Diagnostic value of PSA-based markers in prostate cancer: a retrospective analysis stratified by body mass index Authors: Tao Yang, Qiang Zhang, Ziyang Liu, Jia Tian, Xiaohua Hu, Zhiyong Lv Published: 2026 ABSTRACT ObjectiveBody mass index (BMI) influences the accuracy of prostate-specific antigen (PSA) in prostate cancer (PCa) diagnostic performance. This study investigates the effect of BMI on conventional PSA-based markers and evaluates a novel age-adjusted PSA density parameter (A-PSAD) to optimize diagnostic performance strategies across different BMI groups.ResultsThe cohort comprised 307 low-BMI (91 PCa, 216 benign) and 356 high-BMI (109 PCa, 247 benign) patients. In the low-BMI group, A-PSAD yielded the highest diagnostic efficacy (AUC = 0.714), outperforming PSAD (0.685) and total PSA (TPSA, 0.588). In the high-BMI group, A-PSAD also demonstrated superior performance (AUC = 0.793) compared to PSAD (0.774) and TPSA (0.613). A-PSAD’s overall diagnostic value was significantly higher than conventional markers across both groups (P < 0.01). Furthermore, LASSO regression identified four optimal predictive variables for the low-BMI group and seven for the high-BMI group; A-PSAD consistently emerged as a core independent predictor in both.ConclusionBMI significantly modulates the diagnostic efficacy of PCa diagnostic performance indicators. A-PSAD demonstrates robust, superior diagnostic value across varying BMI categories, serving as a promising primary diagnostic performance tool to optimize individualized pre-biopsy risk stratification. Introduction Prostate cancer (PCa) ranks as the second most common malignancy among men worldwide. According to global research data, the incidence of PCa increased from 1.276 million cases in 2018 to 1.4 million cases in 2020, with mortality rising from 359,000 to 375,000 deaths, posing a serious threat to the health of middle-aged and elderly men [1, 2]. Historically, the incidence of PCa in Asian populations has been comparably lower than in Western countries. However, due to the rapid improvement of living standards, dietary westernization, the inevitable trend of population aging, and advancements in diagnostic performance techniques, PCa incidence in China has risen dramatically in recent years [3–5]. The pathogenesis of PCa remains unclear. Some scientists suggest that PCa development may be closely associated with factors such as obesity, race, age, and family history [6, 7]. Studies have shown that obesity may stimulate the production of inflammatory cytokines such as interleukin-6 (IL-6), interleukin-8 (IL-8), and vascular endothelial growth factor (VEGF), which may alter the microenvironment of prostate cells, leading to cellular damage and promoting PCa development [8]. More importantly regarding PCa detection, obese patients typically possess a significantly larger intravascular plasma volume. This creates a hemodilution effect that artificially lowers serum prostate-specific antigen (PSA) concentrations, which may mask the true PSA elevation and cause delayed diagnosis or false-negative results during early diagnostic performance [9]. Given the multifaceted influence of obesity on both PCa progression and PSA hemodilution, the primary objective of this retrospective study is to investigate how BMI influences the diagnostic performance of conventional PSA-based markers, and to evaluate whether a novel age-adjusted PSA density parameter (A-PSAD) can improve PCa detection accuracy in specific BMI subgroups. Materials and methods Clinical Information This study included patients who underwent their first prostate biopsy. Patients included were those who underwent transrectal ultrasound-guided biopsy at the General Hospital of Ningxia Medical University from January 2020 to January 2024, with medical records subsequently analyzed. Exclusion criteria were as follows: 1) patients with acute urinary tract obstruction