Update on the epidemiology and biology of renal cortical neoplasms

Differentiation of benign from malignant solid renal lesions using CT-based radiomics and machine learning: comparison with radiologist interpretation

PURPOSE

To assess the performance of a machine learning model trained with contrast-enhanced CT-based radiomics features in distinguishing benign from malignant solid renal masses and to compare model performance with three abdominal radiologists.

METHODS

Patients who underwent intra-operative ultrasound during a partial nephrectomy were identified within our institutional database, and those who had pre-operative contrast-enhanced CT examinations were selected. The renal masses were segmented from the CT images and radiomics features were derived from the segmentations. The pathology of each mass was identified; masses were labeled as either benign [oncocytoma or angiomyolipoma (AML)] or malignant [clear cell, papillary, or chromophobe renal cell carcinoma (RCC)] depending on the pathology. The data were parsed into a 70/30 train/test split and a random forest machine learning model was developed to distinguish benign from malignant lesions. Three radiologists assessed the cohort of masses and labeled cases as benign or malignant.

RESULTS

148 masses were identified from the cohort, including 50 benign lesions (23 AMLs, 27 oncocytomas) and 98 malignant lesions (23 clear cell RCC, 44 papillary RCC, and 31 chromophobe RCCs). The machine learning algorithm yielded an overall accuracy of 0.82 for distinguishing benign from malignant lesions, with an area under the receiver operating curve of 0.80. In comparison, the three radiologists had significantly lower accuracies (p = 0.02) ranging from 0.67 to 0.75.

CONCLUSION

A machine learning model trained with CT-based radiomics features can provide superior accuracy for distinguishing benign from malignant solid renal masses compared to abdominal radiologists.

Differentiation of benign from malignant solid renal lesions with MRI-based radiomics and machine learning

BACKGROUND

Solid renal masses are often indeterminate for benignity versus malignancy on magnetic resonance imaging. Such masses are typically evaluated with either percutaneous biopsy or surgical resection. Percutaneous biopsy can be non-diagnostic and some surgically resected lesions are inadvertently benign.

PURPOSE

To assess the performance of ten machine learning (ML) algorithms trained with MRI-based radiomics features in distinguishing benign from malignant solid renal masses.

METHODS

Patients with solid renal masses identified on pre-intervention MRI were curated from our institutional database. Masses with a definitive diagnosis via imaging (for angiomyolipomas) or via biopsy or surgical resection (for oncocytomas or renal cell carcinomas) were selected. Each mass was segmented for both T2- and post-contrast T1-weighted images. Radiomics features were derived from the segmented masses for each imaging sequence. Ten ML algorithms were trained with the radiomics features gleaned from each MR sequence, as well as the combination of MR sequences.

RESULTS

In total, 182 renal masses in 160 patients were included in the study. The support vector machine algorithm trained on radiomics features from T2-weighted images performed superiorly, with an accuracy of 0.80 and an area under the curve (AUC) of 0.79. Linear discriminant analysis (accuracy = 0.84 and AUC = 0.77) and logistic regression (accuracy = 0.78 and AUC = 0.78) algorithms trained on T2-based radiomics features performed similarly. ML algorithms trained on radiomics features from post-contrast T1-weighted images or the combination of radiomics features from T2- and post-contrast T1-weighted images yielded lower performance.

CONCLUSION

Machine learning models trained with radiomics features derived from T2-weighted images can provide high accuracy for distinguishing benign from malignant solid renal masses.

CLINICAL IMPACT

Machine learning models derived from MRI-based radiomics features may improve the clinical management of solid renal masses and have the potential to reduce the frequency with which benign solid renal masses are biopsied or surgically resected.

Characterization of solid renal neoplasms using MRI-based quantitative radiomics features

PURPOSE

To assess the diagnostic value of magnetic resonance imaging (MRI)-based radiomics features using machine learning (ML) models in characterizing solid renal neoplasms, in comparison/combination with qualitative radiologic evaluation.

