RENAL nephrometry score is a predictive factor for the annual growth rate of renal mass

Kidney scoring surveillance: predictive machine learning models for clear cell renal cell carcinoma growth using MRI

INTRODUCTION

Classification of clear cell renal cell carcinoma (ccRCC) growth rates in patients with Von Hippel-Lindau (VHL) syndrome has several ramifications for tumor monitoring and surgical planning. Using two separate machine-learning algorithms, we sought to produce models to predict ccRCC growth rate classes based on qualitative MRI-derived characteristics.

MATERIAL AND METHODS

We used a prospectively maintained database of patients with VHL who underwent surgical resection for ccRCC between January 2015 and June 2022. We employed a threshold growth rate of 0.5 cm per year to categorize ccRCC tumors into two distinct groups-'slow-growing' and 'fast-growing'. Utilizing a questionnaire of qualitative imaging features, two radiologists assessed each lesion on different MRI sequences. Two machine-learning models, a stacked ensemble technique and a decision tree algorithm, were used to predict the tumor growth rate classes. Positive predictive value (PPV), sensitivity, and F1-score were used to evaluate the performance of the models.

RESULTS

This study comprises 55 patients with VHL with 128 ccRCC tumors. Patients' median age was 48 years, and 28 patients were males. Each patient had an average of two tumors, with a median size of 2.1 cm and a median growth rate of 0.35 cm/year. The overall performance of the stacked and DT model had 0.77 ± 0.05 and 0.71 ± 0.06 accuracies, respectively. The best stacked model achieved a PPV of 0.92, a sensitivity of 0.91, and an F1-score of 0.90.

CONCLUSION

This study provides valuable insight into the potential of machine-learning analysis for the determination of renal tumor growth rate in patients with VHL. This finding could be utilized as an assistive tool for the individualized screening and follow-up of this population.

The kidney imaging surveillance scoring system (KISSS): using qualitative MRI features to predict growth rate of renal tumors in patients with von-Hippel Lindau (VHL) syndrome

OBJECTIVE

To determine the reliability of an MRI-based qualitative kidney imaging surveillance scoring system (KISSS) and assess which imaging features predict growth rate (GR) of renal tumors in patients with VHL.

MATERIALS AND METHODS

We identified 55 patients with VHL with 128 renal tumors who underwent intervention from 2015 to 2020 at the National Cancer Institute. All patients had 2 preoperative MRIs at least 3 months apart. Two fellowship-trained radiologists scored each tumor on location and MR-sequence-specific imaging parameters from the earlier MRI. Weighted kappa was used to determine the degree of agreement between radiologists for each parameter. GR was calculated as the difference in maximum tumor dimension over time (cm/year). Differences in mean growth rate (MGR) within categories of each imaging variable were assessed by ANOVA.

RESULTS

Apart from tumor margin and renal sinus, reliability was at least moderate (K > 0.40) for imaging parameters. Median initial tumor size was 2.1 cm, with average follow-up of 1.2 years. Tumor MGR was 0.42 cm/year. T2 hypointense, mixed/predominantly solid, and high restricted diffusion tumors grew faster. When comparing different combinations of these variables, the model with the lowest mean error among both radiologists utilized only solid/cystic and restricted diffusion features.

CONCLUSIONS

We demonstrate a novel MR-based scoring system (KISSS) that has good precision with minimal training and can be applied to other qualitative radiology studies. A subset of imaging variables (T2 intensity; restricted diffusion; and solid/cystic) were independently associated with growth rate in VHL renal tumors, with the combination of the latter two most optimal. Additional validation, including in sporadic RCC population, is warranted.

An MRI-based radiomics model to predict clear cell renal cell carcinoma growth rate classes in patients with von Hippel-Lindau syndrome

PURPOSE

Upfront knowledge of tumor growth rates of clear cell renal cell carcinoma in von Hippel-Lindau syndrome (VHL) patients can allow for a more personalized approach to either surveillance imaging frequency or surgical planning. In this study, we implement a machine learning algorithm utilizing radiomic features of renal tumors identified on baseline magnetic resonance imaging (MRI) in VHL patients to predict the volumetric growth rate category of these tumors.

