Holmium laser enucleation of the prostate (HoLEP) has been established as a durable, size-independent surgical option for treating bladder outlet obstruction secondary to benign prostatic hyperplasia (BPH) (1-3). However, despite careful preoperative evaluation, some patients undergoing HoLEP are diagnosed with prostate cancer based on the pathological examination of the resected tissue. Recent cohort studies have reported incidental prostate cancer (iPCa) rates in HoLEP specimens ranging from low single-digit to over 20% (4, 5). Although most iPCa cases are low-grade, clinically insignificant, and suitable for active surveillance (6-8), the rare occurrence of high-grade disease with the potential for rapid progression underscores the clinical importance of identifying patients at higher risk for significant iPCa (9).

To date, several studies have investigated predictive factors for iPCa detected after HoLEP, such as age, prostate-specific antigen (PSA) level, PSA density (PSAD), prostate volume, and biopsy history (10-12). However, no specific risk factors have been established that can predict incidental cancer with a high probability (13).

Magnetic resonance imaging (MRI) is widely used for prostate cancer detection and risk stratification. A Prostate Imaging Reporting and Data System (PI-RADS) was established to estimate the likelihood of clinically significant prostate cancer (14). Attempts to use PI-RADS to predict incidental cancer in HoLEP have also been reported. Porreca et al. reported that the iPCa detection rate was significantly reduced when multiparametric MRI findings were negative (15). Furthermore, Giampaoli et al. reported that a prostate biopsy can be avoided in cases with negative multiparametric MRI findings (16). However, as far as we are aware, no studies have yet focused on the diagnostic accuracy of multiparametric MRI in the transition zone (TZ), the area resected during HoLEP.

This study aimed to inform individualized perioperative management strategies by retrospectively evaluating whether PI-RADS findings in the TZ can predict iPCa detection after HoLEP.

This study was approved by our Institutional Review Board (R07-19) and conducted in accordance with the Declaration of Helsinki. The requirement for informed consent to publish the patient data was waived because of the retrospective nature of the study. Written informed consent was obtained from the patients at the time of treatment and was archived by the Authors in accordance with their privacy rights.

Population and study design. Between June 2022 and April 2025, 136 patients underwent HoLEP for BPH at our hospital. This retrospective study included 114 patients who underwent preoperative MRI scanning and excluded seven patients with prostate cancer diagnosed by preoperative biopsy.

Indications for HoLEP at our Institution included persistent lower urinary tract symptoms due to BPH, acute or chronic urinary retention secondary to bladder outlet obstruction, or failure to respond to alpha-blocker or 5-alpha reductase inhibitor therapy. Preoperative MRI scanning was routinely conducted and evaluated using PI-RADS version 2.1 (17). When the MRI findings suggested the possibility of prostate cancer in either the peripheral or transition zone, a prostate tissue biopsy was performed before HoLEP to rule out prostate cancer. Biopsies were performed under transrectal ultrasound guidance using a combination of systematic and targeted sampling. However, even when MRI findings suggested prostate cancer, if the patient’s quality of life was significantly impaired by severe urinary symptoms or urinary tract infections, HoLEP was prioritized without performing a tissue biopsy after informing the patient of the possibility of iPCa.

MRI assessment. All MRI examinations were performed on a 3.0-T MR scanner (Ingenia; Philips Healthcare, Best, the Netherlands) or a 1.5-T MR scanner (MAGNETOM Sola; Siemens Healthineers, Erlangen, Germany). Although contrast-enhanced imaging is routinely performed at our institution, patients with renal impairment only undergo non-contrast scans. In this study, MRI findings were evaluated based on PI-RADS version 2.1 (17) and focused on the TZ, which is the area resected in HoLEP. The assessment was performed by an experienced urologist and radiologist who were blinded to all other patient data, with a PI-RADS score <3 being considered negative.

Surgical technique. The HoLEP procedure was performed using a Versa Pulse™ Holmium laser system (Lumenis, Yokneam, Israel) at power settings of 75 W (2.5 kJ × 30 Hz) and a 26-Fr continuous flow resectoscope (Olympus Corporation, Tokyo, Japan) equipped with a 550-μm laser fiber (SlimLine™). Enucleation was performed according to the en bloc method previously described by our group under either general or spinal anesthesia (18). Following the completion of the procedure, a 22-Fr three-way catheter was inserted, and continuous bladder irrigation was maintained until the following morning. The catheter was removed on postoperative day 2 or 3. In all cases, anticoagulants were discontinued before surgery and restarted a few days after surgery.

