Kenneth Boucher, Limited Term Instructor
Oncological Sciences

Methodological Research

My current research interests involve a diverse collection of problems in the areas of mathematical and computer modeling of biological systems with focus on carcinogenesis. I am also interested in the statistical analysis of genetic data, particularly from large databases.

Current Projects

1. Biologically-based mathematical models of carcinogenesis and associated statistical methods for carcinogenic risk assessment. The focus of this research is on quantitative insights into the unobservable processes of initiation and promotion using available experimental and clinical data on tumor induction. In particular, we have applied these methods data on adenomas induced by exposing the lungs of mice to urethane. This work is being done in collaboration with department members. Dr. A.Y. Yakovlev and Dr. A.D. Tsodikov, and also Dr. H. Moolgavkar and Dr. G. Luebeck at the Fred Hutchinson Cancer Research Center.

1. The role of random effects in the phenomenon of spontaneous regression of tumors. The phenomenon of spontaneous regression of benign and malignant tumors is well documented in the literature and is commonly attributed to the induction of apoptosis or activation of the immune system. We attempt at evaluating the role of random effect in this phenomenon. To this end, we consider a stochastic model of tumor growth which is descriptive of the fact that tumors are inherently prone to spontaneous regression due to the random nature of their development. The model describes a population of actively proliferating cells which may give rise to differentiated cells. The process of cell differentiation is irreversible and terminates in cell death. We formulate the model in terms of temporally non-homogeneous Markov branching processes with two types of cells so that the expected total number of neoplastic cells is consistent with the observed mean growth kinetics. In other words, a tumor is modeled as a stochastic system and we explore the properties of such a system which are coherent with the phenomenon of spontaneous tumor regression. This work is being done in collaboration with Dr. A.Y. Yakovlev and Dr. J. DiSario.
2. Application of Markov chain Monte Carlo techniques to estimate gene prevalence from genealogical data. Probabilistic models which incorporate genetic inheritance laws provide mathematical tools from which prevalence of hypothetical mutant genes can be calculated. Linkage of the Utah Population Database with the Utah Cancer registry provides a wealth of data for application of these models. In this study statistical sampling methods are being developed to provide model based estimates of the prevalence and penetrance of hypothesized cancer genes from linked data. This work is being done in collaboration with Dr. Richard Kerber.
3. Analysis of National Marrow Donor Program (NMDP) Data. Currently HLA phenotype data is available for 2.6 million donors in the NMDP registry. Goals of this research include development of effective donor recruitment strategies in order to optimize access of bone marrow transplantation to patients of all races. The level of matching which is required (serological or DNA, for example) is rapidly changing, and this may have a profound influence on the optimal size of the registry. Secondary goals include evaluation of genetic diversity within and between races, which is being explored, for example, by analysis of genetic linkage disequilibrium. Both these goals require development of valid estimation techniques which are computationally efficient enough for application to large data sets. This collaborative project involves Dr. Motomi Mori and Dr. Patrick G. Beatty of the University of Utah, and Dr. Edgar Milford of Brigham and Women?s Hospital.
4. Application of spline regression methods to epidemiological data. Spline based methods have been proposed as an alternative to standard regression analysis of epidemiological data. We are evaluating the effectiveness of splines as compared to a more standard analysis of actual data from a case-control study of colon cancer. This research is being done in collaboration with Dr. M. Slattery.

1. Mori M, Beatty PG, Graves M, Boucher KM and Milford EL (1997) HLA Gene and Haplotype Frequencies in the North American Population: The National Marrow Donor Program Registry. Transplantation 64:1017-1027
2. Boucher KM and Yakovlev AY (1997) Estimating the probability of initiated cell death prior to tumor induction. Proc. Of the Natl Acad Of Sciences 94:12776-12779
3. Mansfield JT, Boucher KM, Lyon JL and Stephenson RA (1997) Prostate needle biopsy Gleason score is a poor predictor of the grade of radical prostatectomy specimens: A population-based study. 1997 Meeting of the American Urological Association
4. Mori M and Boucher KM (1997) Generalized Linear Mixed Models for Informatively Censored Longitudinal Data. 1997 Joint Statistical Meeting, Anaheim, CA.
5. Slattery ML, Boucher KM, Caan BJ, Potter JD and Ma K. (1998) Eating Patterns and Risk of Colon Cancer. Am J Epidemiol (in press)
6. Boucher KM, Pavlova LV and Yakovlev AY (1998) A Model of Multiple Tumorigenesis Allowing for Cell Death: Quantitative Insights into Biological Effects of Urethane. Mathematical Biosciences (in press)
7. Boucher KM, Slattery ML, Berry TD, Quesenberry C and Anderson K (1998) A comparison of statistical methods to analyze dose-response and trend analysis in epidemiological studies, accepted by J Clin Epidemiol
8. Yakovlev AY, Tsodikov AD, Boucher K and Kerber R (1998) The shape of the hazard function in breast cancer: curability of the disease revisited. Submitted to J Natl Cancer Inst
9. Yakovlev AY, Boucher K and DiSario J (1998) Modeling insight into spontaneous regression of tumors. Submitted to Mathematical Biosciences