Purpose: Semantic segmentation is a fundamental part of the surgical application of deep learning. Traditionally, segmentation in vision tasks has been performed using convolutional neural networks (CNNs), but the transformer architecture has recently been introduced and widely investigated. We aimed to investigate the performance of deep learning models in segmentation in robot-assisted radical prostatectomy (RARP) and identify which of the architectures is superior for segmentation in robotic surgery. Materials and Methods: Intraoperative images during RARP were obtained. The dataset was randomly split into training and validation data. Segmentation of the surgical instruments, bladder, prostate, vas and seminal vesicle was performed using three CNN models (DeepLabv3, MANet, and U-Net++) and three transformers (SegFormer, BEiT, and DPT), and their performances were analyzed. Results: The overall segmentation performance during RARP varied across different model architectures. For the CNN models, DeepLabV3 achieved a mean Dice score of 0.938, MANet scored 0.944, and U-Net++ reached 0.930. For the transformer architectures, SegFormer attained a mean Dice score of 0.919, BEiT scored 0.916, and DPT achieved 0.940. The performance of CNN models was superior to that of transformer models in segmenting the prostate, vas, and seminal vesicle. Conclusions Deep learning models provided accurate segmentation of the surgical instruments and anatomical structures observed during RARP. Both CNN and transformer models showed reliable predictions in the segmentation task; however, CNN models may be more suitable than transformer models for organ segmentation and may be more applicable in unusual cases. Further research with large datasets is needed.

Purpose: With the recent rising interest in artificial intelligence (AI) in medicine, many studies have explored the potential and usefulness of AI in urological diseases. This study aimed to comprehensively review recent applications of AI in urologic oncology. Materials and Methods: We searched the PubMed-MEDLINE databases for articles in English on machine learning (ML) and deep learning (DL) models related to general surgery and prostate, bladder, and kidney cancer. The search terms were a combination of keywords, including both “urology” and “artificial intelligence” with one of the following: “machine learning,” “deep learning,” “neural network,” “renal cell carcinoma,” “kidney cancer,” “urothelial carcinoma,” “bladder cancer,” “prostate cancer,” and “robotic surgery.” Results: A total of 58 articles were included. The studies on prostate cancer were related to grade prediction, improved diagnosis, and predicting outcomes and recurrence. The studies on bladder cancer mainly used radiomics to identify aggressive tumors and predict treatment outcomes, recurrence, and survival rates. Most studies on the application of ML and DL in kidney cancer were focused on the differentiation of benign and malignant tumors as well as prediction of their grade and subtype. Most studies suggested that methods using AI may be better than or similar to existing traditional methods. Conclusions AI technology is actively being investigated in the field of urological cancers as a tool for diagnosis, prediction of prognosis, and decision-making and is expected to be applied in additional clinical areas soon. Despite technological, legal, and ethical concerns, AI will change the landscape of urological cancer management.

Purpose: In men with metastatic castration-resistant prostate cancer (mCRPC), new bone lesions are sometimes not properly categorized through a confirmatory bone scan, and clinical significance of the test itself remains unclear. This study aimed to demonstrate the performance rate of confirmatory bone scans in a real-world setting and their prognostic impact in enzalutamide-treated mCRPC. Materials and Methods: Patients who received oral enzalutamide for mCRPC during 2014-2017 at 14 tertiary centers in Korea were included. Patients lacking imaging assessment data or insufficient drug exposure were excluded. The primary outcome was overall survival (OS). Secondary outcomes included performance rate of confirmatory bone scans in a real-world setting. Kaplan-Meier analysis and multivariate Cox regression analysis were performed. Results: Overall, 520 patients with mCRPC were enrolled (240 [26.2%] chemotherapy-naïve and 280 [53.2%] after chemotherapy). Among 352 responders, 92 patients (26.1%) showed new bone lesions in their early bone scan. Confirmatory bone scan was performed in 41 patients (44.6%), and it was associated with prolonged OS in the entire population (median, 30.9 vs. 19.7 months; p < 0.001), as well as in the chemotherapy-naïve (median, 47.2 vs. 20.5 months; p=0.011) and post-chemotherapy sub-groups (median, 25.5 vs. 18.0 months; p=0.006). Multivariate Cox regression showed that confirmatory bone scan performance was an independent prognostic factor for OS (hazard ratio 0.35, 95% confidence interval, 0.18 to 0.69; p=0.002). Conclusion Confirmatory bone scan performance was associated with prolonged OS. Thus, the premature discontinuation of enzalutamide without confirmatory bone scans should be discouraged.

