Abstract
Purpose
This study aimed to evaluate the predictive value of the luteinizing hormone to testosterone (LH/T) ratio in postoperative International Society of Urological Pathology (ISUP) grade upgrading following radical prostatectomy.
Materials and Methods
Clinical data from 503 patients who underwent radical prostatectomy (RP) at Jiangsu Provincial People's Hospital between June 2022 and October 2024 were collected. A stratified random sampling method was used to divide the patients into a training set and a validation set at a 7:3 ratio. In the training set, binary logistic regression analysis was applied to identify key predictive factors for postoperative ISUP classification upgrading. A nomogram predictive model and a multivariate forest plot were constructed. The validation set was assessed using the bootstrap method for C‐index, calibration curve, clinical impact curve (CIC) and decision curve analysis (DCA).
Results
The postoperative ISUP upgrading rate was 31.2% (157/503). LH/T, Prostate Imaging Reporting and Data System (PI‐RADS) score, preoperative ISUP grade and biopsy method were identified as key predictors of pathological upgrading. The C‐index of the training set was 0.800, the validation set was 0.776, and the bootstrap resampling (500 times) in the validation set yielded a C‐index of 0.799, indicating high sensitivity and specificity in distinguishing different categories. Calibration curves demonstrated consistency between predicted and actual values, while clinical impact curve (CIC) and DCA confirmed the model's ability to optimize preoperative decision‐making.
Conclusion
A lower LH/T ratio is associated with a higher risk of ISUP grade upgrading. As a novel predictive biomarker, LH/T may enhance preoperative risk assessment, aiding in more precise treatment decisions for prostate cancer patients.
Keywords: decision curve analysis, ISUP grade upgrading, logistic regression, luteinizing hormone to testosterone ratio, nomogram, prostate cancer, radical prostatectomy, ROC curve
1. INTRODUCTION
Prostate cancer has emerged as the most frequently diagnosed malignancy of the male genitourinary system worldwide. 1 Radical prostatectomy (RP) remains the gold standard surgical approach for managing localized prostate cancer. 2 However, despite preoperative prostate biopsy for pathological assessment, a subset of patients experiences an upgrade in their postoperative International Society of Urological Pathology (ISUP) grade group. This phenomenon, where the final histopathological evaluation reveals a higher malignancy grade than initially estimated via biopsy, can lead to inaccuracies in prognostic assessment and potentially influence clinical decision‐making.
The reported incidence of ISUP grade upgrading varies across different studies and institutions, with literature suggesting an approximate rate of 35%. 3 In recent years, the widespread adoption of multiparametric magnetic resonance imaging (mpMRI) has significantly enhanced the anatomical delineation of the prostate and the characterization of malignant lesions. 4 The combination of MRI‐ultrasound fusion‐targeted biopsy with systematic biopsy has markedly improved diagnostic accuracy and positive detection rates. 5 , 6 Nevertheless, despite these technological advancements, our high‐volume centre—where biopsy techniques are well‐established and quality control is rigorously maintained—continues to observe a substantial incidence of pathological upgrading, occurring in approximately 20%–30% of cases.
Given its potential implications for treatment planning and long‐term oncologic outcomes, accurately predicting postoperative ISUP grade upgrading has become a critical challenge in the optimization of prostate cancer management. Several studies have identified potential predictors of upgrading, including preoperative prostate‐specific antigen (PSA) levels, Prostate Imaging Reporting and Data System (PI‐RADS) scores and biopsy methodology. However, no universally accepted or highly accurate preoperative model currently exists to reliably predict ISUP upgrading. 7 , 8 Therefore, this study aims to evaluate the potential utility of the luteinizing hormone to testosterone (LH/T) ratio as a novel predictive biomarker for ISUP grade upgrading, with the ultimate goal of providing a more precise and clinically applicable risk stratification tool.
2. MATERIALS AND METHODS
2.1. Study population
This study retrospectively collected data from patients who underwent RP at Jiangsu Provincial People's Hospital between June 2022 and October 2024. The inclusion criteria were as follows: (1) patients who underwent RP for prostate cancer; (2) preoperative MRI evaluation with PI‐RADS v2.1 scoring; (3) transperineal systematic biopsy, targeted biopsy or a combination of both, guided by cognitive fusion of MRI and ultrasound; (4) preoperative measurement of luteinizing hormone (LH) and testosterone (T) levels; and (5) histopathological confirmation of prostate adenocarcinoma with complete preoperative and postoperative ISUP grade data. Patients were excluded if they (1) had received androgen deprivation therapy (ADT), radiotherapy or chemotherapy prior to surgery; (2) had incomplete MRI assessments; or (3) had missing clinical data. After applying these criteria, a total of 503 eligible patients were selected. Using a stratified randomization approach, patients were divided into a training cohort (n = 353, 70%) and a validation cohort (n = 150, 30%), (Figure 1).
