Prognostic Value of SOX9, E-Cadherin, and KLF4 Expressions in Clear Cell Renal Carcinoma

By Seyma Ozturk¹, Selma Erdogan Duzcu²

Affiliations

1. Department of Pathology, Sanliurfa Training and Research Hospital, Sanliurfa, Turkiye
2. Department of Pathology, Bolu Abant Izzet Baysal University, Faculty of Medicine, Bolu, Turkiye

doi: 10.29271/jcpsp.2025.04.457

ABSTRACT
Objective: To analyse the immunohistochemical staining of stem cell-related factors SOX9 and KLF4 in radical and partial nephrectomy specimens and to determine their relationship to grade, metastasis, stage, other prognostic parameters, and their predictive value.
Study Design: Descriptive study.
Place and Duration of the Study: Department of Pathology, Bolu Abant Izzet Baysal University, Faculty of Medicine, Bolu, Turkiye, from 2015 to 2022.
Methodology: The study included 92 cases diagnosed with clear cell renal cell carcinoma (ccRCC) from sections and blocks of partial and radical nephrectomies. Formalin-fixed and paraffin-embedded blocks were sliced. A Leica Bond Max staining device stained SOX9, E-cadherin, and KLF4 onto the slides. Nuclear staining for SOX9 and KLF4 and membranous staining for E-cadherin were evaluated. Since E-cadherin loss has an important role in the development of malignancy, the E-cadherin staining pattern was considered when evaluating the staining patterns of SOX9 and KLF4. The Mann-Whitney U and Kruskal-Wallis tests were applied to compare immunohistochemical data with prognostic factors such as gender, tumour size, grade, metastasis, and stage. The statistical analyses were conducted using the IBM SPSS version 20, with a p-value of less than 0.05 considered as statistically significant.
Results: Significant correlations were observed between SOX9, KLF4, and E-cadherin expression levels and high nuclear grade, with p-values of 0.005, <0.001, and 0.002, respectively. In addition, the overexpression of SOX9 was significantly linked to distant metastasis and stage, with p-values of 0.010 and 0.025, respectively.
Conclusion: SOX9, E-cadherin, and KLF4 expressions are linked to unfavourable prognostic factors in ccRCC. Thus, these immunohistochemical stains may serve as potential biomarkers in this cancer and aid in identifying prognosis and treatment response.

Key Words: Clear cell renal carcinoma, SOX9 transcription factor, E-cadherin, Kruppel-like transcription factors, Prognosis.

INTRODUCTION

Renal cell carcinoma (RCC) is the most common malignant kidney tumour in adults, representing about 2% of all adult cancers. The incidence of renal cancer is typically reported as 6.1 per 100,000 in males and 3.2 per 100,000 in females.¹ The most common and aggressive histological type of RCC is clear cell RCC (ccRCC), developing in 60-80% of cases.^2,3

Recent research has shown that cancer stem cells (CSCs) are now recognised as the key factor in tumour formation and metastasis. CSCs have various cell properties, including differentiation capacity, self-renewal, and resistance to apoptosis. Stem cell regulatory proteins are considered potential oncogenes due to their ability to regulate the CSC phenotypes.⁴ CSCs have previously been shown to exist in RCC, contributing to resistance against radiotherapy and chemotherapy.^5,6

SOX9 (Sex-determining region Y (SRY)-related HMG box 9) belongs to the SOX gene superfamily. Recent studies have suggested that SOX9 continues to be expressed in stem cell pools of ectoderm and endoderm-derived tissues, which may regulate CSCs.⁴ SOX9 functions as a transcription factor in the cell. SOX9 gene is identified as an oncogene in lung adenocarcinoma, breast carcinoma, colorectal carcinoma, and prostate carcinoma.^7,8

E-cadherin is a transmembrane glycoprotein involved in cell- cell adhesion found in epithelial tissues.⁹ The loss of E-cadherin homophilic binding disrupts cell adhesion and intercellular connections, enabling cells to detach from the primary tumour, invade nearby tissues, and develop metastatic tumours.¹⁰ E-cadherin has been previously demonstrated to exhibit tumour suppressor properties in various cancers, including hepatocellular carcinoma, head and neck cancers, squamous cell carcinoma of the skin and oesophagus, and melanoma.^11,12

