Establishment and Validation of a Machine Learning Prediction Model Based on Big Data for Predicting the Risk of Bone Metastasis in Renal Cell Carcinoma Patients

Xu, Chan; Liu, Wencai; Yin, Chengliang; Li, Wanying; Liu, Jingjing; Sheng, Wanli; Tang, Haotong; Li, Wenle; Zhang, Qingqing

doi:https://doi.org/10.1155/2022/5676570

Computational and Mathematical Methods in Medicine

On this page

Abstract Introduction Methods Results Discussion Conclusion Abbreviations Data Availability Ethical Approval Conflicts of Interest Authors’ Contributions Acknowledgments References Copyright Related Articles

Special Issue

Machine Learning Approaches Based on Multiscale Data in Diagnosis, Treatment and Prognosis of Diseases

View this Special Issue

Research Article | Open Access

Volume 2022 | Article ID 5676570 | https://doi.org/10.1155/2022/5676570

Establishment and Validation of a Machine Learning Prediction Model Based on Big Data for Predicting the Risk of Bone Metastasis in Renal Cell Carcinoma Patients

Chan Xu,^1,2Wencai Liu,³Chengliang Yin,⁴Wanying Li,²Jingjing Liu,⁵Wanli Sheng,⁶Haotong Tang,⁴Wenle Li,⁷and Qingqing Zhang⁸

Academic Editor: Xiaohan Ren

Received20 Jun 2022

Accepted28 Jul 2022

Published03 Oct 2022

Abstract

Purpose. Since the prognosis of renal cell carcinoma (RCC) patients with bone metastasis (BM) is poor, this study is aimed at using big data to build a machine learning (ML) model to predict the risk of BM in RCC patients. Methods. A retrospective study was conducted on 40,355 RCC patients in the SEER database from 2010 to 2017. LASSO regression and multivariate logistic regression analysis was performed to determine independent risk factors of RCC-BM. Six ML algorithm models, including LR, GBM, XGB, RF, DT, and NBC, were used to establish risk models for predicting RCC-BM. The prediction performance of ML models was weighed by 10-fold cross-validation. Results. The study investigated 40,355 patients diagnosed with RCC in the SEER database, where 1,811 (4.5%) were BM patients. Independent risk factors for BM were tumor grade, T stage, N stage, liver metastasis, lung metastasis, and brain metastasis. Among the RCC-BM risk prediction models established by six ML algorithms, the XGB model showed the best prediction performance (). Therefore, a network calculator based on the XGB model was established to individually assess the risk of BM in patients with RCC. Conclusion. The XGB risk prediction model based on the ML algorithm performed a good prediction effect on BM in RCC patients.

1. Introduction

Renal cell carcinoma (RCC), a renal space-occupying lesion originating from renal tubular epithelial cells, accounts for 4% of all malignant tumors [1]. As the most common type of renal cancer, RCC took up 85%-90% of renal malignant tumors in adults [2]. Since the early clinical manifestations of RCC are hidden, 20%-40% of patients will eventually suffer from metastatic RCC (mRCC) [3]. At present, surgical resection is the first-line treatment for RCC. However, in some RCC patients, distant metastasis after radical nephrectomy still reappeared or even occurred [4]. In addition, most RCC will become resistant to chemotherapy and radiotherapy once they develop into relapse or metastasis [5].

Bone metastasis (BM) is one of the most frequent sites of solid malignant tumors, the occurrence of which reveals the poor prognosis of tumor patients. As RCC has obvious osteotaxis in distant metastasis, bone is the second common distant metastasis site of RCC except the lung [6, 7]. Studies have shown that in about 20-35% of patients with RCC disease progression, renal cell carcinoma bone metastasis (RCC-BM) will emerge, such as the pelvis, spine, and ribs [3]. BM of RCC mainly leads to osteolytic destructive changes such as skeletal-related events (SER) [8], including pathological fracture, spinal cord and/or nerve root compression, and bone pain [9]. More than 70% of BM-RCC patients have experienced at least one SER during their survival, which severely reduced their quality of life and survival time [10]. Previous studies indicated that the prognosis of patients with RCC-BM was poor. Their median overall survival time (OST) was only 12-28 months, and the 5-year overall survival rate was only 11%, while the median OST of mRCC patients without BM was prolonged to 31 months and the 5-year overall survival rate increased to 47% [10, 11]. Therefore, bone metastasis is an important cause of death in patients with advanced RCC and it is crucial to predict the risk of RCC-BM.

