Korean Renal Cell Carcinoma Research Articles

Machine learning models for predicting the onset of chronic kidney disease after surgery in patients with renal cell carcinoma

BackgroundPatients with renal cell carcinoma (RCC) have an elevated risk of chronic kidney disease (CKD) following nephrectomy. Therefore, continuous monitoring and subsequent interventions are necessary. It is recommended to evaluate renal function postoperatively. Therefore, a tool to predict CKD onset is essential for postoperative follow-up and management.MethodsWe constructed a cohort using data from eight tertiary hospitals from the Korean Renal Cell Carcinoma (KORCC) database. A dataset of 4389 patients with RCC was constructed for analysis from the collected data. Nine machine learning (ML) models were used to classify the occurrence and nonoccurrence of CKD after surgery. The final model was selected based on the area under the receiver operating characteristic (AUROC), and the importance of the variables constituting the model was confirmed using the shapley additive explanation (SHAP) value and Kaplan-Meier survival analyses.ResultsThe gradient boost algorithm was the most effective among the various ML models tested. The gradient boost model demonstrated superior performance with an AUROC of 0.826. The SHAP value confirmed that preoperative eGFR, albumin level, and tumor size had a significant impact on the occurrence of CKD after surgery.ConclusionsWe developed a model to predict CKD onset after surgery in patients with RCC. This predictive model is a quantitative approach to evaluate post-surgical CKD risk in patients with RCC, facilitating improved prognosis through personalized postoperative care.

Machine learning based prediction for oncologic outcomes of renal cell carcinoma after surgery using Korean Renal Cell Carcinoma (KORCC) database

We developed a novel prediction model for recurrence and survival in patients with localized renal cell carcinoma (RCC) after surgery and a novel statistical method of machine learning (ML) to improve accuracy in predicting outcomes using a large Asian nationwide dataset, updated KOrean Renal Cell Carcinoma (KORCC) database that covered data for a total of 10,068 patients who had received surgery for RCC. After data pre-processing, feature selection was performed with an elastic net. Nine variables for recurrence and 13 variables for survival were extracted from 206 variables. Synthetic minority oversampling technique (SMOTE) was used for the training data set to solve the imbalance problem. We applied the most of existing ML algorithms introduced so far to evaluate the performance. We also performed subgroup analysis according to the histologic type. Diagnostic performances of all prediction models achieved high accuracy (range, 0.77–0.94) and F1-score (range, 0.77–0.97) in all tested metrics. In an external validation set, high accuracy and F1-score were well maintained in both recurrence and survival. In subgroup analysis of both clear and non-clear cell type RCC group, we also found a good prediction performance.

Real-World Study Evaluating Safety and Effectiveness of Axitinib in Korean Patients with Renal Cell Carcinoma after Failure of One Prior Systemic Therapy.

This post-marketing surveillance (PMS) study was conducted to monitor the usage of axitinib (Inlyta) in clinical practice of Korean patients with advanced renal cell carcinoma (RCC) with disease progression during or after a prior systemic therapy in real world. In this multicenter, observational study, patients indicated for oral axitinib 5 mg twice daily as second-line therapy for advanced RCC were followed up under routine clinical practices, and their safety and effectiveness outcomes were collected. Between 2012 and 2021, 125 patients were enrolled, and data from 111 patients were analyzed. Median age was 65 years (range, 30 to 84 years), 81% was male, and 110 (99%) had clear cell RCC. The median daily dose of axitinib was 10 mg (range, 4.36 to 15.95 mg) with a median administration period of 5.6 months (range, 15 to 750 days). Eighty-three percentage of patients experienced any grade of adverse events, 71% of which were related to study treatment, including diarrhea (36%), hypertension (21%), stomatitis (17%), decreased appetite (14%), palmar-plantar erythrodysesthesia syndrome (12%), and asthenia (11%). Most adverse events were generally well tolerated and manageable, with 13% of grade ≥ 3. Axitinib dose reduction was required in 20% of the adverse events and discontinuation in 8%. Median progression-free survival was 12.4 months (95% confidence interval [CI], 9.6 to 18.9). Objective responses were observed in 30% of patients (95% CI, 21 to 39) with 4% of complete response and 26% of partial response. No new safety signal was found in the present PMS study of Korean RCC patients. Axitinib showed consistent outcomes in terms of effectiveness and safety confirming that the drug is a valid option for second-line therapy in patients with advanced RCC in a real-world setting.