METHODS

Retrospective analysis of 125 patients (mean age 59 years, 67% males) with solid renal neoplasms that underwent MRI before surgery. Qualitative (signal and enhancement characteristics) and quantitative radiomics analyses (histogram and texture features) were performed on T2-weighted imaging (WI), T1-WI pre- and post-contrast, and DWI. Mann-Whitney U test and receiver-operating characteristic analysis were used in a training set (n = 88) to evaluate diagnostic performance of qualitative and radiomics features for differentiation of renal cell carcinomas (RCCs) from benign lesions, and characterization of RCC subtypes (clear cell RCC [ccRCC] and papillary RCC [pRCC]). Random forest ML models were developed for discrimination between tumor types on the training set, and validated on an independent set (n = 37).

RESULTS

We assessed 104 RCCs (51 ccRCC, 29 pRCC, and 24 other subtypes) and 21 benign lesions in 125 patients. Significant qualitative and quantitative radiomics features (area under the curve [AUC] between 0.62 and 0.90) were included for ML analysis. Models with best diagnostic performance on validation sets showed AUC of 0.73 (confidence interval [CI] 0.5-0.96) for differentiating RCC from benign lesions (using combination of qualitative and radiomics features); AUC of 0.77 (CI 0.62-0.92) for diagnosing ccRCC (using radiomics features), and AUC of 0.74 (CI 0.53-0.95) for diagnosing pRCC (using qualitative features).

CONCLUSION

ML models incorporating MRI-based radiomics features and qualitative radiologic assessment can help characterize renal masses.

Quantitative multiparametric MR analysis of small renal lesions: correlation with surgical pathology

PURPOSE

The purpose of the study is to evaluate the utility of apparent diffusion coefficient (ADC), chemical shift signal intensity index (SII), and contrast enhancement in distinguishing between benign lesions and renal cell carcinoma (RCC) and between subtypes of renal lesions.

METHODS

This retrospective study included 98 renal lesions (≤ 3 cm) on MRI with correlative surgical pathology. Scanner field strength, lesion location, and size were recorded. Two readers blinded to surgical pathology independently measured ADC ratio (ADC lesion/ADC non-lesion kidney), SII, and absolute/relative enhancement in the corticomedullary and nephrographic phases of contrast.

RESULTS

There were 76 malignant and 22 benign lesions. 42 RCC were clear cell (ccRCC), 19 papillary (pRCC), 5 chromophobe (cbRCC). Benign lesions included both solid and cystic lesions. Interreader agreement for all variables was good-excellent (ICC 0.70-0.91). There was no difference in ADC or SII between benign and malignant lesions. There was greater absolute corticomedullary enhancement of benign versus malignant lesions (150.0 ± 111.5 vs. 81.1 ± 74.8, p = 0.0115), which did not persist when excluding pRCC. For lesion subtype differentiation, ADC_ratio for pRCC was lower than benign lesions (0.74 ± 0.35 vs. 1.03 ± 0.46, p = 0.0246). ccRCC demonstrated greater SII than other RCC (0.09 ± 0.22 vs. 0.001 ± 0.26, p = 0.0412). Oncocytomas and angiomyolipoma (AML) showed greater absolute corticomedullary enhancement than ccRCC and pRCC (145.6 ± 65.2 vs. 107.2 ± 85.3, p = 0.043 and 186.2 ± 93.9 vs. 37.6 ± 35.3, p = 0.0108), respectively.

CONCLUSIONS

While corticomedullary-phase enhancement was a differentiating feature, quantitative metrics from diffusion and chemical shift imaging cannot reliably differentiate benign from malignant lesions. Quantitative assessment may be useful in differentiating some benign and malignant lesion subtypes.

Application of self-retaining bidirectional barbed absorbable suture in retroperito- neoscopic partial nephrectomy

OBJECTIVE

To investigate the safety and feasibility of self-retaining bidirectional barbed absorbable suture application in retroperitoneoscopic partial nephrectomy.

MATERIALS AND METHODS

From Sep 2011 and Aug 2012, 76 cases of retroperitoneoscopic partial nephrectomy were performed at our hospital. The patients were divided into two groups: self-retaining barbed suture (SRBS) group (n = 36) and non-SRBS group (n = 40). There was no significant difference in age, sex, tumor size and location between the two groups. Clinical data and outcomes were analyzed retrospectively.

