TP53 mutation is one of the most common genetic changes in hepatocellular carcinoma (HCC). It is of great clinical significance to tailor specialized prognostication approach and to explore more therapeutic options for TP53-mutant HCCs. In this study, a total of 1135 HCC patients were retrospectively analyzed. We developed a random forest-based prediction model to estimate TP53 mutational status, tackling the problem of limited sample size in TP53-mutant HCCs. A multi-step process was performed to develop robust poor prognosis-associated signature (PPS). Compared with previous established population-based signatures, PPS manifested superior ability to predict survival in TP53-mutant patients. After in silico screening of 2249 drug targets and 1770 compounds, we found that three targets (CANT1, CBFB and PKM) and two agents (irinotecan and YM-155) might have potential therapeutic implications in high-PPS patients. The results of drug targets prediction and compounds prediction complemented each other, presenting a comprehensive view of potential treatment strategy. Overall, our study has not only provided new insights into personalized prognostication approaches, but also thrown light on integrating tailored risk stratification with precision therapy.

中文翻译：

---

TP53突变型肝细胞癌的预后和个性化治疗预测：精准肿瘤学的计算机策略。

TP53突变是肝细胞癌（HCC）中最常见的遗传变化之一。为TP53 突变型 HCCs 量身定制专门的预后方法和探索更多的治疗选择具有重要的临床意义。本研究回顾性分析了 1135 例 HCC 患者。我们开发了一个基于随机森林的预测模型来估计TP53突变状态，解决了TP53突变HCC中样本量有限的问题。执行多步骤过程以开发稳健的不良预后相关特征（PPS）。与之前建立的基于人群的特征相比，PPS 在预测TP53 突变患者的存活率方面表现出卓越的能力。在silico之后筛选了 2249 种药物靶点和 1770 种化合物，我们发现三个靶点（CANT1、CBFB和PKM）和两种药物（伊立替康和 YM-155）可能对高 PPS 患者具有潜在的治疗意义。药物靶点预测和化合物预测结果相得益彰，全面展示了潜在的治疗策略。总体而言，我们的研究不仅为个性化预测方法提供了新的见解，而且还揭示了将量身定制的风险分层与精准治疗相结合的思路。