Machine learning-based prediction model for late recurrence after surgery in patients with renal cell carcinoma

BackgroundRenal cell carcinoma is characterized by a late recurrence that occurs 5 years after surgery; hence, continuous monitoring and follow-up is necessary. Prognosis of late recurrence of renal cell carcinoma can only be improved if it is detected early and treated appropriately. Therefore, tools for rapid and accurate renal cell carcinoma prediction are essential.MethodsThis study aimed to develop a prediction model for late recurrence after surgery in patients with renal cell carcinoma that can be used as a clinical decision support system for the early detection of late recurrence. We used the KOrean Renal Cell Carcinoma database that contains large-scale cohort data of patients with renal cell carcinoma in Korea. From the collected data, we constructed a dataset of 2956 patients for the analysis. Late recurrence and non-recurrence were classified by applying eight machine learning models, and model performance was evaluated using the area under the receiver operating characteristic curve.ResultsOf the eight models, the AdaBoost model showed the highest performance. The developed algorithm showed a sensitivity of 0.673, specificity of 0.807, accuracy of 0.799, area under the receiver operating characteristic curve of 0.740, and F1-score of 0.609.ConclusionsTo the best of our knowledge, we developed the first algorithm to predict the probability of a late recurrence 5 years after surgery. This algorithm may be used by clinicians to identify patients at high risk of late recurrence that require long-term follow-up and to establish patient-specific treatment strategies.

Machine Learning Approach to Predict the Probability of Recurrence of Renal Cell Carcinoma After Surgery: Prediction Model Development Study.

BackgroundRenal cell carcinoma (RCC) has a high recurrence rate of 20% to 30% after nephrectomy for clinically localized disease, and more than 40% of patients eventually die of the disease, making regular monitoring and constant management of utmost importance.ObjectiveThe objective of this study was to develop an algorithm that predicts the probability of recurrence of RCC within 5 and 10 years of surgery.MethodsData from 6849 Korean patients with RCC were collected from eight tertiary care hospitals listed in the KOrean Renal Cell Carcinoma (KORCC) web-based database. To predict RCC recurrence, analytical data from 2814 patients were extracted from the database. Eight machine learning algorithms were used to predict the probability of RCC recurrence, and the results were compared.ResultsWithin 5 years of surgery, the highest area under the receiver operating characteristic curve (AUROC) was obtained from the naïve Bayes (NB) model, with a value of 0.836. Within 10 years of surgery, the highest AUROC was obtained from the NB model, with a value of 0.784.ConclusionsAn algorithm was developed that predicts the probability of RCC recurrence within 5 and 10 years using the KORCC database, a large-scale RCC cohort in Korea. It is expected that the developed algorithm will help clinicians manage prognosis and establish customized treatment strategies for patients with RCC after surgery.

The prognostic role of preoperative serum albumin/globulin ratio in patients with non-metastatic renal cell carcinoma undergoing partial or radical nephrectomy

This multi-institutional study sought to clarify the association between the preoperative serum albumin/globulin ratio (AGR) and the prognosis of renal cell carcinoma (RCC) in a large cohort. This study encompassed eight institutions and 2,970 non-metastatic RCC patients who underwent a radical or partial nephrectomy from the Korean RCC (KORCC) database. A low AGR (1,143 patients; 38.5%) was defined as a preoperative AGR of less than 1.47 and a high AGR (1,827 patients; 61.5%) was defined as that 1.47 or greater. In the low AGR group, older age, female gender, the incidence of symptom presentation when diagnosed, diabetes, and hypertension was higher than in the high AGR group. Patients with low AGRs showed more progressive tumor stages with higher Fuhrman nuclear grades (all P-values < 0.05). Patients in the low AGR group had a significantly lower overall survival rate (OS) and recurrence-free survival rate (RFS) in the Kaplan–Meier curves (all P-values < 0.05). AGR was an independent prognostic factor for predicting the OS and RFS in the multivariate analysis (all P-values < 0.05). The preoperative AGR is approachable and economical to use clinically for estimating the prognosis of RCC patients treated with surgery.

