Predicting bladder cancer behavior by molecular expression profiling

Created by W.Langdon from gp-bibliography.bib Revision:1.8051

@PhdThesis{Mitra:thesis,

• author = "Anirban Pradip Mitra",
• title = "Predicting bladder cancer behavior by molecular expression profiling",
• school = "Keck School of Medicine, University of Southern California",
• year = "2009",
• type = "Pathobiology",
• address = "Los Angeles, California, USA",
• month = aug,
• keywords = "genetic algorithms, genetic programming, urothelial carcinoma, prognosis, quantitative expression profiling, microarray, immunohistochemistry",
• URL = "http://digitallibrary.usc.edu/cdm/ref/collection/p15799coll127/id/247267",
• size = "223 pages",
• abstract = "Urothelial carcinoma of the urinary bladder is the seventh most common type of cancer worldwide. In the western world, cigarette smoke is the most commonly implicated carcinogen for this disease. Bladder cancer presents itself as two prognostic variants -- the more common noninvasive Ta tumours that frequently recur but rarely invade the basement membrane, and the less common invasive tumors that tend to progress and metastasize. Traditional prognostic metrics, including tumor and nodal stage, are currently the best clinical predictors of subsequent behavior. While lymph node metastasis forebodes a poor prognosis, early detection can allow for radical lymphadenectomy with a curative intent. This manuscript begins by describing a study that used gene expression profiles generated from primary bladder tumours to construct signatures that could identify nodal metastasis. Genetic programming was used to identify classifiers that showed a strong predilection for ICAM1, MAP2K6 and KDR, and could detect nodal metastasis with reasonable sensitivity and specificity. Using similar pathway-based profiling approaches, this manuscript further describes studies that sought to determine if such molecular alterations could supplement traditional pathologic staging to better predict clinical outcome. The manuscript documents the identification and validation of a concise, biologically relevant gene panel comprising of JUN, MAP2K6, STAT3, and ICAM1 that could predict recurrence and survival in bladder cancer. Another study highlights attempts to identify genes profiled from primary noninvasive Ta tumours at first presentation that could predict local recurrence and tumour progression. The final study describes efforts to semi-quantitatively profile expressions of select proteins from primary bladder cancer tissues to analyse associations of their alterations with cigarette smoking, nonsteroidal anti-inflammatory drug use, and clinical outcome across all disease stages in a population-based cohort. These studies underscore the concept that a pathway-specific approach to profiling relevant biomolecules in bladder cancer can identify markers of prognostic significance, and patients who will recur and/or progress despite definitive surgery alone. Such identification of specific molecular alterations in individual tumours will allow for a more accurate and personalized prediction of prognosis, and also identify potential therapeutic targets.",
• notes = "Unrestricted",

}

Genetic Programming entries for Anirban P Mitra

Citations