Several previous studies reported the risk factors and prognostic factors of BM in RCC patients [12–15], and some developed traditional nomograms to predict the risk of RCC-BM [5]. However, there are few studies using machine learning (ML) method to construct the risk prediction model of RCC-BM based on big data. This study is aimed at (1) setting up the RCC-BM risk prediction model through ML and verifying the validity of the model with external data and (2) constructing a network calculator to facilitate clinicians to choose more reasonable diagnosis and treatment for RCC patients.

2. Methods

2.1. Study Population Selection

The training group data analyzed in this study are from the SEER database (http://seer.cancer.gov/about/), where the analysis of anonymous data is exempt from medical ethics review and does not require informed consent from patients.

2.2. Data Collection

All RCC data in the retrospective cohort study from 2010 to 2017 were extracted and subsumed as training group data with SEERStat (version 8.3.6) software. According to the inclusion/exclusion criteria, 40,355 patients were selected into the training group. The inclusion/exclusion criteria were as follows: (1) RCC was the first or primary tumor; (2) patients with RCC diagnosed by pathology (the validation group was diagnosed by at least two pathologists blindly); (3) patients with complete clinicopathological features, demographic data, and follow-up data; and (4) patients with RCC proved by autopsy or death were excluded from this study.

Based on the specific information of RCC patients from the SEER database, 17 variables were selected to determine the independent risk factors of BM in RCC patients, including marital status, age, race, serial number, survival time, survival status, gender, primary location, grade, side, pathological stage, T stage, N stage, tumor size, bone metastasis, brain metastasis, and liver metastasis. The risk prediction models were framed using data of the training group.

2.3. Establishment and Verification of Prediction Models

Six ML models, including logistic regression (LR), gradient boosting machine (GBM), extreme gradient boosting (XGB), Random Forest (RF), Decision Tree (DT), and Naive Bayesian model (NBC), were used to build prediction models, the performance of which was compared by 10-fold cross-validation method [16–19]. The model with the greatest AUC value was regarded as the preferred prediction model, whose corresponding network calculator is designed to individually assess the risk of BM in patients with RCC [20–23].

2.4. Statistical Analysis

The measurement data is expressed in mean (SD), and the counting data is expressed in frequency (percentage). Independent samples -test, chi-square test, LASSO regression analysis, univariate and multivariate logistic regression analysis, 10-fold cross-validation, and other statistical analysis were performed by SPSS 26.0 (SPSS Inc., Chicago, USA) software. values < 0.05 were considered statistically significant. R software (version 4.0.5, https://www.r-project.org/) was applied for drawing the correlation heat map and ROC curve and developing a predictive model which used the “shiny” package to establish a web calculator.

3. Results

3.1. Characteristics of the Study Population

A total of 40,355 RCC patients from the SEER database were included in this study to establish the training group. 4.5% (1811 cases) of RCC patients progressed to BM. Then, six risk factors for predicting BM in RCC patients were screened by LASSO regression, including tumor grade, T stage, N stage, liver metastasis, lung metastasis, and brain metastasis (Figure 1), which were viewed as predictors in the correlation heat map (Figure 2).

(a)

(b)

3.2. Independent Risk Factors of BM in RCC Patients

Univariate and multivariate logistic regression analyses were carried out to value the independent risk factors of BM in RCC patients. Univariate logistic regression analysis displayed that brain metastasis, liver metastasis, lung metastasis, tumor grade, T stage, and N stage were importantly associated with BM in patients with RCC (). Further multivariate logistic regression analysis indicated that brain metastasis (, ), liver metastasis (, ), lung metastasis (, ), tumor grade (poorly differentiated: , ; undifferentiated: , ; undifferentiated: , ), T stage (T2 stage: , ; T3 stage: , ; T4 stage: ,; and TX stage: , ), and N stage (N1 stage: , ; NX stage: , ) were independent risk factors for BM in RCC patients (, Table 1).