FIGURE 1.
Patient grouping flow chart.
2.2. Data collection
Demographic and clinical characteristics, including age, height, weight, PSA levels and prostate volume, were collected for all patients. Body mass index (BMI) and prostate‐specific antigen density (PSAD) were calculated accordingly. PI‐RADS scores were obtained from preoperative multiparametric MRI scans. Hormonal parameters, including testosterone (T) levels, LH levels and the calculated LH/T ratio, were measured. Biochemical markers such as alkaline phosphatase (ALP) and lactate dehydrogenase (LDH) were also recorded. To minimize diurnal variations in hormone levels, blood samples were drawn in a fasting state between 07:00 and 09:00 on the morning following hospital admission. Biopsy techniques included systematic biopsy, targeted biopsy and a combination of both. The number of biopsy cores obtained via each method was documented. Histopathological parameters were assessed based on preoperative and postoperative Gleason scores, which were used to determine ISUP grade groups and evaluate the occurrence of ISUP grade upgrading.
2.3. Statistical analysis
All statistical analyses were conducted using SPSS version 22.0 and R software version 4.4.2. A two‐tailed p‐value <0.05 was considered statistically significant. Continuous variables were expressed as mean ± standard deviation (x̄ ± s) for normally distributed data and as median with interquartile range [M (Q1, Q3)] for nonnormally distributed data. Group comparisons were performed using independent samples t‐tests for normally distributed variables and Wilcoxon rank‐sum tests for nonnormally distributed variables. Categorical variables were presented as n (%) and compared using the chi‐square (χ2) test. In the training cohort, univariate logistic regression analysis was performed to evaluate potential predictors of postoperative ISUP grade upgrading, with variables achieving p < 0.2 being included in the multivariable logistic regression model. A stepwise backward selection approach, guided by the Akaike Information Criterion (AIC), was applied to identify independent predictors. A nomogram prediction model was constructed based on the identified predictors, and internal validation was conducted within the training cohort. The model's performance was assessed using: The concordance index (C‐index) to evaluate discriminative ability. Calibration curves to assess predictive consistency. Decision curve analysis (DCA) and clinical impact curves (CIC) to determine clinical applicability. For external validation, the model was tested in the validation cohort using 500 bootstrap resampling iterations to ensure robustness.
3. RESULTS
3.1. Baseline data analysis
The training cohort comprised 353 patients, while the validation cohort included 150 patients. Comparisons of general characteristics, hormonal markers, biochemical parameters, biopsy methods and pathological indicators between the two cohorts are presented in Table 1. No statistically significant differences were observed between the training and validation cohorts (p > 0.05 for all variables), indicating a well‐balanced distribution of baseline characteristics. The overall postoperative ISUP grade upgrading rate was 31.2% (157/503), which is consistent with previously reported rates in both domestic and international studies.
TABLE 1.
Baseline comparison of training and validation sets.
| Variables | Total (n = 503) | Validation set (n = 150) | Training set (n = 353) | p |
|---|---|---|---|---|
| Group | 0.439 | |||
| Non‐Upgraded Group (n) | 346 (69) | 99 (66) | 247 (70) | |
