PURPOSE: We have investigated the use of human urine as a non-invasive medium to screen for molecular biomarkers of carcinomas of the upper gastrointestinal (uGI) tract using SELDI-TOF-MS. EXPERIMENTAL DESIGN: A total of 120 urine specimens from 60 control and 60 uGI cancer patients were analysed to establish a potential biomarker fingerprint for the weak cation exchanger CM10 chip surface, which was validated by blind testing using a further 59 samples from 33 control and 26 uGI cancer patients. RESULTS: Using Biomarker Pattern software, we established a model with a sensitivity of 98% and specificity of 95% for the learning sample set, and a sensitivity of 96% and specificity of 72% for the validation data set. Model variable importance included six peptides with m/z of 10,230, 10,436, 10,574, 10,311, 10,467, and 10,118 of which the 10, 230 molecular species was the main decider (sensitivity 86% and specificity 80%). Initial protein database searching identified 10,230 as S100-A6, 10,436 as S100-P, 10,467 as S100-A9, and 10,574 as S100-A12 of which S100-A6 and S100-A9 were confirmed by Western blotting. CONCLUSIONS AND CLINICAL RELEVANCE: We have demonstrated that SELDI-TOF-MS as a screening tool is a rapid and valid methodology in the search for urinary cancer biomarkers, and is potentially useful in defining and consolidating biomarker patterns for uGI cancer screening.
PURPOSE: We have investigated the use of human urine as a non-invasive medium to screen for molecular biomarkers of carcinomas of the upper gastrointestinal (uGI) tract using SELDI-TOF-MS. EXPERIMENTAL DESIGN: A total of 120 urine specimens from 60 control and 60 uGI cancerpatients were analysed to establish a potential biomarker fingerprint for the weak cation exchanger CM10 chip surface, which was validated by blind testing using a further 59 samples from 33 control and 26 uGI cancerpatients. RESULTS: Using Biomarker Pattern software, we established a model with a sensitivity of 98% and specificity of 95% for the learning sample set, and a sensitivity of 96% and specificity of 72% for the validation data set. Model variable importance included six peptides with m/z of 10,230, 10,436, 10,574, 10,311, 10,467, and 10,118 of which the 10, 230 molecular species was the main decider (sensitivity 86% and specificity 80%). Initial protein database searching identified 10,230 as S100-A6, 10,436 as S100-P, 10,467 as S100-A9, and 10,574 as S100-A12 of which S100-A6 and S100-A9 were confirmed by Western blotting. CONCLUSIONS AND CLINICAL RELEVANCE: We have demonstrated that SELDI-TOF-MS as a screening tool is a rapid and valid methodology in the search for urinary cancer biomarkers, and is potentially useful in defining and consolidating biomarker patterns for uGI cancer screening.
Authors: Holger Husi; Alisdair MacDonald; Richard J E Skipworth; Janice Miller; Andrew Cronshaw; Kenneth C H Fearon; James A Ross Journal: Biomed Rep Date: 2018-04-25
Authors: Holger Husi; Alisdair MacDonald; Richard J E Skipworth; Janice Miller; Andrew Cronshaw; Carolyn Greig; Kenneth C H Fearon; James A Ross Journal: Biomed Rep Date: 2018-04-25
Authors: Julia Beretov; Valerie C Wasinger; Ewan K A Millar; Peter Schwartz; Peter H Graham; Yong Li Journal: PLoS One Date: 2015-11-06 Impact factor: 3.240
Authors: Holger Husi; Janice B Barr; Richard J E Skipworth; Nathan A Stephens; Carolyn A Greig; Henning Wackerhage; Rona Barron; Kenneth C H Fearon; James A Ross Journal: Int J Proteomics Date: 2013-10-09