Kruppel-like factor 4 (KLF4) is part of the Kruppel-like factor family, which contains zinc fingers that play a role in various physiological processes, including cell cycle regulation, pluripotency, epidermal development, and the maintenance of tissue homeostasis. It acts as a transcription factor in the cell. Depending on the target gene, it can activate or suppress transcription through different mechanisms.¹³ KLF4 downregulation is often observed in gastric, colorectal, hepatic, and prostate cancers.^14-16

The aim of this study was to examine the expressions of SOX9, KLF4, and E-cadherin in ccRCC cases, explore the relationships between these markers and among ccRCC patients, and identify the associations between these markers and clinicopathological parameters with prognostic relevance.

METHODOLOGY

The study included partial and radical nephrectomy specimens from 92 ccRCC patients who underwent surgery at the Department of Pathology, Bolu Abant Izzet Baysal University, Faculty of Medicine, Bolu, Turkiye, from 2015 to 2022. Nephrectomy materials diagnosed with non-clear cell renal cell carcinoma (ccRCC), non-tumour resection materials, and needle biopsies were excluded. The selected blocks were those that best represented the tumour and were most appropriate for immunohistochemical analysis.

Formalin-fixed paraffin-embedded ccRCC tissue blocks were retrieved from the pathology department archives for retrospective analysis. All gross specimens were meticulously examined according to the guidelines established by the College of American Pathologists. Histological diagnosis was made through H&E staining of sections obtained from these blocks. Data on various parameters, such as age, gender, tumour size, grade, stage, perirenal fat invasion, renal sinus invasion, sarcomatoid and rhabdoid differentiation, lymphatic, and vascular invasion, and distant metastasis, were recorded.

Four-micron sections were obtained from the formalin-fixed paraffin-embedded blocks. These sections were stained with primary antibodies against SOX9, E-cadherin, and KLF4 using a Leica Bond-Max fully automated immunohistochemistry system. Immunohistochemical staining was carried out with a standard compact polymer technology kit. Two pathologists independently evaluated the immunohistochemically stained slides.

The sections were stained using the SOX9 antibody (Epitomics, Burlingame, USA, clone: EP317, diluted at 1/150). Colon adenocarcinoma sections were positive controls for SOX9, with nuclear staining considered a positive result. The positive cells’ total percentage (1: 1-10%, 2: 11-40%, 3:41-70%, 4: 71-100 %) and staining intensities (0-3) were determined.

Sections were stained with E-cadherin antibody (Leica, Newcastle, UK, clone: 36B5, diluted: 1/25). Membranous staining for E-cadherin was considered positive, and metastatic invasive ductal carcinoma in the axillary lymph node was used as a positive control. The total percentage of positive cells (0: No staining, 1: 1-24%, 2: 25-49%, 3: 50-74%, 4: 75-100%) and their staining intensities (0-3) were assessed.

Sections were stained with KLF4 antibody (Millipore, Temecula, USA, clone: 1E6, diluted: 1/250). Normal colon tissue sections served as positive controls. Nuclear staining was regarded as a positive indicator for KLF4. The total percentage of positive cells (0: No staining, 1: ≤1%, 2: >1%) and their staining intensities (0-3) were evaluated.

Finally, a combined immune score for each immunohistochemical staining was calculated by summing the staining intensity and percentage scores. The combined immune scores were then analysed.