RESULTS

All 76 cases of retroperitoneoscopic partial nephrectomy were successfully performed, without conversion to open surgery or serious intraoperative complications. In the SRBS group, the suture time, warm ischemia time and operation blood loss were significantly shorter than that of non-SRBS group (p < 0.01), and operation time and hospital stay were shorter than that of non-SRBS group (p < 0.05).

CONCLUSIONS

The application of self-retaining bidirectional barbed absorbable suture in retroperitoneoscopic partial nephrectomy could shorten suture time and warm ischemia time, with good safety and feasibility, worthy of being used in clinic.

Renal cortical tumors: use of multiphasic contrast-enhanced MR imaging to differentiate benign and malignant histologic subtypes

PURPOSE

To investigate the use of quantitative multiphasic contrast material-enhanced magnetic resonance (MR) imaging in differentiating between common benign and malignant histologic subtypes of renal cortical tumors.

MATERIALS AND METHODS

The institutional review board waived informed consent and approved this retrospective HIPAA-compliant study of 138 patients who underwent preoperative contrast-enhanced MR imaging during the period of January 2004-December 2008. At surgery, 152 renal tumors were identified (77 clear cell, 22 papillary, 18 chromophobe, and 10 unclassified carcinomas; 16 oncocytomas; nine angiomyolipomas). Three readers independently identified and measured the most-enhanced area in each tumor and placed corresponding regions of interest in similar positions on images from the precontrast, corticomedullary, nephrographic, and excretory phases. The percentage change in signal intensity (%SI change) between precontrast imaging and each postcontrast phase was calculated. Interreader agreement was evaluated by using the overall concordance correlation coefficient (OCC). A linear mixed-effects model was used to estimate and compare the trajectories of the means of log %SI change across all phases between the six histologic subtypes.

RESULTS

Interreader agreement was substantial to almost perfect (OCC, 0.77-0.88). The %SI change differed significantly between clear cell carcinomas and papillary and chromophobe carcinomas in all phases of enhancement (P < .0001-.0120). In addition, %SI change was significantly higher in angiomyolipomas than in clear cell carcinomas, but only in the corticomedullary phase (P = .0231). Enhancement did not differ significantly between clear cell carcinoma and oncocytoma in any phase (P = .2081-.6000).

CONCLUSION

Quantitative multiphase contrast-enhanced MR imaging offers a widely available, reproducible method to characterize several histologic subtypes of renal cortical tumors, although it does not aid differentiation between clear cell carcinomas and oncocytomas.

MicroRNA target site polymorphisms in the VHL-HIF1α pathway predict renal cell carcinoma risk

Renal cell carcinoma (RCC) accounts for ∼4% of all human malignancies and is the 9th leading cause of male cancer death in the United States. The purpose of this study was to determine the effect of variation within microRNA (miRNA)-binding sites of genes in the VHL-HIF1α pathway on RCC risk. We identified 429 miRNA-binding site single-nucleotide polymorphisms (SNPs) in 102 pathway genes and assessed 53 tagging-SNPs for 31 of these genes for risk in a case-control study consisting of 894 RCC cases and 1,516 controls. Results showed that five SNPs were significantly associated with RCC risk. The most significant finding was rs743409 in MAPK1. Under the additive model, the variant was associated with a 10% risk reduction (OR: 0.90, 95% CI, 0.77-0.98). Other significant findings were for SNPs in CDCP1, TFRC, and DEC1. Cumulative effects analysis showed that subjects carrying four or five unfavorable genotypes had a 2.14-fold increase in risk (95% CI, 1.03-4.43, P = 0.04) than those with no unfavorable genotypes. Potential higher-order gene-gene interactions were identified and categorized subjects into different risk groups. The OR of the high-risk group defined by two SNPs: CDCP1:rs6773576 (GG) and DEC1:rs10982724 (GG) was 4.46 times higher than that of low-risk reference group (95% CI, 1.31-15.08). Overall, our study provides the first evidence supporting a connection between miRNA-binding site SNPs within the VHL-HIF1α pathway and RCC risk. These novel genetic risk factors might help identify individuals at high risk to enable detection of tumors at an early, curable stage.

