Intraobserver and Interobserver Variability of Fuhrman and Modified Fuhrman Grading Systems for Conventional Renal Cell Carcinoma

Recommendations for Performance Evaluation of Machine Learning in Pathology: A Concept Paper From the College of American Pathologists

CONTEXT.—

Machine learning applications in the pathology clinical domain are emerging rapidly. As decision support systems continue to mature, laboratories will increasingly need guidance to evaluate their performance in clinical practice. Currently there are no formal guidelines to assist pathology laboratories in verification and/or validation of such systems. These recommendations are being proposed for the evaluation of machine learning systems in the clinical practice of pathology.

OBJECTIVE.—

To propose recommendations for performance evaluation of in vitro diagnostic tests on patient samples that incorporate machine learning as part of the preanalytical, analytical, or postanalytical phases of the laboratory workflow. Topics described include considerations for machine learning model evaluation including risk assessment, predeployment requirements, data sourcing and curation, verification and validation, change control management, human-computer interaction, practitioner training, and competency evaluation.

DATA SOURCES.—

An expert panel performed a review of the literature, Clinical and Laboratory Standards Institute guidance, and laboratory and government regulatory frameworks.

CONCLUSIONS.—

Review of the literature and existing documents enabled the development of proposed recommendations. This white paper pertains to performance evaluation of machine learning systems intended to be implemented for clinical patient testing. Further studies with real-world clinical data are encouraged to support these proposed recommendations. Performance evaluation of machine learning models is critical to verification and/or validation of in vitro diagnostic tests using machine learning intended for clinical practice.

A Review of the Paradigmatic Role of Adipose Tissue in Renal Cancer: Fat Measurement and Tumor Behavior Features

(1) Background: Renal-cell carcinoma (RCC) incidence has been steadily rising, with obesity identified as a potential risk factor. However, the relationship between obesity and RCC prognosis remains unclear. This systematic review aims to investigate the impact of different adipose tissue measurements on RCC behavior and prognosis. (2) Methods: A search of MEDLINE databases identified 20 eligible studies focusing on various fat measurements, including visceral adipose tissue (VAT), subcutaneous adipose tissue (SAT), perirenal adipose tissue (PRAT), and the Mayo adhesive probability (MAP) score. (3) Results: The review revealed conflicting findings regarding the association between adipose tissue measurements and RCC outcomes. While some studies suggested a protective role of certain fat deposits, particularly VAT, against disease progression and mortality, others reported contradictory results across different adipose metrics and RCC subtypes. (4) Conclusions: Methodological variations and limitations, such as retrospective designs and sample size constraints, pose challenges to standardization and generalizability. Further research is needed to understand these associations better and establish standardized approaches for adiposity assessment in RCC patients, which could inform clinical practice and therapeutic decision-making.

Artificial Intelligence in Renal Cell Carcinoma Histopathology: Current Applications and Future Perspectives

Renal cell carcinoma (RCC) is characterized by its diverse histopathological features, which pose possible challenges to accurate diagnosis and prognosis. A comprehensive literature review was conducted to explore recent advancements in the field of artificial intelligence (AI) in RCC pathology. The aim of this paper is to assess whether these advancements hold promise in improving the precision, efficiency, and objectivity of histopathological analysis for RCC, while also reducing costs and interobserver variability and potentially alleviating the labor and time burden experienced by pathologists. The reviewed AI-powered approaches demonstrate effective identification and classification abilities regarding several histopathological features associated with RCC, facilitating accurate diagnosis, grading, and prognosis prediction and enabling precise and reliable assessments. Nevertheless, implementing AI in renal cell carcinoma generates challenges concerning standardization, generalizability, benchmarking performance, and integration of data into clinical workflows. Developing methodologies that enable pathologists to interpret AI decisions accurately is imperative. Moreover, establishing more robust and standardized validation workflows is crucial to instill confidence in AI-powered systems' outcomes. These efforts are vital for advancing current state-of-the-art practices and enhancing patient care in the future.