Considering the previous studies, taking at least 93 observations with 80% power and 5% margin of error was sufficient, hypothesising that the effect size between the groups would be 0.31. The number of samples was calculated with the PASS 11.0 programme. In descriptive statistics, mean and standard deviation or median and 25^th – 75^th percentile values ​​will be given for numerical variables, and number and percentage values ​​will be given for categorical variables. Kolmogorov-Smirnov test was used to make the normality assumption. Since the assumptions were not met, the Mann-Whitney U test was used to determine the difference between the two groups. The Kruskal-Wallis test was used to determine whether there were differences between three or more groups because the assumptions were unmet. In case of a difference, a pairwise comparison test determined the groups that created the difference. Statistical analysis was performed using the Kruskal-Wallis test to determine if prognostic factor measurements such as tumour grade and stage were based on immunohistochemical staining positivity. A paired samples t-test was performed to reveal whether there was any significant difference between immunohistochemical staining positivity in the tumour and peritumoural kidney parenchyma. All statistical analyses were conducted using the IBM SPSS version 20, with a p-value of less than 0.05 considered as statistically significant.

RESULTS

A total of 92 cases were included in the study, comprising 69 (75%) males and 23 (25%) females, with a mean age of 59.5 ± 10.6 years ranging from 29 to 85 years. The average tumour diameter was 5.46 cm, and the median tumour diameter was 4.50 cm. Based on tumour size, patients were categorised into two groups, ≤4 cm and >4 cm, according to median tumour diameter. Tumour grading was assessed according to the WHO/ISUP classification. Histological grades were distributed with 9 (9.8%) patients in grade 1, 53 (57.6%) patients in grade 2, 23 (25%) patients in grade 3, and 7 (7.6%) patients in grade 4 (Figure 1A-D). Distant metastasis was observed in 14 patients (20%). Regarding tumour stage, 57 (62%) of patients were classified as stage 1, 4 (4.3%) as stage 2, 15 (16.3%) as stage 3, and 16 (17.4%) as stage 4. The clinicopathological characteristics of the cases are presented in Table I.

SOX9 nuclear staining was observed in tumour cells in all cases. Weak staining was detected in 17 (18.5%) cases, moderate staining in 32 (34.8%) cases, and intense staining in 43 (46.7%) cases (Figure 2A-C). The combined immune scores from the staining extent and intensity ranged between 2 and 7. SOX9 staining was found to be more intense at the invasive margins, where it merged with the normal renal parenchyma. Higher SOX9 expression was observed in tumour tissue compared to peritumoural kidney parenchyma (p <0.001).

Table  I:  The  clinicopathological  features  of  the  patients.

Parameters		Total n = 92 n (%)
Age	≤60	47 (51.1%)
	>60	45 (48.9%)
Gender	Female	23 (25%)
	Male	69 (75%)
Tumour diameter	≤4 cm	40 (43.5%)
	>4 cm	52 (56.5%)
Grade	G1	9 (9.8%)
	G2	53 (57.6%)
	G3	23 (25%)
	G4	7 (7.6%)
Lymphatic and vascular invasion	Present	14 (15.2%)
	Absent	78 (84.8%)
Metastasis	M0	56 (80%)
	M1	14 (20%)
Stage	I	57 (62%)
	II	4 (4, 3%)
	III	15 (16, 3%)
	IV	16 (17, 4%)

Figure 1: (A) Grade 1, cells with abundant clear cytoplasm and small hyperchromatic nuclei are seen. Nucleoli are indistinct at 40x magnification. (B) In grade 2, eosinophilic nucleoli are evident at 40x magnification. (C) Grade 3 cells with eosinophilic cytoplasm with prominent eosinophilic nucleoli at 40x magnification. (D) Grade 4, large pleomorphic cells are seen at 40x magnification.

Figure 2: Mild (image (A), x200), moderate (image (B), x200), strong (image (C), x200) SOX9 expressions in tumour cells.

Figure 3: Negative (image (A), x200), mild (image (B), x200), moderate (image (C), x200), strong (image (D), x200) E-cadherin expressions in tumour cells.

Significant associations were found between SOX9 expression and high nuclear grade (p = 0.005), sarcomatoid and rhabdoid differentiation (p = 0.010), distant metastasis (p = 0.025), and stage (p = 0.020). However, no significant association was observed between SOX9 expression and age, gender, tumour size, necrosis, lymphatic, and vascular invasion, perirenal fat invasion, or renal sinus invasion.