Impact of preoperative thrombocytosis on prognosis after surgical treatment in pathological T1 and T2 renal cell carcinoma: results of a multi-institutional comprehensive study.

BackgroundThe prognostic significance of preoperative thrombocytosis (TC) in renal cell carcinoma (RCC) is not without some debate. The aim of the present multi-institutional study was to determine the association of preoperative TC with the clinicopathological features and prognosis of localized RCC patients who underwent surgery in a large cohort.MethodsA study involving 8 institutions, and 4,376 patients with pT1 and pT2 RCC from the Korean renal cell carcinoma (KORCC) database, was conducted. TC was defined as a platelet count ≥400,000/μL. Patients were divided into 2 groups based on the presence of preoperative TC. Clinicopathological variables and survival rates were compared between the 2 groups.ResultsOut of the 4,376 patients in the study, 106 (2.4%) had preoperative TC. Compared to patients without TC, these patients had a lower body mass index. Additionally, these patients had more advanced stage tumors with a higher Fuhrman grade, and higher incidence of symptoms at the time of diagnosis. Kaplan-Meier curves revealed that patients with TC had a significantly lower rate of recurrence-free survival (RFS). Furthermore, a lower rate of overall survival (OS) was exhibited amongst patients with TC. Multivariate analysis revealed that TC was an independent prognostic factor in terms of the RFS and OS.ConclusionsTC appeared to be an important prognostic determinant in localized RCC. Furthermore, preoperative platelet count may be clinically useful for risk stratification of patients with surgically treated localized RCC.

Clinical Significance of Subclassification of Papillary Renal Cell Carcinoma: Comparison of Clinicopathologic Parameters and Oncologic Outcomes Between Papillary Histologic Subtypes 1 and 2 Using the Korean Renal Cell Carcinoma Database

IntroductionThe aims of the present study were to compare the clinicopathologic characteristics between type 1 and type 2 papillary renal cell carcinoma (pRCC) and to evaluate the effect of the subclassification of pRCC on the oncologic outcomes after surgery. Materials and MethodsThe records of 274 patients with pRCC in the Korean renal cell carcinoma (KORCC) database were included for evaluation. Of the 274 patients, 118 had type 1 pRCC and 156 had type 2 pRCC. The patient characteristics, including clinicopathologic parameters, were investigated, and the oncologic outcomes were evaluated. ResultsThe mean patient age was significantly older in the type 2 pRCC group. Compared with type 1 pRCC tumors, type 2 pRCC tumors were more often localized to the renal hilum (P = .030). Patients with type 2 pRCC had a greater incidence of Fuhrman grade 3 and 4 tumors than those with type 1 pRCC (78.8% vs. 22.8; P < .001). Tumor necrosis and capsular invasion were more frequently found in type 2 pRCC (P = .008 and P = .007, respectively). At a mean follow-up period of 38.0 months (interquartile range, 11.8-57.3 months), the subclassification of pRCC did not influence the prognosis of patients with pRCC. ConclusionFrom the information available in the KORCC database, we identified significant differences in clinicopathologic variables, including age, Fuhrman grade, tumor location, tumor necrosis, and capsular invasion between type 1 and 2 pRCC. Although type 2 pRCC had more aggressive clinicopathologic characteristics, subclassification of pRCC did not affect the oncologic outcomes.

The establishment of KORCC (KOrean Renal Cell Carcinoma) database.

