Skip to main content
Insights into Imaging
Download PDF

Radiomics predict the WHO/ISUP nuclear grade and survival in clear cell renal cell carcinoma

Insights into Imaging volume 15, Article number: 175 (2024) Cite this article

Abstract

Objectives

This study aimed to assess the predictive value of radiomics derived from intratumoral and peritumoral regions and to develop a radiomics nomogram to predict preoperative nuclear grade and overall survival (OS) in patients with clear cell renal cell carcinoma (ccRCC).

Methods

The study included 395 patients with ccRCC from our institution. The patients in Center A (anonymous) institution were randomly divided into a training cohort (n = 284) and an internal validation cohort (n = 71). An external validation cohort comprising 40 patients from Center B also was included. Computed tomography (CT) radiomics features were extracted from the internal area of the tumor (IAT) and IAT combined peritumoral areas of the tumor at 3 mm (PAT 3 mm) and 5 mm (PAT 5 mm). Independent predictors from both clinical and radiomics scores (Radscore) were used to construct a radiomics nomogram. Kaplan–Meier analysis with a log-rank test was performed to evaluate the correlation between factors and OS.

Results

The PAT 5-mm radiomics model (RM) exhibited exceptional predictive capability for grading, achieving an area under the curves of 0.80, 0.80, and 0.90 in the training, internal validation, and external validation cohorts. The nomogram and RM gained from the PAT 5-mm region were more clinically useful than the clinical model. The association between OS and predicted nuclear grade derived from the PAT 5-mm Radscore and the nomogram-predicted score was statistically significant (p < 0.05).

Conclusion

The CT-based radiomics and nomograms showed valuable predictive capabilities for the World Health Organization/International Society of Urological Pathology grade and OS in patients with ccRCC.

Critical relevance statement

The intratumoral and peritumoral radiomics are feasible and promising to predict nuclear grade and overall survival in patients with clear cell renal cell carcinoma, which can contribute to the development of personalized preoperative treatment strategies.

Key Points

  • The multi-regional radiomics features are associated with clear cell renal cell carcinoma (ccRCC) grading and prognosis.

  • The combination of intratumoral and peritumoral 5 mm regional features demonstrated superior predictive performance for grading.

  • The nomogram and radiomics models have a broad range of clinical applications.

Graphical Abstract

Introduction

Renal cell carcinoma (RCC) is the most common type of primary kidney cancer, with clear cell RCC (ccRCC) accounting for the majority of cases [1]. The pathological nuclear grade is an important prognostic factor for patients with ccRCC [2]. In 2016, the widely accepted classification system for ccRCC was established by The World Health Organization/International Society of Urological Pathology (WHO/ISUP) criteria. This system categorizes ccRCC into four grades, identifying grades I and II as low-grade and grades III and IV as high-grade [3]. Low-grade ccRCC may be treated with less aggressive options, whereas high-grade ccRCC may require more intensive treatment [4]. This categorization serves to decrease discrepancies between observers, enhance reproducibility, and save time without affecting the ability to predict cancer-specific mortality [5, 6]. Assessment of pathological grading prior to surgery commonly relies on biopsy. However, due to the high spatiotemporal heterogeneity of ccRCC, biopsies may represent only a portion of the lesions that may contribute to selection and may not adequately reflect the nuclear grade of the entire tumor. This invasive procedure also suffers from low reproducibility and a high rate of complications [7]. Therefore, the development of a non-invasive method for predicting ccRCC grading before surgery would greatly benefit clinical practice.

