPSS-vs) of 3. The presumed circle area ratio (PCAR) is defined as the ratio of the maximum cross-sectional area of the prostate to the area of a presumed circle with equal perimeter in each anatomical plane (axial/coronal/sagittal). a, b S(orange solid line), area of the maximum axial MRI prostate section; L, circumference of the maximum axial MRI prostate section; S′(orange dotted line), area of a presumed circle with circumference equal to the circumference of S. c, d S(orange solid line), area of the maximum coronal MRI prostate section; L, circumference of the maximum coronal MRI prostate section; S′(orange dotted line), area of a presumed circle with circumference equal to the circumference of S. e, f S(orange solid line), area of the maximum sagittal MRI prostate section; L, circumference of the maximum sagittal MRI prostate section; S′(orange dotted line), area of a presumed circle with circumference equal to the circumference of S International prostate symptom score The IPSS questionnaire comprises of seven items assessing lower urinary tract symptoms: Q1 - incomplete emptying, Q2 - frequency, Q3 - intermittency, Q4 - urgency, Q5 - weak stream, Q6 - straining, and Q7 - nocturia. Each item is scored on a scale from 0 to 5. The IPSS storage subscore (IPSS-ss) represents the sum of scores for Q2, Q4, and Q7, whereas the IPSS voiding subscore (IPSS-vs) is derived from Q1, Q3, Q5, and Q6. The total IPSS is the sum of all seven items [2]. It should be specifically noted that in our study, the concept of “obstruction” in patients with BPH was operationally defined and assessed using the four voiding symptom related items (incomplete emptying, intermittency, weak stream, and hesitancy). The IPSS questionnaire for each research center was extracted from urological medical records by a urologist with more than 10 years of clinical experience. The total IPSS, IPSS-ss, and IPSS-vs scores were documented separately for each patient. Statistical analysis Statistical analyses were performed using SPSS Statistics (Version 27.0, IBM Corp, Armonk, NY, USA). One-way analysis of variance (ANOVA) was employed to compare MRI parameters across IPSS-defined severity groups. Pearson’s correlation coefficient was used to evaluate associations between the MRI-derived parameters and IPSS (total, storage subscore, and voiding subscore). Variables that demonstrated statistical significance in the univariate linear regression were further analyzed using multivariable linear regression, with multicollinearity testing performed to ensure model validity. To evaluate the incremental value of PCAR beyond conventional MRI parameters, nested regression analyses were performed with a baseline model (age, TPV, TZV, TZI, IPP, and PZ-thickness) and an expanded model additionally including PCAR-axial, PCAR-coronal, and PCAR-sagittal; the increase in explained variance (ΔR²) and F-change were calculated to assess the incremental contribution. All tests were two-tailed, and the results were considered statistically significant at p < 0.05. Results Intergroup comparison of clinical characteristics and MRI parameters of the study population Table 2; Fig. 3 summarize the comparative results of clinical characteristics and MRI-derived parameters among different BPH patient groups. The study cohort comprised 335 BPH patients, stratified into three groups based on symptom severity. The mean IPSS scores were 6, 14, and 26 for Groups 1, 2, and 3, respectively, with statistically significant intergroup differences among the groups (p < 0.05). The overall cohort had a mean age of 66 years, with average total PSA of 6.60 ng/ml and free PSA of 0.83 ng/ml. Significant intergroup differences in mean age and free PSA levels were observed across the three groups (p < 0.05), while total PSA did not differ significantly (p > 0.05). All MRI parameters - including triplanar PCAR (axial/coronal/sagittal), TPV, TZV, TZI, IPP, and PZ-thickness - exhibited statistically significant intergroup variations (p < 0.05). Notably, PCAR (all planes), TPV, TZV, TZI, and IPP demonstrated progressive increases from Group 1 to Group 3 consistent with increasing symptom severity, whereas PZ-thickness demonstrated a decreasing trend. Table 2Clinical characteristics and MRI quantitative parameters of study populationStatistical comparisonTotal cohortGroup 1IPSS (0–7)Group 2IPSS (8–19)Group 3IPSS (20–35)P value(n = 61)(n = 144)(n = 130)IPSS17 (3–35)6 (3–7)14a (8–19)26ab (20–35)< 0.001IPSS-ss19 (1–26)4 (1–7)9a (4–13)15ab (8–26)< 0.001IPSS-vs19 (0–19)2 (0–4)5a (1–10)11ab (5–19)< 0.001Mean age (y)66.3 (17–90)61.9 (17–82)67.3a (33–90)67.1a (26–87)0.009Total PSA (ng/ml)6.60 (0.15–14.81)5.68 (0.15–12.56)7.01 (0.84–14.44)6.58 (0.84–14.81)0.104Free PSA (ng/ml)0.83 (0.11–3.51)0.63 (0.11–1.47)0.88a (0.12–3.27)0.89a (0.16–3.51)0.020TPV (ml)61.03 (16.41–244.11)38.64 (17.60–82.60)56.34a (18.55–198.51)76.74ab (16.41–244.11)< 0.001TZV (ml)39.91 (4.64–220.14)17.63 (4.64–56.28)34.36a (6.77–135.87)56.52ab (6.91–220.14)< 0.001TZI(TZV/TPV)0.59 (0.21–0.99)0.44 (0.25–0.85)0.57a (0.21–0.98)0.67ab (0.31–0.99)< 0.001PCAR-axial0.78 (0.49–0.99)0.67 (0.49–0.84)0.79a (0.52–0.96)0.83ab (0.53–0.99)< 0.001PCAR-coronal0.82 (0.50–0.99)0.70 (0.50–0.84)0.82a (0.60–0.98)0.88ab (0.60–0.99)< 0.001PCAR-sagittal0.79 (0.52–0.98)0.69 (0.54–0.85)0.78a (0.52–0.97)0.84ab (0.56–0.98)< 0.001IPP(mm)0.66 (0–2.86)0.36 (0–1.80)0.61a (0–2.30)0.86ab (0–2.86)< 0.001PZ thickness (mm)1.02 (0–2.02)1.10 (0.26–1.79)1.08 (0.15–2.24)0.93ab (0–2.02)0.002IPSS International prostate symptom score, IPSS-ss IPSS-storage symptom, IPSS-vs IPSS-voiding symptom, PSA Prostate-specific antigen, TPV Total prostate volume, TZV Transition zone volume, TZI Transition zone index(TZI=TZV/TPV), PCAR Presumed circle area ratio, IPP Intravesical prostatic protrusion, PZ Peripheral zonea Significantly different than Group1 (P < 0.05)b Significantly different than Group2 (P < 0.05) Clinical characteristics and MRI quantitative parameters of study population IPSS International prostate symptom score, IPSS-ss IPSS-storage symptom, IPSS-vs IPSS-voiding symptom, PSA Prostate-specific antigen, TPV Total prostate volume, TZV Transition zone volume, TZI Transition zone index(TZI=TZV/TPV), PCAR Presumed circle area ratio, IPP Intravesical prostatic protrusion, PZ Peripheral zone a Significantly different than Group1 (P < 0.05) b Significantly different than Group2 (P < 0.05) MRI scans showing the magnitude of the presumed circle area ratio (PCAR). The progression of prostatic hypertrophy correlates with an increase in PCAR, indicating that the shape of the prostate is approaching a more circle form (i.e. PCAR = 1.0). In a normal prostate without TZ enlargement, the PCAR is approaching 0.60; in a moderately enlarged, it rises to around 0.80; and in a severely enlarged TZ, it exceeds 0.85. PCAR is used to evaluate how closely the shape of horizontal prostate section approaches a circle. As intraprostatic pressure increases, the prostate shape becomes rounder. a–f Each column shows the same patient’s axial, coronal, and sagittal T2WI MRI images (top to bottom), with PCAR values measured for each Pearson correlation coefficients (r) of clinical features and quantitative MRI characteristics vs. IPSS, IPSS-ss, IPSS-vs Correlation analyses between clinical indicators, MRI quantitative parameters, and the total IPSS, IPSS-ss, and IPSS-vs across the three BPH groups are shown in Table S1 and Fig. 4. Total PSA levels did not correlate with IPSS, IPSS-ss, and IPSS-vs (all p > 0.05). Similarly, free PSA was not significantly correlated with IPSS-ss (p > 0.05). Flat-violin plot combined with boxplot comparing presumed circle area ratio-axial (PCAR-axial), presumed circle area ratio-coronal (PCAR-coronal), presumed circle area ratio-sagittal (PCAR-sagittal), peripheral zone (PZ)-thickness, total prostate volume (TPV), transition zone volume (TZV), transition zone index (TZI), and intravesical prostatic protrusion (IPP) between group 1, group 2, and group 3. Post hoc pairwise comparison was performed with Bonferroni correction. *** p < 0.001, ** p < 0.01, * p < 0.05, n.s. p > 0.05 Several MRI parameters demonstrated significant positive correlations with symptom severity. Specifically, TPV (r = 0.421, p < 0.001; r = 0.367, p < 0.001; r = 0.403, p = 0.001), TZV (r = 0.475; r = 0.411; r = 0.457, all p < 0.001), and TZI (r = 0.492; r = 0.442; r = 0.456, all p < 0.001) were all positively associated with total IPSS, IPSS-ss, and IPSS-vs. PCAR-axial (r = 0.485; r = 0.474; r = 0.405), PCAR-coronal (r = 0.592; r = 0.553; r = 0.524), PCAR-sagittal (r = 0.545; r = 0.508; r = 0.484) also demonstrated significant positive correlations with total IPSS, IPSS-ss, and IPSS-vs, respectively (all p < 0.001). IPP showed a mild but significant positive correlation (r = 0.260, p = 0.001; r = 0.187, p = 0.001; r = 0.294, p = 0.007). In contrast, PZ-thickness was negatively correlated with symptom scores showing decreasing values with increasing IPSS severity (r = −0.235, p < 0.001; r = −0.244, p < 0.001; r = −0.180, p = 0.002, respectively). Multiple linear regression analysis To identify the most informative MRI parameter associated with symptom severity, we performed multiple linear regression analysis with total IPSS as the dependent variable and all clinical and MRI parameters as independent variables. The regression model was statistically significant (F = 30.054, p < 0.001), explaining 45.4% of the variance in IPSS scores (see Table S3 for full model details). Among all evaluated predictors, PCAR-axial (standardized β = 0.111, p = 0.047), PCAR-coronal (standardized β = 0.250, p < 0.001), and PCAR-sagittal (standardized β = 0.329, p < 0.001) were positively associated with IPSS, with PCAR-sagittal demonstrating the strongest effect (see Table 3 and Figure S1). Table 3Univariable and multivariable linear regression analyses of clinical features and quantitative MRI characteristics associated with IPSSVariablesUnivariableMultivariableβS.EStd.β P β (95%CI)βS.EStd.β P β (95%CI)Age0.140.040.236 < 0.001 0.14(0.06 ~ 0.23)−0.030.04−0.0340.447−0.03(−0.10 ~ 0.04)tPSA0.120.110.0610.2620.12 (−0.09 ~ 0.34)−0.040.12−0.0230.705−0.04(−0.29 ~ 0.19)fPSA1.610.710.151 0.024 1.61 (0.22 ~ 3.00)0.290.800.0220.7230.29(−1.29 ~ 1.86PCAR-axial37.583.720.501 < 0.001 37.58 (30.29 ~ 44.86)8.634.33 0.111 0.047 8.63(0.13 ~ 17.12)PCAR-coronal47.323.530.648 < 0.001 47.32 (40.40 ~ 54.24)19.985.10 0.250 < 0.001 19.98 (9.99 ~ 29.96)PCAR-sagittal47.634.010.555 < 0.001 47.63 (39.76 ~ 55.49)28.774.50 0.329 < 0.001 28.77 (19.95 ~ 37.59)TZI23.162.240.466 < 0.001 23.16 (18.76 ~ 27.55)4.232.990.0910.1524.23(−1.57 ~ 10.14PZ-thickness−5.601.27−0.280 < 0.001 −5.60 (−8.09 ~ −3.11)−1.931.11−0.0810.082−1.93(−4.11 ~ 0.24IPP3.160.640.259 < 0.001 3.16(1.90 ~ 4.42)0.630.600.0520.2950.63(−0.55 ~ 1.81CI Confidence interval, IPSS International prostate symptom score, IPSS-ss IPSS-storage symptom, IPSS-vs IPSS-voiding symptom, PSA Prostate-specific antigen, TPV Total prostate volume, TZV Transition zone volume, TZI Transition zone index(TZI=TZV/TPV), PCAR Presumed circle area ratio, IPP Intravesical prostatic protrusion, PZ Peripheral zoneBold values indicate statistically significant differences, with P < 0.05 Univariable and multivariable linear regression analyses of clinical features and quantitative MRI characteristics associated with IPSS 0.14 (0.06 ~ 0.23) −0.03 (−0.10 ~ 0.04) −0.04 (−0.29 ~ 0.19) 0.29 (−1.29 ~ 1.86 8.63 (0.13 ~ 17.12) 4.23 (−1.57 ~ 10.14 −1.93 (−4.11 ~ 0.24 3.16 (1.90 ~ 4.42) 0.63 (−0.55 ~ 1.81 CI Confidence interval, IPSS International prostate symptom score, IPSS-ss IPSS-storage symptom, IPSS-vs IPSS-voiding symptom, PSA Prostate-specific antigen, TPV Total prostate volume, TZV Transition zone volume, TZI Transition zone index(TZI=TZV/TPV), PCAR Presumed circle area ratio, IPP Intravesical prostatic protrusion, PZ Peripheral zone Bold values indicate statistically significant differences, with P < 0.05 To further evaluate the incremental value of PCAR beyond conventional MRI parameters, we performed nested regression analyses comparing a baseline model (age, TPV, TZV, TZI, IPP, and PZ-thickness) with an expanded model additionally including PCAR-axial, PCAR-coronal and PCAR-sagittal. The results demonstrated a significant increase in explained variance (ΔR² = 0.205, F-change = 40.755, p < 0.05), supporting the incremental contribution of PCAR (see Table S4). Based on these results, the regression model equation was derived as follows: Discussion This study demonstrates that prostate volume parameters (TPV, TZV, TZI) measured via semi-automatic MRI segmentation exhibit modest but significantly positive correlations with IPSS and its subscales (IPSS-ss, IPSS-vs). Furthermore, morphological parameters such as IPP, 3D-PCAR, and PZ-thickness also demonstrated significant associations with IPSS. Critically, a multivariable regression model integrating 3D-PCAR (axial, coronal, sagittal) effectively predicted IPSS, underscoring the potential of MRI as an objective tool for quantifying BPH severity. In our study, prostate volumetric parameters were assessed using semiautomatic MRI segmentation. The results demonstrated a modest yet statistically significant positive correlation between TPV, TZV, TZI, and IPSS-including