Study suggests testing certain protein combinations can enhance sensitivity of fecal immunochemical test used in colorectal cancer screenings

1. Testing for certain combinations of 4 proteins was found to have a sensitivity for detecting colorectal cancer (CRC) and advanced adenomas higher than that of testing for hemoglobin alone.

2. Identification of such proteins using antibody-based assays and small sample volumes suggests the potential for these protein biomarkers to be incorporated into common CRC screening programs.

Evidence Rating Level: 4 (Below Average)      

Study Rundown: CRC screening aims to decrease mortality through early detection at a treatable stage.  The gold standard for CRC screening is colonoscopy, but cost and patient compliance make non-invasive methods common use as triage to colonoscopy.  One such method is the fecal immunochemical test (FIT), which tests for hemoglobin in the stool.  However, there are concerns about the suboptimal sensitivity of FIT for detecting CRC, especially in cases of advanced adenoma.  Although studies have shown that FIT used in combination with multitarget stool DNA testing has higher sensitivity than using only FIT, this method is cost-prohibitive.  The aim of this case-control study was to find protein biomarkers present in stool samples that would either complement or perform better than hemoglobin for detection of CRC and advanced adenoma.  Researchers found that combinations of 4 proteins allowed them to reach a sensitivity of 80% for detecting CRC and 45% for detecting advanced adenoma. At a specificity of 95%, these sensitivities are better than those testing solely for hemoglobin.  The study serves as proof of concept that detection of these proteins in small FIT samples suggests the potential for these protein biomarkers to be incorporated into common CRC screening programs.

A strength of the study is that a protein-based test allows for the use of a small sample device that is easily sent through the mail and for testing to be done on small sample volumes.  Such a test is also less expensive than the multitarget stool DNA test.  A limitation of the study is that some of the best protein combinations found in FIT samples could not be validated due to the unavailability of certain antibodies.

Click to read the study in Annals of Internal Medicine

Relevant Reading: Multitarget stool DNA testing for colorectal-cancer screening

In-Depth [case-control study]: In this study, researchers recruited patients from a colonoscopy-controlled referral population from multiple centers. A total of 315 stool samples were taken in two series.  In the first series, there were 12 patients with CRC and control samples from 10 people without colorectal neoplasia.  In the second series, there were 81, 40, and 43 patients with CRC, advanced adenomas, and nonadvanced adenomas, respectively, in addition to 129 controls.  There was also a third series of FIT samples from 14, 16, and 18 patients with CRC, advanced adenomas, and nonadvanced adenomas, respectively, in addition to 24 controls.  Using mass spectrometry, logistic regression analysis, classification and regression tree (CART) analysis, researchers identified protein combinations that differed in CRC or advanced adenoma samples in comparison to control samples.  A combination of 4 proteins was found to have higher discriminatory power than ones using 2 or 3 proteins.  Based on areas under the curve (AUCs), the panel that performed best was comprised of C3, lactotransferrin, hemoglobin (HBA1), and HP.  Of the biomarker panels that performed in the top 10, 17 proteins were included.  Amongst these 17 proteins, 11 differentiated CRC from controls (azurocidin 1 [AZU1], C3, C5, CDA, FN1, HBA1, HBB, HP, LTF, MPO, and RBP4), 8 differentiated advanced neoplasia from controls (C3, glutathione-disulfide reductase [GSR], HBB, HP, S100A8, S100A9, SERPINF2, and transferrin [TF]), and 5 differentiated advanced adenoma from controls (AZU1, HPX, LTF, MPO, and SERPINF2).  All 10 panels had a much higher AUC (0.93 to 0.94) than HBA1 alone (0.88).  At 95% specificity, all panels had a statistically significantly higher sensitivity (65% to 80%) compared to HBA1 alone (43%).

Image: CC/Wiki

©2017 2 Minute Medicine, Inc. All rights reserved. No works may be reproduced without expressed written consent from 2 Minute Medicine, Inc. Inquire about licensing here. No article should be construed as medical advice and is not intended as such by the authors or by 2 Minute Medicine, Inc.