3.3. Selection and Verification of the Prediction Models

The prediction performance of six ML algorithm models (LR, NBC, DT, RF, GBM, and XGB) was compared by 10-fold cross-validation method, which indicates that the prediction value of all models above was great (). In descending order, the predictive ability of models is XGB, RF, GBM, NBC, LR, and DT, of which XGB is the best in predicting RCC-BM (average , Figure 3). Therefore, the XGB model is selected as the optimal prediction model finally.

The importance of each risk factor is not identical in different ML prediction models. Among them, lung metastasis is the most important clinical feature in the six models, while brain metastasis is of the least significance feature in RF, GBM, and XGB models, familiar as tumor grade in NBC and DT models and N stage in the LR model. In the XGB model, the independent risk factors are arranged according to their importance, which are lung metastasis, T stage, liver metastasis, tumor grade, N stage, and brain metastasis. The value of risk factors in other models is shown in Figure 4.

(a)

(b)

(c)

(d)

(e)

(f)

3.4. Construction of the Web Calculator

Based on the GBM model possessing the best performance, a risk web calculator was designed in this study (https://share.streamlit.io/liuwencai5/renal_bone/main/renal_bone.py). By inputting the relevant clinicopathological variables of RCC patients, clinicians could predict the risk of BM in patients with RCC (Figure 5).

4. Discussion

As an important marker of poor prognosis in patients with RCC, early detection and intervention for BM are urgently needed. Guo et al. [14] analyzed the data of 45,824 RCC patients recorded in the SEER database from 2010 to 2014 and found that 3.29% patients were diagnosed with BM at the initial diagnosis. In our study, 4.5% RCC patients in the training group developed into BM, while Zekri et al. [24] reported that 30-40% of advanced RCC patients turned into BM. Therefore, the incidence rate of BM may be underestimated since the patients showed no symptoms when the initial diagnosis was made or BM appeared at the advanced stage of RCC disease which was not recorded in the SEER database. At present, the guidelines only recommend bone imaging for patients with uncomfortable symptoms or abnormal alkaline phosphatase level [25]. Thus, consequently, patients with asymptomatic BM could not be treated timely and effectively. At present, bone metastasis has been recognized as one of the adverse prognostic factors of RCC patients [8]. In addition, the resection of whole spinal BM can prolong the survival time potentially for the patients with isolated spinal BM and no visceral metastasis [26], but the prognosis of BM in RCC patients is still poor compared with that of other tumors such as lung cancer. With the improved bone-targeted therapy of BM patients, the average OST of BM patients is 12.0-31.8 months [27–29]. The lack of effective chemotherapeutic drugs may be the main reason for the poor outcome of RCC patients with BM [11]. Due to the occult onset and poor prognosis of BM, it is necessary to study the risk factors of BM in patients with RCC. Additionally, early identification and evaluation of BM are of great significance to improve the precision of the diagnosis, determine the treatment plans, and prevent RCC complications such as SER in patients with symptomless BM.

In this study, multivariate logistic regression analysis revealed that brain metastasis, liver metastasis, lung metastasis, poor tumor differentiation, high T stage, and N stage were independent risk factors for BM in RCC patients. Similarly, Guo et al. [14] found that male gender; higher T stage; lymph node involvement; poor tumor differentiation; presence of lung, liver, and brain metastasis; and the collecting duct type of RCC were positively associated with BM occurrence. Furthermore, Fan et al. [30], using nomogram to quantify the risk of RCC-BM patients, found that the independent factors of RCC complicated with BM include grade, histological type, N stage, operation, brain metastasis, and lung metastasis, which was basically consistent with our research results. Additionally, through a retrospective analysis of 372 RCC patients, Chen et al. [31] discovered that the concentrations of ALP, calcium, and Hb were potential risk factors for bone metastasis in patients with RCC. U/L, mmol/L, and g/L in newly diagnosed RCC patients suggest that BM is more likely to occur in these patients; hence, close monitoring and active bone scanning should be carried out to determine whether bone metastasis has happened. With the in-depth study of RCC-BM, more and more prognostic factors of bone metastasis also have been discovered. Yoshiyama et al. [32] considered that patients’ age, ECOG performance, histology, MSKCC prognosis score, concomitant metastasis, and the time from nephrectomy to bone metastasis were important factors related to the prognosis of RCC-BM. Subsequently, Ruatta et al. [15] tracked 1750 RCC patients and found that MSKCC score, BMs number, and radical resection were essential prognostic factors for RCC patients with BMs.

