Introduction

In 2025, the landscape of prostate cancer and prostatic diseases evolved markedly, with important advances in epidemiology, diagnostic precision, and emerging treatment approaches (Table 1) [1,2,3,4,5,6,7,8,9,10].

Table 1 Best of 2025 in prostate cancer and prostatic diseases.
Full size table

Primary prevention in PCa still represents an unmet need, given the lack of certain preventive measures, notwithstanding the large number of studies on the subject [11]. Emerging evidence suggests that altered lipid metabolism may play a meaningful role in PCa development and progression, linking metabolic health to oncological outcomes. Elevated cholesterol levels and dysregulated lipid pathways appear to influence tumor growth, inflammation, and androgen signaling. In this context, statins have gained attention not only for their cardiovascular benefits but also for their potential protective effects in prostate cancer [12,13,14]. By modulating lipid profiles and exerting anti-inflammatory and anti-proliferative actions, statins may reduce the risk of clinically significant disease and improve long-term outcomes, although definitive causal relationships remain under investigation. Overall, two recent meta-analysis suggest a lower risk of PCa (OR: 0,94) as well as advanced PCa in statin users [15, 16]. Nevertheless, despite the substantial number of studies available, the overall quality of evidence remains insufficient to draw definitive conclusions. Some studies indicate that the potential protective effect of statins may be influenced by genetic variability, which could partly explain the heterogeneity observed across studies regarding this association [1, 17].

In terms of PCa diagnosis, the implementation of multiparametric MRI (mpMRI) in recent years has dramatically transformed diagnostic pathways. However, limited availability, increased workload, variability in MRI quality, and its prognostic role still represent major barriers to widespread adoption. In this context, bi-parametric MRI (bpMRI) is increasingly recognized as a valuable and efficient alternative, offering shorter acquisition times and eliminating the need for contrast agents while maintaining high diagnostic accuracy [18,19,20,21]. This has been further supported by the results of the recently published PRIME trial, which demonstrated comparable performance of bpMRI to mpMRI in detecting clinically significant lesions (csPCa: 29.2% vs 29.6%) [22]. Similarly, the integration of artificial intelligence into MRI protocols is enhancing automation, improving diagnostic accuracy, and reducing interobserver variability [23]. As well, evidence is growing supporting the favorable prognostic role of negative MRI with important implications in patients on active surveillance (AS)protocols [4, 24, 25].

A deeper understanding of MRI’s prognostic value, together with the growing adoption of bpMRI and AI-driven tools, is reshaping early PCa diagnosis toward a faster, more accessible, accurate, and reproducible imaging workflow [2, 3].

Genomic testing has become an essential component of contemporary PCa management, reflecting the growing recognition of hereditary contributions to both disease risk and therapeutic response. Mutations in genes involved in DNA repair pathways—such as BRCA1, BRCA2, ATM, and others are now known to confer higher lifetime risk, earlier onset, and more aggressive disease biology [26]. Identifying these alterations through germline testing not only informs familial counseling and targeted screening but also guides personalized treatment [27]. As a result, germline testing is increasingly integrated into routine care for men with high-risk, metastatic, or familial prostate cancer [28,29,30,31,32]. However, the implementation of genetic testing still has some limitations, including the impact on diagnostic and treatment pathways as well as their effect on patients’ perception of disease [5, 33].

Managing castration-resistant PCa remains a major challenge in urology. The recent introduction of multiple therapeutic strategies has expanded treatment options but also increased the complexity of selecting the optimal sequence [34,35,36]. In this context, balancing patient characteristics, survival outcomes, treatment-related toxicity, and patient-reported outcomes is crucial. Ongoing studies continue to explore different therapeutic combinations to better tailor treatment to the individual patient with CRPCa.

Natural language models (NLMs) are rapidly transforming urology by enhancing both clinical workflows and research productivity [37,38,39]. These models can efficiently process unstructured data, such as clinical notes, operative reports, and pathology narratives, enabling automated information extraction, risk stratification, and clinical decision support [40]. In clinical practice, NLMs are being applied to streamline documentation, assist with guideline-based care pathways and support patient communication through conversational interfaces. In research, they accelerate literature synthesis, data annotation, and hypothesis generation. While issues of validation, transparency, and data governance remain central, NLMs are becoming increasingly integrated across the urologic landscape, offering several tools to improve efficiency and support precision care.