or active urinary infections (acute lower urinary tract symptoms were assessed via detailed medical history and clinical examination, while active infections were stringently ruled out through routine preoperative urinalysis and urine culture); 2) patients who underwent digital rectal examination, prostate massage, cystoscopy, or other related interventions within two weeks before PSA testing. Prostate Biopsy Biopsy indications were required to meet at least one of the guidelines recommended by the Chinese Urological Association (CUA): 1) nodules in the prostate detected by digital rectal examination; 2) abnormal prostate imaging observed through ultrasound, magnetic resonance imaging (MRI); 3) PSA level >10 ng/mL; or 4) PSA level between 4-10 ng/mL, with abnormal (F/T)PSA or PSAD. The procedures were performed by experienced urologists, with antibiotics administered pre- and post-operation to prevent infection. During the procedure, a color Doppler probe (BK Medical ApS, Pro Focus 2202, frequency 50-60 Hz) was inserted into the rectum along with the biopsy needle. The prostate volume (PV) was first measured using the formula: PV = 0.52 × (anteroposterior diameter) × (transverse diameter) × (longitudinal diameter). Subsequently, a standard 10-core systematic biopsy of the prostate was performed. It is important to note that while MRI was utilized as an imaging indication for some patients prior to biopsy, MRI-guided targeted biopsies were not routinely performed in this specific retrospective cohort; all patients uniformly underwent systematic biopsies. Collected samples were examined by trained pathologists. Based on the collected clinical data, the Adjusted Prostate Specific Antigen Density (A-PSAD) was calculated using the following formula: A-PSAD = (Age × TPSA) / PV. Furthermore, patients were stratified into low-BMI and high-BMI groups using a cutoff of 24 kg/m². This threshold was selected based on the official guidelines for overweight and obesity established by the Working Group on Obesity in China (WGOC) [10]. Data collection Collected clinical data included age, height, weight, TPSA, FPSA, PV, and prostate biopsy results. FPSA and TPSA levels were measured using electrochemiluminescence (Roche Diagnostics, Germany). Body mass index (BMI) was calculated as weight in kilograms divided by the square of height in meters (kg/m²). According to the Working Group on Obesity in China (WGOC) criteria for Asian populations, a BMI ≥ 24 kg/m² is defined as overweight, and ≥ 28 kg/m² as obese. Due to the relatively small sample size of strictly obese patients in our cohort, which would significantly limit the statistical power of machine learning algorithms, we dichotomized patients into two groups for robust analysis: Low/Normal-BMI (< 24 kg/m²) and High-BMI (≥ 24 kg/m², encompassing both overweight and obese individuals). Additionally, derived parameters were calculated: F/TPSA ratio = FPSA / TPSA; PSAD = TPSA / PV; and the novel A-PSAD = (Age × TPSA) / PV. Statistical methods Data were analyzed using SPSS 26.0, R software (version 4.1.3, R Foundation for Statistical Computing), GraphPad Prism 8.0, and MedCalc statistical software. For normally distributed continuous data, the mean and standard deviation (SD) were used to describe central tendency and dispersion, with an independent t-test used for comparisons between groups. Categorical data were compared using the χ² test. Notably, because PSAD and A-PSAD are mathematically derived directly from TPSA, PV, and age, extreme multicollinearity existed among these variables, rendering standard multivariate logistic regression models unstable and yielding invalid odds ratios. Therefore, to resolve this multicollinearity, adjust for clinical confounders (including age, PV, diabetes, and hypertension), and select the most robust independent predictors, Least Absolute Shrinkage and Selection Operator (LASSO) regression