Page 53 - 20dynamics of cancer

38 CHAPTER 3

through a sequence of developmental stages. I use the word progres-
sion in this general sense of development from the first to the final
stages. Within the broad sweep of progression, it may sometimes be
useful to distinguish stages of initiation, promotion, and final progres-
sion to metastasis.

3.2 What Is Multistage Progression?

This question has led to confusion. Some people aim for the ordered
list of necessary changes to cellular genomes and to tissues that cause
aggressive cancers. Others emphasize the controversial hypothesis that
two processes occur: initiation by somatic mutation as a first stage, and
promotion by mitotic stimulation as a second stage.
There is no single correct way to pose the question. The listing of spe-
cific changes sets a useful although perhaps rather difficult goal. The
testing of the particular two-stage hypothesis of initiation and promo-
tion has focused on experimental studies of carcinogens in laboratory
animals; the two-stage hypothesis is probably too narrow to provide a
general framework for cancer development.
I focus on how the dynamics of progression within individuals af-
fects the age-onset patterns in populations. Biochemical changes that
do not affect rates of progression can be ignored in dynamical analyses,
even though they may be very important for understanding physiologi-
cal changes and for analyzing which drugs succeed or fail in chemother-
apy.
In focusing on rate processes, I sacrifice comprehensive understand-
ing of all aspects of cancer. In return for that sacrifice, I gain a coherent
framework that gives meaning to the common but often vague asser-
tion that some particular genetic change or biochemical event causes
cancer: in the dynamical framework of multistage progression, causing
cancer means shifting the age-incidence curve. With this quantitative
framework, we can formulate and test hypotheses about how particular
events affect cancer.
My quantitative emphasis on progression and incidence, and on test-
able hypotheses, means that I will not attempt to cover all aspects of
progression in a comprehensive way (see Weinberg 2007). In this chap-
ter, I give just enough background to set the stage for formulating a
quantitative framework and testing simple hypotheses.