MATERIALS AND METHODS

A total of 73 VHL patients with 173 pathologically confirmed Clear Cell Renal Cell Carcinoma (ccRCCs) underwent MRI at least at two different time points between 2015 and 2021. Each tumor was manually segmented in excretory phase contrast T1 weighed MRI and co-registered on pre-contrast, corticomedullary and nephrographic phases. Radiomic features and volumetric data from each tumor were extracted using the PyRadiomics library in Python (4544 total features). Tumor doubling time (DT) was calculated and patients were divided into two groups: DT <  = 1 year and DT > 1 year. Random forest classifier (RFC) was used to predict the DT category. To measure prediction performance, the cohort was randomly divided into 100 training and test sets (80% and 20%). Model performance was evaluated using area under curve of receiver operating characteristic curve (AUC-ROC), as well as accuracy, F1, precision and recall, reported as percentages with 95% confidence intervals (CIs).

RESULTS

The average age of patients was 47.2 ± 10.3 years. Mean interval between MRIs for each patient was 1.3 years. Tumors included in this study were categorized into 155 Grade 2; 16 Grade 3; and 2 Grade 4. Mean accuracy of RFC model was 79.0% [67.4-90.6] and mean AUC-ROC of 0.795 [0.608-0.988]. The accuracy for predicting DT classes was not different among the MRI sequences (P-value = 0.56).

CONCLUSION

Here we demonstrate the utility of machine learning in accurately predicting the renal tumor growth rate category of VHL patients based on radiomic features extracted from different T1-weighted pre- and post-contrast MRI sequences.

Selecting the Best Elements from Previous Kidney Tumor Scoring Systems to Restructure Efficient Predictive Models for Surgery Type

OBJECTIVE

The aim of this work was to select the best elements from previous scoring systems to restructure efficient predictive models for surgery type.

METHODS

Sixteen elements were selected from 7 systems (RENAL, PADUA, DAP, ZS, NephRO, ABC, and CI). They were divided into 6 categories (tumor max. size, exophytic/endophytic, correlation with collecting system or sinus, tumor location, contact situation with the parenchyma, invasion depth). Three elements, selected from 3 different categories, were integrated to establish a total of 320 new models. According to AUC rank, optimized models were developed, and these models were divided into 3 sections. An analysis of the distribution of the 6 categories was made to explore the predictive capacities of the models.

RESULTS

A total of 166 consecutive patients were included. Seventy-five patients underwent radical nephrectomy operations. The AUC of the 7 systems ranged from 0.81 to 0.844. Three optimized models (AUC 0.88) were developed to predict surgery type. These optimized models were composed of DAP (D), PADUA, (sinus), and ABC; DAP (D), RENAL (N), and ABC; NePhRO (O), PADUA (UCS), and ABC. Two categories ("exophytic/endophytic," p < 0.001; "correlation with collecting system or sinus," p = 0.001) were nonuniformly distributed.

CONCLUSIONS

Seven systems held good predictive power for surgery type. Three optimized models were developed. "Correlation with collecting system or sinus" is a critical factor for predicting surgery type.

Renal Tumor Anatomic Complexity: Clinical Implications for Urologists

Anatomic tumor complexity can be objectively measured and reported using nephrometry. Various scoring systems have been developed in an attempt to correlate tumor complexity with intraoperative and postoperative outcomes. Nephrometry may also predict tumor biology in a noninvasive, reproducible manner. Other scoring systems can help predict surgical complexity and the likelihood of complications, independent of tumor characteristics. The accumulated data in this new field provide provocative evidence that objectifying anatomic complexity can consolidate reporting mechanisms and improve metrics of comparisons. Further prospective validation is needed to understand the full descriptive and predictive ability of the various nephrometry scores.