A Weakly Supervised Deep Learning Model and Human-Machine Fusion for Accurate Grading of Renal Cell Carcinoma from Histopathology Slides

(1) Background: The Fuhrman grading (FG) system is widely used in the management of clear cell renal cell carcinoma (ccRCC). However, it is affected by observer variability and irreproducibility in clinical practice. We aimed to use a deep learning multi-class model called SSL-CLAM to assist in diagnosing the FG status of ccRCC patients using digitized whole slide images (WSIs). (2) Methods: We recruited 504 eligible ccRCC patients from The Cancer Genome Atlas (TCGA) cohort and obtained 708 hematoxylin and eosin-stained WSIs for the development and internal validation of the SSL-CLAM model. Additionally, we obtained 445 WSIs from 188 ccRCC eligible patients in the Clinical Proteomic Tumor Analysis Consortium (CPTAC) cohort as an independent external validation set. A human-machine fusion approach was used to validate the added value of the SSL-CLAM model for pathologists. (3) Results: The SSL-CLAM model successfully diagnosed the five FG statuses (Grade-0, 1, 2, 3, and 4) of ccRCC, and achieved AUCs of 0.917 and 0.887 on the internal and external validation sets, respectively, outperforming a junior pathologist. For the normal/tumor classification (Grade-0, Grade-1/2/3/4) task, the SSL-CLAM model yielded AUCs close to 1 on both the internal and external validation sets. The SSL-CLAM model achieved a better performance for the two-tiered FG (Grade-0, Grade-1/2, and Grade-3/4) task, with AUCs of 0.936 and 0.915 on the internal and external validation sets, respectively. The human-machine diagnostic performance was superior to that of the SSL-CLAM model, showing promising prospects. In addition, the high-attention regions of the SSL-CLAM model showed that with an increasing FG status, the cell nuclei in the tumor region become larger, with irregular contours and increased cellular pleomorphism. (4) Conclusions: Our findings support the feasibility of using deep learning and human-machine fusion methods for FG classification on WSIs from ccRCC patients, which may assist pathologists in making diagnostic decisions.

Whole-Slide Images and Patches of Clear Cell Renal Cell Carcinoma Tissue Sections Counterstained with Hoechst 33342, CD3, and CD8 Using Multiple Immunofluorescence

In recent years, there has been an increased effort to digitise whole-slide images of cancer tissue. This effort has opened up a range of new avenues for the application of deep learning in oncology. One such avenue is virtual staining, where a deep learning model is tasked with reproducing the appearance of stained tissue sections, conditioned on a different, often times less expensive, input stain. However, data to train such models in a supervised manner where the input and output stains are aligned on the same tissue sections are scarce. In this work, we introduce a dataset of ten whole-slide images of clear cell renal cell carcinoma tissue sections counterstained with Hoechst 33342, CD3, and CD8 using multiple immunofluorescence. We also provide a set of over 600,000 patches of size 256 × 256 pixels extracted from these images together with cell segmentation masks in a format amenable to training deep learning models. It is our hope that this dataset will be used to further the development of deep learning methods for digital pathology by serving as a dataset for comparing and benchmarking virtual staining models.

Use of High-Plex Data Reveals Novel Insights into the Tumour Microenvironment of Clear Cell Renal Cell Carcinoma

Simple Summary

Cancer is a complex ensemble of morphological and molecular features whose role is still unclear. Moreover, their role may change in different areas of the same tumour. Artificial intelligence (AI) allows pathologists to go beyond human perception and bias and may help better understand how these features influence disease progression. Furthermore, by capturing variation intrinsic to the tumour, AI may improve the accuracy of current prognostic tools, such as Leibovich Score (LS), in predicting patient outcome and response to therapy. For these reasons, we studied clear cell renal cell carcinoma (ccRCC) tissue in which molecular features and their coexpression in the same cell were quantified and mapped using AI-based image analysis software. We demonstrated a novel approach for investigating ccRCC and revealed new potential biomarkers of prognosis which may also be able to direct patients towards the most appropriate personalised therapy.

Abstract

Although immune checkpoint inhibitors (ICIs) have significantly improved the oncological outcomes, about one-third of patients affected by clear cell renal cell carcinoma (ccRCC) still experience recurrence. Current prognostic algorithms, such as the Leibovich score (LS), rely on morphological features manually assessed by pathologists and are therefore subject to bias. Moreover, these tools do not consider the heterogeneous molecular milieu present in the Tumour Microenvironment (TME), which may have prognostic value. We systematically developed a semi-automated method to investigate 62 markers and their combinations in 150 primary ccRCCs using Multiplex Immunofluorescence (mIF), NanoString GeoMx^® Digital Spatial Profiling (DSP) and Artificial Intelligence (AI)-assisted image analysis in order to find novel prognostic signatures and investigate their spatial relationship. We found that coexpression of cancer stem cell (CSC) and epithelial-to-mesenchymal transition (EMT) markers such as OCT4 and ZEB1 are indicative of poor outcome. OCT4 and the immune markers CD8, CD34, and CD163 significantly stratified patients at intermediate LS. Furthermore, augmenting the LS with OCT4 and CD34 improved patient stratification by outcome. Our results support the hypothesis that combining molecular markers has prognostic value and can be integrated with morphological features to improve risk stratification and personalised therapy. To conclude, GeoMx^® DSP and AI image analysis are complementary tools providing high multiplexing capability required to investigate the TME of ccRCC, while reducing observer bias.

