Do Skin Markings Reflect the Location of the Surgical Cavity During Daily Radiotherapy Treatment?

1. Purpose - A feasibility study to determine the accuracy of using skin markings as a surrogate for the underlying surgical cavity (SC) in accelerated partial breast irradiation.
2. Hypothesis - Markings placed on the skin are an accurate representation of the underlying surgical cavity and are adequate to use for patient setup for accelerated partial breast irradiation.
3. Justification Breast conserving surgery with the use of whole breast radiation after partial mastectomy is the preferred treatment for most women with early stage breast cancer. Conventional radiotherapy in this setting delivers radiation to the entire breast, with the consequence that the radiotherapy has to be given over a period of 3 - 6 weeks. However, the majority (70 - 80%) of local recurrences after breast conserving surgery develop in the vicinity of the primary tumor. Partial breast irradiation with the radiation target volume limited to the surgical cavity (SC) in selected patients is therefore potentially feasible if it results in similar local control rates compared to whole breast irradiation. In addition to limiting the radiation to breast tissue most likely to harbor microscopic disease and thereby sparing the normal breast tissue, local breast irradiation offers the distinct advantage of shortening the duration of radiotherapy, thereby minimizing the impact radiotherapy has on the quality of life of breast cancer patients. The current technique for accelerated partial breast irradiation assumes that the cutaneous markers correctly reflect the position of the underlying SC without strong evidence that the relationship between the two does not change. Our study will provide the crucial missing link. Phase II of our study (not included in this REB application) will go further in exploring the use of new imaging modalities in reducing the PTV margins thereby enabling oncologists to better spare normal tissues.
4. The primary objective is to determine whether skin markings truly reflect the surgical cavity location during daily treatment by quantifying any change in relationship between the skin markings \& the surgical cavity. The secondary objective is to investigate the dosimetric consequences of the change in size, shape, and location of the surgical cavity volume (SCV) during the treatment duration.
5. Methods (for details see the attached protocol) To determine whether skin markings truly reflect the surgical cavity location during daily APBI.

Briefly, for Stage 1 of the study, 20 breast cancer patients undergoing adjuvant radiotherapy after breast conserving surgery with a visible SC (cavity visualization score 3 - 5) will be recruited. Skin markers will be placed on the skin around the SC during CT simulation in the manner specified in the RAPID trial. During the first week of their treatment, patients will return to the CT scanner on days 1 and 5 to undergo repeat planning CT. The patients will be set up simulating the actual treatment set up at the linear accelerator. These repeat scans will be compared with the original scan to identify any change in the relationship between the skin markings and the SC. Five oncologists trained in SC identification will then contour the SC on the original as well as the repeat CT scans. These images will be fused for each CT scan to form a representative SC volume (RSCV). The RSCV in planning and subsequent CTs will be related to skin marks to evaluate the reliability of patient positioning. Using these skin marks the beams will be placed on the patient in the CT simulation software package.

The centre of mass (COM) from RSCV will be determined from the contours and variations in x,y, and z coordinates will be averaged. This will allow a comparison between what was planned originally and what has changed over time. Coordinates of previously placed skin markers on the breast surface will be calculated relative to the COM as well as medial and lateral tattoos. This will enable us to verify if setup to external skin marks track the RSCV.

The dosimetric consequences of SCV changes in shape/size/location will be quantified as follows:

1. Repeat CTs with drawn RSCVs will be exported to the planning system and planning fields will be applied to calculate dose distributions in target volumes and organs at risk.
2. Dose-volume histograms (DVH) for CTV, PTV and organs at risk, as defined from CT will be evaluated for meeting planning criteria.
3. The procedure will be repeated with simulated patient realignment using CT data sets.
4. Appropriate PTV margins to achieve desired CTV coverage will be calculated for both the situations, with and without simulated realignment.
5. Statistical Analysis Sample Size: For Stage I, 20 patients can be successfully accrued in 6 months at the Fraser Valley Cancer Centre.

Comparison between skin marks and surgical cavity

The centre of mass (COM) from RSCV will be determined and the mean variations in x,y, and z coordinates will be calculated. In addition, the mean and standard deviation of the change in volume of the RSCV will be calculated. The movement of the COM coordinates will be calculated in reference to both the tattoos and the point guards separately. A paired t-test will be used to assess whether there is a statistically significant difference in mean COM movement in reference to the tattoos and centre spots. This will allow us to verify if setup to external skin marks (tattoos and point guards respectively) track the RSCV.

DVH comparison

The dosimetric consequences of changes in shape, size, and location of the RSCV will be quantified through analysis of DVHs (see section 7.5.). Descriptive statistics will be used to present the change in DVH for target volumes (CTV, PTV, and DEV) and OAR that occurred between the initial simulated plan and the 5 additional simulated plans (see sections 7.2 through 7.5). Furthermore, this will be repeated for the plans that have undergone simulated realignment. A paired t-test will be used to assess whether there is a statistically significant difference in mean change in DVH (for target volumes and OAR) before and after simulated realignment.

Potential decrease in CTV to PTV margin

The margin required on the CTV to create a PTV which ensures it is adequately covered, will be obtained using the DVH criteria in Table 2. Subsequently, this will be repeated for the plans that have undergone simulated realignment. A paired t-test will be used to assess whether there is a statistically significant difference in mean change in the CTV margin needed before and after simulated realignment.

Simulated patient realignment compensates for a portion of the CTV to PTV margin (the size, shape and position of the RSCV) while still leaving some uncertainty (e.g., patient respiration, daily setup). Therefore it is hypothesized that we can decrease the PTV needed with simulated realignment. We will consider a decrease in the PTV margin by ≥ 3 mm with simulated patient realignment justification for progression to Stage II of this study.