All these important improvements in PCa and prostatic diseases diagnosis and management are still limited by the racial and ethnic disparities across the world. Racial and ethnic disparities play a significant role in the epidemiology, diagnosis, and outcomes of prostate cancer (PCa). Genetic susceptibility, variations in tumor biology, unequal access to screening, and delayed diagnosis all contribute to the observed gap. Socioeconomic status, cultural beliefs, and structural barriers further influence treatment choices and cancer outcomes [31, 41,42,43]. The actual literature is making efforts to understand and implement measures to diminish disparities across cultures.

The surgical management of BPH now includes several options, with laser enucleation and minimally invasive techniques (MISTs) gaining increasing attention in recent years [44,45,46,47,48,49,50,51,52,53]. Current guidelines consider all available laser technologies for enucleation to be equivalent; however, the strength of this recommendation remains weak due to the low overall quality of evidence [54, 55]. In the literature, the optimal laser type and surgical technique to maximize outcomes and minimize complications following endoscopic prostate enucleation are still debated. Whether the use of thulium, holmium, or thulium lasers leads to differences in functional results and safety profiles continues to be a subject of ongoing discussion [9, 10, 56]. On the other hand, minimally invasive surgical treatments are gaining a great consensus in the urological community, with the main objectives to improve symptoms and ejaculatory adverse events. However, few studies have evaluated MISTs versus laser prostatectomy or transurethral prostatic resection [54, 55].

In 2025, hundreds of manuscripts were evaluated by our editorial team. In this commentary, we present the best articles selected to highlight the hot topics of this year for “Prostate cancer and prostatic diseases

Statins and PCa: where do we stand?

The role of statins in PCa prevention and diagnosis remains controversial, largely due to the substantial heterogeneity and generally low quality of the available evidence. Amiri et al. conducted a retrospective case–control study of 3,481 patients to examine the relationship between statin use, genetic variation, and prostate cancer (PCa) risk [1]. The authors analysed a biopsy cohort, defining cases as men with positive biopsies and controls as those with negative results. Statin use was recorded at baseline, and single-nucleotide polymorphisms (SNPs) known to influence statin concentrations were assessed. The cohort had a median age of 62 years, with 32% of patients reporting statin use. Overall, PCa was diagnosed in 59% of men, and 45% had high-grade disease. After adjusting for baseline characteristics, statin use was not associated with a lower risk of overall or high-grade PCa. Interestingly, statin users carrying the GG genotype (n = 668; 24%) of rs10276036 demonstrated a significantly reduced PCa risk (HR 0.71, 95% CI 0.51–1.00), although no SNP reached genome-wide significance. This study contributes to the growing body of evidence suggesting a potential chemopreventive role for statins in PCa. Despite plausible biological mechanisms, existing studies often lack critical information, including statin type, exposure duration, concurrent risk factors, and genetic susceptibility. Well-designed prospective studies are still needed to clarify whether targeting lipid metabolism may offer a viable strategy for primary prevention in PCa.

Simplifying MRI pathways and defining its prognostic role

The implementation of multiparametric MRI has dramatically improved the diagnosis of csPCa however, the large number of examinations required, and the associated workload remain significant barriers to its widespread adoption. In this context, biparametric MRI together with AI advancements may help overcome some of these limitations by reducing acquisition time, cost, and complexity. Garcia-Becerra et al. performed an updated systematic review and bivariate meta-analysis to directly compare the diagnostic accuracy of Bp-MRI and Mp-MRI for detecting clinically CsPCa [2]. The review included 19 studies published between 2015 and 2022, encompassing 5173 patients who underwent both imaging modalities. Mp-MRI demonstrated a pooled sensitivity of 0.90 (95% CI 0.87–0.93) and specificity of 0.64 (95% CI 0.50–0.76), whereas Bp-MRI showed a pooled sensitivity of 0.89 (95% CI 0.85–0.92) and a higher specificity of 0.73 (95% CI 0.62–0.82). Meta-regression confirmed no statistically significant difference between Bp-MRI and Mp-MRI in terms of sensitivity or specificity, even when adjusting for covariates. Bp-MRI showed slightly better ability to reduce false positives (absolute reduction 9.8%), translating into fewer unnecessary biopsies. Overall, Bp-MRI appears to provide diagnostic accuracy comparable to Mp-MRI, with the advantage of shorter scan times, lower costs, and elimination of contrast-agent risks, making it a viable alternative, especially when DCE use is undesirable or impractical. The present study supports the recent results of the PRIME study, adding further evidence on the possible implementation of bp-MRI in clinical practice [22]. Impact of its implementation in real-life clinical practice is needed before definitive conclusions.

