Integrating machine learning and single-cell trajectories to analyze T-cell exhaustion to predict prognosis and immunotherapy in colon cancer patients

Integration of machine learning for developing a prognostic signature related to programmed cell death in colorectal cancer

BACKGROUND

Colorectal cancer (CRC) presents a significant global health burden, characterized by a heterogeneous molecular landscape and various genetic and epigenetic alterations. Programmed cell death (PCD) plays a critical role in CRC, offering potential targets for therapy by regulating cell elimination processes that can suppress tumor growth or trigger cancer cell resistance. Understanding the complex interplay between PCD mechanisms and CRC pathogenesis is crucial. This study aims to construct a PCD-related prognostic signature in CRC using machine learning integration, enhancing the precision of CRC prognosis prediction.

METHOD

We retrieved expression data and clinical information from the Cancer Genome Atlas and Gene Expression Omnibus (GEO) datasets. Fifteen forms of PCD were identified, and corresponding gene sets were compiled. Machine learning algorithms, including Lasso, Ridge, Enet, StepCox, survivalSVM, CoxBoost, SuperPC, plsRcox, random survival forest (RSF), and gradient boosting machine, were integrated for model construction. The models were validated using six GEO datasets, and the programmed cell death score (PCDS) was established. Further, the model's effectiveness was compared with 109 transcriptome-based CRC prognostic models.

RESULT

Our integrated model successfully identified differentially expressed PCD-related genes and stratified CRC samples into four subtypes with distinct prognostic implications. The optimal combination of machine learning models, RSF + Ridge, showed superior performance compared with traditional methods. The PCDS effectively stratified patients into high-risk and low-risk groups, with significant survival differences. Further analysis revealed the prognostic relevance of immune cell types and pathways associated with CRC subtypes. The model also identified hub genes and drug sensitivities relevant to CRC prognosis.

CONCLUSION

The current study highlights the potential of integrating machine learning models to enhance the prediction of CRC prognosis. The developed prognostic signature, which is related to PCD, holds promise for personalized and effective therapeutic interventions in CRC.

Unraveling T cell exhaustion in the immune microenvironment of osteosarcoma via single-cell RNA transcriptome

Osteosarcoma (OS) represents a profoundly invasive malignancy of the skeletal system. T cell exhaustion (Tex) is known to facilitate immunosuppression and tumor progression, but its role in OS remains unclear. In this study, single-cell RNA sequencing data was employed to identify exhausted T cells within the tumor immune microenvironment (TIME) of OS. We found that exhausted T cells exhibited substantial infiltration in OS samples. Pseudotime trajectory analysis revealed a progressive increase in the expression of various Tex marker genes, including PDCD1, CTLA4, LAG3, ENTPD1, and HAVCR2 in OS. GSVA showed that apoptosis, fatty acid metabolism, xenobiotic metabolism, and the interferon pathway were significantly activated in exhausted T cells in OS. Subsequently, a prognostic model was constructed using two Tex-specific genes, MYC and FCGR2B, which exhibited exceptional prognostic accuracy in two independent cohorts. Drug sensitivity analysis revealed that OS patients with a low Tex risk were responsive to Dasatinib and Pazopanib. Finally, immunohistochemistry verified that MYC and FCGR2B were significantly upregulated in OS tissues compared with adjacent tissues. This study investigates the role of Tex within the TIME of OS, and offers novel insights into the mechanisms underlying disease progression as well as the potential treatment strategies for OS.