Identification of a Potential PPAR-Related Multigene Signature Predicting Prognosis of Patients with Hepatocellular Carcinoma.
2021
Peroxisome proliferator-activated receptors (PPARs) and part of their target genes have been reported to be related to the progression of hepatocellular carcinoma (HCC). The prognosis of HCC is not optimistic, and more accurate prognostic markers are needed. This study focused on discovering potential prognostic markers from the PPAR-related gene set. The mRNA data and clinical information of HCC were collected from TCGA and GEO platforms. Univariate Cox and lasso Cox regression analyses were used to screen prognostic genes of HCC. Three genes (MMP1, HMGCS2, and SLC27A5) involved in the PPAR signaling pathway were selected as the prognostic signature of HCC. A formula was established based on the expression values and multivariate Cox regression coefficients of selected genes, that was, risk score = 0.1488∗expression value of MMP1 + (-0.0393)∗expression value of HMGCS2 + (-0.0479)∗expression value of SLC27A5. The prognostic ability of the three-gene signature was assessed in the TCGA HCC dataset and verified in three GEO sets (GSE14520, GSE36376, and GSE76427). The results showed that the risk score based on our signature was a risk factor with a HR (hazard ratio) of 2.72 (95%CI (Confidence Interval) = 1.87 ~ 3.95, p < 0.001) for HCC survival. The signature could significantly (p < 0.0001) distinguish high-risk and low-risk patients with poor prognosis for HCC. In addition, we further explored the independence and applicability of the signature with other clinical indicators through multivariate Cox analysis (p < 0.001) and nomogram analysis (C-index = 0.709). The above results indicate that the combination of MMP1, HMGCS2, and SLC27A5 selected from the PPAR signaling pathway could effectively, independently, and applicatively predict the prognosis of HCC. Our research provided new insights to the prognosis of HCC.
Keywords:
- Correction
- Source
- Cite
- Save
- Machine Reading By IdeaReader
26
References
1
Citations
NaN
KQI