Response and toxicity prediction by MALDI-TOF-MS serum peptide profiling in patients with non-small cell lung cancer

2016 ◽  
Vol 10 (7) ◽  
pp. 743-749 ◽  
Author(s):  
Maria Rovithi ◽  
Joline S. W. Lind ◽  
Thang V. Pham ◽  
Johannes Voortman ◽  
Jaco C. Knol ◽  
...  
2012 ◽  
Vol 30 (15_suppl) ◽  
pp. e18094-e18094
Author(s):  
Maria Rovithi ◽  
Joline S. Lind ◽  
Thang V Pham ◽  
Jaco C Knol ◽  
Henk M.W. Verheul ◽  
...  

e18094 Background: Mass spectrometry can be used to generate diagnostic peptide peak profiles "signatures" of serum samples1. Peak profiles can be used to compare different sera and correlate samples (i.e., patient groups) with clinical data to assist in diagnosis, monitoring, and/or prediction. Indeed, in recent studies, we and other researchers have successfully combined serum peptide profiling by mass spectrometry (MS) with bioinformatics and have established distinctive serum polypeptide MS patterns that correlate with cancer types and clinically relevant outcomes2. We performed serum peptide profiling of patients with advanced stage non-small cell lung cancer (NSCLC) treated with erlotinib and sorafenib in a previously reported clinical trial at baseline, after one week of treatment, and after three weeks of treatment, to establish treatment efficacy signatures. Methods: Using automated magnetic C18 bead-assisted serum peptide capture coupled to matrix-assisted laser desorption/ ionization time of flight mass spectrometry (MALDI-TOF MS), serum peptide profiling1 of 50 NSCLC patients was conducted and peptide mass profiles (spectra) obtained. Data analyses of pretreatment serum peptide profiles, as well as dynamic changes in peptide abundance during treatment, were performed to establish support-vector machine-based algorithms that can predict treatment efficacy. Results: A 13 ion-peptide signature could discriminate with a sensitivity of 93% and specificity of 71% a training group of patients with short progression free survival (n=14) and long progression free survival (n=14). The signature shows discriminative power for the remaining non-overlapping set of 22 patients, as well as for the complete dataset of 50 patients. Pattern analysis of overall responses and toxicity is ongoing. Conclusions: Serum peptidome profiling using MALDI-TOF-MS coupled to pattern diagnostics may provide biomarkers predictive of response to targeted treatment in patients with NSCLC, thus enabling pretreatment selection of appropriate subgroups.


PLoS ONE ◽  
2014 ◽  
Vol 9 (6) ◽  
pp. e100566 ◽  
Author(s):  
James L. Sherwood ◽  
Susanne Müller ◽  
Maria C. M. Orr ◽  
Marianne J. Ratcliffe ◽  
Jill Walker

Sign in / Sign up

Export Citation Format

Share Document