Perioperative, oncologic, and functional outcomes of laparoscopic renal cryoablation and open partial nephrectomy: a matched pair analysis

PURPOSE

To directly compare perioperative, oncologic, and functional outcomes of laparoscopic renal cryoablation and open partial nephrectomy using a matched pair analysis.

PATIENTS AND METHODS

A total of 41 patients who underwent laparoscopic cryoablation for an incidental, solid clinical T(1a)N(0)M(0) renal tumor were matched with 82 patients who received partial nephrectomy in cold ischemia, using optimal matching based on propensity scores, which were created on the basis of preoperative aspects and dimensions used for an anatomic classification of renal tumors (PADUA) score, preoperative glomerular filtration rate, age-adjusted Charlson comorbidity index, and sex. Median follow-up was 33.6 months.

RESULTS

No differences in the overall incidence of complications (cryoablation, 20%; partial nephrectomy, 17%; P=0.739) and grade of complications (P=0.424) were observed. After cryoablation, local recurrence developed in four patients with renal-cell carcinoma (n=35) after a median duration of 14 months (range 6-18  mos), but none after partial nephrectomy. The 3-year recurrence-free survival probabilities after laparoscopic renal cryoablation vs open partial nephrectomy were 83% vs 100%, respectively (P=0.015). The average decrease of estimated glomerular filtration rate during follow-up was 7.8±3.1  mL/min/1.73  m(2) after laparoscopic cryoablation and 9.8±2.3  mL/min/1.73  m(2) after open partial nephrectomy, which was not statistically significant (P=0.602).

CONCLUSIONS

Perioperative complications and renal functional outcomes of laparoscopic cryoablation and open partial nephrectomy are similar; however, laparoscopic cryoablation confers a substantially higher local recurrence risk of about 17% after 3 years. Therefore, laparoscopic renal cryoablation should be reserved for high-risk patients with decreased life expectancy. Careful patient counseling is advocated. Study limitations include the small sample size, the lack of randomization, and the short follow-up.

Epidemiology, clinical staging, and presentation of renal cell carcinoma

The increasing incidence of renal cell carcinoma over the past 2 decades can be partly explained by the expanding use of abdominal imaging. As a result, the most incident renal cancers today are small, localized, and asymptomatic. However, the well-documented rise in all stages of RCC calls into question the nature of these asymptomatic lesions. The expected "screening effect" of detecting RCC when it is small and localized, with subsequent decreases in disease-specific mortality, has not been observed. Disease-specific mortality is actually rising, especially in African American patients. Effective interventions aimed at reducing obesity, hypertension, and smoking may help in reducing the incidence of RCC in the future.

Cryoablation and radiofrequency for kidney tumor

The detection of incidental and asymptomatic renal cortical neoplasms has concomitantly increased with radiographic imaging use. The gold standard for treating small renal tumors includes open and laparoscopic partial nephrectomy. Ablative renal procedures intend to duplicate the excellent oncologic outcomes of partial nephrectomy, while decreasing complications and shortening hospitalization time and convalescence. Only short and medium-term data are available, but ablation with cryotherapy or radiofrequency probes may achieve these goals. Ablation also offers the advantages of a minimally invasive surgical approach, with a significantly lower complication rate than partial nephrectomy. Ablated lesions are typically left in situ. Leaving potentially malignant tissues in place, albeit in a nonviable condition, certainly represents a major change in surgical thinking processes. This article reviews the status of cryoablation and radiofrequency ablation, the two ablative modalities currently available for clinical application.

Obstinate cough as a sole presenting symptom of non-metastatic renal cell carcinoma

Renal cell carcinoma (RCC) causes many kinds of symptoms such as hypercalcemia, hypertension, polycythemia and fever. Here we describe a rare case of RCC presenting with a persistent cough. After radical nephrectomy, the obstinate cough disappeared. When the tumor recurred locally, the cough also recurred. Furthermore, the cough disappeared completely again after the removal of the recurrent tumor. Although all the clinical findings suggested that the RCC caused the cough, we could not identify a specific humoral substance responsible for the cough.