Introduction of AI in MRI pathways will soon be part of our daily practice. Zhao et al. conducted a systematic review and meta-analysis to assess the diagnostic performance of machine learning–based MRI models for detecting prostate cancer [3, 33]. Twelve studies published between 2019 and 2023 were included, encompassing 3474 patients and a combination of biparametric and multiparametric MRI sequences such as T2WI, DWI and ADC. Machine-learning MRI showed high accuracy in distinguishing benign from malignant lesions, with pooled sensitivity of 0.92 and specificity of 0.90, and an AUC of 0.96. Diagnostic performance for clinically significant prostate cancer (csPCa, Gleason ≥7) was lower but still robust, with pooled sensitivity 0.83, specificity 0.73, and AUC 0.86. Most studies were retrospective and single-center, and heterogeneity was substantial, reflecting variability in MRI protocols, segmentation methods, machine-learning architectures, and validation strategies. A key limitation highlighted by the authors is the poor generalizability of models trained on single-center datasets: external validation often led to performance drops, emphasizing the need for multi-institutional data harmonization and standardized pipelines. The use of ML in everyday clinical practice is steadily increasing; however, several limitations still hinder its widespread implementation, including issues of availability, reliability, cost, and legal considerations. Overall, AI should currently be viewed as a tool that supports clinicians by improving efficiency and reducing workload, rather than as a substitute for clinical expertise.

On the other side, the role of MRI in AS is always more prominent; however, the prognostic significance of a negative MRI is still a matter of debate [25]. Several studies suggest a negative MRI is associated with favorable genomics and a good prognosis. For instance, Klotz et al. conducted a long-term prospective study investigating the prognostic value of MRI visibility in men undergoing AS for prostate cancer. Among 530 patients who had at least one MRI and a median follow-up of 8.5 years, baseline imaging was classified as MRI-invisible (PI-RADS 1–2, 7.4%), equivocal (PI-RADS 3, 30%), or MRI-visible (PI-RADS 4–5, 62%). MRI invisibility was strongly associated with indolent disease. Only 5% of MRI-invisible cancers were upgraded during follow-up, none progressed to Grade Group ≥3, and no patient required treatment or developed metastases. In contrast, upgrading occurred in 34% of PI-RADS 3 patients and 52% of PI-RADS 4–5 patients, with corresponding treatment rates of 36% and 52%. Overall, MRI visibility was a powerful predictor of grade progression, treatment need, and long-term outcomes, supporting an image-guided AS strategy that restricts biopsies to MRI-targeted regions. This robust long-term dataset reinforces the concept that MRI-invisible cancers represent a biologically indolent subgroup. When integrated with emerging genomic data, these findings may allow better selection and follow-up of patients suitable for active surveillance [24]. However, widespread implementation still depends on access to high-quality imaging and expert radiologic interpretation, factors that remain variable across centers.

Genomics: a permanent shift in prostate cancer care

In the past years, genomics has gained particular attention in the diagnosis and management of PCa. Gebrael et al. conducted a large retrospective study of 276 patients with metastatic hormone-sensitive prostate cancer (mHSPC) treated with androgen deprivation therapy intensification (either an androgen-receptor pathway inhibitor (ARPI) or docetaxel) to evaluate whether baseline tumor genomic alterations predict survival outcomes. All patients underwent comprehensive genomic profiling (CGP) of primary or metastatic tissue using large next-generation sequencing panels (324–648 genes). The most frequent pathogenic alterations were TP53 (33%), TMPRSS2 fusions (31%), PTEN (17%), APC (10%), SPOP (9%), RB1 (6%), CDK12 (6%), and BRCA2 (5%). In multivariable analyses, TP53 (HR: 1,71; p < 0,05), RB1(HR: 2,32; p < 0,05), PTEN (HR: 1,74; p < 0,05), and BRCA2(HR: 2,64; p < 0,05) alterations were all associated with significantly shorter progression-free survival (PFS). For overall survival (OS), TP53, RB1, and PTEN remained independently prognostic, with RB1 showing the highest risk (HR 4.5; p < 0,05). Subgroup analyses demonstrated that these alterations were particularly detrimental among patients treated with ARPI, whereas CDK12 mutations predicted worse outcomes specifically in those receiving docetaxel. Overall, this is one of the largest datasets, suggesting that baseline genomic markers identify patients at higher risk of early progression despite intensified systemic therapy. The present study highlights the importance of genomic characterization to improve the management of patients at high risk of progression. Studies are slowly defining the best strategies based on genomic alteration, heading towards a real personalized medicine. Soon, big data combined with AI will move us toward the concept of digital twins, further enhancing the accuracy of precision medicine [57, 58].

