Page 175 - 20dynamics of cancer

160 CHAPTER 8

anti-apoptotic effects (not shown) also predict the direction of change
in acceleration and age of onset.
Limited sample sizes present the greatest problem in studies that
estimate age-specific incidence for particular genotypes. To get around
this limitation, I formulated the hypotheses as predictions about the
direction of change in comparisons between genotypes. For example,
I predicted that acceleration would decline in a sample with relatively
stronger defects in mismatch repair when compared against a sample
with relatively weaker defects in mismatch repair.
If each key prediction is formulated in a comparative way, laboratory
studies with small sample sizes can be used. Each comparison provides
a single binary outcome that represents either a success or failure of
the theory to predict the direction of change in some attribute of cancer
dynamics. The binary outcomes can be aggregated to form a nonpara-
metric test based on the binomial distribution. This allows my approach
to be applied to small samples of mice in each genotype. The effective
sample size comes from the number of comparisons.
Over the past few years, vast resources have been expended on ani-
mal experiments that compare survival curves for different genotypes.
If these sorts of experiments were designed and analyzed with dynamics
in mind, the research could move to the next level in which the mech-
anistic consequences of particular genetic pathways are related to the
dynamics of carcinogenesis. The data I presented here were not col-
lected to test mechanistic and quantitative hypotheses about dynamics.
A simple reanalysis provided significant insights about how DNA repair
genotypes affect separately the age of onset and the acceleration of can-
cer.

8.3 Polygenic Heterogeneity

The previous sections showed how mutations to the mismatch repair
genes or APC accelerate gastrointestinal cancer, and mutation to Rb ac-
celerates retinoblastoma. Those mutations to single genes have simple
inheritance patterns and cause major changes in incidence, making them
relatively easy to study.
Genetic variation across multiple loci may also strongly affect inci-
dence. However, such polygenic causation creates difficulties in studies,