TCGA数据库助力中国研究团队在乳腺癌和肝癌研究中取得新突破

TCGA数据库助力中国研究团队在乳腺癌和肝癌研究中取得新突破

The Cancer Genome Atlas (TCGA) database, a landmark collection of genomic profiles for 33 cancer types, continues to drive cutting-edge research in oncology. Despite the program's closure in 2018, its legacy lives on through ongoing data updates and new discoveries. Chinese research teams have made significant contributions to this field, leveraging TCGA data to uncover novel insights into breast and liver cancer.

In a recent study published in the Journal of Clinical Hepatology, a team from Peking University and the Fifth Medical Center of PLA General Hospital developed a novel prognostic model for hepatocellular carcinoma (HCC) based on disulfide death-related genes. Disulfide death, a form of programmed cell death, has been implicated in cancer progression and drug resistance.

The researchers analyzed mRNA expression profiles and clinical data from the TCGA-LIHC cohort, identifying seven disulfide death-related genes significantly associated with overall survival (OS). Using LASSO-Cox regression analysis, they constructed a four-gene prognostic model, calculating a risk score (RS-DRG) based on the expression levels of GYS1, LRPPRC, RPN1, and SLC7A1.

The model demonstrated robust prognostic performance, with high-risk patients exhibiting significantly shorter OS compared to low-risk patients (P<0.001). This finding was validated in external cohorts from ICGC and GSE14520, confirming the model's predictive power. Notably, the risk score was an independent prognostic factor in both TCGA and ICGC cohorts, with hazard ratios of 1.869 (P=0.002) and 3.469 (P=0.004), respectively.

Further analysis revealed that the risk score was positively correlated with the infiltration levels of various immune cells in the tumor microenvironment, including B lymphocytes, CD4+ T cells, neutrophils, macrophages, and dendritic cells. Moreover, patients in the high-risk group exhibited increased sensitivity to sorafenib, a commonly used targeted therapy for HCC, as evidenced by lower IC50 values (P<0.001).

KEGG/GO enrichment analysis indicated that differentially expressed genes in the high-risk group were significantly enriched in mitosis-related molecular functions, suggesting a potential mechanism underlying the association between disulfide death and HCC progression.

These findings not only advance our understanding of the molecular mechanisms underlying HCC but also provide a promising tool for prognostic stratification and personalized treatment. The identification of disulfide death-related genes as potential therapeutic targets opens new avenues for the development of targeted therapies, which could improve outcomes for HCC patients.

In another notable study, researchers characterized cancer-driving nucleotides (CDNs) across genes, cancer types, and patients. They found that CDNs tend to code for amino acids with divergent chemical properties and are more widely shared among cancer types than canonical cancer-driving genes. Importantly, each cancer patient is expected to carry 5-8 CDNs, but currently, only 0-2 can be identified. By expanding the sample size to 10^5, most CDNs can be identified, which will facilitate the design of patient-specific targeting against multiple CDN-harboring genes.

These studies exemplify the enduring value of the TCGA database in cancer research. Its rich collection of genomic data continues to fuel innovative discoveries, driving progress towards precision medicine and improved patient outcomes.