its storage (IPSS-ss) and voiding (IPSS-vs) subscales, prior TRUS-based studies reported inconsistent associations: previous studies reported only a weak correlation between TPV and IPSS but identified a stronger association with TZV [13]. These studies also indicated that a TZI of ≥ 0.50 was significantly correlated with clinical symptoms, including IPSS. In contrast, other studies found no significant correlation between IPSS and either TZV or TZI when measured via TRUS [14]. These discrepancies may be attributed to the inherent limitations of TRUS, including operator-dependency, subjective measurement errors, and errors in estimating prostate volume using the ellipsoid volume estimation (*volume = length × width × height × π/6*), which inadequately represents the complex and irregular contours of BPH. Furthermore, TRUS suffers from poor soft-tissue resolution, hindering zonal delineation (TZV/TZI). In contrast, MRI offers high-resolution imaging, accurate zonal segmentation, and has been validated against prostatectomy specimens, guaranteeing measurement precision and reconciling previous contradictions [4, 5, 15, 16]. Taken together, our findings further support the utility of MRI-derived prostate volumetric parameters as objective imaging biomarkers for BPH-related LUTS assessment. Prostate morphological parameters including IPP, prostatic urethral angle (PUA), PCAR, and PZ-thickness have been proposed as useful imaging biomarkers to characterize prostatic morphology and its relationship with LUTS [17–19]. In this study, PUA was excluded from due to the limited sample size and the requirement for indwelling catheters in some patients with severe hyperplasia, which may significantly compromise the authenticity of PUA measurements. IPP, defined as the extent of prostate protrusion into the bladder, has been correlated with IPSS in TRUS-based studies [20]. Owing to the superior soft tissue resolution of MRI, IPP can now be measured more accurately [21, 22]. Consistent with prior TRUS findings, our MRI-based measurements demonstrated a mild yet statistically significant positive correlation between IPP and IPSS, as well as its subscores (IPSS-ss and IPSS-vs). This observation may be attributed to the indwelling catheter balloons in some study subjects: the balloon compression on bladder-protruding prostate tissue might have altered the original IPP degree. PCAR has been established as a useful predictor for assessing the severity of prostatic obstruction and LUTS [9, 23]. As BPH progression is characterized by transition zone enlargement, the PCAR may potentially reflect intraprostatic pressure - induced prostate deformation, increasing in proportional to TZV [8]. However, previous PCAR studies have primarily employed TRUS, with applications limited to prostate roundness assessment solely in the horizontal axial plane. This two-dimensional assessment lacks comprehensive spatial evaluation, thereby compromising its objectivity and clinical relevance. To address this, we proposed a novel 3D PCAR assessment-3D-PCAR-across axial, coronal, and sagittal planes based on MRI. Our results revealed statistically significant positive correlations between PCAR measurements in all three dimensions (PCAR-axial, PCAR-coronal, PCAR-sagittal) and both total IPSS and its subscores (IPSS-ss, IPSS-vs), thereby validating the fundamental premise of PCAR theory. This 3D approach enables precise evaluation of how closely prostate contours approximate a perfect circle in each anatomical plane. Importantly, our findings support the biomechanical principle that increasing intraprostatic pressure from TZ enlargement induces progressive spherical deformation of the prostate gland - the greater the pressure, the more circular the prostate becomes in three dimensions. Furthermore, PZ-thickness demonstrated a mild but significant negative correlation with IPSS, IPSS-ss, and IPSS-vs, respectively. This suggests that as BPH advances and TZ expands, the PZ thins, which is reflected in both higher PCAR and greater symptom burden. As some studies have found that IPSS and obstructive symptoms were most closely associated with PCAR [18, 24–26], which was attributed to the increase in TZ, which led to thinning of the PZ. Our study also further argues for the fundamentals of the PCAR theory described above [8] a normal prostate is triangular in shape, but as BPH progresses, its shape gradually