There is a promising future for the application of radiomics in the prediction of malignant tumors, tumor histopathology, tumor grading, and molecular characteristics [8]. Many radiomics investigations have concentrated on predicting the nuclear grading to assess tumor invasiveness [9,10,11,12,13,14,15,16]. In a study by Cui et al, multiparametric magnetic resonance imaging and multiphase CT-based machine-learning models were created from the radiomics features of a group of 460 patients diagnosed with ccRCC. These models achieved an accuracy of 73% and 79% when distinguishing between high-grade and low-grade ccRCC [13]. Another study demonstrated that a radiomics signature consisting of 20 features displayed excellent performance in differentiating grades in a multi-institutional cohort of 258 patients, with an area under the curve (AUC) of 0.846 in the validation sets [17]. However, these models solely relied on intratumoral radiomics and did not incorporate peritumoral radiomics. Peritumoral and multi-regional radiomics are increasingly used in tumor diagnosis, grading, and prognosis in other cancers [18,19,20,21], but their application in RCC remains relatively rare. Recent research suggests that peritumoral radiomics has a significant role in ccRCC grading [22]. Additionally, most studies have failed to consider the integration of clinical risk factors. It is plausible to hypothesize that models capable of accurately predicting grading could also exhibit strong performance in predicting survival given the close association between grading and prognosis. Unfortunately, most studies have primarily focused on the predictive aspect of grading while disregarding its correlation with prognosis. Thus, the development of a suitable model that not only accurately predicts ccRCC grading but also elucidates the underlying survival mechanisms associated with prognosis presents a significant challenge [23].

This study aimed to evaluate the predictive value of radiomics extracted from intratumoral and peritumoral regions in patients with ccRCC and develop a CT radiomics nomogram for predicting the WHO/ISUP grade. Additionally, we investigated whether the constructed grading prediction model was associated with the overall survival (OS) of ccRCC.

Materials and methods

Patients

This retrospective study complied with the ethical guidelines of the Declaration of Helsinki and was approved by the ethics committee of our hospital, and the requirement for informed consent was waived. We collected data from the hospital’s medical records, including 355 patients with ccRCC who underwent radical or partial nephrectomies between January 2015 and December 2018. The inclusion and exclusion criteria are presented in Appendix S1 and Fig. 1. Additionally, we obtained 40 patients from our branch hospital for external validation in the same inclusion and exclusion criteria. The collected clinical data are presented in Table 1.

Fig. 1
figure 1

Flow chart showing the pathway of patient exclusion

Full size image
Table 1 Characteristics of ccRCC patients in the training set, internal validation set, and external validation set
Full size table

Image acquisition and tumor segmentation

All patients undergo preoperative CT-enhanced examination with 16-detector and 64-detector. Arterial-phase CT images of the patients with ccRCC were analyzed. The detailed CT parameters are listed in Appendix S2. The details about tumor segmentation are illustrated in Appendix S3 and Fig. 2.

Fig. 2
figure 2

Study workflow. a ROI segmentation, b radiomics analysis

Full size image

Radiomics feature selection and reproducibility analysis

Before conducting feature extraction, the CT images went through preprocessing procedures involving resampling and standardization to ensure that the results could be replicated. Pyradiomics v3.0.1 package (https://pypi.org/project/pyradiomics/) was performed to extract radiomics features from the regions of interests (ROIs). The details about the extracted features are illustrated in Appendix S4. Initially, the z-score method was used to standardize these radiomics features. The inter- and intra-observer and correlation coefficients (ICCs) were calculated for reproducibility analysis in Appendix S5.

Feature selection and radiomics model construction

After selecting the remaining features with an ICC > 0.75, the statistical significance of these features was determined by performing the student’s t-test or the Mann–Whitney U-test. Only the features with values of p < 0.05 were considered. Spearman’s correlation coefficient was then used to evaluate the correlations between features that exhibited high repeatability, and any two features with a correlation coefficient > 0.9 were retained. Next, the least absolute shrinkage and selection operator (LASSO) algorithm was used to create the dataset’s signature. A 10-fold cross-validation with minimum criteria was used to find the optimal value of λ, in which the value of λ that resulted in the lowest cross-validation error was chosen. The features with nonzero coefficients from the LASSO algorithm were used to construct radiomics models (RMs) for predicting grading and combined into a radiomics score (Radscore). To assess the discrimination performance of the radiomics signature, the Wilcoxon test was used to compare the difference in the Radscores between the low-grade and high-grade groups. The performance of the RMs in predicting grading was further evaluated using the area under the receiver operating characteristic curve in the training set, internal validation set, and external validation.