Defining the best treatment sequence in castration resistant PCa

In recent years, the management of CRPCa has evolved substantially with the introduction of novel chemotherapy regimens, next-generation androgen receptor inhibitors, and advanced radiotherapy protocols. Despite these advances, the optimal sequencing of available treatments remains uncertain and continues to be a matter of active debate [34, 35]. The ARTO II trial is a prospective, multicenter study designed to further evaluate the role of stereotactic body radiotherapy (SBRT) in combination with androgen receptor pathway inhibitors in patients with oligometastatic or oligoprogressive castration-resistant prostate cancer (omCRPC) [36]. Francolini et al. conducted a subgroup analysis of the ARTO phase II randomized trial to evaluate whether SBRT remains beneficial after progression in patients with oligometastatic castration-resistant prostate cancer (omCRPC) receiving first-line abiraterone acetate plus prednisone. Overall, 63 were eligible for this analysis: 23 received SBRT directed at oligoprogressive sites (Group A), while 40 underwent second-line systemic therapy (Group B). Median follow-up after progression was 14.8 months. No significant difference was found in combined progression-free survival (PFS1 + PFS2): 45.9 months in Group A vs not reached in Group B (HR 0.63, p = 0.489). Overall survival (OS) was also similar, with a nonsignificant trend favoring SBRT (HR 0.50, p = 0.284). Of note, oligoprogression (e.g., 1–3 new lesions detected) was the predominant pattern of progression found in the vast majority of patients undergoing SBRT if compared to poliprogression (95.7% vs 4.3% of cases, respectively). Toxicity and treatment burden were not directly compared, but second-line chemotherapy is known to carry higher AE rates. This analysis suggests that SBRT after oligoprogression may offer outcomes comparable to second-line systemic therapy in carefully selected omCRPC patients, potentially delaying more toxic treatments. Still, patient selection biases and small sample size highlight the need for prospective trials integrating SBRT into modern treatment sequences. Hopefully, future trials will help define the best CRPCa candidates for SBRT.

AI advances in urology vs. persistent racial gaps: a reality check

The introduction of NLM has made AI accessible to everyday clinicians, significantly expanding its potential applications in urology. Zheng et al. developed and validated a natural language processing (NLP) system designed to automatically extract key shared-decision-making (SDM) concepts from PCa treatment consultations and to assess the quality of physician communication [7]. The study analysed 70 recorded consultations, using 50 transcripts (28,927 sentences) for model development and 20 for quality-evaluation validation. Manual coding identified nine SDM-relevant domains: tumor risk, pathology results, life expectancy (LE), cancer prognosis, urinary and erectile function, and three major treatment side effects (ED, UI, LUTS). Several machine-learning models were tested, with Random Forest achieving the highest or near-highest performance across domains, showing AUC values between 0.84 and 0.99 in internal validation. In the external validation set, the system ranked sentences by predicted topic concordance; the top 10 sentences reliably mirrored the highest-quality communication provided in the entire consultation. Accuracy for grading communication quality ranged from 80 to 100%, depending on the domain, with perfect accuracy for tumor risk and urinary incontinence. The authors propose that this technology can support both patient comprehension, by providing curated summaries, and physician feedback, offering a tool for auditing and improving SDM quality in PCa care. This study highlights how NLP can meaningfully support shared decision-making by standardizing complex communication assessments. However, broader external validation and integration with real-time automated transcription remain necessary before such tools can be routinely implemented in clinical practice.

Technological and AI advancements may further exacerbate existing ethnic and racial disparities in PCa management. Masterson et al. conducted a large, crowdsourced conjoint analysis involving 2046 men to explore how LE influences prostate cancer treatment preferences across racial and ethnic groups [8]. Participants, sampled to mirror the demographic distribution of a US prostate cancer population, completed iterative choice tasks comparing aggressive treatment versus conservative management. Decisions were evaluated across four trade-offs: tumor risk reduction, erectile dysfunction, urinary incontinence, and irritative urinary symptoms, while considering each participant’s LE calculated by the Prostate Cancer Comorbidity Index. Overall, longer LE significantly increased the likelihood of choosing aggressive therapy: each additional 5 years of LE increased odds by 17% (OR 1.17; 95% CI 1.12–1.22). Men typically favored aggressive treatment when LE exceeded 13 years, and conservative options when LE was ≤10 years. However, this pattern did not apply uniformly across groups. Among Black men, LE did not influence decisions (OR 0.97; p = 0.48): they consistently preferred aggressive treatment across all LE categories. Hispanic men, in contrast, showed increased preference for aggressive treatment only when LE > 10 years (OR 1.14; p = 0.019). This study highlights important cultural differences in how LE is interpreted in prostate cancer decision-making. Notably, Black men’s preference for aggressive therapy, regardless of LE, suggests that guideline-based “LE thresholds” may not align with patient-perceived risks. Further qualitative work is needed to understand these divergences and to develop culturally tailored communication strategies.