| Upgraded Group (n) | 157 (31) | 51 (34) | 106 (30) | |
| Age (years) | 69 (64, 74) | 69 (63, 74) | 69 (64, 74) | 0.578 |
| Weight (kg) | 69 (64, 75) | 70 (65, 75) | 69 (63, 76) | 0.606 |
| Height (m) | 1.7 (1.65, 1.73) | 1.7 (1.65, 1.74) | 1.7 (1.65, 1.73) | 0.347 |
| BMI (kg/m2) | 24.44 ± 2.92 | 24.43 ± 2.75 | 24.44 ± 2.99 | 0.953 |
| Follicle‐stimulating hormone (FSH, IU/L) | 10.76 (7.16, 15.36) | 11.21 (7.79, 15.03) | 10.73 (6.85, 15.73) | 0.423 |
| Luteinizing hormone (LH, IU/L) | 6.15 (4.47, 8.54) | 6.07 (4.24, 8.38) | 6.15 (4.6, 8.54) | 0.670 |
| Prolactin (mIU/L) | 253.28 (196, 333.44) | 256.74 (194.61, 343.81) | 252.37 (198.04, 331.74) | 0.969 |
| Estradiol (pmol/L) | 90.78 (73.06, 109.3) | 91.03 (73.81, 110.61) | 90.6 (72.7, 109.17) | 0.977 |
| Testosterone (nmol/L) | 12.51 (9.93, 15.13) | 12.54 (9.93, 15.11) | 12.49 (9.94, 15.12) | 0.884 |
| LH/T Ratio (IU/nmol) | 0.49 (0.35, 0.7) | 0.48 (0.34, 0.69) | 0.49 (0.36, 0.7) | 0.775 |
| T/LH Ratio (nmol/IU) | 2.05 (1.44, 2.87) | 2.08 (1.46, 2.94) | 2.04 (1.44, 2.78) | 0.782 |
| Alkaline phosphatase (ALP, U/L) | 69 (59, 81) | 68 (58, 81.75) | 70 (60, 81) | 0.342 |
| Lactate dehydrogenase (LDH, U/L) | 178 (157, 202) | 178 (160.25, 201.5) | 178 (156, 202) | 0.502 |
| Free PSA (fPSA, ng/mL) | 1.33 (0.85, 2.07) | 1.32 (0.84, 2.11) | 1.33 (0.86, 2.05) | 0.625 |
| Total PSA (tPSA, ng/mL) | 13.81 (9.14, 23.64) | 13.03 (9.86, 21.49) | 14.2 (8.9, 25.74) | 0.473 |
| fPSA/tPSA Ratio | 0.09 (0.07, 0.13) | 0.09 (0.07, 0.13) | 0.09 (0.07, 0.13) | 0.467 |
| Prostate volume (mL) | 34.05 (27.12, 46.59) | 34.06 (26.85, 47.82) | 34.05 (27.16, 46.48) | 0.663 |
| PSA density (PSAD, ng/mL/mL) | 0.41 (0.26, 0.72) | 0.4 (0.25, 0.66) | 0.42 (0.26, 0.74) | 0.215 |
| PI‐RADS score | 4 (4, 5) | 4 (4, 5) | 4 (4, 5) | 0.471 |
| Suspicious metastatic lesion | 0.362 | |||
| Yes | 312 (62) | 88 (59) | 224 (63) | |
| No | 191 (38) | 62 (41) | 129 (37) | |
| Target located in MRI peripheral zone | 0.621 | |||
| Yes | 123 (24) | 34 (23) | 89 (25) | |
| No | 380 (76) | 116 (77) | 264 (75) | |
| Target located in MRI Non‐peripheral zone | 0.867 | |||
| Yes | 246 (49) | 72 (48) | 174 (49) | |
| No | 257 (51) | 78 (52) | 179 (51) | |
| Systematic biopsy core count | 12 (12, 12) | 12 (12, 12) | 12 (12, 12) | 0.632 |
| Targeted biopsy core count | 4 (4, 4) | 4 (1, 4) | 4 (4, 6) | 0.429 |
| Total biopsy core count | 16 (12, 16) | 16 (12, 16) | 16 (12, 16) | 0.962 |
| Total positive biopsy core count | 6 (4, 8) | 6 (4, 9) | 6 (4, 8) | 0.260 |
| Total biopsy positivity rate | 0.4 (0.25, 0.58) | 0.44 (0.25, 0.63) | 0.38 (0.25, 0.56) | 0.355 |
| Biopsy method | 0.411 | |||
| 1 (Systematic biopsy) | 117 (23) | 38 (25) | 79 (22) | |
| 2 (Targeted biopsy) | 34 (7) | 7 (5) | 27 (8) | |
| 3 (Combined systematic and targeted biopsy) | 352 (70) | 105 (70) | 247 (70) | |
| Interval between biopsy and radical prostatectomy (days) | 11 (7, 15) | 11 (8, 15) | 11 (7, 15) | 0.387 |
| Preoperative ISUP grade | 2 (2, 3) | 2 (2, 3) | 2 (2, 3) | 0.549 |
| Postoperative ISUP grade | 3 (2, 3) | 3 (2, 3) | 3 (2, 3) | 0.555 |
Note: LH/T represents the ratio of luteinizing hormone (LH) to testosterone (T); T/LH represents the ratio of testosterone (T) to luteinizing hormone (LH).
3.2. Univariate analysis
In the training cohort, univariate logistic regression analysis was performed to calculate the p‐values for each variable (Table 2). Biopsy method was treated as a dummy variable. Notably, LH showed a p‐value of 0.019, while LH/T and T/LH exhibited even lower p‐values, suggesting a stronger association.
TABLE 2.