was performed using the “glmnet” package in R. The optimal penalty parameter (λ) was determined using 10-fold cross-validation based on the minimum mean squared error (MSE). Variables retaining non-zero coefficients at the optimal λ were subsequently selected as key independent predictors. Following variable selection, Receiver Operating Characteristic (ROC) curves for PSA-related clinical biomarkers were constructed using GraphPad Prism 8.0. The Area Under the Curve (AUC) values were calculated, and the differences between ROC curves for each indicator were statistically compared using the DeLong test in MedCalc. A two-sided significance level of P < 0.05 was set for all statistical tests. Results After diagnostic performance according to inclusion and exclusion criteria, a total of 663 patients were included. Subjects were divided into two groups based on a BMI threshold of 24 kg/m². The low-BMI group contained 307 cases (BPH: 216 cases, PCa: 91 cases), while the high-BMI group contained 356 cases (BPH: 247 cases, PCa: 109 cases). Table 1 summarizes relevant clinical data for the BPH and PCa groups. In the low-BMI group, significant differences were observed between BPH and PCa groups in age, TPSA, FPSA, PSAD, PV, and A-PSAD (P < 0.05), but no significant differences were found in hypertension, diabetes, coronary artery disease, and F/TPSA (P > 0.05). Correspondingly, Table 2 presents the clinical characteristics for the high-BMI cohort. In the high-BMI group, significant differences were observed between BPH and PCa groups in age, diabetes, coronary artery disease, TPSA, F/TPSA, PSAD, PV, and A-PSAD (P < 0.05), while no significant difference was observed in hypertension and FPSA. Table 1The comparison of clinical data between BPH and PCa patients in the low BMI grouplevelOverallBPHPCa p n30721691Age(years)(mean (SD))67.95 (7.58)66.87 (7.64)70.53(6.80)< 0.001Hypertension (%)No239 (77.85)173 (80.09)66 (72.53)0.184Yes68 (22.08)43 (19.91)25 (27.47)Diabetes (%)No271(88.27)192 (88.29)79 (86.81)0.747Yes36 (11.73)24 (11.11)12 (13.19)Coronary_heart_disease (%)No284 (92.51)200 (92.59)84 (92.31)1.000Yes23 (7.49)16 (7.41)7 (7.69)TPSA (ng/ml) (mean (SD))11.55(4.39)11.15 (4.29)12.50 (4.50)0.014FPSA (ng/ml) (mean (SD))2.01 (1.97)1.81 (1.21)2.46 (3.07)0.008F/TPSA(mean (SD))0.19 (0.38)0.19 (0.45)0.16 (0.09)0.511PV (ml)(mean (SD))57.60 (30.03)62.04 (28.70)47.07 (30.65)< 0.001PSAD(mean (SD))0.27 (0.22)0.23 (0.17)0.36 (0.28)< 0.001A-PSAD(mean (SD))17.87 (14.98)14.75 (11.03)25.27 (19.84)< 0.001Abbrevation: BPH Benign prostatic hyperplasia, PCa Prostate cancer, BMI Body mass index, TPSA Total prostate-specific antigen, FPSA Free prostate-specific antigen, F/TPSA FPSA/TPSA, PV Prostate volume, PSAD Prostate-specific antigen density, A-PSAD Age-adjusted prostate-specific antigen density Table 1The comparison of clinical data between BPH and PCa patients in the low BMI grouplevelOverallBPHPCa p n30721691Age(years)(mean (SD))67.95 (7.58)66.87 (7.64)70.53(6.80)< 0.001Hypertension (%)No239 (77.85)173 (80.09)66 (72.53)0.184Yes68 (22.08)43 (19.91)25 (27.47)Diabetes (%)No271(88.27)192 (88.29)79 (86.81)0.747Yes36 (11.73)24 (11.11)12 (13.19)Coronary_heart_disease (%)No284 (92.51)200 (92.59)84 (92.31)1.000Yes23 (7.49)16 (7.41)7 (7.69)TPSA (ng/ml) (mean (SD))11.55(4.39)11.15 (4.29)12.50 (4.50)0.014FPSA (ng/ml) (mean (SD))2.01 (1.97)1.81 (1.21)2.46 (3.07)0.008F/TPSA(mean (SD))0.19 (0.38)0.19 (0.45)0.16 (0.09)0.511PV (ml)(mean (SD))57.60 (30.03)62.04 (28.70)47.07 (30.65)< 0.001PSAD(mean (SD))0.27 (0.22)0.23 (0.17)0.36 (0.28)< 0.001A-PSAD(mean (SD))17.87 (14.98)14.75 (11.03)25.27 (19.84)< 0.001Abbrevation: BPH Benign prostatic hyperplasia, PCa Prostate cancer, BMI Body mass index, TPSA Total prostate-specific antigen, FPSA Free prostate-specific antigen, F/TPSA FPSA/TPSA, PV Prostate volume, PSAD Prostate-specific antigen density, A-PSAD Age-adjusted prostate-specific antigen density The comparison