CRC is one of the most common malignancies worldwide. Although surgery, radiation, chemotherapy, and immunotherapy have made great progress in recent years, the mortality rate of CRC is still high, which reflects the importance of predicting prognosis and long-term follow-up20. Therefore, it is particularly important to find biomarkers suitable for the prediction of prognosis in CRC patients, which helps clinicians to develop effective treatment regimens and improve the survival rate of CRC patients21.
Compared with normal cells, iron requirements are increased in order to promote growth in tumor cells. The iron-dependent properties make tumor cells more susceptible to iron catalyzed necrosis and sensible to chemotherapy drugs22,23. Therefore, a complex form of iron-dependent programmed cell death called ferroptosis, has great potential in the field of tumor treatment24. LncRNAs are RNA molecules with a transcript length of more than 200 nt and lack a protein-coding ability. Although lncRNAs do not encode proteins, they bind with some miRNAs in vivo to regulate the expression of related genes as competing endogenous RNAs and participate in various biological processes, including ferroptosis25,26. A study reported that lncRNAs promoted ferroptosis in tumor cells by affecting the p53 signaling pathway related to ferroptosis and inhibiting the expression of the ferroptosis-related gene SLC7A1127. LncRNA has been shown to accelerate ferroptosis in tumor cells by increasing lipid reactive oxygen species (ROS) and iron levels and decreasing the expression of SLC7A1128.These results indicate that LncRNAs play an important role in regulating ferroptosis in tumor cells. LncRNAs have great significance in the diagnosis, prognosis, and treatment of various cancers29. LncRNAs as competitive endogenous RNAs (ceRNAs) could occupy the binding sequence of miRNAs competitively, thus isolating miRNAs and changing the expression of their downstream target genes30. LncRNAs could be used as a potential biomarker for clinical diagnosis, treatment and prognosis of CRC31.
Over the past decade, ML has become a breakthrough technology, and is widely applied in medicine and exhibits excellent efficiency in the field of clinical diagnosis and optimal treatment of various kinds of diseases32,33. In this study, we aimed to investigate the prognostic value and clinical significance of ferroptosis-related lncRNAs in CRC via ML methods. CRC transcriptome data and clinical survival data in the TCGA database were analyzed. A total of 6 ferroptosis-related lncRNAs were screened out, named AP003555.1, AC129492.1, AC010973.2, LINC01857, ITGB1-DT and AP001469.3.The expression of the above lncRNAs in the samples and the results of the LASSO COX regression analysis were analyzed for risk scoring. The data were classified as high or low risk based on the median value of the risk score. The expression of the 6 ferroptosis-related lncRNAs in the high and low risk groups was statistically different, and the expression of the 6 ferroptosis-related lncRNAs in the high risk group was significantly higher than that in the low risk group. AC010973.2, LINC01857, and ITGB1-DT have been shown in studies to be prognostic biomarkers for patients with CRC. This result is consistent with existing studies and proves the reliability of 6 ferroptosis-related lncRNAs screened by bioinformatics methods as prognostic biomarkers.
The results demonstrated that the risk score could be used as an independent prognostic factor to assess patient prognosis, and the ROC curve validated the results of the independent prognostic analysis34. The survival rate of patients with clinical CRC was predicted by using the nomogram. The Kaplan-Meier survival curve and risk curve verified the prognostic value of the 6 ferroptosis-related lncRNAs in CRC patients. Patients in the high-risk group showed a lower survival rate, and the difference between the two groups was statistically significant. Next, this study analyzed immune-related differences among patients in the high and low-risk groups.
Tumor-infiltrating immune cells promote and regulate tumor growth; they are an important part of the complex environment and can affect the sensitivity of diagnosis, survival, and clinical treatment35–37. Tumor-infiltrating immune cells have been linked to CRC progression and prognosis, which could assist clinicians in selecting effective immune therapy for CRC patients and developing an individualized treatment plan38. In this study, different methods were used to analyze the immune infiltration in the low-risk group, and the degree of partial immune infiltration in the high-risk group was significantly lower than that in the low-risk group. The APC Co Stimulation, CCR, Parainflammation, Type II IFN Response, and other immune functions in the low-risk group were significantly higher than those in the high-risk group. In view of the increasing importance of immune checkpoints inhibitors in tumor therapy in recent years39, the differences in immune checkpoints between the high and low risk groups were analyzed in this study.
Immune checkpoints are costimulatory or inhibitory signals that regulate the recognition of T cell receptor antigens in the process of immune response40. Under normal physiological conditions, immune checkpoints can regulate immune functions, prevent the over-activation of immune functions and avoid the occurrence of autoimmune diseases. However, immune functions were suppressed when the immune checkpoints were overexpressed. Conversely, when the expression of immune checkpoints is decreased, the immune function of the body will be abnormal41–43. Up-regulated expression of immune checkpoints like PD-1 and CTLA-4 on T lymphocytes in hepatocellular carcinoma induces immune escape and promotes the occurrence and development of hepatocellular carcinoma44. Tumor cells could affect the immune checkpoints so that antigen cannot be presented effectively and the immune function of the body cannot be activated effectively, inhibiting the function of immune cells, which could cause immune escape45 .Immune checkpoints inhibitors such as PD-1 monoclonal antibody and CTLA-4 antibody have achieved good therapeutic effects in clinical practice, which could remove the effect of tumor cells on immune checkpoints, activate the immune system of the body, and effectively kill tumor cells46–48. It was found that CTLA4, TNFRSF18, and LGALS9 were significantly different in the high and low risk groups.
RNA methylation is a reversible post-transcriptional modification of RNA that exerts significant effects on many biological processes49. N6-methyladenosine (m6A) is considered to be one of the most common types of eukaryotic mRNA methylation modification and exists in a variety of organisms.m6A methylation is involved in many functions, including translation, splicing, storage, transportation, and degradation50. Abnormal regulation of m6A methylation found in CRC tissues is critical for cancer formation, progression, invasion, and metastasis51. In this study, the differences in m6A methylation related genes in the high and low risk groups were analyzed, and significant differences were found in METTL3, YTHDH2, YTHDC1, and other genes in the high and low risk groups. which is consistent with a previous study52. These results could provide a reference for targeting m6A methylation related gene therapy in CRC.