With the development of TNM staging system and pathological classification criteria of renal cell carcinoma, a variety of RCC prognosis analysis systems have appeared. But they have several limitations. TNM staging system depends on three pathological indexes while it ignores other risk factors, which reduces the accuracy of prognosis prediction of RCC patients. MSKCC model (Memorial Sloan Kettering Cancer Center-based poor-risk groups) and IMDC model (International Metastatic Renal Cell Carcinoma Database Consortium-based poor-risk groups) [27] lack the function of comprehensive analysis for patients. GRCC model (Gustave Roussy Cancer Campus) [33] is more accurate and convenient than the MSKCC and IMDC prognostic analysis model, but it is not designed for RCC-BM patients specifically. The B-FOM scoring system (Fujimoto–Owari–Miyake bone score) is characterized by bone metastasis specificity, yet poor tissue source specificity is its short board [34]. The traditional Cox regression or logistic regression analysis is visualized in this article; thus, clinicians can easily calculate the probability of BM in RCC patients without understanding the complex underlying mathematical formula. Distinguished from prediction analysis system or model mentioned above, this study creatively developed and verified six machine learning algorithms models, which were specially performed to estimate the risk of BM in RCC patients. The XGB model with the best prediction performance was selected through 10-fold verification methods, and an online calculator was established to evaluate the individual probability of RCC-BM. The ML-based model can be used to guide clinical treatment decisions, help clinicians better predict the BM risk, and take necessary interventions to improve the survival time and life quality of RCC patients.

The limitations of our study could not be ignored though. Firstly, as a retrospective cohort study, the inevitable selection bias may affect the results considerably. Since the SEER database only collects the initial diagnosis results, BM arising in the advanced stage of RCC may be omitted. Secondly, the deficiency of external validation using the data from the local validation group patients could not assess the accuracy of the selected XGB models in diagnosing BM in RCC patients and therefore establish the other clinical utility analysis such as probability density functions (PDF) and clinical utility curves (CUC). Additionally, we were unable to obtain some effective indicators for predicting RCC-BM from SEER, such as transforming growth factor-β (TGF-β) [35], fibroblast growth factor (FGF) [1], insulin-like growth factor [36], bone morphogenetic protein [37], AFP [38], CA-199 [31], and Fuhrman nuclear grade [39]. Future studies are needed to incorporate tumor characteristics, laboratory results, and treatment regimens to establish a higher dimensional predictive model.

5. Conclusion

This study retrospectively analyzed the independent risk factors of BM in renal cell carcinoma based on the SEER database, including tumor grade, T stage, N stage, liver metastasis, lung metastasis, and brain metastasis. On the foundation of the SEER dataset, we constructed and validated six machine learning models including LR, GBM, XGB, RF, DT, and NBC and subsequently selected XGB as the optimal prediction model. The network calculator designed on the basis of XGB provided important support for clinicians to make accurate treatment decisions.

Abbreviations

BM:	Bone metastasis
RCC:	Renal cell carcinoma
ML:	Machine learning
SEER:	Surveillance, Epidemiology, and End Results
LR:	Logistic regression
GBM:	Gradient boosting machine
XG:	Extreme gradient boosting
RF:	Random Forest
DT:	Decision Tree
NBC:	Naive Bayesian model
AUC:	Area under the curve
mRCC:	Metastatic RCC
SER:	Skeletal-related events
OST:	Overall survival time
PDF:	Probability density functions
CUC:	Clinical utility curves
MSKCC model:	Memorial Sloan Kettering Cancer Center-based poor-risk groups
IMDC model:	International Metastatic Renal Cell Carcinoma Database Consortium-based poor-risk groups
GRCC model:	Gustave Roussy Cancer Campus
B-FOM scoring system:	Fujimoto–Owari–Miyake bone score
TGF-β:	Transforming growth factor-β
FGF:	Fibroblast growth factor.

