Search

Simple Summary: In this study, machine learning models were developed to predict lymph node invasion in Japanese prostate cancer patients, using a structured approach to improve predictive accuracy. Traditional methods, such as the Briganti nomogram, have been widely used but may lack flexibility and adaptability to diverse clinical datasets. Our machine learning models, especially Light Gradient-Boosting Machine and Random Forest, showed higher flexibility and accuracy compared to conventional models. Through a thorough data analysis and feature engineering, including the use of predictive features such as PSA levels and tumor stage, we enhanced model performance and interpretability. A decision curve analysis indicated significant clinical benefits, suggesting that these models can better identify candidates for lymph node dissection, thereby minimizing unnecessary procedures. However, validation with larger and multi-institutional datasets is needed to confirm the utility of these models in broader clinical settings. Background/Objectives: Extended pelvic lymph node dissection is a crucial surgical technique for managing intermediate to high-risk prostate cancer. Accurately predicting lymph node metastasis before surgery can minimize unnecessary lymph node dissections and their associated complications. This study assessed the efficacy of various machine learning models for predicting lymph node metastasis in a cohort of Japanese patients who underwent robot-assisted laparoscopic radical prostatectomy. Methods: Data from 625 patients who underwent extended pelvic lymph node dissection or standard dissection with lymph node metastasis between October 2010 and February 2023 were analyzed. Four machine learning models—Random Forest, Light Gradient-Boosting Machine, Logistic Regression, and Support Vector Machine—were used to predict lymph node metastasis. Their performance was assessed using receiver operating characteristic curves, a decision curve analysis, and predictive values at different thresholds. Results: Lymph node metastasis was observed in 34 patients (5.4%). The Light Gradient-Boosting Machine had the highest AUC of 0.924, followed by the Random Forest model with an AUC of 0.894. The decision curve analysis indicated substantial net benefits for both models, particularly at low threshold probabilities. The Light Gradient-Boosting Machine demonstrated superior accuracy, achieving 95.6% at the 0.05 threshold and 96.7% at the 0.10 threshold, outperforming other models and conventional nomograms in the validation dataset. Conclusion: Machine learning models, especially Light Gradient-Boosting Machine and Random Forest, show significant potential for predicting lymph node metastasis in prostate cancer, thereby aiding in reducing unnecessary surgical interventions. [ABSTRACT FROM AUTHOR]

Background: Surgical resection for pheochromocytoma (PCC) is still challenging. This study assessed the perioperative outcomes of adrenalectomy for PCC and investigated the risk factors for intraoperative hemodynamic instability (HI). Methods: This retrospective study included 571 patients with adrenal tumors who underwent adrenalectomy at Kobe University Hospital and other related hospitals between April 2008 and October 2023. The perioperative outcomes of laparoscopic adrenalectomy were compared between PCC (n = 92) and non‐PCC (n = 464) groups. In addition, we investigated several potential risk factors for intraoperative HI in patients with PCC (n = 107; open, n = 11; laparoscopic, n = 92; robot‐assisted, n = 4). Results: While patients with PCC had a significantly larger amount of blood loss in comparison to those with non‐PCC (mean, 70 and 30 mL, respectively; p = 0.004), no significant difference was observed in the rate of perioperative grade ≥III complications (1.1% vs. 0.6%; p = 0.516), and no perioperative mortality was observed in either group. A tumor size of ≥40 mm, with preoperative hypertension and urinary metanephrines at a level ≥3 times the upper limit of the normal value, were found to be significant predictors of HI, with odds ratios of 2.74 (p = 0.025), 3.91 (p = 0.005), and 3.83 (p = 0.004), respectively. Conclusions: Our data suggest that laparoscopic adrenalectomy for PCC may be as safe as that for other types of adrenal tumors and that large tumors and hormonally active disease may be risk factors for intraoperative HI. The optimal perioperative management for PCC with these risk factors should be established. [ABSTRACT FROM AUTHOR]