Nomogram construction and survival analysis

To create a clinical model (CM), univariable and multivariable logistic regression analyses were performed to identify factors that independently predicted nuclear grade. These analyses included variables, such as age, sex, tumor size, and T stage. A fusion model (FM) was constructed that was based on the Radscore of the best RM and independent clinical predictors. Then, ROC curves were used to evaluate the performances of the CM, RM, and FM models in the three sets. To assess the clinical usefulness of the models, decision curve analysis (DCA) was performed. The Kaplan–Meier method was used to analyze survival outcomes in different groups stratified by nomogram-predicted low and high risk and to create corresponding survival curves. Additionally, the log-rank test was conducted to evaluate the statistical significance of any differences observed between these groups.

Statistical analysis

Student’s t-test, or Mann–Whitney U-test, chi-square, or Fisher’s exact test was used to analyze the difference between low- and high-grade groups. To identify independent predictors for grading ccRCC patients, we performed both univariable and multivariable logistic regression analyses. For statistical significance, we considered two-sided p-values of < 0.05. All statistical analyses were performed in SPSS v.20.0, R software (version 4.2.0), and Pytorch v.1.8.0.

Results

Patients characteristics

The characteristics of the ccRCC patients, including the distribution of patients in the training set, internal validation set, and external validation set, are presented in Table 1. There were significant differences in age, radiological tumor size, and pathological T stage based on CT between low-grade and high-grade patients in the training set. Multivariable logistic regression analysis identified the T stage as an independent predictor of ccRCC grading (Table 2). This valuable information was used to establish the CM.

Table 2 Univariable and multivariable logistic regression analysis of factors in the training
Full size table

Feature selection

Prior to developing RM for grading prediction, we conducted a three-step radiomics feature screening process. This process involved repeatability analysis, t-test or U-test, and Spearman’s analysis (Appendix S6). To further refine our RMs, we used the LASSO method and retained a total of 16 optimal radiomics features for each ROI type. The specific Radscore formulas for these three RMs are shown in Fig. 3.

Fig. 3
figure 3

The retained radiomics features and corresponding coefficients of different models after performing LASSO regression analysis. a IAT model, b PAT 3-mm model, c PAT 5-mm model

Full size image

Radiomics model analysis

In the three cohorts, the peritumoral areas of the tumor (PAT) 5-mm models (RM5) exhibited exceptional performance in predicting grading, achieving AUC values of 0.80, 0.80, and 0.90, respectively (Table 3). These values surpassed those of the internal area of the tumor (IAT) model (RM) (0.79, 0.66, 0.70) and the PAT 3-mm model (RM3) (0.80, 0.77, 0.83). DeLong’s test confirmed that the AUC value of the RM5 and RM3 models significantly differed from the RM models in both the internal validation (0.80 vs. 0.66, p = 0.01; 0.77 vs. 0.66, p = 0.03) and external validation (0.90 vs. 0.70, p = 0.01; 0.80 vs. 0.70, p = 0.04), but the difference between the RM5 and RM3 was not statistically significant in both of the internal validations (p > 0.05). The Wilcoxon test revealed that the RM5-gained Radscore was significantly higher for the high-grade patients than for the low-grade patients in all datasets (p < 0.05) (Supplemental Fig. 1). Consequently, we used the radiomics features of the RM5 model to construct the nomogram for predicting pathological grades.

Table 3 Prediction performance of the established models in the training set, internal validation set, and external validation set
Full size table

Radiomics nomogram construction and survival analysis for OS

A multivariate logistic regression analysis was performed to determine the factors that independently predicted ccRCC grade, namely the radiomics signature and T stage (Table 2). To create an FM model, a nomogram was developed that incorporated both the PAT 5-mm radiomics signature and the T stage. The ROC curves of the PAT 5-mm RM (RM5), CM, and FM model were compared to assess their effectiveness in predicting the grade of ccRCC in the three cohorts (Fig. 4 and Table 3). The AUC was higher in the RM model than in the CM in the internal validation cohort (0.80 vs. 0.57, p = 0.003) and external validation cohort (0.90 vs. 0.61, p < 0.001). The nomogram’s AUC was higher than the CM’s AUC in the internal validation cohort (0.78 vs. 0.57, p = 0.005) and external validation cohort (0.71 vs. 0.61, p = 0.332). The AUC of the nomogram was not improved relative to the RM5 in the internal validation cohort (0.78 vs. 0.80, p = 0.422) and in the external validation cohort (0.71 vs. 0.90, p = 0.033). Additionally, the DCA analysis demonstrated that both the FM and RM5 models had a higher net benefit in predicting grading than that of the CM in the total dataset (Fig. 5). Finally, the Kaplan–Meier survival curves revealed that OS, as determined by the PAT 5-mm Radscore-predicted and nomogram-predicted grading, was significantly worse for the high-grade patients than for the low-grade patients in all cohorts except for the survival curve of the nomogram-predicted score in the external validation set (Fig. 6).