Data Availability

The training group data analyzed in this study are from the SEER database (http://seer.cancer.gov/about/), where the analysis of anonymous data is exempt from medical ethics review and does not require informed consent from patients.

Ethical Approval

This study is based on the SEER database and does not require ethical approval.

Conflicts of Interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as potential conflict of interest.

Authors’ Contributions

CX, WCL, and CLY are all considered as the first authors. QQZ, WLL, and HTT completed the study design. CX and WLL performed the study and collected and analyzed the data. QQZ, STD, and CX drafted the manuscript. WYL, XFX, XMQ, and JJL provided the expert consultations and suggestions. WCL and WLS conceived of the study, participated in its design and coordination, and helped to embellish language. All authors reviewed the final version of the manuscript. Chan Xu, Wencai Liu, and Chengliang Yin contributed equally to this work.

Acknowledgments

This study was supported by Shaanxi Provincial Health and Health Research Fund Project (2022E006).

References

Y. Adachi, H. Kamiyama, K. Ichikawa et al., “Inhibition of FGFR reactivates IFNγ signaling in tumor cells to enhance the combined antitumor activity of lenvatinib with anti-PD-1 antibodies,” Cancer Research, vol. 82, no. 2, pp. 292–306, 2022.
View at: Publisher Site | Google Scholar
Y. Wang, Y. Zhang, P. Wang, X. Fu, and W. Lin, “Circular RNAs in renal cell carcinoma: implications for tumorigenesis, diagnosis, and therapy,” Molecular Cancer, vol. 19, no. 1, p. 149, 2020.
View at: Publisher Site | Google Scholar
S. V. L. Vig, E. Zan, and S. K. Kang, “Imaging for metastatic renal cell carcinoma,” The Urologic Clinics of North America, vol. 47, no. 3, pp. 281–291, 2020.
View at: Publisher Site | Google Scholar
T. Klatte, S. H. Rossi, and G. D. Stewart, “Prognostic factors and prognostic models for renal cell carcinoma: a literature review,” World Journal of Urology, vol. 36, no. 12, pp. 1943–1952, 2018.
View at: Publisher Site | Google Scholar
A. Brozovich, B. Garmezy, T. Pan, L. Wang, M. C. Farach-Carson, and R. L. Satcher, “All bone metastases are not created equal: revisiting treatment resistance in renal cell carcinoma,” Journal of Bone Oncology, vol. 31, article 100399, 2021.
View at: Publisher Site | Google Scholar
J. Wang, H. You, J. Qi, C. Yang, Y. Ren, and H. Cheng, “Autocrine and paracrine STIP1 signaling promote osteolytic bone metastasis in renal cell carcinoma,” Oncotarget, vol. 8, no. 10, pp. 17012–17026, 2017.
View at: Publisher Site | Google Scholar
W. Li, S. Dong, H. Wang et al., “Risk analysis of pulmonary metastasis of chondrosarcoma by establishing and validating a new clinical prediction model: a clinical study based on SEER database,” BMC Musculoskeletal Disorders, vol. 22, no. 1, p. 529, 2021.
View at: Publisher Site | Google Scholar
M. Umer, Y. Mohib, M. Atif, and M. Nazim, “Skeletal metastasis in renal cell carcinoma: a review,” Annals of Medicine and Surgery, vol. 27, pp. 9–16, 2018.
View at: Publisher Site | Google Scholar
T. Negishi, N. Furubayashi, D. Takamatsu et al., “Radiographical efficacy of systemic treatment for bone metastasis from renal cell carcinoma,” Oncology Letters, vol. 20, no. 5, p. 1, 2020.
View at: Publisher Site | Google Scholar
S. C. Chen and P. L. Kuo, “Bone metastasis from renal cell carcinoma,” International Journal of Molecular Sciences, vol. 17, no. 6, p. 987, 2016.
View at: Publisher Site | Google Scholar
E. Woodward, S. Jagdev, L. McParland et al., “Skeletal complications and survival in renal cancer patients with bone metastases,” Bone, vol. 48, no. 1, pp. 160–166, 2011.