The era of laser enucleation and minimally invasive techniques

Laser enucleation techniques represent a cornerstone of contemporary surgical management for BPH. However, complications such as transient urinary incontinence and bleeding still limit their widespread adoption, leading researchers to investigate which laser works best and how to improve the surgical technique. Elmansy H et al. conducted a prospective randomized controlled trial comparing holmium laser enucleation using MOSES™ technology (M-HoLEP) with thulium fiber laser enucleation (ThuFLEP) for the treatment of symptomatic BPH [9]. The study enrolled 104 men with prostates ≥80 g, randomized 1:1, and evaluated perioperative outcomes and 12-month functional results. M-HoLEP showed significantly shorter enucleation times (50 vs 57.5 min; p < 0,05) and higher enucleation efficiency (1.97 vs 1.49 g/min; p < 0,05) than ThuFLEP. Immediate postoperative admissions for hematuria were markedly higher with ThuFLEP (30.8% vs 7.7%; p < 0,05). Early Clavien I complications were also more common in the ThuFLEP arm (42.3% vs 19.2%; p < 0,05). Despite these differences, functional outcomes at 1, 3, 6, and 12 months (IPSS, QoL, Qmax, PVR, and sexual function) were similar across both groups. The present study is clearly in line with the available studies on the subject, in terms of efficacy, reinforcing the concept that enucleation is enucleation. Safety data needs to be interpreted with caution, considering that the available evidence on the subject suggests equal safety of both lasers; a single surgeon design could explain the discrepancies [59,60,61].

Shifting to MISTs, comparative studies are better defining patient profiles for each technique. Manfredi et al. conducted a multicenter prospective comparative study evaluating Rezūm water-vapor therapy versus thulium laser enucleation (ThuLEP) in men with benign prostatic hyperplasia (BPH) and large prostates ≥80 mL [10]. A total of 246 patients were enrolled (126 ThuLEP, 120 Rezūm) with comparable baseline characteristics, aside from slightly higher age and PSA in the ThuLEP group. The primary endpoint was the change in IPSS-Total, assessed at 3, 6, and 12 months; secondary endpoints included IIEF-EF, MSHQ-EjD scores, and Clavien-Dindo ≥III complications. Both procedures produced significant and durable improvements in mean IPSS (12 months: ThuLEP: –14 vs Rezum –11,7; p < 0,05), mean Qmax (12 months: ThuLEP: + 13,2 vs Rezum + 9,7; p < 0,05), and mean PVR (12 months: ThuLEP: 33 cc vs Rezum 35 cc; p > 0,05). ThuLEP yielded statistically superior IPSS and Qmax outcomes at all time points; however, the differences never reached the Minimal Clinically Important Difference, indicating limited clinical relevance. Erectile function improved similarly in both groups, while Rezūm demonstrated clear superiority in ejaculation outcomes, showing marked improvements in MSHQ-EjD Function/Bother and a substantially lower rate of anejaculation at 3 months (4.2% vs. 88.9%; p < 0,05). Major complications were rare, with clot-retention requiring intervention occurring in 5 ThuLEP patients vs. 1 Rezūm patient. Dysuria was more frequent after Rezūm. This study reinforces the emerging role of Rezūm as a viable option for large prostates, particularly for patients prioritizing sexual preservation. In recent years, MISTs have gained substantial popularity, introducing the concept of first-line interventional therapy as an alternative to medical or surgical treatment, particularly for patients who wish to avoid sexual adverse effects [9, 62]. In this context, effective patient communication is of utmost importance to better counsel patients on the pros and cons of the different MISTs vs standard techniques and to optimize clinical outcomes [63].

Conclusions

Prevention in PCa remains the Achilles’ heel of urology, as no strong evidence-based preventive strategies are currently available. Technical innovations, particularly advances in artificial intelligence and MRI protocols, are gradually improving diagnostic pathways and identifying significant PCa. Moreover, genomic testing is reshaping prostate cancer management and treatment, moving the field towards personalized and precision medicine. Management of CRPC remains a challenge: integrating radiotherapy and medical treatment seems a feasible option to improve outcomes. Unfortunately, some of these advancements must be adopted with caution to avoid widening the already existing ethnic and racial disparities in PCa management. Similarly, technological progress in BPH surgery has expanded therapeutic options, although balancing patient preferences and expectations with the efficacy and durability of the available innovative treatments will remain a key challenge in the coming years.