Trends in disagreement with outside genitourinary pathology diagnoses at an academic center

AIMS

To evaluate the frequencies and types of disagreements in a contemporary urological second-opinion consult service in order to improve pathologist awareness.

METHODS

For 7 years ending 30 October 2021, records were kept of our department's total urologic outside consultation and disagreed-upon cases. Disagreements were categorized according to specimen type and nature of conflict. All grading and staging assignments used International Society of Urological Pathology (ISUP) criteria. Statistical analyses for each specimen type included the percent disagreement. Cohen's kappa analysis was done to measure interrater reliability on the prostate biopsies, prostatectomies, and the bladder biopsies/resections. In addition, for the prostate biopsies, the potential for change in treatment candidacy calculation (CTC), was assessed as sum of changes from cancer to non-malignant tissue or the reverse, plus changes from Gleason Grade group (GG)1 to GG ≥ 2 (3 +4 =7) or the reverse.

RESULTS

Overall mean disagreement rate for all specimens was 15.2%. The highest rate was among 1545 prostate biopsy cases, where 410 contained disagreements (26.5%). 118 (7.6%) met criteria for CTC: 10 cases were altered from cancer to non-cancer, 38 cases downgraded from GG≥ 2 to GG1, and 70 upgraded from GG1 to GG≥ 2. Second opinion downgraded the overall highest GG more often than it upgraded it, with downgrade:upgrade ratios of 64:37 for the GG1/GG2 threshold, 79:67 for the GG2/GG3, and 14:0 for the GG3/GG4. 146 specimen parts had disagreements as to cancer vs. suspicious vs. benign, with 85 undercalled and 61 overcalled. Other rates of disagreement included: prostatectomy 34/198 (17.2%); bladder resection or biopsy 68/591 (11.5%); kidney 27/175 (15.4%); and orchiectomy 9/82 (11.0%). In bladder specimens, overgrading was 6X more frequent than undergrading; and overstaging muscularis propria invasion was 6X more frequent than understaging.

CONCLUSIONS

The review of uropathologic materials before definitive therapy can lead to changes that impact clinical decisions significantly. As an example, for prostate biopsies, candidacy for active surveillance versus definitive treatment hinges on GG1 versus 2 and this distinction constituted most CTC cases. The above findings highlight aspects of urological pathology to be emphasized to residents in training, and pathologists in practice.

Long Non-Coding RNAs as Novel Biomarkers in the Clinical Management of Papillary Renal Cell Carcinoma Patients: A Promise or a Pledge?

Papillary renal cell carcinoma (pRCC) represents the second most common subtype of renal cell carcinoma, following clear cell carcinoma and accounting for 10-15% of cases. For around 20 years, pRCCs have been classified according to their mere histopathologic appearance, unsupported by genetic and molecular evidence, with an unmet need for clinically relevant classification. Moreover, patients with non-clear cell renal cell carcinomas have been seldom included in large clinical trials; therefore, the therapeutic landscape is less defined than in the clear cell subtype. However, in the last decades, the evolving comprehension of pRCC molecular features has led to a growing use of target therapy and to better oncological outcomes. Nonetheless, a reliable molecular biomarker able to detect the aggressiveness of pRCC is not yet available in clinical practice. As a result, the pRCC correct prognosis remains cumbersome, and new biomarkers able to stratify patients upon risk of recurrence are strongly needed. Non-coding RNAs (ncRNAs) are functional elements which play critical roles in gene expression, at the epigenetic, transcriptional, and post-transcriptional levels. In the last decade, ncRNAs have gained importance as possible biomarkers for several types of diseases, especially in the cancer universe. In this review, we analyzed the role of long non-coding RNAs (lncRNAs) in the prognosis of pRCC, with a particular focus on their networking. In fact, in the competing endogenous RNA hypothesis, lncRNAs can bind miRNAs, resulting in the modulation of the mRNA levels targeted by the sponged miRNA, leading to additional regulation of the target gene expression and increasing complexity in the biological processes.