Univariate logistic regression analysis.
| Variables | B | SE | OR(95%CI) | Z | p |
|---|---|---|---|---|---|
| Preoperative ISUP grade | −0.843 | 0.13391 | 0.431(0.328–0.555) | −6.294 | <0.001 |
| Interval between biopsy and radical prostatectomy (days) | 0.011 | 0.0061 | 1.012(1–1.025) | 1.885 | 0.059 |
| Biopsy Method 2 (targeted biopsy) | −0.515 | 0.46658 | 0.597(0.231–1.462) | −1.105 | 0.269 |
| Biopsy Method 3 (combined systematic and targeted biopsy) | −0.937 | 0.26982 | 0.392(0.231–0.666) | −3.473 | 0.001 |
| Total biopsy positivity rate | −0.567 | 0.48053 | 0.567(0.217–1.435) | −1.181 | 0.238 |
| Total positive biopsy core count | −0.064 | 0.03426 | 0.938(0.876–1.002) | −1.859 | 0.063 |
| Total biopsy core count | −0.052 | 0.03138 | 0.949(0.892–1.01) | −1.653 | 0.098 |
| Targeted biopsy core count | −0.088 | 0.04216 | 0.916(0.842–0.994) | −2.085 | 0.037 |
| Systematic biopsy core count | −0.001 | 0.03458 | 0.999(0.935–1.072) | −0.041 | 0.967 |
| Target located in MRI non‐peripheral zone | 0.284 | 0.23331 | 1.329(0.842–2.105) | 1.218 | 0.223 |
| Target located in MRI peripheral zone | 0.199 | 0.27339 | 1.22(0.721–2.114) | 0.728 | 0.467 |
| Suspicious metastatic lesions | 0.245 | 0.23868 | 1.278(0.798–2.037) | 1.027 | 0.304 |
| PI‐RADS score | 0.28 | 0.1371 | 1.323(1.019–1.747) | 2.044 | 0.041 |
| PSA density (PSAD, ng/mL/mL) | 0.004 | 0.21357 | 1.004(0.642–1.508) | 0.02 | 0.984 |
| Prostate volume (mL) | 0.008 | 0.00596 | 1.008(0.996–1.02) | 1.319 | 0.187 |
| Total PSA (tPSA, ng/mL) | 0 | 0.00683 | 1(0.986–1.013) | 0.011 | 0.992 |
| Free PSA (fPSA, ng/mL) | 0.098 | 0.06659 | 1.103(0.967–1.26) | 1.477 | 0.140 |
| Lactate dehydrogenase (LDH, U/L) | −0.002 | 0.00343 | 0.998(0.991–1.004) | −0.662 | 0.508 |
| Alkaline phosphatase (ALP, U/L) | −0.004 | 0.00653 | 0.996(0.983–1.008) | −0.644 | 0.519 |
| T/LH ratio (nmol/IU) | 0.249 | 0.08639 | 1.282(1.085–1.526) | 2.878 | 0.004 |
| LH/T ratio (IU/nmol) | −1.095 | 0.40262 | 0.335(0.143–0.695) | −2.719 | 0.007 |
| Testosterone (nmol/L) | 0.035 | 0.02759 | 1.036(0.981–1.094) | 1.282 | 0.200 |
| Estradiol (pmol/L) | 0.007 | 0.00376 | 1.007(1–1.015) | 1.912 | 0.056 |
| Prolactin (mIU/L) | 0 | 0.00045 | 1(0.999–1.001) | −0.055 | 0.956 |
| Luteinizing hormone (LH, IU/L) | −0.086 | 0.03649 | 0.918(0.851–0.983) | −2.347 | 0.019 |
| Follicle‐stimulating hormone (FSH, IU/L) | −0.042 | 0.01571 | 0.959(0.928–0.987) | −2.654 | 0.008 |
| BMI (kg/m2) | 0.017 | 0.03892 | 1.017(0.942–1.098) | 0.441 | 0.659 |
| Height (m) | 0.218 | 2.10107 | 1.244(0.02–77.13) | 0.104 | 0.917 |
| Weight (kg) | 0.005 | 0.01179 | 1.005(0.982–1.029) | 0.457 | 0.648 |
| Age (years) | −0.027 | 0.01647 | 0.974(0.942–1.005) | −1.631 | 0.103 |
Note: The biopsy method has been treated as a dummy variable, including (1) systematic biopsy group, (2) targeted biopsy group and (3) combined systematic and targeted biopsy group, with the systematic biopsy group serving as the reference. The p‐values of LH/T and T/LH are significantly lower than that of luteinizing hormone.