E-cadherin exhibited positive membrane-associated staining in 88 (95.7%) cases. No staining was observed in 4 (4.3%) cases, weak staining in 28 (30.4%) cases, moderate staining in 44 (47.8%) cases, and intense staining in 16 (17.4%) cases (Figure 3A-D). The combined immune score from the staining extent and intensity ranged between 0 and 7. Higher E-cadherin expression was seen in peritumoural kidney parenchyma compared to tumour tissue (p <0.001).

Significant correlations were found between decreased E-cadherin expression and high nuclear grade (p = 0.003), sarcomatoid and rhabdoid differentiation (p = 0.036), and distant meta- stasis (p = 0.033).

Considering combined immune scores by clinical staging, E-cadherin expression was lowest in Stage 4 tumours and highest in Stage 3 tumours. No significant differences were found between the E-cadherin combined immune scores and stage groups (p = 0.304).

E-cadherin expression did not differ significantly in age, gender, tumour diameter, necrosis, lymphatic and vascular invasion, perirenal fat invasion, and renal sinus invasion.

KLF4 demonstrated positive nuclear staining in 21 (22.8%) cases. No staining was observed in 71 (77.2%) cases, weak staining in 5 (5.4%) cases, moderate staining in 9 (9.8%) cases, and intense staining in 7 (7.6%) cases (Figure 4A-D). The combined immune score from the staining extent and intensity ranged from 0 to 5. In the peritumoural kidney parenchyma, nuclear staining was observed in the tubules, mainly in the distal and collecting tubules. Mild positive staining was observed in 10 (10.9%) of 92 cases. Higher KLF4 expression was seen in peritumoural kidney parenchyma compared to tumour tissue (p <0.001).

Table II: SOX9, E-cadherin, and KLF4 expression by prognostic factors.

CCRCC		SOX9		E-cadherin		KLF4
		Mean ± SD	p-value	Mean ± SD	p-value	Mean ± SD	p-value
Age	≤60	5.00 ± 1.69	0.392	4.13 ± 1.81	0.456	0.64 ± 1.49	0.234
	>60	4.69 ± 1.83		4.42 ± 1.43		0.93 ± 1.58
Gender	Female	4.78 ± 1.85	0.727	4.09 ± 1.73	0.360	1.04 ± 1.69	0.317
	Male	4.87 ± 1.74		4.33 ± 1.61		0.70 ± 1.48
Tumour diameter	≤4 cm	5.09 ± 1.63	0.198	4.17 ± 1.51	0.616	1.26 ± 1.84	0.004
	>4 cm	4.61 ± 1.86		4.37 ± 1.76		0.30 ± 0.96
Grade	G1	5.56 ± 1.50	0.005	4.78 ± 1.30	0.002	3.00 ± 2.06	0.000
	G2	4.51 ± 1.76		4.68 ± 1.49		0.74 ± 1.44
	G3	4.74 ± 1.68		3.78 ± 1.38		0.26 ± 0.91
	G4	6.86 ± 0.37		2.14 ± 2.03		0.00 ± 0.00
Lymphatic and vascular invasion	Present	5.14 ± 1.83	0.464	3.57 ± 2.10	0.178	0.36 ± 0.92	0.336
	Absent	4.79 ± 1.75		4.40 ± 1.52		0.86 ± 1.61
Metastatis	M0	4.68 ± 1.71	0.025	4.44 ± 1.51	0.040	0.90 ± 1.63	0.107
	M1	5.79 ± 1.76		3.36 ± 2.02		0.14 ± 0.53
Stage	I	4.98 ± 1.67	0.020	4.33 ± 1.36	0.200	1.18 ± 1.79	0.020
	II	3.50 ± 1.29		5.50 ± 1.29		0.00 ± 0.00
	III	3.93 ± 1.79		4.40 ± 1.99		0.20 ± 0.77
	IV	5.56 ± 1.75		3.63 ± 2.09		0.13 ± 0.50

Significant associations existed between reduced KLF4 expression and high nuclear grade (p <0.001), high stage (p = 0.020), tumour diameter (p = 0.004), and renal sinus invasion (p = 0.024).