View at: Publisher Site | Google Scholar
C. R. Goodwin, A. K. Ahmed, C. Boone et al., “The challenges of renal cell carcinoma metastatic to the spine: a systematic review of survival and treatment,” Global Spine Journal, vol. 8, no. 5, pp. 517–526, 2018.
View at: Publisher Site | Google Scholar
A. Guida, B. Escudier, and L. Albiges, “Treating patients with renal cell carcinoma and bone metastases,” Expert Review of Anticancer Therapy, vol. 18, no. 11, pp. 1135–1143, 2018.
View at: Publisher Site | Google Scholar
Q. Guo, C. Zhang, X. Guo et al., “Incidence of bone metastasis and factors contributing to its development and prognosis in newly diagnosed renal cell carcinoma: a population-based study,” Cancer Management and Research, vol. 10, pp. 2935–2944, 2018.
View at: Publisher Site | Google Scholar
F. Ruatta, L. Derosa, B. Escudier et al., “Prognosis of renal cell carcinoma with bone metastases: experience from a large cancer centre,” European Journal of Cancer, vol. 107, pp. 79–85, 2019.
View at: Publisher Site | Google Scholar
W. Li, Y. Liu, W. Liu et al., “Machine learning-based prediction of lymph node metastasis among osteosarcoma patients,” Frontiers in Oncology, vol. 12, article 797103, 2022.
View at: Publisher Site | Google Scholar
W. Li, W. Liu, M. F. Hussain et al., “An external-validated prediction model to predict lung metastasis among osteosarcoma: a multicenter analysis based on machine learning,” Computational Intelligence and Neuroscience, vol. 2022, Article ID 2220527, 10 pages, 2022.
View at: Publisher Site | Google Scholar
S. Jiang, X. Ren, S. Liu et al., “Integrated analysis of the prognosis-associated RNA-binding protein genes and candidate drugs in renal papillary cell carcinoma,” Frontiers in Genetics, vol. 12, article 627508, 2021.
View at: Publisher Site | Google Scholar
X. Ren, X. Chen, X. Zhang et al., “Immune microenvironment and response in prostate cancer using large population cohorts,” Frontiers in Immunology, vol. 12, article 686809, 2021.
View at: Publisher Site | Google Scholar
W. Li, H. Wang, S. Dong et al., “Establishment and validation of a nomogram and web calculator for the risk of new vertebral compression fractures and cement leakage after percutaneous vertebroplasty in patients with osteoporotic vertebral compression fractures,” European Spine Journal, vol. 31, no. 5, pp. 1108–1121, 2022.
View at: Publisher Site | Google Scholar
W. Li, J. Wang, W. Liu et al., “Machine learning applications for the prediction of bone cement leakage in percutaneous vertebroplasty,” Frontiers in Public Health, vol. 9, article 812023, 2021.
View at: Publisher Site | Google Scholar
J. C. Wang, R. Fu, X. W. Tao et al., “A radiomics-based model on non-contrast CT for predicting cirrhosis: make the most of image data,” Biomarker Research, vol. 8, no. 1, p. 47, 2020.
View at: Publisher Site | Google Scholar
J. Wang, S. Tang, Y. Mao et al., “Radiomics analysis of contrast-enhanced CT for staging liver fibrosis: an update for image biomarker,” Hepatology International, vol. 16, no. 3, pp. 627–639, 2022.
View at: Publisher Site | Google Scholar
J. Zekri, N. Ahmed, R. E. Coleman, and B. W. Hancock, “The skeletal metastatic complications of renal cell carcinoma,” International Journal of Oncology, vol. 2001, pp. 379–382, 2001.
View at: Publisher Site | Google Scholar
R. J. Motzer, E. Jonasch, N. Agarwal et al., “Kidney cancer, version 2.2017, NCCN clinical practice guidelines in oncology,” Journal of the National Comprehensive Cancer Network, vol. 15, no. 6, pp. 804–834, 2017.
View at: Publisher Site | Google Scholar