Fig. 4
figure 4

ROC comparison of the clinical model, radiomics model, and fusion model in (a) the training set, (b) internal validation set, and (c) external validation set

Full size image
Fig. 5
figure 5

Demonstration of the nomogram (fusion model). a A fusion model incorporating the PAT 5-mm-gained Radscore and T stage was constructed. b The decision curve analysis of the clinical model, radiomics model, and fusion model based on the whole cohort. Radiomics model based on the PAT 5-mm region

Full size image
Fig. 6
figure 6

Kaplan–Meier survival curves were generated for the RM5 and FM5 models in the training (a, d), internal validation (b, e), and external validation cohorts (c, f). RM5, Radiomics model based on the PAT 5-mm region; FM, a fusion model incorporating the PAT 5-mm-gained Radscore and T stage

Full size image

Pathological characteristics and association with radiomics features

We detected cases with pathological features like perirenal or sinus fat invasion in peritumoral tissues and examined their relationship with radiomics features chosen from the PAT 5-mm model. Our analysis revealed that a significant number of these peritumoral parameters (7 out of 16) are connected to invasive biological behaviors (p < 0.05) (See Supplemental Fig. 2).

Discussion

The study findings indicate that radiomics features, particularly peritumoral and tumoral signatures, serve as more precise indicators for grading and predicting survival. Additionally, the study demonstrated that the RM and nomogram performed better than the CM.

Invasion of tumor cells often disrupts the normal structure of the surrounding parenchymal tissues and leads to alterations in the peritumoral microenvironment. Unfortunately, the significance of this peritumoral microenvironment is sometimes overlooked in research that primarily focuses on the internal aspects of the tumor [20, 24]. Conventional imaging techniques struggle to accurately depict the microenvironment around a tumor. However, peritumoral characteristics offer the potential for quantitatively analyzing the heterogeneity of the microenvironment [25, 26]. In the present study, we constructed RMs for IAT and PAT with the goal of predicting the grading of ccRCC and investigating its association with survival prognosis. Our results revealed that the PAT model (PAT 3 mm and PAT 5 mm) had higher predictive values than that of the IAT model. This finding confirms that incorporating incremental information from both the internal tumor components and peritumoral features into a combined model can significantly enhance predictive performance. Interestingly, our findings align with those of previous reports showing that all peritumoral-feature models yielded grading performance superior to that based on features solely within the tumor [22]. Pathological research has demonstrated the crucial role of the peritumoral microenvironment in evaluating tumor invasiveness [27]. Pathological grading is correlated with tumor invasiveness [28]. Tumors of higher grades show increased invasiveness, leading to the invasion of surrounding tissues. As a consequence, the tumor’s surrounding environment undergoes heterogeneous alterations. Consequently, the evaluation of intratumoral grading heavily relies on peritumoral characteristics, which capture the heterogeneity within the tumor and its surroundings [17]. Our study revealed an association between the histological characteristics of perirenal fat invasion and radiomics features. This finding provides evidence that the radiomics features obtained from the PAT region can reflect the biological behavior of tumors. Therefore, when delineating ROIs, it is important not to overlook areas within the tumor and those outside the tumor, as this is associated with tumor proliferation and heterogeneity. Hence, the PAT model developed in this study automatically extends to the surrounding region based on the delineation of the IAT, not only eliminating the need for time-consuming operations to remove the internal tumor mask but also improving the clinical workflow.