KLF4 expression did not significantly differ based on age, gender, necrosis, lymphatic and vascular invasion, perirenal fat invasion, or distant metastasis.

Table II shows the relationship between clinicopathological factors and combined immune scores of SOX9, KLF4, and E-cadherin staining.

DISCUSSION

ccRCC is the most prevalent subtype of RCC, and various prognostic factors have been described for it to date. The most prominent one may be the tumour’s pathological stage. Among tumours of the same stage, poor prognostic parameters include tumour grade, presence of necrosis, and sarcomatoid and rhabdoid differentiation.¹⁷

Numerous immunohistochemical and molecular studies have been conducted on the clinical use of CSCs-related factors in determining the prognosis of RCC and ccRCC.⁶ This study explored the roles of SOX9 and KLF4, along with recognised factors, in regulating the CSC phenotype and influencing the development and prognosis of ccRCC.

There are very few studies on SOX9 in RCC.^2,8,18 Two of these studies examined the immunohistochemical expression of SOX9 and its association with prognostic factors.² Two studies found a significant association between SOX9 expression and high tumour grade.^8,18

Li et al. found that 16 of 38 cases with metastasis were SOX9 positive.¹⁸ On the other hand, Wan et al. found SOX9 expression to be significantly associated with increased stage.⁸ Increased SOX9 expression was detected in grade 4 tumours compared to grade 2 and 3 tumours. Additionally, elevated SOX9 expression was observed in 14 cases with distant metastases.

Activation of epithelial-mesenchymal transition (EMT) results in the development of cancer stem cells and metastasis in malignant epithelial cells. The primary factor involved in EMT is E-cadherin downregulation.³ Previous research has shown that E-cadherin expression strongly correlates with advanced pathological stages⁹ and grades.^3,11,19,20 Therefore, decreased E-cadherin expression may be seen in the sarcomatoid component of the tumour.²¹ This study identified a significant reduction in E-cadherin expression in high-grade tumours, with notably lower expression in the sarcomatoid and rhabdoid components.

Previous studies have demonstrated that the loss of E-cadherin expression is associated with tumour invasion, metastasis, and poor survival. Metastatic RCCs may also show significantly low E-cadherin expression.^3,9 The study discovered a significant relationship between distant metastasis and low E-cadherin expression. Loss of E-cadherin expression is often expected to affect EMT and lead to metastasis directly, and these findings align with those reported in the literature.³

Three studies have inquired about the role of KLF4 in developing RCC and ccRCC. Their findings revealed that KLF4 has a tumour suppressor role in developing RCC.^6,16,22,23 Song et al. showed that KLF4 overexpression significantly inhibits proliferation in human ccRCC cell lines.¹⁶

Song et al. observed moderate-to-intense positive staining in tumour tissue in 40% of ccRCC cases, whereas Li et al. reported this rate as 13%.^16,22 Liu et al. investigated KLF4 expression through molecular techniques.⁶ They found KLF4 downregulation in 16% of the cases. Three studies showed significantly lower KLF4 expression in tumour tissue than in normal renal parenchyma.^6,16,22 In this study, KLF4 stained positive in 22.8% of the cases, with positivity found in 10.9% of normal renal parenchyma.

Li et al. found significant associations between low KLF4 expression, stage, and lymph node metastasis.²² Significantly higher KLF4 expression was found in Stage 1 tumours than Stage 3 and 4 tumours. Additionally, KLF4 expression was lower in patients with distant metastasis.

Song et al. found significantly lower KLF4 expression in cases with a high tumour diameter (above 7 cm).¹⁶ Substantially lower KLF4 expression was observed in cases with a tumour diameter over 4 cm. Similarly, when the tumour diameter was categorised as either less than or greater than 7 cm, significantly lower KLF4 expression was found in the group with larger tumour diameters.