Evaluation of the Clinical Use of Robot-Assisted Retroperitoneal Laparoscopy and Preoperative RENAL Scoring for Nephron Sparing Surgery in Renal Tumor Patients

The present study aims to compare the operative outcomes following the use of robot-assisted retroperitoneal partial nephrectomy (RARPN) with radius, exophytic/endophytic, nearness to sinus, anterior/posterior, and location (RENAL) scoring or laparoscopic retroperitoneal partial nephrectomy (LRPN) for the treatment of renal tumors. Eighty-three nephron-sparing surgery (NSS) procedures performed between January 2013 and December 2015 were reviewed. The study set consisted of 26 robot-assisted retroperitoneal laparoscopes, of which 3 were high risk (RENAL score ≥10), 11 were medium risk (RENAL score ≥7 < 9), and 12 were low risk (RENAL score <7) and 57 laparoscopic retroperitoneal partial nephrectomy procedures (7 high, 22 medium, and 28 low risk). All surgeries were successful in the absence of conversion or transfusion. Operative times were 96.0 ± 16.9 and 110.0 ± 19.4 min for RARPN and LRPN, respectively (P < 0.05). Warm ischemia times (WITs) were 17.6 ± 3.1 and 22.8 ± 3.5 min, respectively (P < 0.05). Estimated blood losses (EBLs) were 45 ± 15 and 97 ± 25 mL, respectively (P < 0.05). No statistical significance was found in duration of drainage, intestinal recovery time, hospital stay, serum creatinine, and perioperative complications (P > 0.05). RARPN affords significant advantages in outcomes of WIT, EBL, and recovery time over conventional LRPN owing to an increased accuracy in excision and suturing. Patients bearing high-risk renal tumors (RENAL score ≥10) are suitable candidates for RARPN.

Update on active surveillance for clinical T1 renal tumors

PURPOSE OF REVIEW

The purpose is to discuss current data on the utilization and outcomes of active surveillance for T1a renal masses. Specifically, to address which patients are optimal for active surveillance and how their outcomes differ from those undergoing immediate treatment.

RECENT FINDINGS

Although nephron sparing surgery is the standard of care for small renal masses (SRMs), active surveillance is becoming a more popular intervention given the results of prospective studies revealing active surveillance to be safe and have excellent cancer-specific survival with intermediate follow-up. Older and sicker patients have competing risk of death from other causes when diagnosed with a SRM.

SUMMARY

Active surveillance is becoming a more popular treatment modality for SRMs given the increasing number of incidental diagnoses and better understanding of their often indolent course. Active surveillance with delayed intervention is a well-tolerated treatment modality and appears to have the most benefit for those patients that are older with more comorbidities.

R.E.N.A.L. nephrometry score predicts postoperative recurrence of localized renal cell carcinoma treated by radical nephrectomy

Background

We investigated the association between the R.E.N.A.L. nephrometry score (RNS) and the postoperative recurrence of localized renal cell carcinoma (RCC).

Methods

We retrospectively analyzed a database comprising 91 patients with non-small localized RCC (pT1b–T2b) treated by radical nephrectomy at our hospital from January 2002 to March 2010. RNS was scored based on imaging findings at diagnosis. The Cox proportional hazards model was used to predict recurrence-free survival (RFS) and to calculate hazard ratio (HR).

Results

The median age at operation was 63 years (range, 30–85 years). Postoperative recurrence occurred in 19 patients (21 %). Median RNS sum was 9 (range, 5–11). High RNS sum (10–12) was significantly associated with RFS (P = 0.0012). Multivariate analysis revealed that high RNS sum [HR, 9.05; 95 % confidence interval (CI), 2.11–63.9; P = 0.0019] were significantly associated with RFS. Regarding each component of RNS, only the L component, which referred to tumor location relative to the polar line, was associated with RFS (HR, 15.0; 95 % CI, 2.68–396; P = 0.0006).

Conclusions

RNS was associated with RFS in cases of non-small localized RCC (pT1b–2b), thus supporting its utility as a prognostic factor.