The analysis of feature selection revealed that the first-order texture feature “ exponential_firstorder_Kurtosis,” which was obtained by exponentiating the “firstorder_Kurtosis,” had the strongest correlation with the grade in the peritumoral 5 mm area. The relationship between kurtosis and tumors can be inferred from the morphological characteristics of tumors [29, 30]. A higher kurtosis value indicates greater cell density within the tumor, implying a uniform and densely structured tumor. Kurtosis is frequently employed in radiomics to assess tumor heterogeneity [31]. Higher values of kurtosis may indicate increased tumor heterogeneity, which has a potential correlation with tumor malignancy. Our research supports this relationship, as we have discovered a close association between higher kurtosis and tumor malignancy grade. “Entropy” and “JointEntropy” are two terms that measure the uncertainty or disorder in a random variable or joint probability distribution, respectively. Higher values indicate greater heterogeneity [32]. Previous studies [33, 34] have also highlighted the close association between entropy and tumor invasiveness, which aligns with the findings of our research. From a pathologic perspective, the grades of ccRCC are determined by nuclear diameter, nuclear shape, and nucleoli. A higher grade is characterized by a larger nuclear diameter, a more irregular nuclear shape, and greater irregularity in the arrangement of histological internal components in pathological sections. “Busyness” is an additional crucial feature in the PAT 5 mm model, which measures the intensity change between a pixel and its neighboring pixels. A high value for busyness indicates a “busy” image with rapid changes in CT intensity between a pixel and its neighboring pixels, often signifying the rate of tumor growth or progression. Busyness is linked to tumor prognosis [35, 36]. Our study indicates a positive correlation between this parameter and tumor grade, as well as a close association with perirenal or perirenal sinus fat invasion. To summarize, quantitative analysis of radiomics features is a non-intrusive strategy to better understand the biological properties of tumors. This strategy can provide a substantial foundation for achieving personalized precision treatment of ccRCC.

We then analyzed the clinical risk factors, which have the potential to offer supplemental data and enhance the accuracy of the predictive model [37, 38]. Previous research has prioritized the examination of radiomic characteristics while neglecting clinical risk factors [9,10,11,12,13,14,15,16]. In this investigation, we performed multivariable logistic regression analysis and identified the T stage as an independent predictive factor. This finding aligns with the results of Zheng et al [17] who demonstrated a strong association between T stage and renal cancer grade. The AUC of the nomogram was improved relative to that of the RM in the training set (0.93 vs. 0.80). However, there was no improvement observed in the internal validation (0.78 vs. 0.80) and external validation (0.61 vs. 0.71), indicating that clinical risk factors may not have a strong predictive value for grade. Nevertheless, the nomogram-gained score was able to predict OS, highlighting the importance of both the RM and the nomogram in clinical applications.

Our study had several limitations. Initially, this study was conducted retrospectively and did not encompass chromophobe or papillary tumors and those who will not have nephrectomy, thus potentially introducing selection bias. We collected data on a limited number of cases of chromophobe and papillary RCC (approximately 20–30 cases). The small sample size could potentially impact the reliability of the model and statistical results, affecting the statistical power. In the future, we will expand the sample size and add more cases of other types. To assess the robustness and consistency of our models, we used a dataset from a different institution to evaluate their effectiveness. However, it is important to note that the predictive accuracy of the external validation cohort may be limited by the small sample size and relatively short follow-up period. Consequently, further validation through additional studies incorporating larger sample sizes is a necessary step for future research.

To summarize, the performance of the PAT RM outperformed the IAT model for predicting ccRCC grade before surgery. Additionally, both nomograms and RMs improved predictive accuracy and survival prediction. This advancement can help in adjusting treatment strategies promptly and establishing prognostic risk stratification.

Data availability

The datasets generated and analyzed during the current study are not publicly available but are available from the corresponding author on reasonable request.

Abbreviations

AUC:

Area under the curve

ccRCC:

Clear cell renal cell carcinoma

CM:

Clinical model

DCA:

Decision curve analysis

FM:

Fusion model

IAT:

Internal area of the tumor

ICC:

Intraclass correlation coefficient

ISUP:

International Society of Urological Pathology

LASSO:

Least absolute shrinkage and selection operator

OS:

Overall survival

PAT:

Peritumoral areas of the tumor

RM:

Radiomics model

Radscore:

Radiomics scores

ROI:

Regions of interest

RCC:

Renal cell carcinoma

WHO:

World Health Organization

References

  1. Capitanio U, Bensalah K, Bex A et al (2019) Epidemiology of renal cell carcinoma. Eur Urol 75:74–84

    Article  PubMed  Google Scholar 

  2. Kuthi L, Jenei A, Hajdu A et al (2017) Prognostic factors for renal cell carcinoma subtypes diagnosed according to the 2016 WHO Renal Tumor Classification: a study involving 928 patients. Pathol Oncol Res 23:689–698

    Article  PubMed  Google Scholar 

  3. Moch H, Cubilla AL, Humphrey PA et al (2016) The 2016 WHO Classification of Tumours of the Urinary System and Male Genital Organs-Part A: renal, penile, and testicular tumours. Eur Urol 70:93–105

    Article  PubMed  Google Scholar 

  4. Vetterlein MW, Jindal T, Becker A et al (2016) Small renal masses in the elderly: Contemporary treatment approaches and comparative oncological outcomes of nonsurgical and surgical strategies. Investig Clin Urol 57:231–239

    Article  PubMed  PubMed Central  Google Scholar 

  5. Sun M, Lughezzani G, Jeldres C et al (2009) A proposal for reclassification of the Fuhrman grading system in patients with clear cell renal cell carcinoma. Eur Urol 56:775–781

    Article  PubMed  Google Scholar 

  6. Becker A, Hickmann D, Hansen J et al (2016) 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. Eur J Surg Oncol 42:419–425

    Article  CAS  PubMed  Google Scholar 

  7. Marconi L, Dabestani S, Lam TB et al (2016) Systematic review and meta-analysis of diagnostic accuracy of percutaneous renal tumour biopsy. Eur Urol 69:660–673

    Article  PubMed  Google Scholar 

  8. Roussel E, Capitanio U, Kutikov A et al (2022) Novel imaging methods for renal mass characterization: a collaborative review. Eur Urol 81:476–488

    Article  PubMed  PubMed Central  Google Scholar 

  9. Han D, Yu Y, Yu N et al (2020) Prediction models for clear cell renal cell carcinoma ISUP/WHO grade: comparison between CT radiomics and conventional contrast-enhanced CT. Br J Radiol 93:20200131

    Article  PubMed  PubMed Central  Google Scholar 

  10. Nazari M, Shiri I, Hajianfar G et al (2020) Noninvasive Fuhrman grading of clear cell renal cell carcinoma using computed tomography radiomic features and machine learning. Radiol Med 125:754–762

    Article  PubMed  Google Scholar 

  11. Sun X, Liu L, Xu K et al (2019) Prediction of ISUP grading of clear cell renal cell carcinoma using support vector machine model based on CT images. Medicine 98:e15022

    Article  PubMed  PubMed Central  Google Scholar 

  12. Shu J, Tang Y, Cui J et al (2018) Clear cell renal cell carcinoma: CT-based radiomics features for the prediction of Fuhrman grade. Eur J Radiol 109:8–12

    Article  PubMed  Google Scholar 

  13. Cui E, Li Z, Ma C et al (2020) Predicting the ISUP grade of clear cell renal cell carcinoma with multiparametric MR and multiphase CT radiomics. Eur Radiol 30:2912–2921

    Article  PubMed  Google Scholar 

  14. Sun J, Pan L, Zha T et al (2021) The role of MRI texture analysis based on susceptibility-weighted imaging in predicting Fuhrman grade of clear cell renal cell carcinoma. Acta Radiol 62:1104–1111

    Article  PubMed  Google Scholar 

  15. Stanzione A, Ricciardi C, Cuocolo R et al (2020) MRI radiomics for the prediction of Fuhrman grade in clear cell renal cell carcinoma: a machine learning exploratory study. J Digit Imaging 33:879–887

    Article  PubMed  PubMed Central  Google Scholar 

  16. Yaşar S, Voyvoda N, Voyvoda B et al (2020) Using texture analysis as a predictive factor of subtype, grade and stage of renal cell carcinoma. Abdom Radiol (NY) 45:3821–3830

    Article  PubMed  Google Scholar 

  17. Zheng Z, Chen Z, Xie Y et al (2021) Development and validation of a CT-based nomogram for preoperative prediction of clear cell renal cell carcinoma grades. Eur Radiol 31:6078–6086

