Identification of a distinct luminal subgroup diagnosing and stratifying early stage prostate cancer by tissue-based single-cell RNA sequencing

The highly intra-tumoral heterogeneity and complex cell origination of PCa greatly limited the significance of traditional bulk RNA sequencing in finding better biomarker for disease stratification. In the context of these sequencing strategies, significant gene expression differences in specific cell population might be normalized or even shaded by genes in those less “important” but large in amount cells. Tissue specimens based single-cell RNA sequencing holds great promise for identification of novel stratification biomarkers. However, the preparation of enough viable cells for sequencing, in particular for prostate cancer (PCa), is very challenging when using this technique. In the current study, we, for the first time, employed single-cell RNA sequencing of PCa cells isolated from cancerous prostate tissues of two patients and identified fifteen cell groups including three luminal clusters with different expression profiles. One of these luminal clusters had the highest copy number variation level and marker genes mainly enriched in PCa-related metabolic process, thereafter, was considered as the malignant luminal cells in PCa. Further analysis indicates that the fifth subgroup of these cells highly expressed genes significantly correlated with PCa initiation and early development. In addition, we found this subgroup highly expressed PCa prognosis biomarkers such as AMACR and PCA3. These findings suggest that this cell subgroup might be a critical cell population for PCa diagnosis and stratification. Moreover, by further deciphering the gene expression of this subgroup, we identified another marker gene, HPN, with a 0.930 area under the curve (AUC) score distinguishing normal tissue from PCa lesion using the RNA-seq data from The Cancer Genome Atlas (TCGA). This finding was further validated by immunostaining of HPN in PCa tissue array. The expression of HPN in PCa with Gleason score > 6 is significantly higher than those in Gleason score = 6. Taken together, we provide a valuable resource for interpreting the tumor heterogeneity in PCa, and a novel candidate marker for PCa management. Examination of intra-tumoral heterogeneity of prostate cancer by single-cell RNA sequencing of cancerous prostate tissues from two patients.

前列腺癌（Prostate Cancer, PCa）极高的瘤内异质性与复杂的细胞起源，极大限制了传统批量RNA测序（bulk RNA sequencing）在筛选更优质的疾病分层（disease stratification）生物标志物（biomarker）方面的应用价值。在这类测序策略下，特定细胞群中的显著基因表达差异，可能会被那些虽非核心但数量庞大的细胞中的基因所归一化甚至掩盖。基于组织标本的单细胞RNA测序（single-cell RNA sequencing），在鉴定新型分层生物标志物方面展现出巨大潜力。然而，利用该技术获取足量活细胞用于测序极具挑战，针对前列腺癌的相关研究尤为如此。
本研究首次对两名患者癌性前列腺组织中分离得到的前列腺癌细胞开展单细胞RNA测序，成功鉴定出15个细胞群，其中包含3个具有不同表达谱的腔上皮细胞簇（luminal clusters）。其中一个腔上皮细胞簇拥有最高的拷贝数变异（copy number variation）水平，其标志物基因主要富集于前列腺癌相关代谢过程，因此被认定为前列腺癌恶性腔上皮细胞。进一步分析显示，该类细胞的第五亚群高表达与前列腺癌发生及早期发展显著相关的基因。此外，本研究发现该亚群高表达AMACR、PCA3等前列腺癌预后生物标志物。上述结果提示，该细胞亚群或许是前列腺癌诊断与疾病分层的关键细胞群体。
不仅如此，通过进一步解析该亚群的基因表达谱，本研究还鉴定出另一个标志物基因HPN：利用癌症基因组图谱（The Cancer Genome Atlas, TCGA）的RNA测序数据，该基因区分正常组织与前列腺癌病灶的曲线下面积（Area Under the Curve, AUC）得分为0.930。该发现通过前列腺癌组织芯片（tissue array）的HPN免疫染色（immunostaining）实验得到了进一步验证。格里森评分（Gleason score）大于6的前列腺癌组织中，HPN的表达量显著高于格里森评分等于6的组织。
综上，本研究为解析前列腺癌的瘤内异质性提供了宝贵的研究资源，同时也为前列腺癌的临床管理提供了新型候选标志物。本研究通过对两名患者癌性前列腺组织分离得到的癌细胞进行单细胞RNA测序，实现了对前列腺癌瘤内异质性的系统解析。