TY - JOUR
T1 - Integrating biomedical knowledge to model pathways of prostate cancer progression
AU - Morris, David S.
AU - Tomlins, Scott A.
AU - Rhodes, Daniel R.
AU - Mehra, Rohit
AU - Shah, Rajal B.
AU - Chinnaiyan, Arul M.
N1 - Funding Information:
In order to analyze the contamination of stroma in grossly dissected tissue, the results of a previous meta-analysis were exam- ined in the context of profiling signatures obtained from LCM samples of pure cancer epithelia. Although many of the same overexpressed genes were present in both signatures from grossly dissected and LCM tissues, the majority of under-expressed genes identified in the meta-analysis were not under-expressed in LCM isolated epithelia. This was consistent with the proposal of stromal masking due to a decreasing percentage of stromal tissue compared with epithelial elements in higher grade cancer. For example, MME, a gene known to be lost during PCA progression and thought to function as a tumor suppressor,88 was masked by stromal signatures in progression models based on grossly dissected tissue. MME ranks 8th in our “underexpressed in progression” signature based on LCM while ranking no higher than 639th in three studies using grossly dissected tissue.89-91To further evaluate stromal masking, we gener- ated a “stromal signature” consisting of 2,152 features overexpressed (q < 0.05) in stroma compared with epithelia by analyzing twelve stromal and eighty-nine epithelial samples obtained by LCM. When the elements underexpressed in the meta-analysis were referenced with this “stromal signature,” thirty-five of the top forty features (87.5%) underexpressed during progression were significantly overexpressed (p < 0.05) in the stromal signature, suggesting the downregulation of genes during progression in the meta-analysis was attributable to a decline in stromal features. We then applied this “stromal signature” to previous profiling studies to determine the extent of stromal bias (Fig. 3). Similar to our meta-analysis, genes underexpressed in progression signatures from previous studies using grossly dissected tissue were enriched in stromal features.92-94 Furthermore, signatures related to Gleason Acknowledgements grade,PSA recurrence,and outcomes in non-prostate 95-9798-102 Supported in part by Department of Defense (DAMD17-03-profiling studies such as breast cancer103-107 were also significantly 2-0033 to A.M.C., PC040517 to R.M., W81XWH-06-1-0224 to enriched in stromal concepts. While some studies, such as Glinsky A.M.C., D.M.), the American Cancer Society (RSG-02-179-MGO et al., showed enrichment of genes underexpressed in high Gleason to A.M.C.), the National Institutes of Health (U54 DA021519- grade cancer in the stromal signature, other studies, such as Lapointe 01A1 to A.M.C., Prostate SPORE P50CA69568 to A.M.C. and et al., showed enrichment of overexpressed genes from stromal R.B.S.), the Early Detection Research Network (UO1 CA111275- ©2007 LANDES BIOSCIENCE01 to A.M.C.), the Department of Defense (PC040517 to R.M., samples with varying amounts of stroma, possibly accounting for PC020322 to A.M.C.). and the Cancer Center Bioinformatics some of the low concordance between studies. In addition, genes Core (Support Grant 5P30 CA46592). S.A.T. is supported by over-expressed in the signatures of patients without PSA recurrence a Rackham Predoctoral Fellowship. A.M.C. is supported by a also showed significant enrichment in the stromal signature, likely Clinical Translational Research Award from the Burroughs Welcome due to indolent tumors of lower grade containing a relatively high Foundation and S.A.T. and D.R.R. are Fellows of the Medical amount of stroma when compared with high Gleason grade tumors. Scientist Training Program. These interaction networks suggest the stromal signature may serve as a marker of features predictive of positive outcome, such as low grade or small tumor volume.
PY - 2007/5/15
Y1 - 2007/5/15
N2 - Due to pathologic, histologic, and biologic variation within prostate cancers, profiling the genetic changes associated with disease progression has been difficult. Although initial integration of data from profiling studies had been limited by platform variation, bioinformatic tools and analytic techniques have enabled integrative analysis of profiling studies and the identification of more robust and valid profiles. The identification of key transition points in the progression of prostate cancer relies on profiling precursor lesions and "pure" cell populations. Utilizing laser-capture microdissection to isolate 101 cell populations, a more specific genetic profile of progression from benign epithelium to metastatic disease was obtained. This laser-capture profile was analyzed in the context of the molecular concepts map (MCM), a compendium of over 20,000 molecular concepts including other expression profiles of prostate cancer, to obtain an integrative molecular model of progression. The conceptual connections associated with progression confirm that prostate cancer biology is largely driven by pathways related to androgen signaling and epithelial cell biology; however, further analysis of concepts associated with progression suggests stromal factors are highly associated with progression of prostate cancer. The effect of stromal signatures on the progression model suggests the impact of stromal signature downregulation may reflect both a change in the epithelia:stroma ratio within higher grade tumors and also a microenvironment influence on prostate epithelia. Analyzing complex gene expression signatures in the context of molecular concepts improves integrative models and may improve detection, prognostication, or targeted therapy.
AB - Due to pathologic, histologic, and biologic variation within prostate cancers, profiling the genetic changes associated with disease progression has been difficult. Although initial integration of data from profiling studies had been limited by platform variation, bioinformatic tools and analytic techniques have enabled integrative analysis of profiling studies and the identification of more robust and valid profiles. The identification of key transition points in the progression of prostate cancer relies on profiling precursor lesions and "pure" cell populations. Utilizing laser-capture microdissection to isolate 101 cell populations, a more specific genetic profile of progression from benign epithelium to metastatic disease was obtained. This laser-capture profile was analyzed in the context of the molecular concepts map (MCM), a compendium of over 20,000 molecular concepts including other expression profiles of prostate cancer, to obtain an integrative molecular model of progression. The conceptual connections associated with progression confirm that prostate cancer biology is largely driven by pathways related to androgen signaling and epithelial cell biology; however, further analysis of concepts associated with progression suggests stromal factors are highly associated with progression of prostate cancer. The effect of stromal signatures on the progression model suggests the impact of stromal signature downregulation may reflect both a change in the epithelia:stroma ratio within higher grade tumors and also a microenvironment influence on prostate epithelia. Analyzing complex gene expression signatures in the context of molecular concepts improves integrative models and may improve detection, prognostication, or targeted therapy.
KW - Expression profiling
KW - Integrative analysis
KW - Laser capture microdissection
KW - Prostate cancer progression
UR - http://www.scopus.com/inward/record.url?scp=34548337033&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=34548337033&partnerID=8YFLogxK
U2 - 10.4161/cc.6.10.4247
DO - 10.4161/cc.6.10.4247
M3 - Review article
C2 - 17495538
AN - SCOPUS:34548337033
SN - 1538-4101
VL - 6
SP - 1177
EP - 1187
JO - Cell Cycle
JF - Cell Cycle
IS - 10
ER -