More reading for the more curious minded! I apologize for the numerous posts.
Extracted the below from 66th American Association of Clinical Chemistry. (2016)
Development and Initial Evaluation of a Multi-Protein Biomarker Blood Test for Organ Confined Prostate Cancer Diagnosis (OCProDx) C. Rooney 1 , Y. Fan 1 , R. Inzitari 1 , B. Hernandez 1 , A. Parnell 1 , P. J. Twomey 2 , S. R. Pennington 1 . 1
University College Dublin, Dublin, Ireland, 2 St. Vincent’s University Hospital, Dublin, Ireland Background: About one in six men will get a diagnosis of prostate cancer during their lives. Generally, prostate cancer is treated effectively, but for many men the disease is not life threatening and they will die with prostate cancer rather than because of it. Too many men are treated unnecessarily. For them active surveillance of the disease would be a better option. Unfortunately, the existing readily available tools for disease diagnosis (PSA test, digital rectal examination and trans-rectal ultrasound guided biopsy), do not adequately guide this key decision of whether to pursue active surveillance or invasive treatment.
Through analysis of the key decisions in prostate cancer patient management we highlighted that establishing whether the disease is organ confined (localized, OC) or has spread beyond the extracellular capsule of the organ (non-organ confined, NOC) would provide important information to guide this decision [Oon SF, Pennington SR, Fitzpatrick JM, Watson RW. Nature Reviews Urololgy (2011) 8:131-8.]. Our objective was to identify serum protein biomarkers to determine disease status in terms of organ confinement. Methods: We undertook unbiased protein discovery experiments using gel and LC-MS based proteomics. Discovery was undertaken with affinity depleted (MARS14) serum samples (n>50 for gel and n=30 for LC-MS) taken from patients at time of diagnosis and for whom OC or NOC status was determined following radical prostatectomy.
Statistical analysis of differentially expressed proteins was undertaken at univariate (Student t-test) and multivariate levels to assemble a panel of 59 candidate proteins. We supplemented this panel of 59 proteins with 5 proteins identified from the literature and developed a multiplexed MRM assay to support the simultaneous measurement of 63 of the proteins. The protein panel was evaluated its by undertaking two initial validation studies in which first 31/63 and then 63/64 of the candidate proteins were measured using patient samples distinct from those used for the discovery experiments.
Serum samples were from the Irish Prostate Cancer Research Consortium.
Results: Initially, the relative abundance of the highest MRM transition from 50 peptides was used to measure 31 proteins in 63 clinical samples. The data, extracted using Skyline, were fitted into a PLS-DA model and the predicted performance was assessed through 200 times bootstrapping. The predictions in the out-of-bag samples were compared with the true group information and ROC curves were generated. The AUC for differentiating between OC and NOC was 0.824.
Subsequently, 63 candidate proteins were evaluated with total of 116 patient samples and data analysed using a range of different statistical approaches. The AUC values for distinguishing organ confined from non-organ confined disease were >0.8. It was notable that proteins within the second phase of MRM development (n=32) made a contribution to these AUC values.
Conclusions: This initial evaluation data clearly demonstrates the potential of the 63 protein multiplexed MRM assay to discriminate OC from NOC prostate cancer. With incorporation of appropriate QC methods we suggest the OCProDx MRM assay may be capable of translation to diagnostic use to support the discrimination between OC and NOC prostate cancer.
Source:
https://play.google.com/store/books...ead&pcampaignid=books_booksearch_viewport
(20min delay)