Renal Oncocytoma: The Diagnostic Challenge to Unmask the Double of Renal Cancer

Renal oncocytoma represents the most common type of benign neoplasm that is an increasing concern for urologists, oncologists, and nephrologists due to its difficult differential diagnosis and frequent overtreatment. It displays a variable neoplastic parenchymal and stromal architecture, and the defining cellular element is a large polygonal, granular, eosinophilic, mitochondria-rich cell known as an oncocyte. The real challenge in the oncocytoma treatment algorithm is related to the misdiagnosis due to its resemblance, at an initial radiological assessment, to malignant renal cancers with a completely different prognosis and medical treatment. Unfortunately, percutaneous renal biopsy is not frequently performed due to the possible side effects related to the procedure. Therefore, the majority of oncocytoma are diagnosed after the surgical operation via partial or radical nephrectomy. For this reason, new reliable strategies to solve this issue are needed. In our review, we will discuss the clinical implications of renal oncocytoma in daily clinical practice with a particular focus on the medical diagnosis and treatment and on the potential of novel promising molecular biomarkers such as circulating microRNAs to distinguish between a benign and a malignant lesion.

Active Surveillance of Renal Masses: The Role of Radiology

Active surveillance of renal masses, which includes serial imaging with the possibility of delayed treatment, has emerged as a viable alternative to immediate therapeutic intervention in selected patients. Active surveillance is supported by evidence that many benign masses are resected unnecessarily, and treatment of small cancers has not substantially reduced cancer-specific mortality. These data are a call to radiologists to improve the diagnosis of benign renal masses and differentiate cancers that are biologically aggressive (prompting treatment) from those that are indolent (allowing treatment deferral). Current evidence suggests that active surveillance results in comparable cancer-specific survival with a low risk of developing metastasis. Radiology is central in this. Imaging is used at the outset to estimate the probability of malignancy and degree of aggressiveness in malignant masses and to follow up masses for growth and morphologic change. Percutaneous biopsy is used to provide a more definitive histologic diagnosis and to guide treatment decisions, including whether active surveillance is appropriate. Emerging applications that may improve imaging assessment of renal masses include standardized assessment of cystic and solid masses and radiomic analysis. This article reviews the current and future role of radiology in the care of patients with renal masses undergoing active surveillance.

Comparison of Different Machine Models Based on Contrast-Enhanced Computed Tomography Radiomic Features to Differentiate High From Low Grade Clear Cell Renal Cell Carcinomas

Purpose

This study was to investigate the role of different radiomics models with enhanced computed tomography (CT) scan in differentiating low from high grade renal clear cell carcinomas.

Materials and Methods

CT data of 190 cases with pathologically confirmed renal cell carcinomas were collected and divided into the training set and testing set according to different time periods, with 122 cases in the training set and 68 cases in the testing set. The region of interest (ROI) was delineated layer by layer.

Results

A total of 402 radiomics features were extracted for analysis. Six of the radiomic parameters were deemed very valuable by univariate analysis, rank sum test, LASSO cross validation and correlation analysis. From these six features, multivariate logistic regression model, support vector machine (SVM), and decision tree model were established for analysis. The performance of each model was evaluated by AUC value on the ROC curve and decision curve analysis (DCA). Among the three prediction models, the SVM model showed a high predictive efficiency. The AUC values of the training set and the testing set were 0.84 and 0.83, respectively, which were significantly higher than those of the decision tree model and the multivariate logistic regression model. The DCA revealed a better predictive performance in the SVM model that possessed the highest degree of coincidence.

Conclusion

Radiomics analysis using the SVM radiomics model has highly efficiency in discriminating high- and low-grade clear cell renal cell carcinomas.

Is preoperative neutrophil-to-lymphocyte ratio a red flag which can predict high-risk pathological characteristics in renal cell carcinoma?

Introduction

Renal cell carcinoma (RCC) is known to invoke both immunological and inflammatory responses. While the neutrophils mediate the tumor-induced inflammatory response, the lymphocytes bring about the various immunological events associated with it. The neutrophil-to-lymphocyte ratio (NLR) is a simple indicator of this dual response. We investigated the association between preoperative NLR and histopathological prognostic variables of RCC intending to find out whether it can be of value as a red flag capable of alerting the clinician as to the biological character of the tumor under consideration.

