Abstract
Background/Aim: Renal cell carcinoma (RCC) is highly heterogeneous, with distinct patient management between clear cell RCC (ccRCC) and non-ccRCC groups. Previous bioinformatics and machine learning techniques identified fatty acid binding protein 7 (FABP7) as a potential ccRCC biomarker. However, FABP7 expression studies between ccRCC and non-ccRCC were incomplete. This study aimed to assess FABP7 as a biomarker for distinguishing between ccRCC and non-ccRCC tissue samples. Patients and Methods: FABP7 expression was evaluated via immunohistochemical staining in 58 RCC cases, including 43 ccRCC and 15 non-ccRCC cases. Staining results were interpreted using H-scores; scores above the cut-off were deemed positive. The correlation between FABP7 expression and clinicopathological RCC features was investigated. Results: FABP7 positivity was 48.8% in ccRCC and only 13.3% in non-ccRCC cases, with weak positivity in non-ccRCC tissues. FABP7 expression significantly differed between ccRCC and non-ccRCC (p<0.05). This finding was confirmed in a TCGA dataset. However, FABP7 expression was not correlated with other RCC clinicopathological features in our dataset. TCGA results linked FABP7 expression to tumor stage and disease-free survival in patients with ccRCC. Conclusion: This study preliminarily evaluated FABP7 as a differential diagnostic biomarker in RCC subtyping, showing higher expression in ccRCC than non-ccRCC. FABP7 may serve as a potential diagnostic and prognostic biomarker for ccRCC.
- Clear cell renal cell carcinoma
- non-clear cell renal cell carcinoma
- FABP7
- diagnostic marker
- immunohistochemistry
Kidney cancer is the second most common urinary system cancer. Renal cell carcinoma (RCC) arises from renal tubular epithelial cells and accounts for ~90% of kidney cancer cases (1, 2). RCCs are diverse malignant tumors that include a variety of subtypes. These subtypes are distinguished by their unique cytoplasmic features, architectural structures, anatomical locations, and specific molecular alterations. In addition, RCC has been suggested to be a metabolic disease involving energy metabolic alteration, anaerobic glycolysis shift, and remodeling of fatty acids metabolism, especially in clear cell RCC (ccRCC) (3). RCCs exhibit diverse subtypes according to WHO classification guidelines. The approach to treatment and patient care based on National Comprehensive Cancer Network (NCCN) guidelines is broadly categorized into two main groups: the majority of metastatic tumors with clear cell histology (83–88%) and tumors with non-clear cell histology, collectively referred to as non-ccRCC (1, 4, 5). The RCC classification and subgrouping have become so intricate and necessary that they require ancillary investigation, mainly immunohistochemistry, to support a precise diagnosis (6). Although carbonic anhydrase 9 (CA9) is currently used as a standard diagnosis marker for ccRCC, it is not an entirely specific marker for ccRCC. CA9 expression can be seen in some subtypes of RCCs other than ccRCC (7). Consequently, identifying novel and efficient markers for distinguishing ccRCC and non-ccRCC groups is crucial.
Our previous bioinformatics and machine learning analysis study identified NDUFA4L2, FABP7, DAT, ANGPTL4, and NPTX2 as potential biomarkers for ccRCC. Fatty acid binding protein 7 (FABP7) was determined to be a second-rank gene that has the highest false discovery rate (FDR) values and log fold change (8). FABP7 is a member of the FABP family, and it plays a vital role in regulating the actions of polyunsaturated fatty acids (PUFAs) within cells. The expression of FABP7 is observed in various cell types, including glial cells like astrocytes and oligodendrocyte progenitor cells, during brain development (9). In a previous study, the expression of FABP7 was correlated with tumor progression in many cancers, such as glioblastoma (10), RCC (11), and melanoma (12). Especially in ccRCC, FABP7 expression was associated with advanced clinical stage, distant metastasis, and poor cancer-specific survival in patients with ccRCC (13). The few studies of FABP7 expression in ccRCC (13-15) suggest that FABP7 could serve as a prognostic biomarker for ccRCC. However, the investigation of the potential of FABP7 to differentiate between ccRCC and non-ccRCC diagnoses has not yet been explored.