3.3. Multivariate logistic regression analysis and prediction model construction
During the selection process, variables with p < 0.2 in the univariate logistic regression analysis were included in the multivariate logistic regression model. A stepwise backward selection method was applied, with AIC used to identify independent predictors of postoperative ISUP grade upgrading (Table 3). The results showed that fPSA, LH/T, PI‐RADS score, preoperative ISUP grade and Biopsy Method 3 were significantly associated with ISUP upgrading (p < 0.05). Although Biopsy Method 2 had a p‐value >0.05, it was retained in the model based on the principle that dummy variables should be entered and removed simultaneously, ensuring better model interpretability and robustness. A nomogram and a multivariate regression forest plot were constructed based on these predictors (Figure 2A,B). The preoperative ISUP grade, Biopsy Method 3, fPSA, LH/T and PI‐RADS score were identified as key predictors of postoperative pathological upgrading in prostate cancer patients. Among these, preoperative ISUP grade, Biopsy Method 3 and LH/T acted as negative predictors, while the others were positive predictors.
TABLE 3.
Multivariate logistic regression analysis.
| Variables | B | SE | OR(95%CI) | Z | p |
|---|---|---|---|---|---|
| Intercept | 0.261 | 0.79927 | 1.298(0.266–6.182) | 0.327 | 0.744 |
| Preoperative ISUP grade | −1.091 | 0.15538 | 0.335(0.244–0.450) | −7.021 | <0.001 |
| Biopsy Method 2 | −0.403 | 0.53411 | 0.668(0.228–1.879) | −0.755 | 0.450 |
| Biopsy Method 3 | −0.953 | 0.32948 | 0.385(0.200–0.732) | −2.892 | 0.004 |
| fPSA (ng/mL) | 0.16 | 0.07842 | 1.173(1.003–1.372) | 2.043 | 0.041 |
| LH/T Ratio (IU/nmol) | −1.105 | 0.42754 | 0.331(0.133–0.721) | −2.583 | 0.010 |
| PI‐RADS score | 0.585 | 0.1737 | 1.795(1.291–2.555) | 3.37 | 0.001 |
Note: The p‐value of LH/T is 0.01, indicating significant statistical relevance and a substantial contribution to the model. The variance inflation factor (VIF) values for all independent variables were close to 1, indicating the absence of multicollinearity.
FIGURE 2.
(A) Nomogram. (B) Multivariate regression Forest plot (Biopsy Method: (1) Systematic, (2) Targeted, (3) Combined; Reference: 1)
3.4. Prediction model evaluation
The C‐index for the training cohort was 0.800, while that for the validation cohort was 0.776. After 500 bootstrap resampling iterations, the validation cohort C‐index reached 0.799 (95% CI: 0.726–0.871), indicating good discriminative ability (Figure 3A,B,C). Calibration curves for the training cohort, validation cohort and bootstrap validation (500 resampling iterations) demonstrated good agreement between predicted and observed outcomes (Figure 3D,E,F). The Hosmer–Lemeshow test yielded p‐values of 0.338 and 0.430 for the training and validation cohorts, respectively, both well above 0.05, confirming good model fit with no significant deviation from expectations. In the clinical DCA, the net benefit range was observed between 0.03 and 0.74 in the validation cohort, supporting clinical utility of the model (Figure 3G). The CIC for the training cohort showed convergence at 0.34, further reinforcing the model's clinical applicability (Figure 3H). In the ROC analysis based on predicted probabilities in the training set, the current model demonstrated the highest AUC compared to individual predictors, suggesting superior discriminative ability and model robustness (Figure 3I).
FIGURE 3.
(A,B,C) Receiver operating characteristic (ROC) curve: (A) Training set; (B) Validation set; (C) Bootstrap resampling of validation set (500 iterations). (D,E,F) Calibration curves for training and validation sets: (D) Training set; (E) validation set; (F) bootstrap resampling of validation set (500 iterations). (G,H) Clinical evaluation of the predictive model: (G) Decision curve analysis (DCA) for validation set with bootstrap resampling (500 iterations); (H) clinical impact curve (CIC) for validation set with bootstrap resampling (500 iterations). (I) ROC curves of the model and individual predictors in the training set. The model showed the highest AUC.
4. DISCUSSION
Postoperative ISUP grade upgrading remains a critical issue in the clinical management of prostate cancer. 9 Due to the multifocal and heterogeneous nature of prostate cancer, 10 , 11 one of the most common reasons for pathological upgrading is the widespread use of ultrasound‐guided systematic biopsy, which primarily delineates the prostate's contour and benign nodules. This approach is inherently blind, often missing prostate lesions, particularly those with the highest malignancy. Moreover, the experience of the performing physician and variations in biopsy equipment further limits the accuracy of pathological grading.
Although mpMRI and MRI‐ultrasound fusion‐targeted biopsy have significantly improved biopsy accuracy and cancer detection rates, the risk of pathological upgrading remains. Developing more precise predictive models and optimizing treatment strategies are key challenges in the clinical management of prostate cancer.
