Background
Clear cell renal cell carcinoma(ccRCC) is one of the most common malignancies. However, there are still many barriers to its underlying causes, early diagnostic techniques and therapeutic approaches. Materials and
Conclusions
We developed a risk signature constructed based on the five LLPS-related genes and can have a high ability to predict the prognosis of ccRCC patients, further providing a strong support for clinical decision-making.
Methods
The Cancer Genome Atlas (TCGA)- Kidney renal clear cell (KIRC) cohort differentially analysed liquid-liquid phase separation (LLPS)-related genes from the DrLLPS website. Univariate and multivariate Cox regression analyses and LASSO regression analyses were used to construct prognostic models. The E-MTAB-1980 cohort was used for external validation. Then, potential functions, immune infiltration analysis, and mutational landscapes were analysed for the high-risk and low-risk groups. Finally, quantitative real-time polymerase chain reaction (qRT-PCR) experiments as well as single-cell analyses validated the genes key to the model.
Results
We screened 174 LLPS-related genes in ccRCC and constructed a risk signature consisting of five genes (CLIC5, MXD3, NUF2, PABPC1L, PLK1). The high-risk group was found to be associated with worse prognosis in different subgroups. A nomogram constructed by combining age and tumour stage had a strong predictive power for the prognosis of ccRCC patients. In addition, there were differences in pathway enrichment, immune cell infiltration, and mutational landscapes between the two groups. The results of qRT-PCR in renal cancer cell lines and renal cancer tissues were consistent with the biosignature prediction. Three single-cell data of GSE159115, GSE139555, and GSE121636 were analysed for differences in the presence of these five genes in different cells. Conclusions: We developed a risk signature constructed based on the five LLPS-related genes and can have a high ability to predict the prognosis of ccRCC patients, further providing a strong support for clinical decision-making.