DataSheet_1_Mitochondria Dysfunction-Mediated Molecular Subtypes and Gene Prognostic Index for Prostate Cancer Patients Undergoing Radical Prostatectomy or Radiotherapy.docx

BackgroundGiven the age relevance of prostate cancer (PCa) and the role of mitochondrial dysfunction (MIDS) in aging, we orchestrated molecular subtypes and identified key genes for PCa from the perspective of MIDS.
MethodsCluster analysis, COX regression analysis, function analysis, and tumor immune environment were conducted. We performed all analyses using software R 3.6.3 and its suitable packages.
ResultsCXCL14, SFRP4, and CD38 were eventually identified to classify the PCa patients in The Cancer Genome Atlas (TCGA) database and the Gene Expression Omnibus (GEO) dataset into two distinct clusters. Patients in the cluster 2 had shorter BCR-free survival than those in the cluster 1 in terms of both TCGA database and GEO dataset. We divided the patients from the TCGA database and the GEO dataset into high- and low-risk groups according to the median of MIDS-related genetic prognostic index. For patients in the TCGA database, the biochemical recurrence (BCR) risk in high-risk group was 2.34 times higher than that in low-risk group. Similarly, for patients in the GEO dataset, the risk of BCR and metastasis in high-risk group was 2.35 and 3.04 times higher than that in low-risk group, respectively. Cluster 2 was closely associated with advanced T stage and higher Gleason score for patients undergoing radical prostatectomy or radiotherapy. For patients undergoing radical prostatectomy, the number of CD8+ T cells was significantly lower in cluster 2 than in cluster 1, while cluster 2 had significantly higher stromal score than cluster 1. For patients undergoing radical radiotherapy, cluster 2 had significantly higher level of CD8+ T cells, neutrophils, macrophages, dendritic cells, stromal score, immune score, and estimate score, but showed lower level of tumor purity than cluster 1.
ConclusionsWe proposed distinctly prognosis-related molecular subtypes at genetic level and related formula for PCa patients undergoing radical prostatectomy or radiotherapy, mainly to provide a roadmap for precision medicine.

**背景** 鉴于前列腺癌（prostate cancer, PCa）的年龄相关性，以及线粒体功能障碍（mitochondrial dysfunction, MIDS）在衰老进程中的作用，本研究从线粒体功能障碍视角出发，对前列腺癌开展分子分型并筛选其关键基因。
**方法** 本研究采用聚类分析、COX回归分析、功能富集分析及肿瘤免疫微环境分析等方法，所有分析均通过R 3.6.3软件及其适配包完成。
**结果** 最终筛选出CXCL14、SFRP4及CD38作为分型基因，将癌症基因组图谱（The Cancer Genome Atlas, TCGA）数据库与基因表达综合（Gene Expression Omnibus, GEO）数据集内的前列腺癌患者划分为两个独立亚型。在TCGA数据库及GEO数据集中，聚类2组患者的无生化复发生存期均短于聚类1组患者。本研究根据线粒体功能障碍相关基因预后指数的中位数，将TCGA数据库及GEO数据集内的患者划分为高风险组与低风险组。对于TCGA数据库中的患者，高风险组的生化复发（biochemical recurrence, BCR）风险是低风险组的2.34倍。同样地，在GEO数据集中，高风险组的生化复发与转移风险分别为低风险组的2.35倍与3.04倍。对于接受根治性前列腺切除术或放射治疗的患者，聚类2组与更高的T分期及Gleason评分密切相关。在接受根治性前列腺切除术的患者中，聚类2组的CD8+ T细胞数量显著低于聚类1组，而聚类2组的基质评分显著高于聚类1组。对于接受根治性放射治疗的患者，聚类2组的CD8+ T细胞、中性粒细胞、巨噬细胞、树突状细胞数量，以及基质评分、免疫评分与ESTIMATE评分均显著高于聚类1组，而肿瘤纯度显著低于聚类1组。
**结论** 本研究从基因层面提出了与预后显著相关的前列腺癌分子分型模型及对应预后计算公式，可为接受根治性前列腺切除术或放射治疗的前列腺癌患者提供精准医学指导框架。