PurposeThe purpose of this article is to report establishment of the 1st Web-based database (DB) system to collect renal cell carcinoma (RCC) data in Korea.Materials and MethodsThe new Web-based DB system was established to collect basic demographic and clinicopahtological characteristics of a large cohort of patients with RCC in Korea. Data from a total of 6,849 patients were collected from 8 tertiary care hospitals that agreed to participate in organizing the Korean Renal Cell Carcinoma (KORCC) study group as of 1 July 2015. Basic demographic and clinicopathological characteristics were collected. The data of patients who underwent surgical treatments were analyzed to characterize Korean RCC.ResultsWe established the 1st Web-based DB of Korean RCC, a database comprising renal mass management cases from multiple centers in Korea. The data of 5,281 patients who underwent surgical management (mean follow-up, 32 months) were analyzed. The most common symptom was incidentally detected renal mass (76.9%). Clinical T1a was the most common (54.3%) stage and mean tumor size was 4.8±4.2 cm. Radical nephrectomy accounted for 62.7% of cases and an open approach was used in 50.7% and 52.2% of radical and partial nephrectomies, respectively. The 5-year overall, cancer-specific and recurrence-free survival rates were 88.1%, 92.2%, and 88.0%, respectively.ConclusionsWe report the 1st establishment of a Web-based DB system to collect RCC data in Korea. This DB system will provide a solid basis for the characterization of Korean RCC.

ALK-Positive Renal Cell Carcinoma in a Large Series of Consecutively Resected Korean Renal Cell Carcinoma Patients

BackgroundRecently, there have been a few reports of renal cell carcinoma (RCC) cases with anaplastic lymphoma kinase (ALK) gene fusion. In this study, we screened consecutively resected RCCs from a single institution for ALK protein expression by immunohistochemistry, and then we performed fluorescence in situ hybridization to confirm the ALK gene alteration in ALK immunohistochemistry-positive cases.MethodsWe screened 829 RCCs by ALK immunohistochemistry, and performed fluorescence in situ hybridization analysis using ALK dual-color break-apart rearrangement probe. Histological review and additional immunohistochemistry analyses were done in positive cases.ResultsOne ALK-positive case was found. Initial diagnosis of this case was papillary RCC type 2. This comprises 0.12% of all RCCs (1/829) and 1.9% of papillary RCCs (1/53). This patient was a 44-year-old male with RCC found during routine health check-up. He was alive without evidence of disease 12 years after surgery. The tumor showed a papillary and tubular pattern, and showed positivity for CD10 (focal), epithelial membrane antigen, cytokeratin 7, pan-cytokeratin, PAX-2, and vimentin.ConclusionsWe found the first RCC case with ALK gene rearrangement in Korean patients by ALK immunohistochemistry among 829 RCCs. This case showed similar histological and immunohistochemical features to those of previous adult cases with ALK rearrangement, and showed relatively good prognosis.

855P - Everolimus-Associated Non-Infections Pneumonitis (NIP) in Korean Patients with Renal Cell Carcinoma (RCC)

ABSTRACT Background Everolimus prolonged survival in advanced RCC patients who failed to VEGF inhibitors. However, nearly 13.5 to 30% of patients experienced NIP. Although a lower incidence of NIP was observed in Japanese patients from RECORD-1, the actual incidence of everolimus-associated NIP was unknown in Asian patients. Therefore, this study was undertaken to evaluate everolimus-associated NIP in Korean RCC patients treated with everolimus. Patients and methods Total 36 patients were enrolled at Seoul National University Hospital including advanced RCC patients who participated in REACT (N = 20) and in a trial for non-clear cell RCC (N = 16). NIP associated with everolimus was diagnosed using sequential computed tomography (CT) of the chest or bronchoscopy after an infectious origin and other causes of radiographic infiltrates were excluded. Clinico-pathologic findings were compared between NIP and non-NIP groups. Factors associated with NIP were identified using a binary logistic regression analysis. Results Overall, 10 (27.8%) of 36 RCC patients treated with everolimus developed NIP that was radiologically proven: ≥ grade 3, 6 (16.7%) patients; and grade 1, 4 (11.1%). Two patients died within 1 week after NIP diagnosis, while 8 recovered from NIP after cessation of everolimus (N = 4) and steroid use with drug interruption (N = 4). The mean cumulative dose of everolimus was 1,293 mg (range, 590-2420mg) at NIP diagnosis. There were no differences in NIP occurrence according to age, performance status, subtype (clear vs. non-clear), and underlying lung disease (Ps > .05). However, patients who experienced NIP had a better response (9 of 10 patients) than those without NIP (17 of 26) (P = .015). The response to everolimus remained predictive of NIP occurrence in multivariate analysis (P = .018). Conclusion NIP is relatively common (27.8%) in Korean patients with RCC who receive everolimus. Response to everolimus is an independent predictor for NIP development in advanced RCC patients. Disclosure All authors have declared no conflicts of interest.

