Recurrence prediction in clear cell renal cell carcinoma using machine learning of quantitative nuclear features.

The recurrence of non-metastatic renal cell carcinoma (RCC) may occur early or late after surgery. This study aimed to develop a recurrence prediction machine learning model based on quantitative nuclear morphologic features of clear cell RCC (ccRCC). We investigated 131 ccRCC patients who underwent nephrectomy (T1-3N0M0). Forty had recurrence within 5 years and 22 between 5 and 10 years; thirty-seven were recurrence-free during 5-10 years and 32 were for more than 10 years. We extracted nuclear features from regions of interest (ROIs) using a digital pathology technique and used them to train 5- and 10-year Support Vector Machine models for recurrence prediction. The models predicted recurrence at 5/10 years after surgery with accuracies of 86.4%/74.1% for each ROI and 100%/100% for each case, respectively. By combining the two models, the accuracy of the recurrence prediction within 5 years was 100%. However, recurrence between 5 and 10 years was correctly predicted for only 5 of the 12 test cases. The machine learning models showed good accuracy for recurrence prediction within 5 years after surgery and may be useful for the design of follow-up protocols and patient selection for adjuvant therapy.

Prediction of non-muscle invasive bladder cancer recurrence using machine learning of quantitative nuclear features.

Non-muscle invasive bladder cancer (NMIBC) generally has a good prognosis; however, recurrence after transurethral resection (TUR), the standard primary treatment, is a major problem. Clinical management after TUR has been based on risk classification using clinicopathological factors, but these classifications are not complete. In this study, we attempted to predict early recurrence of NMIBC based on machine learning of quantitative morphological features. In general, structural, cellular, and nuclear atypia are evaluated to determine cancer atypia. However, since it is difficult to accurately quantify structural atypia from TUR specimens, in this study, we used only nuclear atypia and analyzed it using feature extraction followed by classification using Support Vector Machine and Random Forest machine learning algorithms. For the analysis, 125 patients diagnosed with NMIBC were used; data from 95 patients were randomly selected for the training set, and data from 30 patients were randomly selected for the test set. The results showed that the support vector machine-based model predicted recurrence within 2 years after TUR with a probability of 90% and the random forest-based model with probability of 86.7%. In the future, the system can be used to objectively predict NMIBC recurrence after TUR.

Distinct effect of body mass index by sex as a prognostic factor in localized renal cell carcinoma treated with nephrectomy ~ data from a multi-institutional study in Japan ~.

BACKGROUND: We assessed the prognostic value of body mass index (BMI) in Asian patients with localized RCC who underwent nephrectomy. METHODS: A total of 665 patients who underwent nephrectomy for localized RCC were enrolled in the present study and divided into the two BMI groups: i.e., BMI < 25 in 463 (69.6%) and BMI > 25 in 202 (30.4%) patients. RESULTS: In total, there were 482 (72.5%) males and 183 (27.5%) females. Five-year cancer-specific survival (CSS) rates were significantly higher in increased BMI than the lower BMI group (97.1 and 92.5%: P = 0.007). When stratified by sex, significantly longer CSS in higher BMI was confirmed in males (5-year CSS of 92.7% in BMI < 25 and 98.1% in BMI > 25, p = 0.005), while there was no difference in CSS between BMI groups for female patients. Multivariable analysis exhibited that higher BMI was an independent predictor for favorable CSS in male (cox model: p = 0.041, Fine & Gray regression model: p = 0.014), but not in the female. Subgroup analysis for CSS revealed that favorable CSS with higher BMI was observed in patient subgroups of age < 65 (p = 0.019), clear cell histology (p = 0.018), and tumor size > 4 cm, p = 0.020) as well as male (p = 0.020). CONCLUSION: Our findings collected from the multi-institutional Japanese dataset demonstrated longer survival in patients with higher BMI than lower BMI for non-metastatic RCC treated with nephrectomy. Intriguingly, this finding was restricted to males, but not to females.

Adjuvant chemotherapy improves overall survival in patients with localized upper tract urothelial carcinoma harboring pathologic vascular invasion: a propensity score-matched analysis of multi-institutional cohort.

OBJECTIVE: Whether adjuvant chemotherapy (AC) for patients with upper tract urothelial carcinoma (UTUC) offers survival benefit is still controversial. To explore the impact of AC on overall survival (OS) of cN0M0 UTUC patients, we conducted a propensity score-matched analysis using the regression model, including pathologic features such as lymphatic and vascular invasion. METHODS: A multi-institutional cohort of 413 UTUC patient record was used. Propensity score matching was performed to reduce bias by potential confounding factors for survival, including pathologic features from the specimen of radical nephroureterectomy (RNU), RESULTS: Ninety-eight patients were identified as pair-matched groups (49 patients in RNU and 49 patients in RNU + AC). Kaplan-Meier curves demonstrated that a 5-year OS rate of 72.7% for patients treated with RNU + AC was significantly higher than 51.6% for those treated with RNU (p = 0.0156). On multivariate analysis, pathologic vascular invasion (HR 3.41, 95% CI 1.24-10.66, p = 0.0166) and administration of AC (HR 0.45, 95% CI 0.19-0.98, p = 0.0438) still remained as the significant predictors for OS. In patients with pathologic vascular invasion (51 of 98 patients), a significantly longer OS in RNU + AC groups was observed (median OS of 30 and 70 months in RNU and RNU + AC groups, respectively: p = 0.0432), whereas there was no significant difference in the OS between RNU (median OS: not reached) and RNU + AC (median OS: not reached) groups in patients without the invasion (p = 0.4549). CONCLUSION: The result indicates a significant benefit for OS by the administration of AC, and pathologic vascular invasion in the specimen of RNU could help the patient selection to better predict the effect of AC.

