Machine learning-based integration develops a multiple programmed cell death signature for predicting the clinical outcome and drug sensitivity in colorectal cancer.

Tumorigenesis and treatment are closely associated with various programmed cell death (PCD) patterns. However, the coregulatory role of multiple PCD patterns in colorectal cancer (CRC) remains unknown. In this study, we developed a multiple PCD index (MPCDI) based on 19 PCD patterns using two machine learning algorithms for risk stratification, prognostic prediction, construction of nomograms, immune cell infiltration analysis, and chemotherapeutic drug sensitivity analysis. As a result, in the TCGA-COAD, GSE17536, and GSE29621 cohorts, the MPCDI can effectively distinguished survival outcomes in CRC patients and served as an independent factor for CRC patients. We then explored the immune infiltration landscape in two groups using the nine algorithms and found more overall immune infiltration in the high-MPCDI group. TIDE scores suggested that the increased immune evasion potential and immune checkpoint inhibition therapy may be less effective in the high-MPCDI group. Immunophenoscores indicated that anti-PD1, anti-cytotoxic T-lymphocyte associated antigen 4 (anti-CTLA4), and anti-PD1-CTLA4 combination therapies are less effective in the high-MPCDI group. In addition, the high-MPCDI group was more sensitive to AZD1332, Foretinib, and IGF1R_3801, and insensitive to AZD3759, AZD5438, AZD6482, Erlotinib, GSK591, IAP_5620, and Picolinici-acid, which suggests that the MPCDI can guide drug selection for CRC patients. As a new clinical classifier, the MPCDI can more accurately distinguish CRC patients who benefit from immunotherapy and develop personalized treatment strategies for CRC patients.

Integrating machine learning and multi-omics analysis to develop an asparagine metabolism immunity index for improving clinical outcome and drug sensitivity in lung adenocarcinoma.

Lung adenocarcinoma (LUAD) is a malignancy affecting the respiratory system. Most patients are diagnosed with advanced or metastatic lung cancer due to the fact that most of their clinical symptoms are insidious, resulting in a bleak prognosis. Given that abnormal reprogramming of asparagine metabolism (AM) has emerged as an emerging therapeutic target for anti-tumor therapy. However, the clinical significance of abnormal reprogramming of AM in LUAD patients is unclear. In this study, we collected 864 asparagine metabolism-related genes (AMGs) and used a machine-learning computational framework to develop an asparagine metabolism immunity index (AMII) for LUAD patients. Through the utilization of median AMII scores, LUAD patients were segregated into either a low-AMII group or a high-AMII group. We observed outstanding performance of AMII in predicting survival prognosis in LUAD patients in the TCGA-LUAD cohort and in three externally independently validated GEO cohorts (GSE72094, GSE37745, and GSE30219), and poorer prognosis for LUAD patients in the high-AMII group. The results of univariate and multivariate analyses showed that AMII can be used as an independent risk factor for LUAD patients. In addition, the results of C-index analysis and decision analysis showed that AMII-based nomograms had a robust performance in terms of accuracy of prognostic prediction and net clinical benefit in patients with LUAD. Excitingly, LUAD patients in the low-AMII group were more sensitive to commonly used chemotherapeutic drugs. Consequently, AMII is expected to be a novel diagnostic tool for clinical classification, providing valuable insights for clinical decision-making and personalized management of LUAD patients.

Identification and validation of an immune-derived multiple programmed cell death index for predicting clinical outcomes, molecular subtyping, and drug sensitivity in lung adenocarcinoma.

OBJECTIVES: Comprehensive cross-interaction of multiple programmed cell death (PCD) patterns in the patients with lung adenocarcinoma (LUAD) have not yet been thoroughly investigated. METHODS: Here, we collected 19 different PCD patterns, including 1911 PCD-related genes, and developed an immune-derived multiple programmed cell death index (MPCDI) based on machine learning methods. RESULTS: Using the median MPCDI scores, we categorized the LUAD patients into two groups: low-MPCDI and high-MPCDI. Our analysis of the TCGA-LUAD training cohort and three external GEO cohorts (GSE37745, GSE30219, and GSE68465) revealed that patients with high-MPCDI experienced a more unfavorable prognosis, whereas those with low-MPCDI had a better prognosis. Furthermore, the results of both univariate and multivariate Cox regression analyses further confirmed that MPCDI serves as a novel independent risk factor. By combining clinical characteristics with the MPCDI, we constructed a nomogram that provides an accurate and reliable quantitative tool for personalized clinical management of LUAD patients. The findings obtained from the analysis of C-index and the decision curve revealed that the nomogram outperformed various clinical variables in terms of net clinical benefit. Encouragingly, the low-MPCDI patients are more sensitive to commonly used chemotherapy drugs, which suggests that MPCDI scores have a guiding role in chemotherapy for LUAD patients. CONCLUSION: Therefore, MPCDI can be used as a novel clinical diagnostic classifier, providing valuable insights into the clinical management and clinical decision-making for LUAD patients.