J. Zhai, N. Liu, H. Wang, G. Huang, and L. Man, “Clinical characteristics and prognosis of renal cell carcinoma with spinal bone metastases,” Frontiers in Oncology, vol. 11, article 659779, 2021.
View at: Publisher Site | Google Scholar
Y. Dong, Z. Wang, X. Lu et al., “Clinical outcomes of 168 Chinese patients after local surgery for bone metastases arising from advanced renal cell carcinoma,” Cancer, vol. 126, Supplement9, pp. 2079–2085, 2020.
View at: Publisher Site | Google Scholar
Y. Du, S. Pahernik, B. Hadaschik et al., “Survival and prognostic factors of patients with renal cell cancer with bone metastasis in the era of targeted therapy: a single-institution analysis,” Urologic Oncology, vol. 34, no. 10, pp. 433.e1–433.e8, 2016.
View at: Publisher Site | Google Scholar
M. Santoni, A. Conti, G. Procopio et al., “Bone metastases in patients with metastatic renal cell carcinoma: are they always associated with poor prognosis?” Journal of Experimental & Clinical Cancer Research, vol. 34, no. 1, 2015.
View at: Publisher Site | Google Scholar
Z. Fan, Z. Huang, and X. Huang, “Bone metastasis in renal cell carcinoma patients: risk and prognostic factors and nomograms,” Journal of Oncology, vol. 2021, Article ID 5575295, 17 pages, 2021.
View at: Publisher Site | Google Scholar
X. Y. Chen, M. Lan, Y. Zhou et al., “Risk factors for bone metastasis from renal cell cancer,” Journal of Bone Oncology, vol. 9, pp. 29–33, 2017.
View at: Publisher Site | Google Scholar
A. Yoshiyama, T. Morii, M. Susa et al., “Preoperative evaluation of renal cell carcinoma patients with bone metastases on risks for blood loss, performance status and lethal event,” Journal of Orthopaedic Science, vol. 22, no. 5, pp. 924–930, 2017.
View at: Publisher Site | Google Scholar
L. Derosa, M. A. Bayar, L. Albiges, G. Le Teuff, and B. Escudier, “A new prognostic model for survival in second line for metastatic renal cell carcinoma: development and external validation,” Angiogenesis, vol. 22, no. 3, pp. 383–395, 2019.
View at: Publisher Site | Google Scholar
T. Owari, M. Miyake, Y. Nakai et al., “A genitourinary cancer-specific scoring system for the prediction of survival in patients with bone metastasis: a retrospective analysis of prostate cancer, renal cell carcinoma, and urothelial carcinoma,” Anticancer Research, vol. 38, no. 5, pp. 3097–3103, 2018.
View at: Publisher Site | Google Scholar
Y. Hong, L. Gong, B. Yu, and Y. Dong, “PPM1A suppresses the proliferation and invasiveness of RCC cells via Smad2/3 signaling inhibition,” Journal of Receptor and Signal Transduction Research, vol. 41, no. 3, pp. 245–254, 2021.
View at: Publisher Site | Google Scholar
W. Solarek, M. Koper, S. Lewicki, C. Szczylik, and A. M. Czarnecka, “Insulin and insulin-like growth factors act as renal cell cancer intratumoral regulators,” Journal of Cell Communication and Signaling, vol. 13, no. 3, pp. 381–394, 2019.
View at: Publisher Site | Google Scholar
V. Brand, C. Lehmann, C. Umkehrer et al., “Impact of selective anti-BMP9 treatment on tumor cells and tumor angiogenesis,” Molecular Oncology, vol. 10, no. 10, pp. 1603–1620, 2016.
View at: Publisher Site | Google Scholar
H. Z. Zhang and S. Y. Wang, “AFP-producing Xp11 translocation renal cell carcinoma: case report and review of the literature,” Current Problems in Cancer, vol. 45, no. 3, article 100689, 2021.
View at: Publisher Site | Google Scholar
T. F. Lin, W. R. Lin, M. Chen et al., “Compare Fuhrman nuclear and chromophobe tumor grade on chromophobe RCC,” Open Medicine, vol. 14, pp. 336–342, 2019.
View at: Publisher Site | Google Scholar

Copyright

Copyright © 2022 Chan Xu et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

PDF Download Citation

Download other formats

Order printed copies

Views

366

Downloads

551

Citations