简介：基因组分析彻底改变了肝癌的治疗干预和临床管理。然而，肝癌的发病机制、复发的分子决定因素和一线治疗（抗 PD-(L)1 加贝伐珠单抗）的预测生物标志物仍不完全清楚。材料和方法：采用靶向下一代测序（tNGS）（603个癌症基因组）对232名肝细胞癌（HCC）和22名肝内胆管癌（ICC）患者进行基因组分析，其中47名不可切除/转移性HCC患者接受了抗 PD-1 加贝伐珠单抗治疗。估计基因组改变与血管侵犯 (VI)、发病位置、复发、总生存期 (OS)、无复发生存期 (RFS) 和抗 PD-1 加贝伐单抗治疗反应的关系。结果：基因组图谱显示，HCC 中最常改变的基因是 TP53、FAT3、PDE4DIP、KMT2C、FAT1 和 MYO18A，而 ICC 中 TP53、FAT1、FAT3、PDE4DIP、ROS1 和 GALNT11 经常改变；值得注意的是，KRAS（18.18% vs. 1.29%）和 BAP1（13.64% vs. 1.29%）突变在 ICC 中更为普遍。比较分析证明了中国和西方 HCC 队列之间不同的临床病理学/基因组特征。 VI 相关 HCC 的基因组分析显示，与没有 VI 的患者相比，VI 组的 LDLR、MSH2、KDM5D、PDE3A 和 FOXO1 经常发生改变。与 HCC 患者的肝右叶相比，HCC 患者的左肝叶具有更高的 OS（中位 OS：36.77 个月 vs 未达到，p < 0.05）。通过进一步比较，Notch信号通路相关的改变在HCC患者的右肝叶中显着普遍。值得注意的是，多变量 Cox 回归分析显示，RB1、NOTCH3、MGA、SYNE1 和 ZFHX3 的改变作为独立预后因素，与 HCC 患者的 OS 显着相关。此外，改变的 LATS1 在 HCC 复发组中大量富集，令人印象深刻的是，它在预测 RFS 方面与临床病理特征无关（改变型与野生型的中位 RFS：5.57 个月与 22.47 个月，p < 0.01）。对于那些接受治疗的 HCC 患者，TMB 值、PTPRZ1 改变和细胞周期相关改变被确定与客观缓解率 (ORR) 呈正相关，但 KMT2D 改变与 ORR 呈负相关。此外，KMT2D 和细胞周期信号传导的改变分别与无进展生存期 (PFS) 时间的缩短和延长显着相关。结论：全面的基因组分析破译了肝癌 VI、发病部位、复发和生存时间的不同分子特征。 HCC 中抗 PD-1 加贝伐单抗反应的新型遗传预测因子的鉴定促进了循证治疗方法的开发。版权所有 © 2024 Wang、Shang、Xu、Dong、Zhang、Xia、Sui 和 Yang。

Introduction: Genomic profiling has revolutionized therapeutic interventions and the clinical management of liver cancer. However, pathogenetic mechanisms, molecular determinants of recurrence, and predictive biomarkers for first-line treatment (anti-PD-(L)1 plus bevacizumab) in liver cancer remain incompletely understood. Materials and methods: Targeted next-generation sequencing (tNGS) (a 603-cancer-gene panel) was applied for the genomic profiling of 232 hepatocellular carcinoma (HCC) and 22 intrahepatic cholangiocarcinoma (ICC) patients, among which 47 unresectable/metastatic HCC patients underwent anti-PD-1 plus bevacizumab therapy. Genomic alterations were estimated for their association with vascular invasion (VI), location of onset, recurrence, overall survival (OS), recurrence-free survival (RFS), and anti-PD-1 plus bevacizumab therapy response. Results: The genomic landscape exhibited that the most commonly altered genes in HCC were TP53, FAT3, PDE4DIP, KMT2C, FAT1, and MYO18A, while TP53, FAT1, FAT3, PDE4DIP, ROS1, and GALNT11 were frequently altered in ICC; notably, KRAS (18.18% vs. 1.29%) and BAP1 (13.64% vs. 1.29%) alterations were significantly more prevalent in ICC. Comparison analysis demonstrated the distinct clinicopathological/genomic characterizations between Chinese and Western HCC cohorts. Genomic profiling of HCC underlying VI showed that LDLR, MSH2, KDM5D, PDE3A, and FOXO1 were frequently altered in the VI group compared to patients without VIs. Compared to the right hepatic lobes of HCC patients, the left hepatic lobe of HCC patients had superior OS (median OS: 36.77 months vs. unreached, p < 0.05). By further comparison, Notch signaling pathway-related alterations were significantly prevalent among the right hepatic lobes of HCC patients. Of note, multivariate Cox regression analysis showed that altered RB1, NOTCH3, MGA, SYNE1, and ZFHX3, as independent prognostic factors, were significantly correlated with the OS of HCC patients. Furthermore, altered LATS1 was abundantly enriched in the HCC-recurrent group, and impressively, it was independent of clinicopathological features in predicting RFS (median RFS of altered type vs. wild-type: 5.57 months vs. 22.47 months, p < 0.01). Regarding those treated HCC patients, TMB value, altered PTPRZ1, and cell cycle-related alterations were identified to be positively associated with the objective response rate (ORR), but KMT2D alterations were negatively correlated with ORR. In addition, altered KMT2D and cell cycle signaling were significantly associated with reduced and increased time to progression-free survival (PFS), respectively. Conclusion: Comprehensive genomic profiling deciphered distinct molecular characterizations underlying VI, location of onset, recurrence, and survival time in liver cancer. The identification of novel genetic predictors of response to anti-PD-1 plus bevacizumab in HCC facilitated the development of an evidence-based approach to therapy.Copyright © 2024 Wang, Shang, Xu, Dong, Zhang, Xia, Sui and Yang.