    Article  PubMed  Google Scholar 

  18. Li H, Chen XL, Liu H et al (2022) MRI-based multiregional radiomics for predicting lymph nodes status and prognosis in patients with resectable rectal cancer. Front Oncol 12:1087882

    Article  PubMed  Google Scholar 

  19. Li H, Chen XL, Liu H et al (2023) MRI-based multiregional radiomics for preoperative prediction of tumor deposit and prognosis in resectable rectal cancer: a bicenter study. Eur Radiol 33:7561–7572

    Article  PubMed  Google Scholar 

  20. Wang X, Zhao X, Li Q et al (2019) Can peritumoral radiomics increase the efficiency of the prediction for lymph node metastasis in clinical stage T1 lung adenocarcinoma on CT? Eur Radiol 29:6049–6058

    Article  PubMed  Google Scholar 

  21. Shi J, Dong Y, Jiang W et al (2022) MRI-based peritumoral radiomics analysis for preoperative prediction of lymph node metastasis in early-stage cervical cancer: A multi-center study. Magn Reson Imaging 88:1–8

    Article  CAS  PubMed  Google Scholar 

  22. Zhou Z, Qian X, Hu J et al (2021) CT-based peritumoral radiomics signatures for malignancy grading of clear cell renal cell carcinoma. Abdom Radiol (NY) 46:2690–2698

    Article  PubMed  Google Scholar 

  23. Li ZC, Zhai G, Zhang J et al (2019) Differentiation of clear cell and non-clear cell renal cell carcinomas by all-relevant radiomics features from multiphase CT: a VHL mutation perspective. Eur Radiol 29:3996–4007

    Article  PubMed  Google Scholar 

  24. Zhang Z, Wan X, Lei X et al (2023) Intra- and peri-tumoral MRI radiomics features for preoperative lymph node metastasis prediction in early-stage cervical cancer. Insights Imaging 14:65

    Article  PubMed  PubMed Central  Google Scholar 

  25. Lin Z, Wang T, Li Q et al (2023) Development and validation of MRI-based radiomics model to predict recurrence risk in patients with endometrial cancer: a multicenter study. Eur Radiol 33:5814–5824

    Article  PubMed  Google Scholar 

  26. Wang T, She Y, Yang Y et al (2022) Radiomics for survival risk stratification of clinical and pathologic stage IA pure-solid non-small cell lung cancer. Radiology 302:425–434

    Article  PubMed  Google Scholar 

  27. Joyce JA, Pollard JW (2009) Microenvironmental regulation of metastasis. Nat Rev Cancer 9:239–252

    Article  CAS  PubMed  Google Scholar 

  28. Kierans AS, Rusinek H, Lee A et al (2014) Textural differences in apparent diffusion coefficient between low- and high-stage clear cell renal cell carcinoma. AJR Am J Roentgenol 203:W637–W644

    Article  PubMed  Google Scholar 

  29. Chae HD, Park CM, Park SJ et al (2014) Computerized texture analysis of persistent part-solid ground-glass nodules: differentiation of preinvasive lesions from invasive pulmonary adenocarcinomas. Radiology 273:285–293

    Article  PubMed  Google Scholar 

  30. Kamiya A, Murayama S, Kamiya H et al (2014) Kurtosis and skewness assessments of solid lung nodule density histograms: differentiating malignant from benign nodules on CT. Jpn J Radiol 32:14–21

    Article  PubMed  Google Scholar 

  31. Liang CH, Liu YC, Wan YL et al (2021) Quantification of cancer-developing idiopathic pulmonary fibrosis using whole-lung texture analysis of HRCT images. Cancers 13:5600

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  32. Liu S, Zheng H, Zhang Y et al (2018) Whole-volume apparent diffusion coefficient-based entropy parameters for assessment of gastric cancer aggressiveness. J Magn Reson Imaging 47:168–175

    Article  PubMed  Google Scholar 

  33. Suo S, Zhang K, Cao M et al (2016) Characterization of breast masses as benign or malignant at 3.0T MRI with whole-lesion histogram analysis of the apparent diffusion coefficient. J Magn Reson Imaging 43:894–902