becomes more rounded. Notably, when the prostate approaches a circular shape, intraprostatic pressure increases, and the PZ-thickness becomes thinner. Therefore, PZ-thickness is inversely proportional to intraprostatic pressure. A recent 2025 study by Ji et al. on prostate spherical volume ratio (PSVR) offers a highly relevant contribution to this field [11]. While we recognize the importance of the PSVR study, our preliminary findings suggest that 3D-PCAR may correlate with both IPSS storage and voiding subscores—an aspect not explored in the PSVR work—and appears to remain independently associated with IPSS even after adjusting for TZI and PZ thickness. This might indicate that multi-planar PCAR could offer complementary morphological insights beyond those provided by a single global sphericity metric. In this study, we evaluated the correlations between clinical indicators and MRI-derived quantitative parameters with IPSS. Among our clinical indicators, age and free PSA significantly differed among the three groups, whereas total PSA was not significantly correlated with IPSS. Regarding MRI parameters, PCAR-axial, PCAR-coronal, PCAR-sagittal, TPV, TZV, TZI and IPP were positively correlated with IPSS, while PZ-thickness was negatively correlated with IPSS. However, in multiple linear regression analysis, only PCAR-axial, PCAR-coronal and PCAR-sagittal remained significant predictors of IPSS. Based on the above results, a regression model was constructed. Several factors may explain these findings. First, in some patients, the presence of a retained urinary catheter balloon likely altered the anatomical configuration of prostatic protrusion, thereby compromising the accuracy of IPP measurements. Second, TPV and TZV are components used to derive TZI [4], which are not as diagnostically useful as TZI and may introduce collinearity; their correlation with IPSS also shows some overlap [27]. Moreover, it has been increasingly recognized that prostate morphology, rather than simply volume, is a key determinant of LUTS [8, 24]. PCAR serves as a quantitative metric to evaluate how closely the prostate morphology approximates a sphere. A normal prostate exhibits a triangular shape, with its transverse diameter greater than its anteroposterior diameter. As BPH, which predominantly occurs in the transition zone, progresses, both TZV and TZI increase, thereby elevating intraprostatic pressure. This pressure preferentially drives anteroposterior expansion, resulting in a more spherical (circular) cross-sectional shape of the gland [9]. Concurrently, compression of the peripheral zone results in progressive thinning. Our previous study (Yang et al., Ref 27) demonstrated that combining volumetric, morphological, and compositional MRI parameters better reflects IPSS correlation. The current study extends that work by introducing three-dimensional PCAR assessment across axial, coronal, and sagittal planes. The results confirmed that among all evaluated predictors, PCAR-axial, PCAR-coronal, and PCAR-sagittal each demonstrated significant positive associations with IPSS. Furthermore, nested regression analysis confirmed that adding the three PCAR parameters significantly increased the explained variance, supporting their incremental value over conventional MRI parameters. The present study has several limitations. First, our cohort was enriched for moderate-to-severe BPH due to the histopathological confirmation requirement, introducing selection bias toward patients with elevated PSA or significant LUTS. This may overestimate correlations and limit generalizability to early-stage or community-based populations. Second, bladder catheter balloons compromised IPP measurements in some participants. Third, the cross-sectional design precludes longitudinal evaluation of treatment outcomes and prognosis. Fourth, although IPSS is the standard tool for LUTS assessment, it remains subjective; we partially addressed this by analyzing storage and voiding subscores separately, yet correlation with IPSS does not equate to clinical utility. Fifth, although post-hoc analysis showed no significant field-strength-related differences in morphological parameters, the use of a 5.0T prototype alongside 3.0T scanners introduces technical heterogeneity, limiting transferability; future multi‑center validation with uniform scanners is needed. These limitations warrant future studies with larger samples, multicenter