The present study explored the potential of using FABP7 as a biomarker for distinguishing between ccRCC and non-ccRCC. The immunohistochemistry study was performed to investigate the expression of FABP7 in ccRCC and non-ccRCC patient tissues. The expression levels of FABP7 were scored according to the H-score. Subsequently, the correlation between clinicopathological findings and FABP7 expression was examined. The expression level and clinical association of FABP7 in patients with RCC were also analyzed using transcriptomics data from the The Cancer Genome Atlas (TCGA) database.
Materials and Methods
Patients and tissue samples. 58 tissue samples were obtained from patients with RCC who had undergone nephrectomy. Of these, 43 were diagnosed with ccRCC, while the remaining 15 were diagnosed with non-ccRCC (including 5 cases of pRCC, 4 cases of chRCC, and 6 cases of other rare RCC subtypes). The clinicopathological features of these tissue samples are described in Table I.
The correlation of clinicopathologic features and FABP7 expression.
The tissue samples and clinical data were gathered from the Department of Pathology at Srinagarind Hospital, Faculty of Medicine, Khon Kaen University, Thailand, from January 2017 to December 2021. The study was conducted under the Declaration of Helsinki and the International Council for Harmonization (ICH-GCP) Good Clinical Practice Guidelines, and the protocol (HE641597) was approved by the Khon Kaen University Ethics Committee for Human Research.
Immunohistochemistry (IHC) technique. The IHC technique was used to investigate the expression level of FABP7 in FFPE tissues from patients with RCC. The FFPE blocks were sectioned using a microtome into slices with a thickness of 4 μm that were attached to glass slides. The samples were stained with a FABP7 polyclonal antibody (dilution 1:100, Proteintech, Chicago, IL, USA). As a secondary antibody, we used an ultraView Universal DAB Detection Kit stainer (Ventana Medical Systems, Tucson, AZ, USA). The process was performed on a Ventana Benchmark XT automated stainer (Ventana Medical Systems) according to the manufacturer’s recommendations for visualization.
One pathologist with ten years of experience in practicing pathology evaluated the expression levels of FABP7 using the H-score method. This assessment involved examining slides in ten randomly chosen fields at 400X magnifications. The staining intensity in the cytoplasm, cell membrane, and nuclear region of the malignant cells was graded from 0 to 3, with 0 indicating negative staining, 1 weak staining, 2 moderate staining, and 3 intense staining. The calculation of the H-score involved determining the average positive percentage of each intensity grade using the following formula:
Expression levels higher than a cut-off value were interpreted as a positive stain. This methodology provides a quantitative measure of FABP7 expression, considering both the intensity and the proportion of cells exhibiting specific staining characteristics.
The expression and clinical association of FABP7 based on transcriptomics data from the TCGA database. We further investigated the correlation between FABP7 expression and clinical characteristics, such as tumor stage, overall survival (OS), and disease-free survival (DFS), using independent transcriptomics data from the TCGA via the user-friendly platform Gene Expression Profiling Interactive Analysis (GEPIA) (16).
Statistical analysis. The statistical analysis was performed using IBM SPSS Statistics v.28 (SPSS Inc., Chicago, IL, USA). The correlation between FABP7 expression and clinicopathological characteristics, including age, sex, tumor size, tumor grade, lymphovascular invasion, prognostic stage (TMN stage), pT stage, sarcomatoid feature, and rhabdoid feature, was performed using the Chi-square test (χ2-test) and Fisher’s exact test. A p-value <0.05 was set as the level for statistical significance. A receiver operating characteristic (ROC) curve and an area under the curve (AUC) were analyzed.
Results
FABP7 was over-expressed in ccRCC compared to non-ccRCC groups. FABP7 expression was validated in 43 ccRCC and 15 non-ccRCC samples, each compared with adjacent non-cancerous tissue, to categorize the FABP7 staining as positive or negative. As demonstrated in Figure 1, FABP7 protein was over-expressed in ccRCC samples compared with non-ccRCC and adjacent normal tissues. Among the 58 RCC samples, FABP7 showed a significant differential expression between ccRCC and non-ccRCC (p<0.05) (Table I). Twenty-one out of 43 ccRCC cases (48.8%) exhibited positive staining, whereas 22 out of 43 ccRCC (51.2%) showed negative staining. Furthermore, FABP7 showed heterogeneous staining in FABP7-positive ccRCC (Figure 1D). In contrast, 13.3% of non-ccRCC cases showed positive staining, whereas 86.7% were utterly FABP7-negative (Figure 1F, H, and J). As shown on the box plot, the expression of FABP7 was significantly higher in ccRCC than in the non-ccRCC (Figure 2A). In addition, the ROC curve analysis indicated moderate diagnostic accuracy of FABP7 in ccRCC, with an AUC of 0.71 (Figure 2B).