Statistical analysis. Patients (n=114) were divided into iPCa (n=17) and BPH (n=97) groups. Perioperative clinical characteristics and postoperative outcomes at 3 months were compared. Urinary function was assessed using the International Prostate Symptom Score (IPSS), quality of life (QoL) score (19); maximum urinary flow rate (Qmax) measured by uroflowmetry; and post-void residual (PVR). Prostate volume was calculated from MRI measurements as follows: volume (ml)=0.524 × anteroposterior diameter × transverse diameter × craniocaudal diameter (cm). Age, PSA level, TZ-adjusted PSA level (PSAD-TZ), history of dutasteride (5α-reductase inhibitor) administration, history of prostate biopsy, and PI-RADS score for MRI-TZ findings were selected as potential predictors of prostate cancer. Univariate and multivariate logistic regression analyses were performed. Continuous variables were dichotomized based on optimal cut-off values determined using receiver operating characteristic curve analysis.

Statistical analyses were performed using MedCalc Statistical Software, version 23.3.7 (MedCalc Software Ltd., Ostend, Belgium), employing nonparametric tests when necessary. Statistical significance was set at p<0.05.

Clinical characteristics. In total, 114 patients were included in the study: 97 in the BPH group and 17 in the iPCa group. Preoperative characteristics are listed in Table I. No significant differences were identified in the median age at surgery, PSA level, total prostate volume, TZ volume, or urinary function. In addition, no significant differences were observed in the frequency of dutasteride use or prostate biopsy history between the two groups. MRI-TZ PI-RADS score of ≥3 was more prevalent in the iPCa group than in the BPH group (64.7% vs. 16.5%; p<0.0001). Figure 1 shows the frequency distributions of the two groups.

Background of MRI-TZ findings. Table II shows the differences between the two groups based on MRI-TZ abnormality. There were no significant differences in age, urinary function, or dutasteride use between the two groups. However, in the MRI-TZ-positive group, the median (range) PSA level [7.9 (2.0-24.3) vs. 5.1 (0.2-28.4) ng/ml; p=0.020] and PSAD-TZ [0.15 (0.04-0.65) vs. 0.11 (0.002-0.50) ng/ml/ml; p=0.031] were significantly higher. Additionally, previous prostate biopsies were significantly more frequent in the MRI-TZ-positive group (44.4% vs. 13.8%; p=0.0007), reflecting the clinical practice in which patients with MRI-TZ abnormalities underwent prostate biopsy to exclude prostate cancer before HoLEP (Figure 2).

Clinical outcomes. Three months after HoLEP, no significant differences were observed in postoperative PSA levels, PSA reduction rate, International Prostate Symptom Score, quality of life score, maximum flow rate, or post-void residual volume (Table III).

Predictive factors. In the univariate analysis, MRI-TZ PI-RADS score of ≥3 [odds ratio (OR)=9.28, p=0.0001], PSAD-TZ of 0.18 ng/ml/ml (OR=3.19, p=0.033), and PSA level greater than 6.2 ng/ml (OR=3.36, p=0.026) were identified as significant predictors of iPCa. However, in multivariate analysis, only MRI-TZ abnormality remained a significant predictor (OR=8.96, p=0.0004) (Table IV).

HoLEP has emerged as the gold standard surgical treatment for BPH, offering superior outcomes compared to traditional transurethral resection of the prostate or open prostatectomy (1). Multiple studies have demonstrated that HoLEP provides excellent efficacy regardless of prostate size, with improved urinary outcomes, lower morbidity, reduced reoperation rates, and fewer perioperative complications (1-3). Concurrently, increasing attention has been directed toward iPCa detected following large-volume adenoma resection (4, 5). The detection rate of iPCa after HoLEP varies across studies, but most cases involve low-risk tumors that can be managed with active surveillance (6-8). However, as severe cases occasionally require immediate treatment (9), it is important to predict oncological risks before surgery.

Tissue diagnosis using prostate biopsy is necessary to exclude prostate cancer preoperatively in patients who are candidates for HoLEP. However, in patients with severe urinary retention, urinary tract infection, or catheter-dependent obstruction, complications associated with prostate biopsy may delay surgical intervention (20-22). Under these circumstances, a strategy allowing early HoLEP without prostate biopsy, provided the oncological risk is acceptable, holds clinical value.

Several reports have suggested that older age, elevated total PSA level, and increased PSAD may be potential risk factors for prostate cancer (10-12). However, no established predictive factors exist, and the significance of preoperative prostate biopsy remains unclear (13).

