" Advancing Geometrical and Biological Targeting
in Radiation Therapy "
-pdf
Abstract: It is well established that increasing the radiation
dose improves local tumor control. Radiation treatment delivery
necessarily irradiates normal tissue and may also dose organs
and other sensitive structures. Thus, the goal of radiation therapy
has always been to maximize the tumor dose while minimizing dosing
of normal tissues in order to achieve the highest survival with
the fewest complications. Radiotherapy methods have evolved with
advances in imaging. Volumetric CT and MR studies made the construction
of 3D patient models possible, enabling the design of radiation
beam apertures to conform to the tumor shape and the selection
of beam directions to avoid critical normal structures. The recent
technological development of intensity modulated radiation therapy
(IMRT) provides the ability to control radiation treatment delivery
as precisely as the best imaging technologies, making voxel by
voxel dose prescription possible.
Current planning of IMRT is based on dose optimization and not
on biological outcomes such as tumor control and normal organ
complications. It is widely recognized that biologically based
treatment plan optimization is required to realize the full potential
of IMRT technology. A few years ago, I began work on the determination
of radiation dose distributions that maximize tumor control probability
(TCP) in heterogeneous tumors for fractionated radiation therapy
protocols will be described. This was done for tumor volumes
whose radiation response is characterized by the linear quadratic
model, with spatially and temporally varying radiosensitivities,
repopulation rates and cell densities. Recently obtained results
on the necessary condition for achieving TCP that increases with
each fraction that was found will be described. Under certain
conditions an acceptable TCP cannot be achieved, and this is
highly dependent on the temporal behaviors of the repopulation
and radiosensitivity. This effect was not considered before,
and may provide a partial explanation of treatment failures.
Radiotherapy planning is based on imaging studies taken several
days prior to the initiation of therapy and patients undergo daily
treatment sessions for up to several weeks. Successful treatment
requires positioning a patient so that the geometric relation between
target volume and delivery unit is identical to the 3D computer
model. Typically, a margin is added to the target volume to account
for known differences. Although this insures that the target is
hit, addition of a margin increases the treatment toxicity. However,
certain tumors move with the respiratory cycle. This has significant
implications for imaging studies and has prompted 4DCT acquisition.
Treatment rooms now incorporate planar imaging, fluoroscopy, and
cone beam CT with the patient positioned on the treatment unit.
Techniques to best utilize these technologies are evolving rapidly.
Preliminary results of our investigation of methods for monitoring
respiratory motion, registering in room kV x-ray images to 4DCT,
and localization with implantable markers will be described.
|