Expression of FABP7 in clear cell renal cell carcinoma (ccRCC) tissues compared with normal renal and non-ccRCC tissues (20′). A, C, E, G, and I) The illustration depicts histomorphology features with H&E staining of normal renal tissue, ccRCC, papillary RCC (pRCC), chromophobe RCC (chRCC), and rare subtypes RCC (rRCC). B, D, F, H, and J) The expression of FABP7 in normal renal tissue, ccRCC, pRCC, chRCC, and rRCC. B) The weak membranous and cytoplasmic staining of FABP7 in normal renal tubes. D) In ccRCC, FABP7 is expressed differently between low nuclear grade (low expression) and high nuclear grade (high expression). F, H, and J) FABP7 showed negative staining in non-ccRCC groups, including pRCC, chRCC, and rRCC.
The gene distribution and the receiver operating characteristic curve of FABP7 expression between clear cell renal cell carcinoma (ccRCC) and non-ccRCC. A) The box plot diagram of FABP7 protein expression comparing ccRCC and non-ccRCC. The x-axis represents the groups of RCC, and the y-axis represents the gene expression values based on the immunohistology score (H-score). FABP7 increased in ccRCC groups compared with non-ccRCC. B) In the ROC curve and AUC analysis of FABP7 expression, the AUC value was moderate, 0.71. C) The box plots of FABP7 expression in three common subtypes of RCC compared with normal tissues. The x-axis represents the three common subtypes of RCC, including ccRCC, pRCC, and chRCC. The y-axis represents the FABBP7 expression based on log2 values. Expression of FABP7 mRNA in 523 ccRCC compared to 100 normal tissues was statistically significant. In pRCC, there is minimal FABP7 mRNA expression, although lower than that observed in ccRCC. There is no expression of FABP7 mRNA in chRCC. *p<0.05.
Correlation between FABP7 expression and clinicopathologic features. We also examined the relationship between gene expression and clinicopathological features and found that FABP7 expression was not significantly correlated with any clinicopathological features, except patient grouping by tumor subtype. Additionally, we observed that the down-regulation of FABP7 was associated with tumors larger than 7 cm in diameter. The statistical data clearly showed a trend towards significance (p=0.063).
Validation of FABP7 expression and its clinical association using transcriptomics data from the TCGA database. According to our IHC findings, FABP7 exhibited significantly higher expression in ccRCC than in non-ccRCC. However, our study was limited by the small sample size of RCC tissues. We further validated the expression of FABP7 and its clinical associations, including tumor stage, OS, and DFS, using transcriptomics data from the TCGA database via the GEPIA online tool. Transcriptomics data from the TCGA database consist of 523 ccRCC, 286 papillary RCC (pRCC), and 66 chromophobe RCC (chRCC) cases. The results from GEPIA confirmed that FABP7 was significantly over-expressed in only ccRCC but not in non-ccRCC, which included pRCC and chRCC (p<0.05) (Figure 2C).
Our research also revealed a significant correlation between FABP7 and the pathological stage of ccRCC, particularly advanced stages (Figure 3A). However, FABP7 showed no such correlation in pRCC and chRCC (Figure 3B and C).
The violin plots of the correlation between FABP7 and the pathological stage of renal cell carcinoma (RCC) in each subtype. A) ccRCC; the level of FABP7 expression was closely associated with the advanced pathological stages of clear cell RCC (ccRCC). B and C) pRCC and chRCC; the expression of FABP7 was not statistically significant. The x-axis represents the FABP7 expression; the y-axis represents the pathological stages. F Value: F-value is derived from an ANOVA test (Analysis of Variance) and is used to compare the means of the distributions to examined if they are statistically different from each other. An F-value greater than 1 suggests a variability between the group means. Pr(>F): This is the p-value derived from the ANOVA test. A p-value less than 0.05 indicates a statistically significant difference between the group means. The smaller the p-value, the stronger the evidence against the null hypothesis (which would be that there is no difference between the group means). White dot: This indicates the mean of the data.