Objectives: We investigated poor prognosticators in advanced or unresectable urothelial carcinoma, focusing on renal parenchymal invasion (RPI). Methods: This study included 48 bladder cancer (BC) and 67 upper tract urothelial carcinoma (UTUC) patients treated with pembrolizumab from December 2017 to September 2022 at Kobe University Hospital. Medical records were retrospectively reviewed for clinical characteristics, objective response rate (ORR), progression‐free survival (PFS), and overall survival (OS). Multivariate analyses were performed using the Cox proportional hazard regression model to identify parameters associated with either PFS or OS. Results: Of 67 UTUC patients, 23 had RPI and 41 patients did not, while 3 cases could not be evaluated. Patients with RPI were predominantly elderly and had liver metastases. ORR for patients with RPI was 8.7%, while it was 19.5% for those without RPI. PFS was significantly shorter for patients with RPI compared with those without RPI. Patients with RPI had significantly shorter OS than those without RPI. On multivariate analysis, performance status (PS) ≥ 2, neutrophil‐lymphocyte ratio (NLR) ≥ 3, C‐reactive protein ≥0.3 mg/dL and RPI were independent prognostic factors for PFS. PS ≥ 2, NLR ≥ 3, visceral metastasis and RPI were independent prognostic factors for OS. UTUC patient OS was significantly shorter than BC patient OS, while no significant difference in PFS or OS was observed between BC patients and UTUC patients without RPI. Conclusions: RPI was a poor prognostic factor in advanced urothelial carcinoma treated with pembrolizumab, possibly resulting in a poorer prognosis for UTUC compared with BC. [ABSTRACT FROM AUTHOR]

Objectives: The objective of this study was to assess the clinical outcomes following combined treatment with pembrolizumab and axitinib as first‐line therapy for patients with advanced RCC. Methods: This study retrospectively included 47 consecutive Japanese patients who were diagnosed with advanced RCC and subsequently received pembrolizumab and axitinib between February 2020 and January 2022. Efficacy and safety of this combined therapy in these patients were comprehensively investigated. Results: The 47 included patients were classified into the following 3 groups by the IMDC system: favorable, 7 (14.9%); intermediate, 24 (51.1%) and poor, 16 (34.0%). Responses to this combined therapy in the 47 patients were as follows: CR, 8 (17.0%); PR, 20 (42.6%); SD, 16 (34.0%) and PD, 3 (6.4%); thus, the ORR was 59.6%. During the observation period, disease progression and death occurred in 19 (40.4%) and 9 (19.1%) patients, respectively, and the median PFS and OS were 18 months and not reached, respectively. Univariate analyses identified the following significant predictors for poor prognostic outcomes: lack of nephrectomy, liver metastasis, bone metastasis, elevated CRP and IMDC poor risk for PFS; and lack of nephrectomy, non‐CCC and elevated CRP for OS. AEs and those corresponding to grade ≥ 3 occurred in all (100%) and 30 (63.8%) patients, respectively. Conclusions: To our knowledge, this is the first study focusing on real‐world outcomes following pembrolizumab and axitinib for treatment‐naïve advanced Japanese RCC patients, which showed the efficacy and safety of this combined therapy being similar or even superior to those in clinical trial. [ABSTRACT FROM AUTHOR]

The introduction of tyrosine kinase inhibitors has revolutionized treatment strategies for metastatic renal cell carcinoma (RCC) and has improved survival rates. The number of patients with bone metastases from RCC requiring surgery will increase as survival rates improve. However, there is insufficient evidence to standardize the treatment of bone metastases after the introduction of targeted therapy for metastatic RCC. We aimed to determine the outcomes of palliative surgical treatment of bone metastases in the extremities of patients with metastatic RCC. We retrospectively reviewed 26 lesions from 17 patients who underwent surgery for extremity and acetabular bone metastases and were treated with targeted therapies for advanced RCC between 2008 and 2020. The median follow-up duration was 19 months (range, 4–76). We assessed the patients' activities of daily living, quality of life, and pain and analyzed their postoperative values relative to preoperative values. Postoperative overall survival (OS), local progression-free survival (LPFS), and the factors affecting them were evaluated using the Kaplan-Meier method and log-rank test. The 5-year OS and LPFS rates were 39.5% and 65.6%, respectively. The factors affecting OS were sex, Katagiri score, visceral metastases, and preoperative targeted therapy, while the factors affecting LPFS were pathologic fractures and surgical technique. In this study, the postoperative outcomes of palliative surgery for bone metastases from metastatic RCC were good. We suggest that systemic treatment should be prioritized over local control for advanced bone metastasis in RCC and surgery before pathological fracture should be performed for local control. [ABSTRACT FROM AUTHOR]