Methods

Preoperative NLR and clinicopathological variables, namely histological subtype, nuclear grade, staging, lymphovascular invasion, capsular invasion, tumor necrosis, renal sinus invasion, and sarcomatoid differentiation of 60 patients who underwent radical or partial nephrectomy, were analyzed to detect the association between the two.

Results

We found that mean preoperative NLR was significantly higher in clear-cell carcinomas (3.25 ± 0.29) when compared with nonclear-cell carcinomas (2.25 ± 0.63). There was a linear trend of NLR rise as the stage of the disease advanced. A significant rise in preoperative NLR was noted in tumors with various high-risk histopathological features such as tumor size, capsular invasion, tumor necrosis, and sarcomatoid differentiation.

Conclusion

Preoperative measurement of NLR is a simple test which may provide an early clue of high-risk pathological features of renal cell cancer.

Preoperative Predicting the WHO/ISUP Nuclear Grade of Clear Cell Renal Cell Carcinoma by Computed Tomography-Based Radiomics Features

Nuclear grade is important for treatment selection and prognosis in patients with clear cell renal cell carcinoma (ccRCC). This study aimed to determine the ability of preoperative four-phase multiphasic multidetector computed tomography (MDCT)-based radiomics features to predict the WHO/ISUP nuclear grade. In all 102 patients with histologically confirmed ccRCC, the training set (n = 62) and validation set (n = 40) were randomly assigned. In both datasets, patients were categorized according to the WHO/ISUP grading system into low-grade ccRCC (grades 1 and 2) and high-grade ccRCC (grades 3 and 4). The feature selection process consisted of three steps, including least absolute shrinkage and selection operator (LASSO) regression analysis, and the radiomics scores were developed using 48 radiomics features (10 in the unenhanced phase, 17 in the corticomedullary (CM) phase, 14 in the nephrographic (NP) phase, and 7 in the excretory phase). The radiomics score (Rad-Score) derived from the CM phase achieved the best predictive ability, with a sensitivity, specificity, and an area under the curve (AUC) of 90.91%, 95.00%, and 0.97 in the training set. In the validation set, the Rad-Score derived from the NP phase achieved the best predictive ability, with a sensitivity, specificity, and an AUC of 72.73%, 85.30%, and 0.84. We constructed a complex model, adding the radiomics score for each of the phases to the clinicoradiological characteristics, and found significantly better performance in the discrimination of the nuclear grades of ccRCCs in all MDCT phases. The highest AUC of 0.99 (95% CI, 0.92-1.00, p < 0.0001) was demonstrated for the CM phase. Our results showed that the MDCT radiomics features may play a role as potential imaging biomarkers to preoperatively predict the WHO/ISUP grade of ccRCCs.

Tumor Necrosis Adds Prognostically Significant Information to Grade in Clear Cell Renal Cell Carcinoma: A Study of 842 Consecutive Cases From a Single Institution

Tumor necrosis has been shown to be an independent predictor of adverse outcome in renal cell carcinoma. A modification of the International Society of Urological Pathology (ISUP) grading system for renal cell carcinomas has recently been proposed, which incorporates the presence of tumor necrosis into grade. The investigators proposing this system found that necrosis added significant prognostic information to ISUP grade. We attempted to describe our experience with the effect of tumor necrosis in relationship to nuclear grade by reviewing the slides from a large consecutive series of localized clear cell renal cell carcinomas from our institution and obtaining long-term clinical follow-up information (overall survival). Of the 842 clear cell renal cell carcinomas reviewed, 265 (31.5%) were ISUP grade 1 or 2, 437 (51.9%) were ISUP grade 3, and 140 (16.6%) were ISUP grade 4. Tumor necrosis was present in 177 (21%) cases. Five hundred and forty-seven (64.9%) cases were stage pT1, 83 (9.9%) were stage pT2, 193 (22.9%) were stage pT3a, and 19 (2.3%) were pT3b or higher. Median follow-up was 73.2 months (range 0.12 to 273.6), and 310 (36.8%) patients died. On univariable analysis, there was no significant difference in outcome for tumors of ISUP grades 1 to 3. After adjustment for age, tumor stage, and tumor size, ISUP grade 4 and necrosis were significant predictors of overall survival on multivariable analysis. When the recently proposed modified grading system incorporating tumor necrosis was applied to our data, there was no significant difference in overall survival between patients with modified grade 1 tumors and those with modified grade 2 tumors (P=0.31); however, there was a statistically significant difference between patients with modified grade 1 or 2 tumors and those with modified grade 3 tumors (P=0.04),and a substantial difference in outcome between those with modified grade 3 and modified grade 4 tumors (P<0.001). When a recursive partitioning approach was applied to our data, patients of a given ISUP grade could be further prognostically separated according to the presence or absence of necrosis and could be divided into 3 statistically significant prognostic groups: (1) non-necrotic ISUP grade 1 to 3 tumors, (2) ISUP grade 1 to 3 tumors with necrosis and ISUP grade 4 tumors with <10% necrosis, and (3) ISUP grade 4 tumors with >10% necrosis. In conclusion, our study shows that tumor necrosis adds additional prognostic information to ISUP grade and that quantification of necrosis can further stratify patients with ISUP grade 4 tumors.