The survival analysis was conducted using a Kaplan-Meier plotter. Although the OS analysis showed no statistically significant difference between ccRCC patients with high versus low FABP7 expression [logrank p=0.086; hazard ratio (HR)=1.3, p(HR)=0.087] (Figure 4A), the high expression of FABP7 was significantly associated with shortened DFS of patients with ccRCC compared to patients with ccRCC who have low expression [log-rank p=0.008; HR=1.6, p(HR)=0.0087] (Figure 4B).
The result of the Kaplan-Meier analysis of FABP7 expression in clear cell renal cell carcinoma (ccRCC) from the TCGA cohort: A) The overall survival analysis indicated no significant difference between the low and high expression of FABP7 in ccRCC patients; however, there was a trend toward significance [log-rank p=0.086; hazard ratio (HR)=1.3, p(HR)=0.087]. B) The disease-free survival analysis showed that high expression of FABP7 was associated with shortened survival in patients with ccRCC compared to those with low expression [log-rank p=0.008; HR=1.6; p(HR)=0.0087].
Discussion
RCCs are a diverse group of malignancies with the exact anatomical origin but essentially distinct metabolic profiles and clinical characteristics (6). According to the NCCN guideline, the treatment protocol of patients with advanced RCC differs between ccRCC and non-ccRCC groups. Advanced RCC tumors with high WHO/ISUP nuclear grade often loss the conventional or typical histomorphology of each specific subtype or sometimes even display overlapping patterns of histology in the same tumor (e.g., clear cell plus papillary or other), necessitating the use of ancillary techniques like immunohistochemistry for diagnosis, which also requires the expertise of experienced uropathologists. Furthermore, Choi et al. reported that metastatic RCC patients with WHO/ISUP nuclear high grade had a higher Eastern Cooperative Oncology Group (ECOG) score than patients with low grade. This study also revealed that patients with a high grade had a worse OS rate than those with a low grade (17). However, nuclear grading can only be applied to certain ccRCC and pRCC subtypes. Therefore, definitive RCC subclassification is crucial for subsequent nuclear grading. Thus, we investigated diagnostic markers for differentiating ccRCC from non-ccRCC. Our previous study identified five top genes with particular expression in ccRCC subtypes based on a silico analysis; FABP7 was the second top gene in those panels (8). FABP7, which is also known as a brain fatty acid-binding protein (BFABP), belongs to the fatty acid-binding protein (FABP) family and plays a pivotal role in cellular activity regulation through the modulation of lipid raft functions, interaction with peroxisome proliferator-activated receptors (PPARs), and facilitation of fatty acid (FA) uptake and lipid storage in hypoxic conditions (18). Zheng and co-workers showed that the up-regulation of four fatty acid transport-related proteins (FATRP), including FABP1, FABP5, FABP7, and FATP2, was positively associated with tumor progression and poor prognosis in patients with ccRCC. However, the potential role of an individual single gene in this panel in predicting the progression and prognosis has not been determined (14).
This study evaluated the expression of FABP7 protein using IHC techniques on RCC tissues, including ccRCC and non-ccRCC groups. The results showed that FABP7 was significantly over-expressed in ccRCC compared with non-ccRCC groups (pRCC and chRCC) (p<0.05).
Although this study faced significant limitations due to the small sample size and an imbalance in the distribution between the two RCC groups on the IHC technique, the validation of FABP7 expression and its clinical association using transcriptomic data from the TCGA database helped substantiate the findings acquired using IHC. Independent transcriptomic analysis using GEPIA showed that FABP7 mRNA was over-expressed in ccRCC compared with pRCC and chRCC subtypes. Similarly, previous studies demonstrated that FABP7 was over-expressed explicitly in RCC tumors of clear cell types but low to non-detected in normal tissues (13, 19, 20). Tello et al. demonstrated that FABP7 mRNA expression was correlated with tumor grade. High-grade (G3 and G4) RCC tumors showed significantly lower FABP7 expression than low-grade tumors (20). Conversely, our study found that FABP7 protein expression was higher in high-grade areas of ccRCC than in low-grade areas. Moreover, we found that FABP7 expression was correlated with advanced tumor stage and shortened DFS of patients with ccRCC. This finding was similar to that of Zhou et al., who found that FABP7 was related to advanced clinical stages, tumor volume, and poor survival of patients with ccRCC. Patients with FABP7-positive tumors showed a higher tumor volume than those with FABP7-negative tumors. FABP7 also promoted tumor cell growth by activating ERK and STAT3 signaling pathways. It is possible that FABP7 is a proto-oncogene in RCC (13). Wu et al., showed that the over-expression of FABP5, FABP6, and FABP7, along with the low expression of FABP1, were correlated with lower OS and DFS in patients with ccRCC. Additionally, the co-expression of FABP1, FABP5, FABP6, and FABP7 was related to transforming growth factor-β1 (TGF-β1), peroxisome proliferator-activated receptor alpha (PPARA), and lipoprotein lipase (LPL) (21). Interestingly, Teratani et al. reported that FABP7 could be detected in the urine of patients with RCC before receiving nephrectomy compared with healthy controls. FABP7 may be a urine diagnostic marker for RCC (22).
