Introduction: A long-term goal of our group is to develop proteomic-based approaches to the detection and use of protein biomarkers for improvement in diagnosis, prognosis, and tailoring of treatment for head and neck squamous cell cancer (HNSCC). We have previously demonstrated that protein expression profiling of serum can identify multiple protein biomarker events that can serve as molecular fingerprints for the assessment of HNSCC disease state and prognosis.
Methods: An automated Bruker Daltonics (Billerica, MA) ClinProt matrix-assisted laser desorption/ionization time-of-flight (MALDI-TOF) mass spectrometer was used. Magnetic chemical affinity beads were used to differentially capture serum proteins prior to MALDI-TOF analysis. The resulting spectra were analyzed using postprocessing software and a pattern recognition genetic algorithm (ClinProt 2.0). An HNSCC cohort of 48 sera samples from 24 patients consisting of matched pretreatment and 6 to 12 month posttreatment samples was used for further analysis. Low-mass differentially expressed peptides were identified using MALDI-TOF/TOF.
Results: In the working mass range of 1,000 to 10,000 m/z, approximately 200 peaks were resolved for ionic bead capture approaches. For spectra generated from weak cation bead capture, a k-nearest neighbor genetic algorithm was able to correctly classify 94% normal from pretreatment HNSCC samples, 80% of pretreatment from posttreatment samples, and 87% of normal from posttreatment samples. These peptides were then analyzed by MALDI-TOF/TOF mass spectometry for sequence identification directly from serum processed with the same magnetic bead chemistry or alternatively after gel electrophoresis separation of the captured proteins. We were able to compare this with similar studies using surface-enhanced laser desorption ionization (SELDI)-TOF to show this method as a valid tool for this process with some improvement in the identification of our groups.
Conclusions: This initial study using new high-resolution MALDI-TOF mass spectrometry coupled with bead fractionation is suitable for automated protein profiling and has the capability to simultaneously identify potential biomarker proteins for HNSCC. In addition, we were able to show improvement with the MALDI-TOF in identifying groups with HNSCC when compared with our prior data using SELDI-TOF. Using this MALDI-TOF technology as a discovery platform, we anticipate generating biomarker panels for use in more accurate prediction of prognosis and treatment efficacies for HNSCC.