4.1. The role of sex hormones in prostate cancer
Sex hormones play a pivotal role in the initiation, progression and metastasis of prostate cancer. 12 , 13 Androgens such as testosterone (T) bind to androgen receptors (AR) on the cell membrane, mediating a cascade of cellular functions and gene expression. 14 , 15 The prostate is a primary target organ for testosterone, where 5α‐reductase converts testosterone into dihydrotestosterone (DHT), a more potent androgen that regulates prostate cell proliferation and differentiation. 16 , 17 Abnormal testosterone levels may lead to malignant transformation of prostate cells, and elevated testosterone is significantly associated with the development of advanced prostate cancer.
LH stimulates testicular testosterone production, further promoting prostate cancer cell proliferation. 17 Multiple studies have demonstrated that LH levels are typically elevated in prostate cancer patients compared to healthy controls, 18 , 19 , 20 and LH levels correlate with both the risk and aggressiveness of prostate cancer. Therefore, we hypothesize that sex hormones may be a key factor in ISUP grade upgrading. 21
4.2. LH/T as a novel predictor of pathological upgrading
To date, no studies have incorporated sex hormones as a key factor in predicting postoperative pathological upgrading. Our model revealed that for every unit increase in the LH/T ratio, the likelihood of postoperative ISUP grade upgrading decreased by approximately 66.9%, indicating a protective effect. In other words, a lower amount of LH relative to testosterone corresponds to a higher risk of upgrading. This suggests an imbalance in the role of LH in testosterone regulation. 22 , 23 , 24
When LH/T is high, prostate cancer cells may remain relatively stable, making further deterioration and pathological upgrading less likely. Conversely, when LH/T is low, the T/LH ratio increases, meaning that each unit of LH corresponds to a higher level of testosterone, which may destabilize prostate cancer cells, leading to further malignant progression and pathological upgrading. Notably, traditional testosterone measurements often remain within the normal range, making them ineffective predictors of pathological upgrading. This underscores the clinical significance of LH/T in our study.
Initially, we considered incorporating T/LH into the model, but LH/T was ultimately selected based on the lowest AIC. However, both ratios demonstrated clinical value.
4.3. Biopsy method and its impact on pathological upgrading
Our model incorporated biopsy method, preoperative ISUP grade, PI‐RADS score and fPSA levels to enhance predictive accuracy. We found that Biopsy Method 2 (targeted biopsy) did not significantly differ from systematic biopsy in predicting pathological upgrading. 25 , 26 This finding highlights the impact of prostate cancer's heterogeneous distribution on biopsy outcomes. 27
However, Biopsy Method 3 (combined systematic and targeted biopsy) significantly reduced sampling errors and decreased the risk of upgrading. 28 Additionally, lower preoperative ISUP grades, higher PI‐RADS scores and elevated fPSA levels were significant predictors of postoperative upgrading.
4.4. Clinical implications of the model
Our nomogram‐based predictive model enables clinicians to assess the risk of pathological upgrading at the biopsy stage, facilitating more informed treatment decisions.
For instance, ISUP grade 1 (Gleason score ≤6) is often considered clinically insignificant prostate cancer. If a patient is at low risk of upgrading, surgery, radiotherapy or endocrine therapy may not be necessary, and active surveillance could be a viable alternative. This approach minimizes treatment‐related side effects and preserves quality of life, particularly in elderly or comorbid patients. 29
Conversely, ISUP grade ≥2 (Gleason score >7) represents clinically significant prostate cancer. For patients with a high risk of upgrading, aggressive intervention may be warranted. In non‐surgical candidates or those requiring neoadjuvant therapy, intensified treatment strategies—such as oral next‐generation androgen receptor inhibitors—could be considered. For surgical patients, extended pelvic lymph node dissection may improve outcomes and optimize oncologic control.
By reducing overtreatment of low‐risk patients and undertreatment of high‐risk patients, this model may enhance risk stratification, minimize patient burden, optimize healthcare resource utilization and improve clinical outcomes.
5. CONCLUSIONS
This study is the first to identify a significant association between a lower LH/T ratio and postoperative ISUP grade upgrading (p < 0.05), suggesting LH/T as a novel predictive biomarker. A low LH/T ratio (or a high T/LH ratio) may indicate an increased risk of pathological upgrading, further reinforcing the impact of androgen axis dysregulation on tumour biology. 30
However, the underlying mechanisms by which LH/T influences tumour progression remain to be fully elucidated. Further multicenter validation is necessary to enhance the model's accuracy, stability and clinical applicability.