The present study has several limitations. Due to its single-centre design and small sample size, the findings could not be generalised. Additionally, extensive research on the expression of SOX9 and KLF4, particularly with immunohisto-chemical data, is limited in ccRCC of the kidney. The limited availability of comparable data makes it challenging to interpret the existing information.

CONCLUSION

Analysis of CSC-related factors in cases with ccRCC suggests that changes in the expression of these factors may be one of the oncogenic mechanisms underlying the pathogenesis of ccRCC. Specifically, findings highlight that SOX9 and KLF4 expressions contribute to the carcinogenesis and progression of ccRCC, indicating that CSC-related factors are likely involved in its pathogenesis.

FUNDING:
This study was supported by the Bolu Abant Izzet Baysal University, Scientific Research projects (2021.08.19.1490).

ETHICAL APPROVAL:
This study was approved by the Bolu Abant Izzet Baysal University’s Ethics Committee for Clinical Research (No: 2020/226).

PATIENTS’ CONSENT:
Patients’ consent was not required as the patients’ identities were not disclosed or compromised.

COMPETING INTEREST:
The authors declared no conflict of interest.

AUTHORS’ CONTRIBUTION:
SO: Designed the study, collected and analysed the data, and prepared the manuscript.
SED: Designed the study, reviewed and edited the manuscript.
Both authors approved the final version of the manuscript to be published.
 