of clinical data between BPH and PCa patients in the low BMI group Abbrevation: BPH Benign prostatic hyperplasia, PCa Prostate cancer, BMI Body mass index, TPSA Total prostate-specific antigen, FPSA Free prostate-specific antigen, F/TPSA FPSA/TPSA, PV Prostate volume, PSAD Prostate-specific antigen density, A-PSAD Age-adjusted prostate-specific antigen density Table 2The comparison of clinical data between BPH and PCa patients in the high BMI grouplevelOverallBPHPCa P n356247109Age(mean (SD))66.97 (7.73)65.75 (7.83)69.74 (6.74)< 0.001Hypertension (%)No215 (60.39)151 (61.13)64 (58.72)0.755Yes141 (39.61)96 (38.87)45 (41.28)Diabetes (%)No309 (86.80)224 (90.69)85 (77.98)0.002Yes47 (13.20)23 (9.31)24 (22.02)Coronary_heart_disease (%)No314 (88.20)225 (91.09)89 (81.65)0.018Yes42 (11.80)22 (8.91)20 (18.35)TPSA (ng/ml) (mean (SD))10.99 (4.37)10.48 (4.35)12.16 (4.21)< 0.001FPSA (ng/ml) (mean (SD))2.05 (3.35)1.84 (1.13)2.51 (5.80)0.080F/TPSA(mean (SD))0.18 (0.11)0.19 (0.12)0.149 (0.081)0.002PV (ml)(mean (SD))67.61 (43.65)76.90 (45.46)46.57 (30.19)< 0.001PSAD (mean (SD))0.25 (0.35)0.18 (0.14)0.39 (0.57)< 0.001A-PSAD (mean (SD))16.58 (23.01)11.97 (9.34)27.04 (37.18)< 0.001Abbrevation: BPH Benign prostatic hyperplasia, PCa Prostate cancer, BMI Body mass index, TPSA Total prostate–specific antigen, FPSA Free prostate–specific antigen, F/TPSA FPSA/TPSA, PV Prostate volume, PSAD Prostate–specific antigen density, A-PSAD Age-adjusted prostate-specific antigen density Table 2The comparison of clinical data between BPH and PCa patients in the high BMI grouplevelOverallBPHPCa P n356247109Age(mean (SD))66.97 (7.73)65.75 (7.83)69.74 (6.74)< 0.001Hypertension (%)No215 (60.39)151 (61.13)64 (58.72)0.755Yes141 (39.61)96 (38.87)45 (41.28)Diabetes (%)No309 (86.80)224 (90.69)85 (77.98)0.002Yes47 (13.20)23 (9.31)24 (22.02)Coronary_heart_disease (%)No314 (88.20)225 (91.09)89 (81.65)0.018Yes42 (11.80)22 (8.91)20 (18.35)TPSA (ng/ml) (mean (SD))10.99 (4.37)10.48 (4.35)12.16 (4.21)< 0.001FPSA (ng/ml) (mean (SD))2.05 (3.35)1.84 (1.13)2.51 (5.80)0.080F/TPSA(mean (SD))0.18 (0.11)0.19 (0.12)0.149 (0.081)0.002PV (ml)(mean (SD))67.61 (43.65)76.90 (45.46)46.57 (30.19)< 0.001PSAD (mean (SD))0.25 (0.35)0.18 (0.14)0.39 (0.57)< 0.001A-PSAD (mean (SD))16.58 (23.01)11.97 (9.34)27.04 (37.18)< 0.001Abbrevation: BPH Benign prostatic hyperplasia, PCa Prostate cancer, BMI Body mass index, TPSA Total prostate–specific antigen, FPSA Free prostate–specific antigen, F/TPSA FPSA/TPSA, PV Prostate volume, PSAD Prostate–specific antigen density, A-PSAD Age-adjusted prostate-specific antigen density The comparison of clinical data between BPH and PCa patients in the high BMI group Abbrevation: BPH Benign prostatic hyperplasia, PCa Prostate cancer, BMI Body mass index, TPSA Total prostate–specific antigen, FPSA Free prostate–specific antigen, F/TPSA FPSA/TPSA, PV Prostate volume, PSAD Prostate–specific antigen density, A-PSAD Age-adjusted prostate-specific antigen density We constructed ROC curves based on clinical indicators for each group (Fig. 1). In the low-BMI group, age (AUC = 0.632), TPSA (AUC = 0.588), FPSA (AUC = 0.573), F/TPSA (AUC = 0.470), PSAD (AUC = 0.685), and PV (AUC = 0.692) demonstrated relatively low diagnostic value for PCa, whereas A-PSAD (AUC = 0.714) showed modest diagnostic efficacy. In the high-BMI group, age (AUC = 0.650), TPSA (AUC = 0.613), FPSA (AUC = 0.514), and F/TPSA (AUC = 0.624) exhibited low diagnostic significance, while PSAD (AUC = 0.774), PV (AUC = 0.725), and A-PSAD (AUC = 0.793) demonstrated superior diagnostic value. Fig. 1ROC curves A B: The ROC curves of TPSA, FPSA, F/TPSA, PSAD, PV, and A-PSAD for the diagnosis of PCa in the low-BMI group. ROC curves C D: ROC curves of TPSA, FPSA, F/TPSA, PSAD, PV,and A-PSAD for the diagnosis of PCa in the high-BMI group. Key: TPSA: Total prostate-specific antigen; FPSA: Free prostate-specific antigen; F/TPSA: FPSA/TPSA; PSAD: Prostate-specific