As two novel cell death modalities, mounting evidence has demonstrated that ferroptosis and necroptosis are intimately linked with tumor progression[17–19]. However, few studies have thoroughly explored the potential value of combining differentially expressed FRLRs and NRLRs as a prognostic signature of LIH to predict immunotherapy responses.
Dysfunction during necroptosis, a process is regulated by lncRNAs, is involved in the development of HCC[20–22]. Thus, this study first identified lncRNAs associated with ferroptosis and necroptosis and investigated their value in LIHC immunotherapy[23, 24]. The expression and prognostic significance of 319 FRLRs and 174 NRLRs in LIHC were explored. Taking these findings into consideration, a prognostic gene expression signature was constructed based on ferroptosis and necroptosis in LIHC. A total of 120 F-NLRs were involved in the regulation of protein stability, cellular protein localization, neuronal death, CD40 receptor complex, the cytoplasmic side of the plasma membrane, ubiquitin-like protein ligase binding, and ubiquitin protein ligase binding. These genes were closely linked to necrotic apoptosis and the immune response.
Among the F-NLRs identified in the study, five lncRNAs were significantly associated with OS in LIHC, including KDM4A-AS1, ZFPM2-AS1, AC099850.3, MKLN1-AS, and BACE1-AS. The expression of proteins encoded by these genes was further validated by qRT-PCR, demonstrating that all five F-NLRs were overexpressed in liver cancer cells. Expression levels of KDM4A-AS1 were significantly elevated in both LIHC tissues and cell lines, and high KDM4A-AS1 expression was associated with advanced TNM staging and lymphatic metastasis. A previous study reported that KDM4A-AS1 expression was closely linked to OS and prognosis of LIHC. Thus, KDM4A-AS1 is considered to be an important prognostic factor for patients with HCC[25]. Additionally, KDM4A-AS1 has been shown to promote proliferation, migration, invasion, and epithelial–mesenchymal transition in HCC cells under loss- and gain-of-function conditions. Specifically, KDM4A-AS1 acted via the miR-411-5p/KPNA2/AKT pathway to promote the growth and metastasis of HCC[26]. Moreover, KDM4A-AS1 was shown to play an important role in the progression of castration-resistant prostate cancer (CRPC) and enzalutamide resistance by regulating the deubiquitination of androgen receptors (AR) and AR splice variants, thus providing a potential therapeutic target for CRPC[27]. These findings suggested that KDM4A-AS1 may be an independent potential prognostic biomarker for patients with LIHC. Furthermore, KDM4A-AS1 may be important for controlling the occurrence of LIHC since it is closely associated with the TME. However, the role of KDM4A-AS1 in carcinogenesis remains unclear. Meanwhile, the expression of ZFPM2-AS1 was upregulated in HCC and high ZFPM2-AS1 expression was associated with age, T stage, and pathological stage of the patient. A previous study reported that STAT1 activated the translational expression of ZFPM2-AS1 in HCC, thus modulating levels of the protein[28]. Knockdown of ZFPM2-AS1 inhibited cell proliferation, migration, invasion, and apoptosis[29]. Furthermore, ZFPM2-AS1 overexpression predicted poor prognosis in lung adenocarcinoma (LUAD) and boosted cell proliferation in these tumors[30]. Previous studies have indicated that AC099850.3 plays a role in cancer. A study by Zhou demonstrated that AC099850.3 was associated with prognosis and competition between endogenous RNAs in tongue squamous cell carcinoma[31]. AC099850.3 was also significantly upregulated in non-small cell lung cancer cells, and its depletion markedly inhibited cell proliferation and migration in LUAD cells[32]. Because AC099850.3 is significantly associated with the TME in HCC, it may provide a potential immunotherapy target[33]. Research demonstrated that MKLN1-AS upregulation was associated with vascular invasion, suggesting that MKLN1-AS was involved in tumor progression[34]. As a competitive endogenous RNA, MKLN1-AS increased the expression of hepatoma-derived growth factor in HCC cells by competitively binding to miR-654-3p[35]. Patients with HCC may benefit from the use of MKLN1-AS as a therapeutic target in the future. Finally, BACE1-AS is considered to be an effective biomarker for predicting the prognosis of patients with LIHC and has been used in a risk model to assess LIHC prognosis[25]. M2 macrophage abundance was negatively correlated with BACE1-AS levels in tenosynovial giant cell tumors, LIHC, lung squamous cell carcinoma, colon adenocarcinoma, prostate adenocarcinoma, and KIRC, and was positively correlated with BACE1-AS levels in thyroid carcinoma, cervical squamous cell carcinoma and endocervical adenocarcinoma, and acute myeloid leukemia. The BACE1-AS protein may promote tumor antigen presentation in cancer cells, thereby suppressing immune responses[36]. Moreover, BACE1-AS may inhibit the proliferation and invasion of human ovarian cancer stem cells, providing a potential new mechanism for anisomycin treatment in LIHC[37].
In the present study, patients with LIHC were divided into low- and high-risk groups according to the median risk score, revealing that patients in the low-risk group had better outcomes than those in the high-risk group. Risk score and stage were independent predictors of OS in patients with LIHC. Further, survival prediction for patients with LIHC was improved using the combined F-NLR signature model compared to that achieved using conventional clinical characteristics. An AUC value closer to 1.0 indicates a more accurate prediction of diagnosis, with an AUC value > 0.7 indicating a highly accurate model. High statistical significance was established for OS at 1, 3, and 5 years, suggesting that the combined F-NLR signature can accurately predict the prognosis of patients with LIHC.
Subsequently, potential candidate drugs for LIHC were identified using the F-NLR model. Assessment of IC50 values, the TME, and immunotherapy responses indicated that patients in the high-risk group responded more positively to immunotherapeutic agents. TMB scores were calculated using the cellular mutation data from the TCGA-LIHC cohort, and the low-risk group did not outperform the high-risk group. A high TMB score in LIHC was associated with poorer outcome and could be used as a prognostic marker. The results indicated that the combined F-NLR signature model had a higher prognostic value than TMB. KDM4A-AS1, ZFPM2-AS1, AC099850.3, MKLN1-AS, and BACE1-AS may provide biomarkers for drug therapy at different tumor stages and gene mutation loads, thus facilitating precise treatment for patients with LIHC. However, the mechanisms of these five lncRNAs in liver cancer warrant further exploration.
According to the combined F-NLR signature model, patients in the low- and high-risk groups displayed great differences in drug sensitivity, which were associated with necroptosis and the TME. These results suggest that patients with LIHC who differentially express F-NLRs may have different drug sensitivities for different targets, reflecting individual differences in treatment responses. These results are relevant for future LIHC research investigating therapeutic targets related to immunotherapy and the TME. Due to differences in immune responses, targeted therapy can be combined with immunotherapy to provide personalized and precise treatment for patients with LIHC based on lncRNA expression levels. Although KDM4A-AS1, ZFPM2-AS1, AC099850.3, MKLN1-AS, and BACE1-AS were highly expressed in LIHC cell lines, their expression differed between most tumor types and normal tissues.
Nevertheless, this study had some limitations. First, data from clinical studies are required for specific validation of the established model. Second, further experimental studies are warranted to elucidate the molecular mechanisms of F-NLRs in LIHC, including regulation of FRGs and NRGs by the five key lncRNAs. These results will provide further insight for the development of immunotherapy drugs for LIHC based on F-NLRs.