Renal cell carcinoma in South Korea: A multicenter study

The incidence of renal cell carcinoma (RCC) in South Korea is steadily becoming similar to that in Western countries. This study summarizes the results of a 3-year multicenter survey of RCC in South Korea, conducted by the Korean Genitourinary Pathology Study Group. A total of 795 cases of RCC were collected from 20 institutes between 1995 and 1997, including 686 clear cell RCCs (86.3%), 58 papillary RCCS (7.30%), 49 chromphobe RCCs (6.16%), and 2 collecting duct RCCs (0.25%). At least 5 years of follow-up was available for 627 clear cell, 54 papillary, and 49 chromophobe RCCs. All subtypes presented most frequently with stage T3aN0M0 at the time of operation, and papillary RCCs demonstrated more frequent lymph node metastasis. Overall survival was not significantly related to the histological subtype (clear cell vs papillary, P = 0.8651; clear cell vs chromophobe, P = 0.0584; papillary vs chromophobe, P = 0.0743). For clear cell RCCs, statistically significant associations were found between overall survival and sex ( P = 0.0153), multiplicity ( P = 0.0461), necrosis ( P = 0.0191), age, sarcomatoid change, TNM stage, nuclear grade, and modality of treatment (all P <0.0001). Overall survival was significantly associated with tumor size ( P = 0.0307), nuclear grade ( P = 0.0235), multiplicity, sarcomatoid change, and TNM stage (all P <0.0001) for papillary RCCs and with the presence of sarcomatoid change ( P = 0.0281), nuclear grade ( P = 0.0015), treatment modality ( P = 0.0328), and TNM stage ( P <0.0001) for chromophobe RCCs. Age ( P = 0.0125), nodal stage ( P = 0.0010), and treatment modality ( P = 0.0001) were significant independent prognostic indicators for clear cell RCC on multivariate analysis. This is the first multicenter study of RCC in South Korea, demonstrating the general patterns and prognostic factors of Korean RCCs.

Renal cell carcinoma in South Korea: A multicenter study*1

The incidence of renal cell carcinoma (RCC) in South Korea is steadily becoming similar to that in Western countries. This study summarizes the results of a 3-year multicenter survey of RCC in South Korea, conducted by the Korean Genitourinary Pathology Study Group. A total of 795 cases of RCC were collected from 20 institutes between 1995 and 1997, including 686 clear cell RCCs (86.3%), 58 papillary RCCS (7.30%), 49 chromphobe RCCs (6.16%), and 2 collecting duct RCCs (0.25%). At least 5 years of follow-up was available for 627 clear cell, 54 papillary, and 49 chromophobe RCCs. All subtypes presented most frequently with stage T3aN0M0 at the time of operation, and papillary RCCs demonstrated more frequent lymph node metastasis. Overall survival was not significantly related to the histological subtype (clear cell vs papillary, P = 0.8651; clear cell vs chromophobe, P = 0.0584; papillary vs chromophobe, P = 0.0743). For clear cell RCCs, statistically significant associations were found between overall survival and sex (P = 0.0153), multiplicity (P = 0.0461), necrosis (P = 0.0191), age, sarcomatoid change, TNM stage, nuclear grade, and modality of treatment (all P <0.0001). Overall survival was significantly associated with tumor size (P = 0.0307), nuclear grade (P = 0.0235), multiplicity, sarcomatoid change, and TNM stage (all P <0.0001) for papillary RCCs and with the presence of sarcomatoid change (P = 0.0281), nuclear grade (P = 0.0015), treatment modality (P = 0.0328), and TNM stage (P <0.0001) for chromophobe RCCs. Age (P = 0.0125), nodal stage (P = 0.0010), and treatment modality (P = 0.0001) were significant independent prognostic indicators for clear cell RCC on multivariate analysis. This is the first multicenter study of RCC in South Korea, demonstrating the general patterns and prognostic factors of Korean RCCs.