antigen density; A-PSAD: Age multiplied by PSA and divided by PV; BMI: Body mass index; AUC: Area under the curve; ROC:The receiver operating characteristic ROC curves A B: The ROC curves of TPSA, FPSA, F/TPSA, PSAD, PV, and A-PSAD for the diagnosis of PCa in the low-BMI group. ROC curves C D: ROC curves of TPSA, FPSA, F/TPSA, PSAD, PV,and A-PSAD for the diagnosis of PCa in the high-BMI group. Key: TPSA: Total prostate-specific antigen; FPSA: Free prostate-specific antigen; F/TPSA: FPSA/TPSA; PSAD: Prostate-specific antigen density; A-PSAD: Age multiplied by PSA and divided by PV; BMI: Body mass index; AUC: Area under the curve; ROC:The receiver operating characteristic Tables 3 and 4 summarize AUC comparisons for each clinical indicator across different BMI groups. In the low-BMI group, the diagnostic AUC of A-PSAD was significantly higher than those of TPSA, FPSA, F/TPSA, and PV (P < 0.01). In the high-BMI group, A-PSAD’s diagnostic AUC was significantly higher than those of TPSA, FPSA, F/TPSA, PSAD, and PV (P < 0.01). Table 3Comparison of the AUC of A-PSAD and other related parameters in the diagnosis of prostate cancer in the low BMI groupTPSAFPSAF/TPSAA-PSADPSADPVAUC = 0.588AUC = 0.573AUC = 0.530AUC = 0.714AUC = 0.685AUC = 0.692TPSA------FPSA0.670-----F/TPSA0.2420.544----A-PSAD0.0010.019< 0.001---PSAD< 0.0010.019< 0.001< 0.001--PV0.0300.003< 0.0010.4190.769-Abbrevation: PCa Prostate cancer, AUC Area under the ROC curves, TPSA Total prostate-specific antigen, FPSA Free prostate-specific antigen, (F/T) PSA FPSA/TPSA, PV Prostate volume, PSAD Prostate-specific antigen density, A-PSAD Age-adjusted prostate-specific antigen density Table 3Comparison of the AUC of A-PSAD and other related parameters in the diagnosis of prostate cancer in the low BMI groupTPSAFPSAF/TPSAA-PSADPSADPVAUC = 0.588AUC = 0.573AUC = 0.530AUC = 0.714AUC = 0.685AUC = 0.692TPSA------FPSA0.670-----F/TPSA0.2420.544----A-PSAD0.0010.019< 0.001---PSAD< 0.0010.019< 0.001< 0.001--PV0.0300.003< 0.0010.4190.769-Abbrevation: PCa Prostate cancer, AUC Area under the ROC curves, TPSA Total prostate-specific antigen, FPSA Free prostate-specific antigen, (F/T) PSA FPSA/TPSA, PV Prostate volume, PSAD Prostate-specific antigen density, A-PSAD Age-adjusted prostate-specific antigen density Comparison of the AUC of A-PSAD and other related parameters in the diagnosis of prostate cancer in the low BMI group Abbrevation: PCa Prostate cancer, AUC Area under the ROC curves, TPSA Total prostate-specific antigen, FPSA Free prostate-specific antigen, (F/T) PSA FPSA/TPSA, PV Prostate volume, PSAD Prostate-specific antigen density, A-PSAD Age-adjusted prostate-specific antigen density Table 4Comparison of the AUC of A-PSAD and other related parameters in the diagnosis of prostate cancer in the high BMI groupTPSAFPSAF/TPSAA-PSADPSADPVAUC = 0.613AUC = 0.514AUC = 0.624AUC = 0.793AUC = 0.774AUC = 0.725TPSA------FPSA0.082-----F/TPSA0.781< 0.001----A-PSAD0.001< 0.001< 0.001---PSAD< 0.001< 0.001< 0.001< 0.001--PV0.013< 0.0010.0040.0010.011-Abbrevation: PCa Prostate cancer, AUC Area under the ROC curves, TPSA Total prostate-specific antigen, FPSA Free prostate-specific antigen, F/TPSA FPSA/TPSA, PV Prostate volume, PSAD Prostate-specific antigen density, A-PSAD A age-adjusted prostate-specific antigen density Table 4Comparison of the AUC of A-PSAD and other related parameters in the diagnosis of prostate cancer in the high BMI groupTPSAFPSAF/TPSAA-PSADPSADPVAUC = 0.613AUC = 0.514AUC = 0.624AUC = 0.793AUC = 0.774AUC = 0.725TPSA------FPSA0.082-----F/TPSA0.781< 0.001----A-PSAD0.001< 0.001< 0.001---PSAD< 0.001< 0.001< 0.001< 0.001--PV0.013< 0.0010.0040.0010.011-Abbrevation: PCa Prostate cancer, AUC Area under the ROC curves, TPSA Total prostate-specific antigen, FPSA Free prostate-specific antigen, F/TPSA FPSA/TPSA, PV Prostate volume, PSAD Prostate-specific antigen density, A-PSAD A age-adjusted prostate-specific antigen density