The CANLPH Score, an Integrative Model of Systemic Inflammation and Nutrition Status (SINS), Predicts Clinical Outcomes After Surgery in Renal Cell Carcinoma: Data From a Multicenter Cohort in Japan.

BACKGROUND: A myriad of studies have demonstrated the clinical association of systemic inflammatory and nutrition status (SINS) including C-reactive protein/albumin ratio (CAR), the neutrophil/lymphocyte ratio (NLR), and the platelet/hemoglobin ratio (PHR). This study aimed to investigate the predictive value of the score integrating these variables (CANLPH) in patients with renal cell carcinoma (RCC). METHODS: Using cohort data from a multi-institutional study, 757 of 1109 patients were retrospectively analyzed. The optimal cutoff value for outcome prediction of continuous variables in CAR, NLR, and PHR was determined and the CANLPH score was then calculated as the sum score of 0 or 1 by the cutoff value in each ratio. RESULTS: The median follow-up time was 76 months for the patients who survived (n = 585) and 31 months for those who died (n = 172). The Youden Index offered an optimal cutoff of 1.5 for CAR and 2.8 for NLR, and a higher value from the cutoff was assigned as a score of 1. The cutoff value of the PHR was defined as 2.1 for males and 2.3 for females. The patients were assigned a CANLPH score of 0 (47.2%), 1 (31.3%), 2 (13.1%), or 3 (8.5%). In the multivariate analysis, the CANLPH score served as an independent predictor of cancer-specific mortality in both localized and metastatic RCC. CONCLUSION: The score was well-correlated with clinical outcome for the RCC patients. Because this score can be concisely measured at the point of diagnosis, physicians may be encouraged to incorporate this model into the treatment for RCC.

C-reactive protein-albumin ratio as a prognostic factor in renal cell carcinoma - A data from multi-institutional study in Japan.

INTRODUCTION: The C-reactive protein to albumin ratio (CAR) has been shown to provide prognostic information in several cancers. The objective in the study is to examine the prognostic value of CAR in patients with RCC who underwent nephrectomy. MATERIAL AND METHODS: The record data from multi-institutional study of 1,028 patients was analyzed in the study. The cut-off value of the CAR was defined by receive operating characteristic (ROC) analysis. Overall survival (OS), cancer-specific survival (CSS), and recurrence-free survival (RFS) were evaluated, and univariate and multivariate analyses were conducted to assess the predictive value of the variables including CAR. RESULT: The optimal cut-off value of 0.073 in CAR was defined according to the ROC analysis. The AUC in CAR for CSS was greater than that of NLR and PLR, and that for RFS was also greater than GPS and mGPS. Multivariate analysis demonstrated that the CAR was an independent prognostic factor for OS (P < 0.001), CSS (P < 0.001) in total cohort and RFS (P = 0.029) in nonmetastatic cohort. CONCLUSION: The findings of the present study suggested that the preoperative CAR is an independent prognostic indicator of OS, CSS and RFS for patients with RCC. Since CAR can be assessed prior to surgery, clinicians should this take into account for the treatment decision making.

Correlation between number of renal cysts and aortic circumferences measured using autopsy material.

Although the presence of renal cysts has been reported to be associated with aortic aneurysm or dissection by imaging studies, an autopsy study has not been performed. Therefore, in our institute, recent consecutive adult autopsy cases (n=108, 64 males and 44 females) were reviewed. The circumferences and atherosclerosis ratios of both thoracic and abdominal aorta were individually measured and graded. The number of renal cysts was scored and graded. Age of subjects along with histories of smoking, hypertension, and diabetes mellitus were confirmed. Multiple linear regression analyses demonstrated that severity of atherosclerosis and the number of renal cysts were correlated with thoracic aortic circumference, while only the number of renal cysts was correlated with abdominal aortic circumference (p<0.05), which was more predominant in female subjects (p<0.05). Microscopically, significantly more dilated renal tubules (by Student's t-test, p<0.05) along with decreased stainability of basement membrane by Periodic acid-Schiff staining and immunostaining of type IV collagen were noted in background renal tissues in cases with numerous renal cysts than in age- and sex-matched controls without renal cysts (n=10 vs. 10). The present study suggests that a syndrome that affects both aorta and renal tubules may exist.