Kidney cancer is the seventh most prevalent cancer in men and the 10th most common cancer in women, accounting for 5% and 3% of all adult malignancies in men and women, respectively. The average age for diagnosis is 60 years old, with a notable gender disparity existing for receiving a diagnosis. Renal cell carcinoma (RCC) is a sneaky neoplasm that accounts for 2% of all cancer diagnoses and fatalities globally and is expected to become more prevalent (1). More developed regions like Europe and North America have a higher incidence of kidney cancer than less developed ones like Africa and South America. An estimated 372,000 persons worldwide are affected by kidney cancer each year (2).

Of all primary renal neoplasms, 80% to 85% are RCCs, which develop within the renal cortex whereas 8% of cancers are transitional cell carcinomas, which develop in the renal pelvis. RCCs, often known as kidney cancers, are a class of histologically identified tumors that can be characterized by several genetic alterations. Renal cancer has several diverse subtypes, each of which has a unique histology, distinctive genetic and molecular changes, and hence unique response to treatment. The three main RCC subtypes (with ≥5% incidence), which account for 75%, 15%, and 5% of RCCs, are clear cell RCC (ccRCC), papillary RCC (pRCC), and chromophobe RCC (chRCC), respectively (2). Other subtypes of RCC like transitional cell carcinoma, Wilms tumor, and renal sarcoma have ≤1% total incidence and are very rare. Approximately 30% of individuals experience recurrence following the complete removal of the initial tumor. All RCC subtypes are largely unresponsive to conventional chemotherapy and radiation therapy (3).

The pathologic stage and nuclear grade largely determine the prognosis for ccRCC (4). Clear cell tumors feature well-defined cell boundaries and translucent, empty (water-clear) cytoplasm. Variable levels of lipid and glycogen can be seen in the cytoplasm. Radial or partial nephrectomy is the main form of treatment. Despite having mixed results, chemotherapy and immunotherapy are utilized on individuals with metastatic illness. Its clinical symptoms usually appear in later stages, and due to its high morbidity and mortality, early detection and accurate prognosis are critical for improving patient treatment and outcomes, but there are several challenges associated with these goals.

The widespread adoption of RNA-seq has heightened the demand for bioinformatics skills and computational resources to analyze large-scale biological data efficiently and unraveling molecular complexities in genomics research. For this study, RNA-seq data were analyzed. It initially involves isolation of total RNA from target tissues or cells, which then entails building DNA libraries followed by sequencing with a next-generation sequencing approach. RNA-seq data have several advantages over other transcriptome methods, like RNA-seq offers a better resolution and dynamic range, making it more sensitive and can reliably estimate gene expression at a wider range of levels. By RNA-seq data, one can find novel transcripts, alternative splicing events, and non-coding RNAs, and detect known transcripts and quantify these. It also does not possess any bias in the selection of probes, in contrast to microarrays, which depend on pre-designed probes. The availability of such type of data in publicly accessible databases, such as The Cancer Genome Atlas (TCGA) (https://www.cancer.gov/tcga) and Genotype-Tissue Expression (GTEx) databases (https://www.gtexportal.org/home/), has provided researchers with comprehensive datasets for various cancer types, including ccRCC.

A massive amount of research on various aspects of ccRCC has been done (5, 6), including analysis of the cross-model features of ccRCC for prognosis prediction (7) or indulgence of artificial intelligence for studying the pathomics and genomics of RCC (8). However, in this study, we have taken a computational approach to analyzing the DEGs and carrying out the network analysis and pathways involved in ccRCC. Here, we have identified potential prognostic biomarkers for ccRCC using data from the TCGA by selecting the RNA-seq data under the project ID, TCGA-KIRC (kidney renal clear cell carcinoma). We uncovered differentially expressed genes (DEGs) associated with ccRCC using the R programming language. These identified DEGs were analyzed for pathways enrichment using the Qiagen Ingenuity Pathway Analysis (IPA) tool to gain insights into the dysregulation of key pathways involved in ccRCC. We selected the top 10 candidate genes from the gene interaction networks of the identified DEGs. These genes were further investigated by gene expression profiling and survival analysis. The identified candidate genes including the overall survival (OS) and relapse-free survival (RFS) along the Cox multivariate analysis led to a set of three genes as potential biomarkers for ccRCC in the present study.