AUTHOR CONTRIBUTIONS
Liu Bianjiang and Pan Zhihua: Study design and manuscript writing. Pan Zhihua, Zhao Ruizhe and Zhang Shaobo: Data analysis and visualization. Pan Zhihua and Fan Jinjiang: Data processing. Li Jie and Liu Bianjiang: Research supervision and manuscript review.
CONFLICT OF INTEREST STATEMENT
All authors declare no conflicts of interest.
ACKNOWLEDGEMENTS
The authors would like to express their gratitude to their mentors and colleagues for their guidance and strong support throughout the research and manuscript preparation process.
Pan Z, Zhao R, Fan J, Zhang S, Li J, Liu B. ISUP grade upgrade prediction after radical prostatectomy: Role of Luteinizing Hormone to Testosterone ratio. BJUI Compass. 2025;6(7):e70043. 10.1002/bco2.70043
DATA AVAILABILITY STATEMENT
The data supporting the findings of this study are not publicly available due to patient confidentiality and institutional restrictions but can be obtained from the corresponding author upon reasonable request.
REFERENCES
- 1. Netto GJ, Amin MB, Berney DM, et al. The 2022 world health organization classification of tumors of the urinary system and male genital organs‐part B: prostate and urinary tract tumors. Eur Urol. 2022;82(5):469–482. [DOI] [PubMed] [Google Scholar]
- 2. Zhang X, Li X, Yu Q, Ma J, Zeng X, Xue L. Morbidity and all‐cause mortality following radical prostatectomy compared with observation for localized prostate cancer in chinese men: a non‐randomized retrospective study. Med Sci Monit: Int Med J Exp Clin Res. 2021;27:e928596. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Li Q, Yang Z, Wang Z, et al. The influence of prostate volume on pathological outcomes after radical prostatectomy: a single‐center retrospective study. Medicine (Baltimore). 2023;102(49):e36526. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Chatterjee A, Thomas S, Oto A. Prostate MR: pitfalls and benign lesions. Abdom Radiol (N Y). 2020;45(7):2154–2164. [DOI] [PubMed] [Google Scholar]
- 5. Rodríguez Socarrás ME, Gomez Rivas J, Cuadros Rivera V, et al. Prostate mapping for cancer diagnosis: the Madrid protocol. Transperineal prostate biopsies using multiparametric magnetic resonance imaging fusion and micro‐ultrasound guided biopsies. J Urol. 2020;204(4):726–733. [DOI] [PubMed] [Google Scholar]
- 6. Rud E, Baco E, Eggesbø HB. MRI and ultrasound‐guided prostate biopsy using soft image fusion. Anticancer Res. 2012;32(8):3383–3389. [PubMed] [Google Scholar]
- 7. Preisser F, Würnschimmel C, Pose RM, Heinze A, Steuber T, Michl U, et al. Concordance of biopsy and pathologic ISUP grading in salvage radical prostatectomy patients for recurrent prostate cancer. Prostate. 2022;82(2):254–259. [DOI] [PubMed] [Google Scholar]
- 8. Liakos N, Witt JH, Rachubinski P, Leyh‐Bannurah SR. The dilemma of misclassification rates in senior patients with prostate cancer, who were treated with robot‐assisted radical prostatectomy: implications for patient counseling and diagnostics. Front Surg. 2022;9:838477. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Pedraza AM, Ferguson EL, Ramos‐Carpinteyro R, et al. Single‐port robotic transvesical partial prostatectomy: a novel approach for focal treatment in prostate cancer. J Endourol. 39(3):261–270. [DOI] [PubMed] [Google Scholar]
- 10. Ibeawuchi C, Schmidt H, Voss R, et al. Genome‐wide investigation of multifocal and unifocal prostate cancer‐are they genetically different? Int J Mol Sci. 2013;14(6):11816–11829. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Salami SS, Hovelson DH, Kaplan JB, Mathieu R, Udager AM, Curci NE, et al. Transcriptomic heterogeneity in multifocal prostate cancer. JCI Insight. 2018;3(21):e123468. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Wu YH, Gugala Z, Barry MM, Shen Y, Dasgupta S, Wang H. Optimization and characterization of a bone culture model to study prostate cancer bone metastasis. Mol Cancer Ther. 2022;21(8):1360–1368. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Wegiel B, Evans S, Hellsten R, Otterbein LE, Bjartell A, Persson JL. Molecular pathways in the progression of hormone‐independent and metastatic prostate cancer. Curr Cancer Drug Targets. 2010;10(4):392–401. [DOI] [PubMed] [Google Scholar]
