Deep plasma proteomics identifies and validates an eight-protein biomarker panel that separate benign from malignant tumors in ovarian cancer

Ovarian cancer has the highest mortality of all gynecological cancers and in symptomatic women, surgery is commonly used as final diagnostic. Available literature indicates that women with benign tumors could often be conservatively managed but accurate molecular tests are needed for triaging where gold-standard imaging techniques are inconclusive or lacking. Here, we analyzed 5416 plasma proteins in two independent cohorts (N=171+233) with symptomatic women that have been surgically diagnosed with benign or malignant tumors. Using one cohort as discovery, we compared protein levels of benign tumors with early stage (I-II), late stage (III-IV) or any stage (I-IV) ovarian cancer. In this analysis, 327 associations, corresponding to 191 unique proteins, were identified out of which 326 (99.7%) replicated. The 191 proteins were compared with their corresponding tumor gene expression in the replication cohort and only 11% (21/191) were found to have significant correlation. Protein-protein correlation networks were generated and 62 of the 191 proteins were highly correlated with at least one other protein, suggesting that many of the observed associations could be secondary effects. Multivariate models were trained using the discovery cohort including a fixed cut-off for malignancy. In the replication cohort, an eight-protein model achieved an AUC of 0.96 corresponding to 97% sensitivity at 68% specificity. For early-stage tumors, the sensitivity was estimated at 91% at 68% specificity compared to 85% and 54% for CA-125 alone. Our results indicate that up to one third of benign cases could be identified by molecular measures thereby reducing the need for diagnostic surgery.