Multi-omic serum biomarkers for prognosis of disease progression in prostate cancer
Michael A. Kiebish,Jennifer Cullen,Prachi Mishra,Amina Ali,Eric Milliman,Leonardo O. Rodrigues,Emily Y. Chen,Vladimir Tolstikov,Lixia Zhang,Kiki Panagopoulos,Punit Shah,Yongmei Chen,Gyorgy Petrovics,Inger L. Rosner,Isabell A. Sesterhenn,David G. McLeod,Elder Granger,Rangaprasad Sarangarajan,Viatcheslav Akmaev,Alagarsamy Srinivasan,Shiv Srivastava,Niven R. Narain,Albert Dobi
DOI: https://doi.org/10.1186/s12967-019-02185-y
IF: 8.44
2020-01-07
Journal of Translational Medicine
Abstract:Abstract Background Predicting the clinical course of prostate cancer is challenging due to the wide biological spectrum of the disease. The objective of our study was to identify prostate cancer prognostic markers in patients ‘sera using a multi-omics discovery platform. Methods Pre-surgical serum samples collected from a longitudinal, racially diverse, prostate cancer patient cohort (N = 382) were examined. Linear Regression and Bayesian computational approaches integrated with multi-omics, were used to select markers to predict biochemical recurrence (BCR). BCR-free survival was modeled using unadjusted Kaplan–Meier estimation curves and multivariable Cox proportional hazards analysis, adjusted for key pathologic variables. Receiver operating characteristic (ROC) curve statistics were used to examine the predictive value of markers in discriminating BCR events from non-events. The findings were further validated by creating a training set (N = 267) and testing set (N = 115) from the cohort. Results Among 382 patients, 72 (19%) experienced a BCR event in a median follow-up time of 6.9 years. Two proteins—Tenascin C (TNC) and Apolipoprotein A1V (Apo-AIV), one metabolite—1-Methyladenosine (1-MA) and one phospholipid molecular species phosphatidic acid (PA) 18:0-22:0 showed a cumulative predictive performance of AUC = 0.78 [OR (95% CI) = 6.56 (2.98–14.40), P 80%) for BCR. The combination of pTstage and Gleason score with the analytes, further increased the sensitivity [AUC = 0.89, 95% (CI) = 4.45–32.05, P < 0.05], with an increased NPV (0.96) and OR (12.4) for BCR. The panel of markers combined with the pathological parameters demonstrated a more accurate prediction of BCR than the pathological parameters alone in prostate cancer. Conclusions In this study, a panel of serum analytes were identified that complemented pathologic patient features in predicting prostate cancer progression. This panel offers a new opportunity to complement current prognostic markers and to monitor the potential impact of primary treatment versus surveillance on patient oncological outcome.
medicine, research & experimental