- 14. Dudkowska M, Jaworski T, Grzelakowska‐Sztabert B, Manteuffel‐Cymborowska M. Androgen receptor and c‐myc transcription factors as putative partners in the in vivo cross‐talk between androgen receptor‐mediated and c‐met‐mediated signalling pathways. Acta Biochim Pol. 2007;54(2):253–259. [PubMed] [Google Scholar]
- 15. Papadopoulou N, Charalampopoulos I, Anagnostopoulou V, Konstantinidis G, Föller M, Gravanis A, et al. Membrane androgen receptor activation triggers down‐regulation of PI‐3K/akt/NF‐kappaB activity and induces apoptotic responses via bad, FasL and caspase‐3 in DU145 prostate cancer cells. Mol Cancer. 2008;7:88. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Zahalka AH, Brodin NP, Maryanovich M, Wang X, Watts KL, Pinho S, et al. Using CT‐guided stereotactic prostate radiation therapy (CT‐SPRT) to assess sustained murine prostate ablation. Sci Rep. 2021;11(1):6571. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Welén K, Damber JE. Androgens, aging, and prostate health. Rev Endocr Metab Disord. 2022;23(6):1221–1231. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Soyupek F, Soyupek S, Perk H, Ozorak A. Androgen deprivation therapy for prostate cancer: effects on hand function. Urol Oncol. 2008;26(2):141–146. [DOI] [PubMed] [Google Scholar]
- 19. Hilz H, Graefen M, Noldus J, Hammerer P, Knabbe C, Huland E, et al. Advanced prostate cancer is associated with a decrease in serum luteinizing hormone. Eur Urol. 2000;38(3):243–249. [DOI] [PubMed] [Google Scholar]
- 20. Fabre B, Machulsky NF, Grosman H, et al. Life events are positively associated with luteinizing hormone in middle age adult men: role of cortisol as a third variable. Stress (Amst Neth). 2014;17(4):328–333. [DOI] [PubMed] [Google Scholar]
- 21. Martin‐Caraballo M. Regulation of molecular biomarkers associated with the progression of prostate cancer. Int J Mol Sci. 2024;25(8):4171. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Yu X, Li X, Yang H. Unraveling intestinal microbiota's dominance in polycystic ovary syndrome pathogenesis over vaginal microbiota. Front Cell Infect Microbiol. 2024;14:1364097. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Abedi H, Zarrin‐Mehr A, Ebrahimi B, Haghshenas H, Parvin N, Kargar Jahromi H. The effect of aqueous extract of orchid root on the structure of ovary and hypothalamic‐pituitary‐gonadal hormones in polycystic ovary syndrome rat model. An Experimental Study Int J Reprod Biomed. 2023;22(3):203–210. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Tao Y. Endocrine aberrations of human nonobstructive azoospermia. Asian J Androl. 2022;24(3):274–286. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Diamand R, Hollans M, Lefebvre Y, Sirtaine N, Limani K, Hawaux E, et al. The role of perilesional and multiparametric resonance imaging‐targeted biopsies to reduce the risk of upgrading at radical prostatectomy pathology: a retrospective monocentric study. Urol Oncol. 2022;40(5):e11–e192. [DOI] [PubMed] [Google Scholar]
- 26. Hsiang W, Ghabili K, Syed JS, Holder J, Nguyen KA, Suarez‐Sarmiento A, et al. Outcomes of serial multiparametric magnetic resonance imaging and subsequent biopsy in men with low‐risk prostate cancer managed with active surveillance Eur Urol. Focus. 2021;7(1):47–54. [DOI] [PubMed] [Google Scholar]
- 27. Liss MA, Zeltser N, Zheng Y, Lopez C, Liu M, Patel Y, et al. Upgrading of grade group 1 prostate cancer at prostatectomy: germline risk factors in a prospective cohort. Cancer Epidemiol Biomarkers Prev. 2024;33(11):1500–1511. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Smani S, Sundaresan V, Lokeshwar SD, Choksi AU, Carbonella J, Brito J, et al. Risk factors for Gleason score upgrade from prostate biopsy to radical prostatectomy. Explor Target Anti‐Tumor Ther. 2024;5(4):981–996. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Fitch M, Pang K, Ouellet V, Loiselle C, Alibhai S, Chevalier S, et al. Canadian men's perspectives about active surveillance in prostate cancer: need for guidance and resources. BMC Urol. 2017;17(1):98. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Chen H, Zhou L, Wu X, Li R, Wen J, Sha J, et al. The PI3K/AKT pathway in the pathogenesis of prostate cancer. Front Biosci (Landmark Ed). 2016;21(5):1084–1091. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The data supporting the findings of this study are not publicly available due to patient confidentiality and institutional restrictions but can be obtained from the corresponding author upon reasonable request.