The complete summarized workflow adopted for the RNA-seq analysis of the TCGA-KIRC data has been described in Figure 1 . Each method used is described in detail below.

ccRCC is the most prevalent subtype of RCC, with a 5-year survival rate of less than 10% (15). However, post-treatment survival with Lenvatinib and Pembrolizumab showed an increased median OS of 53.7 months (16). ccRCC is very rare to be diagnosed at an early stage, hence lowering the survival rate of patients. Therefore, more research for early diagnosis of this disease and more precise treatment needs to be provided, lowering the death rate.

Pathway analysis from IPA showed that the pathways highly downregulated have been reported in Phagosome Formation, CREB Signaling in Neurons, and S100 Family Signaling Pathway, while pathways related to immune profiling like the Pathogen-Induced Cytokine Storm Signaling Pathway, IL-17 Signaling, and IL-4 Signaling pathway are upregulated by our identified 492 upregulated genes and 955 downregulated DEGs. Cyclic adenosine monophosphate (cAMP) is an important second messenger; the proliferation, differentiation, death, and immunological responses of cells are just a few of the biological processes that cAMP can control. Any dysregulation or modification of cAMP signaling, i.e., CREB signaling, may result in immunological dysfunction, illness, or cancer as well as cell metabolic disorders (14). It is important to note that phagosomes are cellular compartments responsible for engulfing and digesting foreign particles. The downregulation of the Phagosome formation pathway in the context of cancer may have several implications, such as immune evasion and affecting tumor-associated macrophages, creating an immunosuppressive environment. In vivo and in vitro experimentation on tumor-associated markers and their phagocytic function have elucidated multiple outcomes that are plausible in the tumor microenvironment (17). The proteins in the S100 family perform a variety of intracellular and extracellular functions. They work with a variety of receptors and signal transducers to control pathways that govern inflammation, cell differentiation, proliferation, and apoptosis (12). Dysregulation of the cytokine signaling pathway brought on by pathogens is a major factor in the initiation and spread of cancer. There is an abnormal cytokine and growth factor expression in a variety of malignancies, including lymphomas and solid tumors, including IL-6 and IL-10 (13, 18). These improperly functioning cytokines can serve as autocrine growth factors that encourage the survival, multiplication, and invasion of tumor cells (19). There is evidence that IL-17 mediates the release of additional proinflammatory cytokines and chemokines, particularly from the smooth muscle and pulmonary epithelium (20).

The results obtained from the survival analysis showed that the genes MMP9, PIK3R6, and IFNG are associated with decreased OS, while MMP9 and PGF are associated with increased chances of recurrence (RFS) of KIRC. These genes had an HR value ≥ 1.5 and a log-rank p-value of less than 0.05. Moreover, the Cox multivariate analysis of these candidate genes where the age at initial diagnosis was considered as a covariate results in the exclusion of the PIK3R6 gene due to its low statistical significance. Hence, the genes MMP9, IFNG, and PGF are identified as potential prognostic biomarkers for the KIRC.

MMPs are fascinating genes associated with the development of cancer, functional angiogenesis enhancement, invasion, metastasis, and evasion of immune surveillance. These genes are typically elevated in cancer, and murine studies suggest a role in tumor-associated tissue remodeling (21). According to the research carried out from a pan-cancer perspective, it has been shown that the prognostic value of MMPs is limited to ccRCC (22). As seen in the results, the MMP9 gene, which is upregulated in this dataset, had HR values of 1.7 and 2.4 in the OS and RFS, respectively, with high significance rates.