Purpose: Pathologic T3b (pT3b) prostate cancer, characterized by seminal vesicle invasion (SVI), exhibits variable oncological outcomes post–radical prostatectomy (RP). Identifying prognostic factors is crucial for patient-specific management. This study investigates the impact of bilateral SVI on prognosis in pT3b prostate cancer. Materials and Methods: We evaluated the medical records of a multi-institutional cohort of men who underwent RP for prostate cancer with SVI between 2000 and 2012. Univariate and multivariable analyses were performed using Kaplan-Meier analysis and covariate-adjusted Cox proportional hazard regression for biochemical recurrence (BCR), clinical progression (CP), and cancer-specific survival (CSS). Results: Among 770 men who underwent RP without neo-adjuvant treatment, median follow-up was 85.7 months. Patients with bilateral SVI had higher preoperative prostate-specific antigen levels and clinical T category (all p < 0.001). Extracapsular extension, tumor volume, lymph node metastasis (p < 0.001), pathologic Gleason grade group (p < 0.001), and resection margin positivity (p < 0.001) were also higher in patients with bilateral SVI. The 5-, 10-, and 15-year BCR-free survival rates were 23.9%, 11.7%, and 8.5%; CP-free survival rates were 82.8%, 62.5%, and 33.4%; and CSS rates were 96.4%, 88.1%, and 69.5%, respectively. The bilateral SVI group demonstrated significantly lower BCR-free survival rates, CP-free survival rates, and CSS rates (all p < 0.001). Bilateral SVI was independently associated with BCR (hazard ratio, 1.197; 95% confidence interval, p=0.049), CP (p=0.022), and CSS (p=0.038) in covariate-adjusted Cox regression. Conclusion Bilateral SVI is a robust, independent prognostic factor for poor oncological outcomes in pT3b prostate cancer.

Purpose: To find out the incidence and predictors for late high-grade genitourinary (GU) toxicity following salvage radiotherapy (SRT), we investigated the consecutive patients who were treated with SRT after radical prostatectomy. Materials and Methods: Patients who underwent SRT for biochemical recurrence after radical prostatectomy were reviewed. The incidence of GU toxicity was assessed and risk factors for grade ≥2 and ≥3 GU toxicity were evaluated. The STROBE (STrengthening the Reporting of OBservational studies in Epidemiology) guided the reporting of this study. Results: Among the total of 217 patients, 88 patients (40.5%) showed late grade ≥2 GU toxicity. The incidence of late grade ≥3 GU toxicity was 11.5%. The presence of grade ≥2 baseline GU dysfunction (hazard ratio [HR], 6.097; 95% confidence interval [CI], 3.280–11.333; p<0.001) and short interval (<1 year) from surgery to SRT (HR, 1.994; 95% CI, 1.182–3.365; p=0.01) were associated with late grade ≥2 GU toxicity. A short interval from surgery to SRT was an independent predictor of late grade ≥3 GU toxicity (HR, 2.975; 95% CI, 1.135–7.794; p=0.027). Conclusions The incidence of late high-grade GU toxicity was not uncommon after SRT. Thus, care should be taken when we consider SRT in patients with baseline urinary dysfunction and a short interval from surgery to SRT, to determine an optimal treatment strategy with balancing quality of life and oncologic outcome of patients.

Purpose: This study investigated the prognostic factors and cancer-specific survival (CSS) of patients who had renal cell carcinoma (RCC) with venous thrombus and underwent radical nephrectomy with thrombectomy (RNTx). Materials and Methods: From January 1990 to December 2022, we retrospectively reviewed the medical records of patients diagnosed with RCC with venous thrombus who underwent RNTx at a single tertiary medical center. Univariate and multivariable Cox proportional hazard regression analyses were conducted to identify significant prognostic factors affecting CSS. A Kaplan-Meier model was used to calculate CSS rates at 1, 3, and 5 years after RNTx. Results: We included 262 patients in the final analysis (median age, 59 years) with a median follow-up of 28 months. The 1-, 3-, and 5-year CSS rates were 84.1%, 62.5%, and 46.4%, respectively. Multivariable analysis revealed that pathologic T4 stage (hazard ratio [HR], 3.711; 95% confidence interval [CI], 1.599–8.611, p=0.002), pathologic N1 stage (HR, 2.371; 95% CI, 1.231–4.567; p=0.01), sarcomatoid differentiation (HR, 1.89; 95% CI, 1.027–3.477; p=0.041), and tumor necrosis (HR, 2.993; 95% CI, 1.132–7.914; p=0.027) were associated with CSS. Conclusions Approximately one-third of all RCC patients with venous thrombus remained disease-free, and half survived 5 years after RNTx. Sarcomatoid differentiation and the presence of tumor necrosis in pathology predicted poorer CSS outcomes in our study. Further retrospective studies are required to validate these findings.