Critical analysis of a simplified Fuhrman grading scheme for prediction of cancer specific mortality in patients with clear cell renal cell carcinoma--Impact on prognosis

INTRODUCTION AND OBJECTIVES

The traditional 4-tiered Fuhrman grading system (FGS) is widely accepted as histopathological classification for clear cell renal cell carcinoma (ccRCC) and has shown prognostic value. As intra- and inter-observer agreement are sub-optimal, simplified 2- or 3-tiered FGSs have been proposed. We aimed to validate these simplified 2- or 3-tiered FGSs for prediction of cancer-specific mortality (CSM) in a large study population from 2 European tertiary care centers.

METHODS

We identified and followed-up 2415 patients with ccRCC who underwent radical or partial nephrectomy in 2 European tertiary care centers. Univariable and multivariable analyses and prognostic accuracy analyses were performed to evaluate the ability of several simplified FGSs (i.e. grades I + II vs., grades III + IV, grades I + II vs. grade III and grade IV) to predict CSM.

RESULTS

Independent predictor status in multivariate analyses was proved for the simplified 2-tiered FGS (high-grade vs. low-grade), for the simplified 3-tiered FGS (grades I + II vs. grade III and grade IV) as well as for the traditional 4-tiered FGS. The prognostic accuracy of multivariable models of 77% was identical for all tested models. Prognostic accuracy of the model without FG was 75%.

CONCLUSIONS

A simplified 2- or 3-tiered FGS could predict CSM as accurate as the traditional 4-tiered FGS in a large European study population. Application of new simplified 2- or 3-tiered FGS may reduce inter-observer-variability and facilitate clinical practice without compromising the ability to predict CSM in ccRCC patients after radical or partial nephrectomy.

Comparison of utility of tumor size and apparent diffusion coefficient for differentiation of low- and high-grade clear-cell renal cell carcinoma

BACKGROUND

There is a significant correlation between tumor size and tumor grade for clear-cell renal cell carcinoma (RCC) in pathology. Thus, apparent diffusion coefficient (ADC) of clear-cell RCC might be influenced by tumor size.

PURPOSE

To compare the utility of tumor size and ADC for distinguishing low-grade from high-grade clear-cell RCC.

MATERIAL AND METHODS

Forty-nine patients undergoing preoperative magnetic resonance imaging were retrospectively assessed. ADC values were calculated using b-value combinations of 0 and 800 s/mm(2) at 1.5 T. Two radiologists in consensus measured ADC values via small region of interest (ROI) (mean ROI area, 88.8 mm(2); range, 80-108 mm(2)) placement on an area of solid tumor on a single slice. Maximum tumor diameter was measured at the maximum tumor area. A single pathologist reviewed all pathological slides to determine the nuclear grade according to the Fuhrman classification. The utility of ADC, tumor size, and ADC/size ratio for distinguishing low-grade from high-grade tumors was assessed. Receiver-operating characteristic (ROC) analysis and regression analysis of the each index were performed. The correlation between ADC and tumor size was also investigated.