Additionally, FABP7 has also been reported to be over-expressed in gliomas, breast cancer, melanoma, adenoid cystic carcinoma, and subependymal giant cell astrocytoma (18). Kaloshi et al. reported that FABP7 expression in glioblastoma (GBM) was associated with invasive tumor behavior. Particularly, the presence of nuclear FABP7 in EGFR-positive GBM was correlated with a worse prognosis. Additionally, inhibiting FABP7 was demonstrated to reduce tumor cell migration in GBM (23, 24). Similarly, FABP7 has been shown to be involved in cell proliferation and invasion, leading to poor outcomes in metastasis melanoma (25, 26) and in adenoid cystic carcinoma of the salivary glands (27). In basal-like breast cancer (BLBC), FABP7 expression (especially nuclear staining) was associated with favorable survival (28). It is possible that FABP7 is a marker that differentiate between a good and poor prognosis for BLBC patients (29). Conversely, FABP7 over-expression in triple-negative breast cancer (TNBC) was associated with an unfavorable prognosis; it acted as an adverse factor promoting tumor growth and progression. Depletion of FABP7 decreased the rate of cell growth and made the cells more responsive to growth inhibition by omega-3 docosahexaenoic acid (DHA) (30). Additionally, FABP7 and three other genes correlate with infiltrating immune cells in breast cancer (31). Indeed, the sub-cellular localization of FABP7 was associated with different biological behaviors in each cancer (23, 28). In those studies, FABP7 was suggested as a potential biomarker for prognostic and therapeutic applications.
Our study suggested the potential of using FABP7 as a diagnostic biomarker for distinguishing between ccRCC and non-ccRCC samples. The feasibility of utilizing FABP7 as a prognostic biomarker for ccRCC has also been demonstrated.
Conclusion
In conclusion, our study demonstrated differential expression of the FABP7 protein between the ccRCC and non-ccRCC groups, with FABP7 showing elevated levels in the ccRCC group, as supported by both IHC and mRNA analysis. These results provide a new strategy for differential diagnosis between ccRCC and non-ccRCC. However, our study has significant limitations, including a small sample size and class imbalance issues among renal cell carcinoma (RCC) patient samples. These challenges may impact the reliability of our results. Further studies with a large sample size or mRNA analysis techniques (such as PCR) are needed to enhance the robustness of our conclusions. Additionally, in vitro models should be used to understand this gene’s function.
Acknowledgements
The Authors are grateful to all the staff members of the Pathological Laboratory at the Department of Pathology, Srinagarind Hospital, Faculty of Medicine, Khon Kaen University, for their invaluable assistance during this study.
Footnotes
Authors’ Contributions
CP, SS, and WS: developed the concept and designed the study. CP: performed experiments and contributed to drafting the paper. SS: was involved in immunohistochemistry interpretation and approving the final version of the article. WS: performed bioinformatics and machine learning analyses and approved the final article. MT: was involved in the interpretation of immunohistochemistry results and contributed reagents, materials, tools, and approved the final article.
Funding
The study was supported by a Postgraduate Study Support Grant from the Faculty of Medicine, Khon Kaen University, and an Invitation Research Fund from the Faculty of Medicine, Khon Kaen University, Thailand (grant no. IN65204). The funders had no role in study design, data collection and analysis, manuscript preparation, or publication decisions.
Conflicts of Interest
The Authors declare that there are no conflicts of interest in relation to this study.
- Received June 13, 2024.
- Revision received July 23, 2024.
- Accepted July 24, 2024.
- Copyright © 2024 The Author(s). Published by the International Institute of Anticancer Research.
This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY-NC-ND) 4.0 international license (https://creativecommons.org/licenses/by-nc-nd/4.0).