MRI has recently become a widely used tool for detecting prostate cancer and for risk stratification. PI-RADS was established to estimate the likelihood of clinically significant prostate cancer. It was introduced in 2012 (14), updated to version 2.0 in 2015 (23), and progressed to version 2.1 in 2019 (17), thereby improving the detection rates of clinically significant prostate cancer (24, 25). Recent reports examining the role of MRI in HoLEP candidates suggest that negative multiparametric MRI findings may be associated with lower iPCa incidence than evaluations based solely on prostate biopsy (15, 16). Furthermore, some studies have suggested that MRI lesions with high PI-RADS scores (of 4-5) are strongly associated with prostate cancer detection in patients scheduled for HoLEP (26). In contrast, more recent data suggest that the presence or location of PI-RADS lesions alone does not clearly correlate with the incidence or grade of iPCa in HoLEP pathology, implying that MRI findings alone may be insufficient to accurately predict incidental diseases (27). Several studies have attempted to stratify risk using other clinical parameters, such as age, PSA level, PSAD, and prostate volume, to improve the diagnostic accuracy of MRI. Li et al. reported that combining PSAD with the PI-RADS v2.1 score helped improve the diagnostic accuracy of prostate cancer and avoid unnecessary biopsies (28). Zheng et al. demonstrated that the PSAD in the TZ is more accurate than that of the total prostate volume for diagnosing prostate cancer during prostate biopsy. Accordingly, the present study adopted PSAD-TZ as a predictor of incidental prostate cancer (29). Regarding measurement of prostate volume, Lin et al. reported that a segmentation-based method offers greater accuracy for irregularly shaped prostate with hyperplastic nodules (30). Furthermore, Engel et al. demonstrated that quantitative diffusion-weighted imaging analysis using apparent diffusion coefficient values was useful for risk stratification of PI-RADS score 3 lesions in the TZ (31). Taken together, the incorporation of new MRI-based quantitative parameters has the potential to further improve the diagnostic accuracy of incidental prostate cancer on MRI.

In this study, univariate analysis demonstrated that an MRI-TZ PI-RADS score of ≥3, PSAD greater than 0.18 ng/ml/ml, and PSA level greater than 6.2 ng/ml were significant predictors of iPCa. Notably, multivariate analysis identified an MRI-TZ PI-RADS score of ≥3 as the sole independent predictor of iPCa. These results suggest that preoperative MRI evaluation provides valuable risk stratification beyond that given by conventional clinical parameters. Additionally, these findings suggest the importance of a combined model using PSAD-TZ, because the discriminatory power of MRI scans alone is limited (28). Furthermore, our analysis found no significant differences in postoperative urinary function or short-term oncological outcomes between patients with and without iPCa. For patients with BPH, it is often necessary to urgently treat severe lower urinary tract symptoms or complications, while excluding clinically significant prostate cancer. In this study, most prostate cancer lesions detected using HoLEP were low grade and did not adversely affect postoperative recovery or short-term outcomes. However, larger studies with longer follow-up periods are needed to rule out high-risk incidental cancers that progress rapidly.

The results of this study support the use of preoperative MRI scanning as a tool for the risk stratification of patients undergoing HoLEP. By predicting the likelihood of iPCa, preoperative MRI enables individualized decision-making regarding whether to prioritize prostate biopsy for cancer diagnosis or proceed directly with surgical intervention for symptom relief.

Limitations. This study has several limitations. Firstly, it was based on a retrospective analysis, introducing selection bias due to the exclusion of patients with prostate cancer. Specifically, because only TZ tissue obtained from HoLEP was evaluated, cases with abnormal findings in the peripheral zone on MRI or suspected prostate cancer may have already been diagnosed with prostate cancer and were excluded from the study. Therefore, further evaluation of the long-term oncological outcomes is required. Secondly, the limited number of observations and the lack of matching between the two groups compromised the statistical power of the analysis. Thirdly, it was not possible to eliminate interobserver variability in all PI-RADS assessments. Furthermore, non-contrast MRI in patients with renal impairment may also have affected the accuracy of the PI-RADS score. Finally, the variable selection in this multivariate logistic regression model was subjective, potentially omitting important variables.

Despite these limitations, this study demonstrated that an MRI-TZ PI-RADS score of ≥3 is an important predictor of iPCa before HoLEP and are readily applicable in clinical practice. Future large-scale prospective cohort studies with longer follow-up periods are needed to establish a comprehensive oncological risk stratification. Such studies may facilitate the simultaneous improvement of severe lower urinary tract symptoms with HoLEP and optimal prostate cancer management in BPH and prostate cancer.

Our findings show that preoperative MRI can predict iPCa detection in the TZ prior to HoLEP. Although the early follow-up data did not reveal significant differences in oncological outcomes, our findings suggest that a risk-stratified surveillance strategy based on preoperative MRI findings should be considered to optimize long-term cancer management in patients undergoing HoLEP.

Our findings show that preoperative MRI can predict iPCa detection in the TZ prior to HoLEP. Although the early follow-up data did not reveal significant differences in oncological outcomes, our findings suggest that a risk-stratified surveillance strategy based on preoperative MRI findings should be considered to optimize long-term cancer management in patients undergoing HoLEP.

The Authors have no conflicts of interest to declare regarding this study.

Toshiki Ito: Conceptualization, Methodology, Project administration, Data curation, Formal analysis, Writing – original draft. Yusaku Hisamatsu, Shuhei Mizutani, Takashi Kodama, Shinya Watanabe and Hiroyuki Amano: Investigation (data collection). Teruo Inamoto: Writing, review, and editing, Supervision. All Authors read and approved the final manuscript.

No artificial intelligence (AI) tools, including large language models or machine-learning software, were used in the preparation, analysis, or presentation of this manuscript.