collaboration, and harder endpoints such as retention, surgical intervention, or urodynamic parameters. In summary, 3D-PCAR provides a non-invasive MRI-derived morphological biomarker that reflects intraprostatic pressure and shape remodeling during BPH progression. Its significant correlation with IPSS supports its potential clinical value in objectively assessing symptom severity in BPH. However, validation in prospective randomized controlled studies remains essential before clinical implementation. Supplementary Information Below is the link to the electronic supplementary material. Supplementary Material 1 Acknowledgements The authors express their sincere gratitude to all patients who participated in this mult-institutional study. We sincerely thank the radiologists and clinical staff at the Third Medical Center of the Chinese People’s Liberation Army General Hospital, the Changji Branch of the First Affiliated Hospital of Xinjiang Medical University, and Zhongda Hospital Affiliated to Southeast University for their invaluable contributions to data collection and image acquisition. We are grateful to the imaging technologists and medical physics teams at all participating centers for maintaining high-quality imaging standards throughout the study period. We acknowledge the statistical consultation provided during the study design and analysis phases. The authors also thank the information technology teams at each institution for their support in data management and secure inter-institutional data transfer. We appreciate the thorough ethical review and approval of this research protocol by the institutional review boards at all participating centers. Finally, this work was supported by the Natural Science Foundation Project of Xinjiang Uygur Autonomous Region (Grant No. 2025D01A131). We are grateful for this funding and for the collaborative spirit of all researchers, whose dedication and efforts made the successful completion of this multi-center retrospective cohort study possible. Author contributions J. Yang : Funding acquisition, Resources, Writing - Original Draft, Writing - Review & Editing. W. Shi : Resources, Methodology, Writing - Review & Editing. H. Xie : Software, Validation. Y. Zhang : Supervision, Resources. J. Li : Supervision, Resources. G. Guo : Clinical data curation. M. Yuan : Data curation. Z. Zou : Methodology, Formal analysis. X. Mu : Project administration, Methodology, Writing - Review & Editing. All authors read and approved the final manuscript. Funding This work was supported by the Natural Science Foundation Project of Xinjiang Uygur Autonomous Region(Grant No.2025D01A131). Data availability The datasets generated and/or analysed during the current study are not publicly available due to patient privacy protection and institutional data sharing policies, but are available from the corresponding author on reasonable request and with appropriate ethical approval. Declarations Conflict of interest The authors declare no competing interests. Ethical approval and consent to participate This study was approved by the Institutional Review Board of the Third Medical Center of the Chinese People’s Liberation Army General Hospital (approval number: KY2025-001) and conducted in accordance with the Declaration of Helsinki. Ethics approval was also obtained from all participating centers: Changji Branch of the First Affiliated Hospital of Xinjiang Medical University and Zhongda Hospital Affiliated to Southeast University. The requirement for informed consent was waived by the ethics committees due to the retrospective nature of this study and complete anonymization of patient data. Consent for publication Not applicable. This manuscript does not contain any individual person’s data in any form (including individual details, images, or videos). Footnotes Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Jianli Yang and Wenwei Shi contributed equally to this work. Contributor Information Zhenyu Zou, Email: 704088872@qq.com. Xuetao Mu, Email: muxuetao091@163.com. References Associated Data Supplementary Materials Supplementary Material 1 Data Availability Statement The datasets generated and/or analysed during the current study are not publicly available due to patient privacy protection and institutional data sharing policies, but are available from the corresponding author on reasonable request and with appropriate ethical approval.