Biallelic TP53 inactivation is the most important high-risk factor associated with poor survival in multiple myeloma. Classical biallelic TP53 inactivation has been defined as simultaneous mutation and copy number loss in most studies; however, numerous studies have demonstrated that other factors could lead to the inactivation of TP53. Here, we hypothesized that novel biallelic TP53 inactivated samples existed in the multiple myeloma population. A random forest regression model that exploited an expression signature of 16 differentially expressed genes between classical biallelic TP53 and TP53 wild-type samples was subsequently established and used to identify novel biallelic TP53 samples from monoallelic TP53 groups. The model reflected high accuracy and robust performance in newly diagnosed relapsed and refractory populations. Patient survival of classical and novel biallelic TP53 samples was consistently much worse than those with mono-allelic or wild-type TP53 status. We also demonstrated that some predicted biallelic TP53 samples simultaneously had copy number loss and aberrant splicing, resulting in overexpression of high-risk transcript variants, leading to biallelic inactivation. We discovered that splice site mutation and overexpression of the splicing factor MED18 were reasons for aberrant splicing. Taken together, our study unveiled the complex transcriptome of TP53, some of which might benefit future studies targeting abnormal TP53.
双等位基因TP53失活是多发性骨髓瘤中与不良生存相关的最重要高风险因素。在多数研究中，经典的双等位基因TP53失活被定义为同时存在突变和拷贝数缺失；然而，大量研究表明其他因素也可能导致TP53失活。本研究提出假说：在多发性骨髓瘤群体中存在新型双等位基因TP53失活样本。随后，我们建立了一个随机森林回归模型，该模型利用了经典双等位基因TP53样本与TP53野生型样本之间16个差异表达基因的表达特征，并用于从单等位基因TP53组中识别新型双等位基因TP53样本。该模型在新诊断、复发和难治性患者群体中均表现出高准确性和稳健性能。经典与新型双等位基因TP53样本的患者生存率始终显著差于单等位基因或野生型TP53状态的患者。研究还证实，部分预测的双等位基因TP53样本同时存在拷贝数缺失和异常剪接，导致高风险转录变体过表达，从而引发双等位基因失活。我们发现剪接位点突变及剪接因子MED18的过表达是导致异常剪接的原因。综上所述，本研究揭示了TP53复杂的转录组特征，其中部分发现可能为未来靶向异常TP53的研究提供有益参考。
肿瘤(癌症)患者之家