Comparison of the AUC of A-PSAD and other related parameters in the diagnosis of prostate cancer in the high BMI group Abbrevation: PCa Prostate cancer, AUC Area under the ROC curves, TPSA Total prostate-specific antigen, FPSA Free prostate-specific antigen, F/TPSA FPSA/TPSA, PV Prostate volume, PSAD Prostate-specific antigen density, A-PSAD A age-adjusted prostate-specific antigen density To address potential collinearity among clinical parameters, LASSO regression was employed for optimal variable selection (Fig. 2). For the low-BMI group, at the optimal penalty factor (λ = 0.025, based on lambda.1se), four predictors with non-zero coefficients were identified: age, FPSA, PV, and A-PSAD. For the high-BMI group, at the optimal λ = 0.053, seven predictors with non-zero coefficients were retained: age, diabetes, TPSA, FPSA, F/TPSA, PV, and A-PSAD. Fig. 2Variable selection using LASSO regression. The x-axis represents the log value of the penalty parameter (λ), and the y-axis represents the coefficients of the variables. The optimal λ was determined by 10-fold cross-validation based on the minimum mean squared error (MSE). Variables with non-zero coefficients at the optimal Log(λ) value (indicated by the vertical dashed line) were selected as significant predictors for the final diagnostic models in different BMI groups. Key: TPSA: Total prostate-specific antigen; FPSA: Free prostate-specific antigen; F/TPSA: FPSA/TPSA; PSAD: Prostate-specific antigen density; A-PSAD: Age multiplied by PSA and divided by PV; BMI: Body mass index; MSE:mean squared error Variable selection using LASSO regression. The x-axis represents the log value of the penalty parameter (λ), and the y-axis represents the coefficients of the variables. The optimal λ was determined by 10-fold cross-validation based on the minimum mean squared error (MSE). Variables with non-zero coefficients at the optimal Log(λ) value (indicated by the vertical dashed line) were selected as significant predictors for the final diagnostic models in different BMI groups. Key: TPSA: Total prostate-specific antigen; FPSA: Free prostate-specific antigen; F/TPSA: FPSA/TPSA; PSAD: Prostate-specific antigen density; A-PSAD: Age multiplied by PSA and divided by PV; BMI: Body mass index; MSE:mean squared error Discussion With the general improvement in quality of life, the impact of obesity on malignancies, including the incidence and progression of PCa, has attracted significant attention [7, 11, 12]. While obesity may affect Gleason scores and malignancy levels [13], some studies have offered contrary opinions [14, 15]. Importantly, obesity can lead to decreased serum PSA levels [9, 16, 17],, thereby affecting its value in the early diagnostic performance of PCa [18, 19]. This finding is frequently attributed to the larger plasma volume often seen in obese patients, which dilutes PSA in the bloodstream [20], alongside potential hormonal metabolism disorders affecting PSA secretion [21, 22]. Recent evidence also strongly suggests that PSA-related parameters and prostate tissue microenvironments are significantly modulated by metabolic factors, inflammation, and hormonal alterations [23, 24]. Although the speed and safety of serum PSA testing facilitate timely diagnosis, unnecessary biopsies increase both the financial and psychological burden on patients [25, 26]. BMI is one of the key indicators for assessing obesity [27]. To reduce the impact of BMI on serum PSA and improve diagnostic accuracy, indices such as PSAM [20] and PSAMR [28] have been proposed for PCa diagnostic performance [29]. The use of these indices can further help avoid repeat biopsies in patients with negative resultse [30]. Building on this, our study explored the diagnostic value of PSA-based clinical markers across different BMI groups to select the most appropriate diagnostic performance tool. Currently, although TPSA is the primary marker for