Geometric accuracy of accelerated partial breast irradiation on the Cyberknife using standard surgical clips as a fiducial marker

Over the last decades, the prognosis of breast cancer patients has dramatically
improved. Recent SEER data shows a 5-year cancer specific survival for
localized breast cancer patients of 98.9%. For this reason, decreasing the
burden of treatment has gained attention.
Radiobiological data showed that it is favorable to hypofractionate breast
cancer radiotherapy. Most local recurrences of early stage breast cancer occur
in the vicinity of the original tumor. This means that it is sufficient to
treat only the region of the tumor bed instead of the whole breast. Combining
these two strategies results in accelerated partial breast irradiation (APBI).
Several multicenter randomized trials showed non-inferior local control of APBI
compared to whole breast irradiation. There are several techniques available to
deliver APBI. The Cyberknife has the advantages that it is not invasive, can
create a highly conformal dose distribution and can track the target real-time
during treatment delivery. Therefore, the safety margin needed to compensate
for geometric uncertainties during treatment delivery is smaller than with
conventional linacs. The exact value of the optimal safety margin is unknown.
The aim of our study is to fully assess the geographical uncertainties during
CK-APBI, and to calculate the optimal PTV margin required. We will investigate
whether we can use the surgical clips instead of implanted interstitial gold
markers as fiducials to track the target. These clips are inserted during
lumpectomy as a standard procedure to aid the delineation of the tumor bed.
These clips are hence per definition within the delineated target volume.
Another advantage of using the surgical clips as fiducials instead of
interstitial gold markers is that a separate invasive procedure to implant the
fiducials would not be necessary. The third type of fiducial that we will test
is a gold marker taped onto the breast surface.
After the calculation of the optimal PTV margin, treatment plans will be
generated using the new margin, and dose to organs at risk will be compared
with the plans using the conventional margin.

The primary objective is to prospectively quantify the geometric uncertainty of
CK-APBI and to propose a validated PTV margin.

Secondary objectives are:
- To evaluate the success rate of using surgical clips as fiducial markers
- To evaluate the success rate of using a gold marker taped on the breast
surface as fiducial marker
- To evaluate the success rate of using implanted interstitial gold markers as
fiducial marker
- To calculate the dosimetric gain of PTV margin reduction
- To report acute toxicity and pain of CK-APBI

The proposed study is a single center cohort study. APBI is an accepted
treatment outside clinical trials for highly selected, low-risk breast cancer
patients, according to international guidelines [12-16]. Our aim is to improve
APBI delivered on the Cyberknife stereotactic system. We will treat a cohort of
patients with CK-APBI, using conventional large safety margins. To assess the
geometric accuracy, CT scans will be made before each fraction in treatment
position. We will test the surgical clips, superficial gold markers an
implanted interstitial gold markers as fiducials to track the target.
Every year, about 240 patients are referred to our radiotherapy department that
are eligible for APBI. About a third have large titanium or tantalum clips
implanted in the tumor bed, so we plan to complete accrual within one year.

The risks of this study are estimated to be very small. APBI is an accepted
treatment for early stage low-risk breast cancer patients. The inclusion
criteria of this study are in accordance to the international guidelines for
selection of appropriate candidates for APBI outside clinical trials (see
chapter 4).
During the study, a CT scan will be acquired before each treatment fraction. As
there are 5 fractions, a total of 5 extra CT scans will be made. Radiation dose
of a CT-scan of the breast is estimated to be around 5mSv. This is equivalent
to the background yearly radiation dose. Compared to the dose due to the
radiation (30,000 mSv), the extra radiation dose by five CT-scans is
negligible. It takes about 10 minutes of extra time to acquire the pretreatment
CT scan. There are no extra hospital visits.
In normal practice, a new planning CT scan is made in case of anatomical
changes or treatment delivery problems. In this study, we will be able to use
the daily pretreatment CT scan for this purpose and avoid the extra delay in
treatment delivery until the new planning CT scan is performed.
There is no direct benefit of study participation for the patient and the
treatment will be accessible for patient not willing to participate to the
study.