RESULTS

The 49 clear-cell RCC included 34 low-grade and 15 high-grade tumors. The differences of ADC, tumor size, and ADC/size ratio between high-grade and low-grade tumors were statistically significant (P <0.05). The area under the ROC curve of ADC, tumor size, and ADC/size ratio were 0.802, 0.763, and 0.804 respectively. However, using regression analysis, only ADC (P <0.05) was statistically significant index as independent risk factors for high-grade clear-cell RCC. Moreover, weak significant correlation was observed between tumor size and ADC (R(2) = 0.3865, P <0.01).

CONCLUSION

There was a weak significant correlation between tumor size and ADC value of clear-cell RCC. Using ROC and regression analysis, ADC was statistically significant index for distinguishing low-grade from high-grade clear-cell RCC more than tumor size and ADC/size ratio.

Clinical prognostic factors and survival outcome in renal cell carcinoma patients--a malaysian single centre perspective

BACKGROUND

This study concerns clinical characteristics and survival of renal cell carcinoma (RCC) patients in University Malaya Medical Centre (UMMC), as well as the prognostic significance of presenting symptoms.

MATERIALS AND METHODS

The clinical characteristics, presenting symptoms and survival of RCC patients (n=151) treated at UMMC from 2003-2012 were analysed. Symptoms evaluated were macrohaematuria, flank pain, palpable abdominal mass, fever, lethargy, loss of weight, anaemia, elevated ALP, hypoalbuminemia and thrombocytosis. Univariate and multivariate Cox regression analyses were performed to determine the prognostic significance of these presenting symptoms. Kaplan Meier and log rank tests were employed for survival analysis.

RESULTS

The 2002 TNM staging was a prognostic factor (p<0.001) but Fuhrman grading was not significantly correlated with survival (p=0.088). At presentation, 76.8% of the patients were symptomatic. Generally, symptomatic tumours had a worse survival prognosis compared to asymptomatic cases (p=0.009; HR 4.74). All symptoms significantly affect disease specific survival except frank haematuria and loin pain on univariate Cox regression analysis. On multivariate analysis adjusted for stage, only clinically palpable abdominal mass remained statistically significant (p=0.027). The mean tumour size of palpable abdominal masses, 9.5±4.3cm, was larger than non palpable masses, 5.3±2.7cm (p<0.001).

CONCLUSIONS

This is the first report which includes survival information of RCC patients from Malaysia. Here the TNM stage and a palpable abdominal mass were independent predictors for survival. Further investigations using a multicentre cohort to analyse mortality and survival rates may aid in improving management of these patients.

Automated grading of renal cell carcinoma using whole slide imaging

INTRODUCTION

Recent technology developments have demonstrated the benefit of using whole slide imaging (WSI) in computer-aided diagnosis. In this paper, we explore the feasibility of using automatic WSI analysis to assist grading of clear cell renal cell carcinoma (RCC), which is a manual task traditionally performed by pathologists.

MATERIALS AND METHODS

Automatic WSI analysis was applied to 39 hematoxylin and eosin-stained digitized slides of clear cell RCC with varying grades. Kernel regression was used to estimate the spatial distribution of nuclear size across the entire slides. The analysis results were correlated with Fuhrman nuclear grades determined by pathologists.

RESULTS

The spatial distribution of nuclear size provided a panoramic view of the tissue sections. The distribution images facilitated locating regions of interest, such as high-grade regions and areas with necrosis. The statistical analysis showed that the maximum nuclear size was significantly different (P < 0.001) between low-grade (Grades I and II) and high-grade tumors (Grades III and IV). The receiver operating characteristics analysis showed that the maximum nuclear size distinguished high-grade and low-grade tumors with a false positive rate of 0.2 and a true positive rate of 1.0. The area under the curve is 0.97.

CONCLUSION

The automatic WSI analysis allows pathologists to see the spatial distribution of nuclei size inside the tumors. The maximum nuclear size can also be used to differentiate low-grade and high-grade clear cell RCC with good sensitivity and specificity. These data suggest that automatic WSI analysis may facilitate pathologic grading of renal tumors and reduce variability encountered with manual grading.

[Usefulness of new schemes to group Fuhrman grades in clinical practice for clear cell renal tumour]

OBJECTIVE

To evaluate if re-grading renal cell carcinoma (CRCC) in two or three-tiered grading schemes versus the traditional Fuhrman classification maintains the same prognostic value.

