Liljedahl Leena, Malmström Johan, Kristjansdottir Björg, Waldemarson Sofia, Sundfeldt Karin
CREATE Health Translational Cancer Center, Department of Immunotechnology, Lund University, Medicon Village, 223 81 Lund, Sweden.
Department of Clinical Sciences, Lund University, BMC D13, 221 84 Lund, Sweden.
Cancers (Basel). 2021 Nov 15;13(22):5713. doi: 10.3390/cancers13225713.
Epithelial ovarian cancer (OC) is a disease with high mortality due to vague early clinical symptoms. Benign ovarian cysts are common and accurate diagnosis remains a challenge because of the molecular heterogeneity of OC. We set out to investigate whether the disease diversity seen in ovarian cyst fluids and tumor tissue could be detected in plasma. Using existing mass spectrometry (MS)-based proteomics data, we constructed a selected reaction monitoring (SRM) assay targeting peptides from 177 cancer-related and classical proteins associated with OC. Plasma from benign, borderline, and malignant ovarian tumors were used to verify expression ( = 74). Unsupervised and supervised multivariate analyses were used for comparisons. The peptide signatures revealed by the supervised multivariate analysis contained 55 to 77 peptides each. The predictive (Q2) values were higher for benign vs. low-grade serous Q2 = 0.615, mucinous Q2 = 0.611, endometrioid Q2 = 0.428 and high-grade serous Q2 = 0.375 (stage I-II Q2 = 0.515; stage III Q2 = 0.43) OC compared to benign vs. all malignant Q2 = 0.226. With targeted SRM MS we constructed a multiplexed assay for simultaneous detection and relative quantification of 185 peptides from 177 proteins in only 20 µL of plasma. With the approach of histology-specific peptide patterns, derived from pre-selected proteins, we may be able to detect not only high-grade serous OC but also the less common OC subtypes.
上皮性卵巢癌（OC）因早期临床症状不明显，死亡率较高。良性卵巢囊肿很常见，由于OC的分子异质性，准确诊断仍然是一项挑战。我们着手研究能否在血浆中检测到卵巢囊肿液和肿瘤组织中存在的疾病多样性。利用现有的基于质谱（MS）的蛋白质组学数据，我们构建了一种选择反应监测（SRM）分析方法，针对与OC相关的177种癌症相关蛋白和经典蛋白的肽段。使用来自良性、交界性和恶性卵巢肿瘤的血浆来验证表达情况（n = 74）。采用无监督和有监督的多变量分析进行比较。有监督多变量分析揭示的肽段特征每组包含55至77个肽段。与良性vs.所有恶性的Q2 = 0.226相比，良性vs.低级别浆液性Q2 = 0.615、黏液性Q2 = 0.611、子宫内膜样Q2 = 0.428和高级别浆液性Q2 = 0.375（I-II期Q2 = 0.515；III期Q2 = 0.43）OC的预测（Q2）值更高。通过靶向SRM MS，我们构建了一种多重分析方法，仅需20微升血浆就能同时检测和相对定量来自177种蛋白质的185个肽段。通过源自预先选择蛋白质的组织学特异性肽段模式方法，我们或许不仅能够检测高级别浆液性OC，还能检测较罕见的OC亚型。