    Article  PubMed  Google Scholar 

  34. Wibmer A, Hricak H, Gondo T et al (2015) Haralick texture analysis of prostate MRI: utility for differentiating non-cancerous prostate from prostate cancer and differentiating prostate cancers with different Gleason scores. Eur Radiol 25:2840–2850

    Article  PubMed  PubMed Central  Google Scholar 

  35. Vuong D, Tanadini-Lang S, Wu Z et al (2020) Radiomics feature activation maps as a new tool for signature interpretability. Front Oncol 10:578895

    Article  PubMed  PubMed Central  Google Scholar 

  36. Tang YY, Zhao YN, Zhang T et al (2021) Comprehensive radiomics nomogram for predicting survival of patients with combined hepatocellular carcinoma and cholangiocarcinoma. World J Gastroenterol 27:7173–7189

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  37. Khodabakhshi Z, Amini M, Mostafaei S et al (2021) Overall survival prediction in renal cell carcinoma patients using computed tomography radiomic and clinical information. J Digit Imaging 34:1086–1098

    Article  PubMed  PubMed Central  Google Scholar 

  38. Deniffel D, McAlpine K, Harder FN et al (2023) Predicting the recurrence risk of renal cell carcinoma after nephrectomy: potential role of CT-radiomics for adjuvant treatment decisions. Eur Radiol 33:5840–5850

    Article  PubMed  Google Scholar 

Download references

Funding

The research reported in this publication was supported by the Scientific Research Project of Fujian for Youth (no. 2020QNB062 and no. 2020QNB061).

Author information

Author notes

  1. Xiaoxia Li, Jinglai Lin and Hongliang Qi contributed equally to this work.

  2. Dengqiang Lin and Jianjun Zhou jointly supervised to this work.

Authors and Affiliations

  1. Department of Radiology, Zhongshan Hospital (Xiamen), Fudan University, Xiamen, 361015, China

    Xiaoxia Li, Yi Guo & Jianjun Zhou

  2. Department of Urology, Zhongshan Hospital (Xiamen), Fudan University, Xiamen, 361015, China

    Jinglai Lin & Dengqiang Lin

  3. Department of Clinical Engineering, Southern Medical University, Nanfang Hospital, Guangzhou, 510515, China

    Hongliang Qi

  4. Department of Radiology, Zhongshan Hospital, Fudan University, No 180, Fenglin Road, Xuhui District, Shanghai, 200032, China

    Chenchen Dai & Jianjun Zhou

  5. Xiamen Municipal Clinical Research Center for Medical Imaging, Xiamen, 361015, China

    Jianjun Zhou

  6. Fujian Province Key Clinical Specialty for Medical Imaging, Xiamen, 361015, China

    Jianjun Zhou

  7. Xiamen Key Laboratory of Clinical Transformation of Imaging Big Data and Artificial Intelligence, Xiamen, 361015, China

    Jianjun Zhou

Authors
  1. Xiaoxia Li
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Jinglai Lin
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Hongliang Qi
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Chenchen Dai
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Yi Guo
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Dengqiang Lin
    View author publications

    You can also search for this author in PubMed Google Scholar

  7. Jianjun Zhou
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

X.L. and L.J. conceived the project. X.L., Y.G., C.D., Q.H., and L.J. collected the data. X.L. analyzed the data; X.L. wrote the paper. D.L. and J.Z. revised this paper. All authors read and approved the final manuscript.

Corresponding authors

Correspondence to Dengqiang Lin or Jianjun Zhou.

Ethics declarations

Ethics approval and consent to participate

The study was conducted in accordance with the principles of the Declaration of Helsinki and approved by the institutional Ethics Committee in Zhongshan Hospital Fudan University, and the committee’s reference number was B2021-608R.

Consent for publication

The need for written informed consent was waived with the confirmation of patient data confidentiality by the institutional Ethics Committee for this retrospective study.

Competing interests

The authors declare that they have no competing interests.

Additional information

Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Li, X., Lin, J., Qi, H. et al. Radiomics predict the WHO/ISUP nuclear grade and survival in clear cell renal cell carcinoma. Insights Imaging 15, 175 (2024). https://doi.org/10.1186/s13244-024-01739-z

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1186/s13244-024-01739-z

Keywords