We present a three-step method to predict prostate cancer (PCa) regions on biopsy tissue samples based on high-confidence, low-resolution PCa regions marked by a pathologist. First, we will apply a texture-analysis technique on a high-magnification optical image to predict PCa regions on an adjacent tissue slice. Second, we will design a prediction model for the same purpose, using matrix-assisted laser desorption/ionization mass spectrometry (MALDI-MS) tissue-imaging data from the adjacent slice. Finally, we will fuse those two results to obtain the PCa regions that will assist MALDI imaging biomarker identification. Experiment results show that the texture analysis–based prediction is sensitive (87.45%) but less specific (75%), and the prediction based on the MALDI spectra data processing is not sensitive (50.98%) but supremely specific (100%). By combining these two results, an optimized prediction for PCa regions on the adjacent slice can be achieved (sensitivity: 80.39%, specificity: 93.09%).
JemalASiegelRWardE. Cancer statistics, 2008. CA Cancer J Clin2008; 58:71–96.
2.
U.S.Department of Health and Human Services, National Institutes of Health (NIH), National Cancer Institute (NCI). Fact sheet on proteomics and cancer. August2009.
3.
CatalonaWJSmithDSRatliffTLDoddsKM. Measurement of prostate-specific antigen in serum as a screening test for prostate cancer. N. Engl. J. Med. 1991; 324: 1156–1161.
4.
DrakeRRCazaresLSemmesOJ. Mining the low molecular weight proteome of blood. Proteomics Clin. Appl., 2007; 1: 758–768.
5.
LiJManryMTNarasimhaPL. Feature selection using a piecewise linear network. IEEE Transaction on Networks2006; 17: 1101–1115 (2006).
6.
LiJManryMTLiuLM. Iterative improvement of neural classifiers. FLAIRS2004; 700–705.
7.
Wun-Jae KimSPKimYJ. Biomarkers in bladder cancer: present status and perspectives. Biomarker Insights2007; 2: 95–105.
8.
SemmesOJFengZAdamB. Evaluation of serum protein profiling by surface-enhanced laser desorption/ionization time-of-flight mass spectrometry for the detection of prostate cancer. Clin. Chem. 2005; 51: 102–112.
9.
Van ’t VeerLJDaiHVan de VijverMJ. Gene expression profiling predicts clinical outcome of breast cancer. Nature2002; 415: 530–536.
10.
PritzkerKP. Cancer biomarkers: easier said than done, Clin. Chem. 2002; 48: 1147–1150.
11.
DuffyMJEstevaFJHarbeckN. National Academy of Clinical Biochemistry guidelines for the use of tumor markers in breast cancer. NACB: Practice Guidelines and Recommendations for Use of Tumor Markers in the Clinic, 2006.
12.
GastMCSchellensJHBeijnenJH. Clinical proteomics in breast cancer: a review. Breast Cancer Res Treat. 2009; 116:17–29.
13.
CaprioliRMFarmerTBGileJ. Molecular imaging of biological samples: localization of peptides and proteins using MALDI-TOF MS. Anal. Chem. 1997; 69: 4751–4760.
14.
ChaurandPCornettDSCaprioliRM. Molecular imaging of thin mammalian tissue sections by mass spectrometry. Curr. Opin. Biotechnol. 2006; 17: 431–436.
15.
CaldwellRLCaprioliRM. Tissue profiling by mass spectrometry: a review of methodology and applicationsMol. Cell. Proteomics2005; 4: 394–401.
16.
ChaurandPSandersMEJensenRA. Proteomics in diagnostic pathology: profiling and imaging proteins directly in tissue sections. Am. J. Pathol. 2004; 165: 1057–1068.
17.
TabeshATeverovskiyMPangH. Multifeature prostate cancer diagnosis and gleason grading of histological images. IEEE Trans. on Medical Imaging2007; 26: 1366–1378.
18.
HuangPHsiungC. Automatic classification for pathological prostate images based on fractal analysis. IEEE Trans. on Medical Imaging2009; 28: 1037–1050.
19.
TahirMABouridaneAKurugolluF, An FPGA-based coprocessor for GLCM and Haralick texture features and their application in prostate cancer classification. Analog Integrated Circuits Signal Process2005; 43: 205–215.
20.
GallowayMM. Texture analysis using gray level run lengths. Computer Graphics and Image Processing1975; 4: 172–179.
21.
ChaurandPStoeckliMCaprioliRM. Direct profiling of proteins in biological tissue sections by MALDI mass spectrometry. Anal. Chem. 1999; 71: 5263–5270.
22.
CaldwellRLCaprioliRM. Tissue profiling by mass spectrometry: a review of methodology and applications. Mol. Cell. Proteomics2005; 4: 394–401.
23.
RauschM. MSI additions by Markus Stoeckli.software – BioMap: www.maldi.msi.orgAccessed: January2008; Novartis.
24.
VapnikV. The Nature of Statistical Learning Theory. Berlin: Springer-Verlag, 1995.
25.
MantenaVJiangWLiJ. Prostate cancer biomarker identification using MALDI-MS data: initial results. IEEE/NIH LiSSA2009; 116–119.
26.
MyersSRAliY. Determination of tobacco specific hemoglobin adducts in smoking mothers and new born babies by mass spectrometry. Biomarker Insights2007; 2: 269–282.
27.
OldWMMeyer-ArendtKAveline-WolfLPierceKG. Comparison of label-free methods for quantifying human proteins by shotgun proteomics. Mol. Cell. Proteo-mics2005; 4: 1487–1502.
28.
JordanKWHeWHalpernEF. Evaluation of tissue metabolites with high-resolution magic angle spinning MR spectroscopy human prostate samples after three-year storage at −80ºC. Biomarker Insights2007; 2: 147–154.
29.
Biomarkers Definitions Working Group. Biomarkers and surrogate endpoints: preferred definitions and conceptual framework. Clin. Pharmacol. Ther. 2001; 69: 89–95.
30.
GewinV. Missing the mark. Nature2007; 449: 770–771.
31.
McLerranDGrizzleWESemmesOJ. Analytical validation of serum proteomic profiling for diagnosis of prostate cancer: sources of sample bias. Clin. Chem. 2008; 54: 44–52, 2008.
32.
McLerranDGrizzleWEFengZ. SELDI-TOF MS whole serum proteomic profiling with IMAC surface does not reliably detect prostate cancer. Clin. Chem. 2008; 54: 53–60.
33.
TracyMBChenHWeaverDM. Precision enhancement of MALDI-TOF MS using high-resolution peak detection and label-free alignment. Proteomics2008; 8: 1530–1538.
34.
LiJVadlamudiAChuangS. Combining prostate cancer region predictions from MALDI spectra processing and texture analysis. IEEE International Conference on Bioinformatics and Bioengineering (BIBE), Philadelphia, PA, May 31-June 3, 2010: 73–78.
