Biomedical Engineering Reference
In-Depth Information
efforts on understanding and modeling dose-volume effects in the
belief that they were somewhat easier to assess (especially as regards
dose distributions within tumors), and that they were in need of
attention because their effects were largely disregarded at the time I
entered the field of radiotherapy. My opinion hasn't changed and,
in what follows, I focus entirely on dose-volume effects. A good
review of dose-volume models can be found in York (2003), and data
on partial organ irradiation can be found in Seminars in Radiation
Oncology (2001).
Skepticism concerning models
There are a number of biophysical models that purport to describe
dose-volume effects in tumors and normal tissues, and I will briefly
address some of them below and also point to some problems with
them. However, I first want to make a general point.
When I began giving talks on biophysical modeling, I noticed very
different reactions from audiences consisting mainly of physicists and
audiences consisting mainly of physicians. The former tended to
embrace the ideas and models I presented enthusiastically; the latter
were highly skeptical if not downright opposed. So I included some
comments concerning the credibility of the models in my intro-
ductions. But, I made different comments to the two audiences.To
physicists, I advised great caution and skepticism; when talking to
clinicians, I invited them to be a bit open-minded and to consider if
there wasn't at least something to the ideas I was presenting.
My point in mentioning this experience is the following. I am
enormously concerned that, at the time of writing, while physicists
have continued to be enthusiastic, clinicians have forgotten to be
skeptical. There is too little critical thought being given to the very
simple ideas which the models embody, and too little concern about
accepting the implications of the models. A number of deviations
from established experience have been either instigated by, or have
been given support from, biophysical models. There is nothing wrong
in deviating from established experience, providing it is done as part
of a carefully controlled clinical trial. But widespread adoption of
untested deviations is very worrying.
In my view, when models do not lead one to deviate far from
established experience, one can proceed with relative confidence.
When the deviations from established experience suggested by a
model are substantial, one should look very long and hard at what is