PCa diagnostic performance, its specificity is inherently limited by factors like age, BMI, PV, and prostatitis [31]. Similarly, while F/TPSA has long been considered an important parameter in distinguishing BPH from PCa [32], it did not exhibit significant diagnostic value in either the low or high BMI group in our study. This suggests that F/TPSA is not an ideal standalone diagnostic performance marker across diverse body mass populations. However, the literature extensively supports that optimizing free/total PSA ratios can significantly improve diagnostic accuracy and reduce the overdiagnosis of insignificant PCa [33]. Furthermore, combining PSA derivatives with novel genetic markers, such as the PCA3 score, has proven highly valuable in enhancing diagnostic accuracy, particularly in repeat biopsy settings [34]. Other studies suggest that PSAD can partially compensate for the limitations of TPSA. For example, Zheng established an optimal PSAD cutoff for Chinese males [35], and Liu et al. defined a cutoff for men in the PSA gray zone [36]. In our study, PSAD demonstrated modest diagnostic value in the high BMI group but lacked prominent efficacy in the low BMI group, further highlighting the complex impact of BMI on traditional PSA parameters. In contrast, A-PSAD significantly outperformed traditional markers (TPSA, F/TPSA, and PSAD) in both groups. By incorporating the age factor and adjusting for PV, A-PSAD neutralizes BPH interference and mitigates the false negatives caused by blood dilution in high-BMI populations [22]. Consistent with literature suggesting that PSAD-related metrics perform better in high-BMI patients [37], A-PSAD maintained robust diagnostic capability across our cohorts. Although predictive values may vary among patients from different racial and regional backgrounds [38], our results strongly support A-PSAD as a stable and advantageous primary diagnostic performance marker. Despite these promising findings, several specific limitations in our study must be explicitly addressed. First, there was a lack of standardized pre-biopsy evaluation; specifically, not all patients underwent multiparametric magnetic resonance imaging (mpMRI) prior to biopsy, which is crucial in modern guidelines. This underscores the inherent limitations of diagnostic performance methods that rely solely on non-imaging diagnostics. Second, our reliance on systematic 10-core biopsies without MRI-guided targeted biopsies introduces the possibility of missed PCa cases, particularly for micro-lesions or tumors located in the anterior or apical regions. Third, our study lacked a uniform reference pathology or a standardized re-biopsy strategy for patients who had initial negative biopsy results but maintained persistently elevated PSA levels. ly, as a single-center analysis, inherent selection biases exist. Moreover, the evolving role of advanced imaging modalities, including mpMRI and recently 68Ga-PSMA PET/CT, offers superior precision in identifying lesions for targeted biopsies [39]. Future multicenter investigations incorporating these modern imaging protocols are essential to comprehensively validate the clinical utility of A-PSAD and optimize biopsy decision-making. Conclusion In conclusion, our study demonstrates that Age-adjusted PSA Density (A-PSAD) is a superior and highly stable marker for early PCa diagnostic performance compared to traditional indices (TPSA, F/TPSA, and PSAD) across varying BMI populations. By effectively integrating age and PV, A-PSAD mitigates the diagnostic interference caused by benign prostatic hyperplasia and the hemodilution effect prevalent in high-BMI individuals. Clinically, incorporating A-PSAD into routine individualized diagnostic performance strategies can significantly enhance diagnostic accuracy, reduce false-positive rates, and ultimately spare patients from the financial and psychological burdens of unnecessary biopsies.