Gene expression analysis of metastatic adrenocortical tumors. Gene expression analysis of metastatic adrenocortical tumors

Background: Adrenocortical carcinoma (ACC) is a rare, often-aggressive neoplasm of the adrenal cortex, with a 14.5-month median overall survival. We asked whether tumors from patients with advanced or metastatic ACC would offer clues as to putative genes that might have critical roles in disease progression or in more aggressive disease biology. Methods: We conducted comprehensive genomic and expression analyses of 43 ACCs. Results: Copy number gains and losses matched that previously reported. We identified a median mutation rate of 3.38 per megabase (Mb), somewhat higher than in a previous study possibly related to the more advanced disease. The mutational signature was characterized by a predominance of C>T, C>A and T>C transitions. As in previously reports, only cancer genes TP53 (26%) and beta-catenin (CTNNB1, 14%) were mutated in more than 10% of samples. The TCGA-identified putative cancer genes MEN1 and PRKAR1A were found in low frequency – 4.7% and 2.3%, respectively. Most of the mutations were in genes not implicated in the etiology or maintenance of cancer. Specifically, amongst the 38 genes that were mutated in more than 9% of samples, only four were represented in Tier 1 of the 576 COSMIC Cancer Gene Census (CCGC). Thus, 82% of genes found to have mutations likely have no role in the etiology or biology of ACC; while the role of the other 18%, if any, remains to be proven. Finally, the transcript length for the 38 most frequently mutated genes in ACC is statistically longer than the average of all coding genes, raising the question of whether transcript length in part determined mutation probability. Conclusions: We conclude that the mutational and expression profiles of advanced and metastatic tumors is very similar to those from newly diagnosed patients –with very little in the way of genomic aberration to explain it. Our data and that in the previous analyses finds the rate of mutations in ACCs lower than that in other cancers and suggests an epigenetic basis for the disease should be the focus of future studies. The Affymetrix PrimeView platform was used for the gene expression profiling. Overall design: Tumor samples embedded in OCT were sectioned, stained with hematoxylin and eosin and reviewed by a pathologist. RNA was extracted from 50-100 mg of tumor using the Qiagen miRNeasy Kit and then used for cDNA array analyses.

### 背景
肾上腺皮质癌（Adrenocortical carcinoma, ACC）是一种罕见且常具有侵袭性的肾上腺皮质源性肿瘤，中位总生存期仅14.5个月。本研究旨在探讨晚期或转移性ACC患者的肿瘤组织，能否为鉴定在疾病进展或更具侵袭性的肿瘤生物学行为中发挥关键作用的候选基因提供线索。

### 方法
本研究对43例ACC样本开展了全面的基因组与表达谱分析。

### 结果
拷贝数变异的扩增与缺失情况与既往报道一致。本研究检测到的中位突变率为每兆碱基（Mb）3.38个突变，略高于既往研究，这可能与本研究纳入的样本疾病分期更晚相关。突变特征以C>T、C>A及T>C转换为主。与既往报道一致，仅肿瘤基因TP53（26%）与β-连环蛋白（beta-catenin, CTNNB1，14%）的突变率超过10%。癌症基因组图谱（TCGA）鉴定出的候选肿瘤基因MEN1与PRKAR1A的突变频率较低，分别为4.7%与2.3%。绝大多数突变发生在与癌症病因或肿瘤维持无关的基因中。具体而言，在突变频率超过9%的38个基因中，仅有4个被纳入癌症体细胞突变目录（COSMIC Cancer Gene Census, CCGC）576个基因的Tier 1范畴。因此，82%被检测到存在突变的基因可能与ACC的病因或肿瘤生物学无关；剩余18%的基因若存在相关作用，仍有待进一步验证。最后，ACC中38个突变频率最高的基因的转录本长度，在统计学上显著长于所有编码基因的平均长度，这一结果引发了一个疑问：转录本长度是否在一定程度上决定了突变发生的概率。

### 结论
本研究结论表明，晚期与转移性ACC的突变谱及表达谱与初诊患者的肿瘤样本高度相似，几乎不存在可解释该现象的基因组异常。本研究数据与既往分析结果均显示，ACC的突变率低于其他癌症类型，提示该疾病的表观遗传调控机制应作为未来研究的重点方向。本研究采用Affymetrix PrimeView平台开展基因表达谱分析。

### 整体实验设计
将OCT包埋的肿瘤样本进行切片，经苏木精-伊红（HE）染色后由病理学家进行阅片复核。使用Qiagen miRNeasy试剂盒从50~100mg肿瘤组织中提取RNA，随后用于cDNA芯片分析。