Table3_Identification of Neoantigens and Construction of Immune Subtypes in Prostate Adenocarcinoma.XLSX

Background: Messenger ribonucleic acid (mRNA) vaccine has been considered as a potential therapeutic strategy and the next research hotspot, but their efficacy against prostate adenocarcinoma (PRAD) remains undefined. This study aimed to find potential antigens of PRAD for mRNA vaccine development and identify suitable patients for vaccination through immunophenotyping.
Methods: Gene expression profiles and clinical information were obtained from TCGA and ICGC. GEPIA2 was used to calculate the prognostic index of the selected antigens. The genetic alterations were compared on cBioPortal and the correlation between potential antigen and immune infiltrating cells was explored by TIMER. ConsensusClusterPlus was used to construct a consistency matrix, and identify the immune subtypes. Graph learning-based dimensional reduction was performed to depict immune landscape. Boruta algorithm and LASSO logistic analysis were used to screen PRAD patients who may benefit from mRNA vaccine.
Results: Seven potential tumor antigens selected were significantly positively associated with poor prognosis and the antigen-presenting immune cells (APCs) in PRAD, including ADA, FYN, HDC, NFKBIZ, RASSF4, SLC6A3, and UPP1. Five immune subtypes of PRAD were identified by differential molecular, cellular, and clinical characteristics in both cohorts. C3 and C5 had immune “hot” and immunosuppressive phenotype, On the contrary, C1&C2 had immune “cold” phenotype. Finally, the immune landscape characterization showed the immune heterogeneity among patients with PRAD.
Conclusions: ADA, FYN, HDC, NFKBIZ, RASSF4, SLC6A3, and UPP1 are potential antigens for mRNA vaccine development against PRAD, and patients in type C1 and C2 are suitable for vaccination.

背景：信使核糖核酸疫苗（Messenger ribonucleic acid, mRNA）被视为一种潜在的治疗策略与前沿研究热点，但其针对前列腺腺癌（prostate adenocarcinoma, PRAD）的疗效仍未明确。本研究旨在筛选可用于mRNA疫苗研发的PRAD潜在肿瘤抗原，并通过免疫表型分析确定适合接种疫苗的患者人群。
方法：从癌症基因组图谱（The Cancer Genome Atlas, TCGA）与国际癌症基因组联盟（International Cancer Genome Consortium, ICGC）获取基因表达谱与临床信息；采用GEPIA2计算筛选出的抗原的预后指数；通过cBioPortal平台对比遗传变异情况，并借助肿瘤免疫评估资源（Tumor Immune Estimation Resource, TIMER）探究潜在抗原与免疫浸润细胞之间的相关性；采用ConsensusClusterPlus构建一致性矩阵，以识别PRAD免疫亚型；采用基于图学习的降维分析刻画PRAD免疫图谱；采用Boruta算法与最小绝对收缩和选择算子（Least Absolute Shrinkage and Selection Operator, LASSO）logistic回归分析，筛选可从mRNA疫苗中获益的PRAD患者。
结果：筛选出的7种潜在肿瘤抗原与PRAD患者不良预后及抗原呈递免疫细胞（antigen-presenting immune cells, APCs）显著正相关，具体包括ADA、FYN、HDC、NFKBIZ、RASSF4、SLC6A3及UPP1。在两个研究队列中，均基于分子、细胞及临床特征的差异识别出5种PRAD免疫亚型：亚型C3与C5呈现免疫"热"表型与免疫抑制表型，与之相反，亚型C1与C2呈现免疫"冷"表型。最终，免疫图谱刻画结果揭示了PRAD患者群体内部的免疫异质性。
结论：ADA、FYN、HDC、NFKBIZ、RASSF4、SLC6A3及UPP1是可用于抗PRAD的mRNA疫苗研发的潜在肿瘤抗原，且亚型C1与C2的患者适合接种该类疫苗。