Purpose: This study aimed to evaluate the impact of serum testosterone level before enzalutamide treatment in metastatic castration-resistant prostate cancer (mCRPC) for antitumor outcomes. Materials and Methods: Single-center, retrospective study including patients that treated with enzalutamide for mCRPC before and after docetaxel chemotherapy. Clinicopathological parameters including serum testosterone at initial enzalutamide use were examined. Prostate-specific antigen (PSA) response, progression-free survival (PFS), and cancer-specific survival (CSS) were the outcomes of interest. Logistic-regression analysis was done for discovering odds for PSA response. Cox-proportional model was applied for risk stratification for progression and cancer-specific death. Results: A total of 228 patients with mCRPC, treated with enzalutamide, both prechemotherapy and postchemotherapy, between 2011 and 2019 were included. One hundred sixty-two of patients (71.1%) experienced PSA decline over 50%. Median PFS and CSS were 5.4 and 13.2 months, respectively. Serum testosterone at initial enzalutamide use was the noble predictor for progression (hazard ratio [HR], 0.409; p=0.020) and cancer-specific death (HR, 0.454; p=0.033) in postchemotherapy group. No significant effect of serum testosterone in prechemotherapy group was detected. Time to CRPC, high-metastatic burden revealed as risk factors for PSA response, PFS, and CSS, both in prechemotherapy and postchemotherapy group. Conclusions High testosterone level at commencement of enzalutamide treatment was associated with a good prognosis in postdocetaxel setting, but not related to oncological outcomes in chemotherapy-naïve patients.

Purpose: To identify the risk factors leading to radical cystectomy in patients who had undergone nephroureterectomy (NUx). Materials and Methods: We retrospectively reviewed the medical records of patients with upper tract urothelial carcinoma who underwent NUx during 2011–2019 and excluded patients with metastatic cancer. In total 646 patients were included in this study; of these, 532 had no previous bladder cancer history. Follow-up was performed every 3 months for 2 years after NUx was administered, and recurrence was confirmed using cystoscopy, urine cytology, computed tomography, and chest radiography. Bladder recurrence was confirmed through biopsy, urine cytology, or radiologic examination. Univariate and multivariate Cox regression analyzes were performed for statistical analysis of risk factors leading to radical cystectomy in patients undergoing NUx. Results: Lymphovascular invasion (LVI) (hazard ratio [HR], 4.728; 95% confidence interval [CI], 1.463–15.570; p=0.011), previous transurethral resection of bladder tumor history (HR, 3.825; 95% CI, 1.164–12.571; p=0.027), and intravesical recurrence (IVR) within 6 months (HR, 3.733; 95% CI, 1.091–12.778; p=0.036) in patients undergoing NUx are predictors of radical cystectomy implementation. In a multivariate analysis of patients without bladder cancer history, bladder recurrence was identified as a predictor of radical cystectomy implementation, if it occurred within 6 months of NUx (HR, 8.608; 95% CI, 1.545–47.976; p=0.014). Conclusions LVI and IVR within 6 months and previous bladder cancer history are factors that can predict the need for radical cystectomy after NUx. Even in patients without bladder cancer history, early bladder recurrence within 6 months is a major predictor of radical cystectomy.

Purpose: To construct a urologic cancer database using a standardized, reproducible method, and to assess preliminary characteristics of this cohort. Materials and Methods: Patients with prostate, bladder, and kidney cancers who were enrolled with diagnostic codes in the electronic medical record (EMR) at Asan Medical Center from 2007–2016 were included. Research Electronic Data Capture (REDCap) was used to design the Asan Medical Center-Urologic Cancer Database (AMC-UCD). The process included developing a data dictionary, applying branching logic, mapping clinical data warehouse structures, alpha testing, clinical record summary testing, creating “standards of procedure,” importing data, and entering data. Descriptive statistics were used to identify rates of surgeries and numbers of patients. Results: Clinical variables (n=407) were selected to develop a data dictionary from REDCap. In total, 20,198 urologic cancer patients visited our institution from 2007–2016 (bladder cancer, 4,616; kidney cancer, 5,750; prostate cancer, 10,330). The overall numbers of patients and surgeries increased over time, with robotic surgeries rapidly growing over a decade. The most common treatment for urologic cancer was surgery, followed by chemotherapy and radiation therapy. Conclusions Using a standardized method, the AMC-UCD fosters multidisciplinary research. This constructed database provides access to clinical statistics to effectively assist research. Preliminary data should be refined through EMR chart review. The successful organization of data from 2007–2016 provides a framework for future periods of investigation and prospective models.

Antineoplastic Agents ; Cardiovascular Diseases ; Cohort Studies ; Comorbidity ; Diagnosis ; Drug Therapy ; Follow-Up Studies ; Gonadotropin-Releasing Hormone ; Humans ; Incidence ; Male ; Morinda ; Multivariate Analysis ; Myocardial Ischemia ; National Health Programs ; Observational Study ; Passive Cutaneous Anaphylaxis ; Prostate ; Prostatic Neoplasms ; Social Class