Case-controlled identification of colorectal cancer based on proteomic profiles and the potential for screening
| Autor: | Wilma E. Mesker, A.M. Deelder, R.A.E.M. Tollenaar, Marco R. Bladergroen, Y.E.M. van der Burgt, A. Huijbers, Bart Mertens |
|---|---|
| Jazyk: | angličtina |
| Rok vydání: | 2014 |
| Předmět: |
Oncology
Adult Male MALDI-TOF Pathology medicine.medical_specialty Colorectal cancer Proteomics Logistic regression Sensitivity and Specificity proteomics Discriminative model Internal medicine medicine Biomarkers Tumor Humans early detection Early Detection of Cancer Aged Aged 80 and over Crc screening business.industry screening Gastroenterology Area under the curve Reproducibility of Results Middle Aged Serum samples medicine.disease Logistic Models Case-Control Studies Spectrometry Mass Matrix-Assisted Laser Desorption-Ionization Biomarker (medicine) Female business Colorectal Neoplasms |
| Zdroj: | Colorectal Disease, 16(11), 907-913 |
| Popis: | Aim Colorectal cancer (CRC) screening programmes detect early cancers but unfortunately have limited sensitivity and specificity. Mass spectrometry-based determination of serum peptide and protein profiles provides a new approach for improved screening. Method Serum samples were obtained from 126 CRC patients before treatment and 277 control individuals. An additional group of samples from 50 CRC patients and 82 controls was used for validation. Peptide and protein enrichments were carried out using reverse-phase C18 and weak-cation exchange magnetic beads in an automated solid-phase extraction and spotting procedure. Profiles were acquired on a matrix-assisted laser desorption/ionization time-of-flight system. Discriminant rules using logistic regression were calibrated for the peptide and protein signatures separately, followed by combining the classifications to obtain double cross-validated predicted class probabilities. Results were validated on an identical patient set. Results A discriminative power was found for patients with CRC representative for all histopathological stages compared with controls with an area under the curve of 0.95 in the test set (0.93 for the validation set) and with a high specificity (94–95%). Conclusion The study has shown that a serum peptide and protein biomarker signature can be used to distinguish CRC patients from healthy controls with high discriminative power. This relatively simple and cheap test is promising for CRC screening. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|