*************************************************************************

                Department of Mathematical Sciences                       
                   The Johns Hopkins University                           
                                                                        
                              SEMINAR                                
                                                                        
*************************************************************************

Giovanni Parmigiani                               April 19, 2001
Departments of Oncology and Biostatistics         304 Whitehead Hall
The Johns Hopkins School of Public Health         Preseminar: 3:00 p.m.
                                                  Refreshments: 3:30 p.m.
                                                  Seminar: 4:00 p.m.

*************************************************************************

     STATISTICAL MODELING OF BREAST CANCER SUSCEPTIBILITY GENES

*************************************************************************

                             ABSTRACT

Recent advances in the understanding of genetic susceptibility to breast
cancer -- notably, identification of the BRCA1 and BRCA2 susceptibility
genes -- raise important questions for clinicians, patients, and
policy-makers.  In this lecture, I will describe examples of how
statistical modeling can contribute to answering some of these questions.

The lecture will revolve around an algorithm and software package named
BRCAPRO, which estimates the probability of carrying a genetic
susceptibility mutation, for a given family pedigree with history of
breast and related cancers.  The model combines biological evidence about
the genes' prevalence, penetrance, and inheritance mechanism, together
with Bayesian modeling and prediction techniques.  I will briefly
outline the basic elements of BRCAPRO and then describe three areas
of clinical application and current research: (i) directly using the model
in genetic counseling: assessing and communicating uncertainty about
whether an individual is a carrier and about the model's own estimated
probability; (ii) validating the model using pedigree data on tested
individuals, accounting for errors in genetic testing; and (iii) using
the model for exploring differences in prognosis between carriers and
noncarriers of the genes, by incorporating detailed pedigree information
in survival analysis.  The last application raises some interesting
questions about inference in survival models with misclassified
categorical covariates.
 
Papers related to this lecture and other information about BRCAPRO are
available from the site
http://biosun01.biostat.jhsph.edu/~gparmigi/brcapro.html/ .

*************************************************************************