MATERIAL AND METHODS

A study of a cohort of 383 treated CRCC with radical or partial nephrectomy between 1990-2009 was made. We analyzed the demographic data, evolution and survival of these patients. An uropathologist reassigned the Fuhrman grades blindly to the first classification. In order to study if the prognostic value was maintained with the different classification, three Cox multivariate regression analysis were performed, classifying the variable of grade into four categories: (I-II-III-IV), into three (I+II-III-IV) and into two (I+II-III+IV). The explanatory variables were: age, gender, tumor size, study stage and grade. The response variables were progression-free survival (local-regional recurrence/metastasis) and cancer specific survival time.

RESULTS

The median overall survival was 125 months (95% CI: 92-159). In the three multivariate analyses carried out, the Fuhrman classification showed independent predictive value (p=:0.0001) compared to progression-free survival and cancer specific survival. The predictive power was maintained in the new classifications. In the three categories, the changing from grade I+II to III meant RR: 2.31 (p=0.0001) and from grade III to IV RR: 2.47 (p=0.0001) and in two-tiered classification an RR: 2.8 (p=0.001) was found when changing from I+II to III+IV.

CONCLUSIONS

Our results show that categorizing the Fuhrman grade into three or two-tiered grading schemes provide the same predictive accuracy on progressive free survival and cancer specific survival. Grades III and IV have different outcomes so that the three-tiered classification seems to be more appropriate to described the course of these patients.

Diffusion-weighted MRI in renal cell carcinoma: a surrogate marker for predicting nuclear grade and histological subtype

BACKGROUND

Though previous investigators have attempted to evaluate its utility in characterization of focal renal lesions, diffusion-weighted MR imaging (DW MRI) in renal diseases is still an evolving field and its role in predicting the aggressiveness of renal cell carcinoma (RCC) is yet to be established.

PURPOSE

To assess whether apparent diffusion coefficient (ADC) values can be used to determine the nuclear grade and histological subtype of RCCs and to identify the tumor attributes contributing to variation in ADC values.

MATERIAL AND METHODS

The institutional ethics committee waived the requirement of informed consent for this retrospective study. The study cohort consisted of 33 patients who underwent MRI (with diffusion-weighted imaging at b values of 0 and 500 s/mm(2)) and were found to have 36 pathologically-proven RCCs. ADC values were determined for solid portions of RCC, cystic/hemorrhagic areas, and normal renal parenchyma. Histological subtype, nuclear grade, and cell count were determined for each lesion. ADC values were compared between different grades and subtypes and correlation with cell count was investigated. Receiver operating characteristic curves were drawn to establish cut-off ADC values.

RESULTS

There were 23 low grade (grades I and II) and 13 high grade tumors (grades III and IV). There were 32 clear-cell and four non-clear-cell RCCs. A decreasing trend of ADC values was seen with increasing grade and mean ADC of high grade RCC was significantly lower than low grade (1.3145 vs 1.6982 × 10(-3) mm(2)/s) (P = 0.005). Mean ADC for clear-cell RCC was significantly higher than non-clear-cell RCC (1.6245 vs. 1.0412 × 10(-3) mm(2)/s) (P = 0.005). ADC values higher than 1.7960 × 10(-3) mm(2)/s were seen only with low grade and values greater than 1.4904 × 10(-3) mm(2)/s were seen only with clear-cell RCC.

CONCLUSION

ADC values provide a non-invasive means to predict the nuclear grade and histological subtype of RCC. Cellularity and morphology are other tumor attributes contributing to the variation in ADC values of RCCs.

Prognostic significance of morphologic parameters in renal cell carcinoma

Purpose

The aim of this study was to determine whether reclassifying the Fuhrman grading system provides further prognostic information.

Materials and methods

We studied the pathological features and cancer specific survival of 237 patients with clear cell cancer undergoing surgery between 1997–2007 in a single centre. The original Fuhrman grading system was investigated as well as various simplified models utilising the original Fuhrman grade.

Results

The median follow up was 69 months. On univariate analysis, the conventional Fuhrman grading system as well various simplified models were predicative of cancer specific survival. On multivariate analysis, only the three tiered modified model in which grades 1 and 2 were combined whilst grades 3 and 4 were kept separate was an independent predictor of cancer specific survival (p=0.001, HR 2.17, 95% CI 1.37-3.43). Furthermore this simplified model demonstrated a stronger relationship to recurrence than the conventional 4 tiered Fuhrman grading system.

Conclusions

A modified, three-tiered Fuhrman grading system has been demonstrated to be an independent predictor of cancer specific survival.