PIK3R6 is highly upregulated and is involved in the dysregulation of 14 different pathways. As per the survival analysis results obtained, this gene resulted in poor prognosis, as the high expression of PIK3R6 decreases the OS rate. Moreover, in a recent study, the relationship between dysregulated lipid metabolism genes (LMGs) and ccRCC progression was investigated. Here, they established a novel risk score model, according to which the PIK3R6 gene was predicted to have a poor prognosis, higher immune pathway activation, and cancer development (23). However, later in this analysis, PIK3R6 was removed from the list due to its low significance rate.

PGF is a member of the VEGFA family and is highly involved in cell proliferation and growth factor activity (24). It has been seen that the plasma levels of PGF can be employed as a standalone prognostic factor for RCC due to their substantial connection with clinical characteristics and VEGF levels (25). The high HR value of 2.3 and the log-rank p-value of 0.0007 make this gene highly significant and a potential prognostic marker for KIRC after MMP9.

IFNG is known for assisting in the prevention of neoplastic disease (26) and is a potent activator of macrophages. Inhibiting growth, making tumor cells more susceptible to death, increasing the expression of MHC class I and class II, and promoting antitumor immune activity are all effects of IFNG. Shorter lung cancer survival has been associated with decreased IFNG serum levels (27). Nevertheless, the opposite is observed in KIRC, where high expression of IFNG, instead of suppressing the tumor, results in lowering the survival rate by almost 1.5 times as compared to its lower expression. This contradiction might be context-dependent, and further research to elucidate the underlying mechanism is needed.

Apart from these four genes, CREB3L3, which is highly upregulated in the KIRC dataset, shows that an increase in its expression increases the OS of KIRC patients; a similar profiling for this gene has been reported in the Human Proteome atlas. According to previous studies, the genes of the CREB family behave differently, most of which have a tumor-suppressing nature like the gene CREB3L3, which is identified as a low-risk gene having high expression in the case of bladder cancer (28). It was also found that high expression of CREB3L3 helps in the bad prognosis of gastric cancer (29). It might be possible that this gene, being a low-risk gene, prevents KIRC progression.

In conclusion, this comprehensive research delves into the identification of potential prognostic biomarkers for ccRCC using transcriptome data analysis. The results shed light on the dysregulation of critical pathways associated with immune response, angiogenesis, and cytokine signaling in ccRCC. Through survival analysis, MMP9 emerged as a robust prognostic marker for both OS and RFS, with high expression levels correlating with adverse outcomes. This is followed by PIK3R6 and IFNG whose high expression levels had poor survival outcomes. PGF also showed promising potential as a prognostic marker for recurrence. The Cox multivariate analysis provided additional support to the prognostic significance of MMP9, IFNG, and PGF, considering the age at diagnosis as a covariate. However, IFNG’s role appeared context-dependent, as IFNG, being a tumor-suppressing gene, promotes tumor progression as observed in the results. Therefore, further research is needed to elucidate the underlying mechanisms.

These findings contribute significantly to our understanding of ccRCC and have important clinical implications. The identification of prognostic biomarkers like MMP9 and PGF can aid in risk stratification and treatment decision-making, potentially leading to more precise and personalized approaches for patients with ccRCC. In the quest for early detection and more effective treatments, this study paves the way for continued investigation, validation, and potential clinical translation of these identified prognostic biomarkers.

Publicly available datasets were analyzed in this study. This data can be found here: TCGA-KIRC through the GDC Data Portal.

Ethical approval was not required for the study involving humans in accordance with the local legislation and institutional requirements. Written informed consent to participate in this study was not required from the participants or the participants’ legal guardians/next of kin in accordance with the national legislation and the institutional requirements.

VV: Formal analysis, Investigation, Methodology, Software, Validation, Writing – review & editing, Data curation, Writing – original draft. SP: Formal analysis, Investigation, Methodology, Software, Writing – review & editing, Conceptualization, Funding acquisition, Resources, Supervision, Visualization. VP: Conceptualization, Formal analysis, Funding acquisition, Investigation, Methodology, Software, Supervision, Visualization, Writing – review & editing, Validation.