开发与自噬相关的长链非编码 RNA 的风险模型和亚型，以增强对肝癌患者的预后预测和治疗方法的精准性。

BACKGROUND: Limited research has been conducted on the influence of autophagy-associated long non-coding RNAs (ARLncRNAs) on the prognosis of hepatocellular carcinoma (HCC). METHODS: We analyzed 371 HCC samples from TCGA, identifying expression networks of ARLncRNAs using autophagy-related genes. Screening for prognostically relevant ARLncRNAs involved univariate Cox regression, Lasso regression, and multivariate Cox regression. A Nomogram was further employed to assess the reliability of Riskscore, calculated from the signatures of screened ARLncRNAs, in predicting outcomes. Additionally, we compared drug sensitivities in patient groups with differing risk levels and investigated potential biological pathways through enrichment analysis, using consensus clustering to identify subgroups related to ARLncRNAs. RESULTS: The screening process identified 27 ARLncRNAs, with 13 being associated with HCC prognosis. Consequently, a set of signatures comprising 8 ARLncRNAs was successfully constructed as independent prognostic factors for HCC. Patients in the high-risk group showed very poor prognoses in most clinical categories. The Riskscore was closely related to immune cell scores, such as macrophages, and the DEGs between different groups were implicated in metabolism, cell cycle, and mitotic processes. Notably, high-risk group patients demonstrated a significantly lower IC50 for Paclitaxel, suggesting that Paclitaxel could be an ideal treatment for those at elevated risk for HCC. We further identified C2 as the Paclitaxel subtype, where patients exhibited higher Riskscores, reduced survival rates, and more severe clinical progression. CONCLUSION: The 8 signatures based on ARLncRNAs present novel targets for prognostic prediction in HCC. The drug candidate Paclitaxel may effectively treat HCC by impacting ARLncRNAs expression. With the identification of ARLncRNAs-related isoforms, these results provide valuable insights for clinical exploration of autophagy mechanisms in HCC pathogenesis and offer potential avenues for precision medicine.

背景：关于自噬相关长链非编码 RNA（ARLncRNAs）对肝细胞癌（HCC）预后影响的研究较少。

方法：我们分析了 TCGA 中的 371 个 HCC 样本，使用自噬相关基因构建了 ARLncRNA 的表达网络。通过单因素 Cox 回归、Lasso 回归和多因素 Cox 回归筛选出与预后相关的 ARLncRNA。进一步采用列线图评估从筛选出的 ARLncRNA 特征计算得到的 Riskscore 预测结果的可靠性。此外，我们比较了不同风险水平患者组的药物敏感性，并通过富集分析研究了潜在的生物学途径，使用共识聚类来识别与 ARLncRNA 相关的亚组。

结果：筛选过程确定了 27 个 ARLncRNA，其中 13 个与 HCC 预后相关。因此，成功构建了由 8 个 ARLncRNA 组成的一组特征，作为 HCC 独立预后因素。高风险组患者在大多数临床类别中预后极差。Riskscore 与巨噬细胞等免疫细胞评分密切相关，不同组之间的差异基因与代谢、细胞周期和有丝分裂过程有关。值得注意的是，高风险组患者对紫杉醇的 IC50 明显降低，提示紫杉醇可能是治疗 HCC 的理想药物。我们进一步鉴定出 C2 是紫杉醇亚型，其中患者的 Riskscore 更高、生存率降低、临床进展更严重。

结论：基于 ARLncRNA 的 8 个特征为 HCC 的预后预测提供了新的靶点。候选药物紫杉醇可能通过影响 ARLncRNA 的表达有效治疗 HCC。通过鉴定与 ARLncRNAs 相关的亚型，这些结果为探索 HCC 发病机制中自噬机制的临床研究提供了有价值的见解，并为精准医学提供了潜在途径。