Chalmers Conferences, 9th European Conference on Mathematical and Theoretical Biology

Uncovering the Natural History of Metastatic Cancer from Autopsy Data
Leonid Hanin

Abstract

The goal of the study is to develop mathematical and statistical methodology
for estimation of important unobservable characteristics of the natural history
of metastatic cancer (such as the rates of growth of the primary tumor and
metastases and mean metastatic latency time) from the autopsy data. The
latter consists of the largest cross-sectional areas of liver metastases obtained
from sections of the liver by parallel planes at certain distance apart from each
other. Estimation of the cancer natural history is based on a previously proposed
comprehensive stochastic model of cancer progression accounting for primary
tumor growth, shedding of metastases, their selection, latency and growth in a
given secondary site. The model was applied to the aforementioned autopsy data
for one breast cancer and one lung cancer patient. Identifiable parameters of the
model were estimated by the method of maximum likelihood. The model with
these parameters provided very good fit to the data. Results of model-based data
analysis will be discussed from biomedical standpoint.

This is a joint work with Jason Rose, a Ph.D. student at the Mathematics
Department of Idaho State University.
(Idaho State University) hanin@isu.edu

Uncovering the Natural History of Metastatic Cancer from Autopsy Data

The goal of the study is to develop mathematical and statistical methodology
for estimation of important unobservable characteristics of the natural history
of metastatic cancer (such as the rates of growth of the primary tumor and
metastases and mean metastatic latency time) from the autopsy data. The
latter consists of the largest cross-sectional areas of liver metastases obtained
from sections of the liver by parallel planes at certain distance apart from each
other. Estimation of the cancer natural history is based on a previously proposed
comprehensive stochastic model of cancer progression accounting for primary
tumor growth, shedding of metastases, their selection, latency and growth in a
given secondary site. The model was applied to the aforementioned autopsy data
for one breast cancer and one lung cancer patient. Identifiable parameters of the
model were estimated by the method of maximum likelihood. The model with
these parameters provided very good fit to the data. Results of model-based data
analysis will be discussed from biomedical standpoint.

This is a joint work with Jason Rose, a Ph.D. student at the Mathematics
Department of Idaho State University.