REFERENCES

1. International Agency for Research on Cancer (World Health Organization). Kidney: Globocan 2020 - The Global Cancer Observatory. Globocan 2020; 419:1-2.
2. Hu B, Wang J, Jin X. MicroRNA-138 suppresses cell proliferation and invasion of renal cell carcinoma by directly targeting SOX9. Oncol Lett 2017; 14(6):7583-8. doi: 10. 3892/ol.2017.7160.
3. Harb OA, Elfeky MA, El Shafaay BS, Taha HF, Osman G, Harera IS, et al. SPOP, ZEB-1 and E-cadherin expression in clear cell renal cell carcinoma (cc-RCC): Clinicopathological and prognostic significance. Pathophysiology 2018; 25(4): 335-45. doi: 10.1016/j.pathophys.2018.05.004.
4. Aguilar-Medina M, Avendano-Felix M, Lizarraga-Verdugo E, Bermudez M, Romero-Quintana JG, Ramos-Payan R, et al. SOX9 Stem-cell factor: Clinical and functional relevance in cancer. J Oncol 2019; 2019:6754040. doi: 10.1155/2019/6754040
5. Siljee S, Milne B, Brasch HD, Bockett N, Patel J, Davis PF, et al. Expression of components of the renin-angiotensin system by cancer stem cells in renal clear cell carcinoma. Biomolecules 2021; 11(4):537. doi: 10.3390/biom11040537.
6. Liu Y, Zhang C, Fan J, Xiao L, Yin B, Zhou L, et al. Comprehensive analysis of clinical significance of stem-cell related factors in renal cell cancer. World J Surg Oncol 2011; 9(1):121. doi: 10.1186/1477-7819-9-121.
7. Chen W, Zhao W, Chen S, Zhang L, Guo Z, Wang L, et al. Expression and correlation of MALAT1 and SOX9 in non-small cell lung cancer. Clin Respir J 2018; 12(7):2284-91. doi: 10.1111/crj.12906.
8. Wan YP, Xi M, He HC, Wan S, Hua W, Zen ZC, et al. Expression and clinical significance of SOX9 in renal cell carcinoma, bladder cancer and penile cancer. Oncol Res Treat 2017; 40(1-2):15-20. doi: 10.1159/000455145.
9. Jang N-R, Choi JH, Gu MJ. Aberrant Expression of E-cadherin, N-cadherin, and P-cadherin in clear cell renal cell carcinoma: Association with adverse clinicopathologic factors and poor prognosis. Appl Immunohistochem Mol Morphol 2021; 29(3):223-30. doi: 10.1097/PAI.0000000000000861.
10. Mendonsa AM, Na TY, Gumbiner BM. E-cadherin in contact inhibition and cancer. Oncogene 2018; 37(35):4769-80. doi: 10.1038/s41388-018-0304-2.
11. Gao Y, Yan Y, Guo J, Zhang Q, Bi D, Wang FEN, et al. HNF ‑ 4 α downregulation promotes tumour migration and invasion by regulating E ‑ cadherin in renal cell carcinoma. Oncol Rep 2019; 42(3):1066-74. doi: 10.3892/or.2019.7214.
12. Wang D, Su L, Huang D, Zhang H, Shin DM, Chen ZG. Downregulation of E-cadherin enhances proliferation of head and neck cancer through transcriptional regulation of EGFR. Mol Cancer 2011; 10:116. doi: 10.1186/1476-4598-10-116.
13. Evans PM, Liu C. Roles of Krupel-like factor 4 in normal homeostasis, cancer and stem cells. Acta Biochim Biophys Sin 2008; 40(7):554-64. doi: 10.1111/j.1745-7270.2008. 00439.x.
14. Zhao W, Hisamuddin IM, Nandan MO, Babbin BA, Lamb NE, Yang VW. Identification of kruppel-like factor 4 as a potential tumour suppressor gene in colorectal cancer. Oncogene 2004; 23(2):395-402. doi: 10.1038/sj.onc.1207067.
15. Zhang N, Su P, Li X, Xi J, Li XU, Xu LUO. Downregulated kruppel-like factor 4 expression is associated with the aggressiveness of prostate cancer. Oncol Rep 2019; 41(3):1789-96. doi: 10.3892/or.2019.6975.
16. Song E, Ma X, Li H, Zhang P, Ni D, Chen W, et al. Attenuation of kruppel-like factor 4 facilitates carcinogenesis by inducing g1/s phase arrest in clear cell renal cell carcinoma. PLoS One 2013; 8(7):e67758. doi: 10.1371/journal.pone.0067758.
17. Moch H, Amin MB, Argani P, Cheville JDB, Martignoni G, Medeiros LJ, et al. TPHTSK. Tumours of the kidney. In: Holger M, Peter HA, Thomas UM. Reuter VE, Eds. WHO Classification of Tumours of the Urinary System and Male Genital Organs. 4^th ed. 2016. p. 12-21.
18. Li XL, Chen XQ, Zhang MN, Chen N, Nie L, Xu M, et al. SOX9 was involved in TKIs resistance in renal cell carcinoma via Raf/MEK/ERK signaling pathway. Int J Clin Exp Pathol 2015; 8(4):3871-81.
19. Andreiana BC, Stepan AE, Margaritescu C, Taisescu O, Osman A, Simionescu C. Snail and E-cadherin immuno-expression in clear cell renal cell carcinoma. Curr Heal Sci J 2019; 45(2):185-9. doi: 10.12865/CHSJ.45.02.09.
20. Cai J. Roles of transcriptional factor snail and adhesion factor E-cadherin in clear cell renal cell carcinoma. Exp Ther Med 2013; 6(6):1489-93. doi: 10.3892/etm.2013.1345
21. Langner C, Ratschek M, Rehak P, Schips L, Zigeuner R. Expression of MUC1 (EMA) and E-cadherin in renal cell carcinoma : A systematic immunohistochemical analysis of 188 cases. Mod Pathol 2004; 17(20):180-8. doi: 10.1038/ modpathol.3800032.
22. Li H, Wang J, Xiao W, Xia D, Lang B, Yu G, et al. Epigenetic alterations of kruppel-like factor 4 and its tumour suppressor function in renal cell carcinoma. Carcinogenesis 2013 Oct; 34(10):2262-70. doi: 10.1093/carcin/bgt189.
23. Chen W, Wang H, Lu Y, Huang Y, Xuan Y, Li X, et al. GTSE1 promotes tumour growth and metastasis by attenuating of KLF4 expression in clear cell renal cell carcinoma. Lab Invest 2022; 102(9):1011-22. doi: 10.1038/s41374-022-00797-5.