Computerized treatment planning in radiation therapy of intact breast: influence of number of CT-cuts

To compare the accuracy of 2D algorithm with an established 3D algorithm, and to define the number of CT-slices needed for treatment planning of intact breast irradiation.

Twenty patients with breast cancer treated with conservative surgery were included in this study, ten of them had right breast cancer and the other 10 patients had cancer of the left breast. For each patient, 3-D calculations (HeLax-TMS) were performed using one CT-slice (central), 3 CT-slices (central, caudal, and cephalic) and full set CT-slices in addition to 2D calculations (Multidata System) on the digitized central cut. All calculations were done using 6MV-photon.

When using 2D planning with lung correction, a large hot area of 105% was found at the medial and lateral subcutaneous (SC) regions. Comparison of 2D-treatment planning using Multidata System (2D-physics) and 2D-planning using HeLax System (3D-physics) showed that the 2D planning using Multidata System gave a large hot area of 105% compared with HeLax-2D at the subcutaneous region. The central axis dose distributions obtained from 2D and 3D calculations using HeLax system were compared. No differences were found in the two planes (central cut plane) and this was because the two planes were based on the same algorithm. The only difference was that the hot area (110%) was found at the superior or the inferior border of the field. Also, a comparison using the 3D-system for the central slice and the multiple slices showed a difference in calculating the maximum dose to the target of 2.19%, which was Statistically significant (p=0.001). For all left sided patients, the maximum dose to the heart was significantly different from one to full CT-cuts. It was 12.0+/-6.0% when using one CT-cut versus 6.6+/-1.3% when using full CT-cuts (p<0.03). We compared isodose distributions using three and full CT-cuts for both small and large breasts. For the large breast patients, larger differences in isodose distributions were observed in the cephalic and the caudal planes than for small breast patients.

Dose distributions based on a single CT cut through the central axis using 2D or even 3D treatment planning system will lead, quite often, to hot volumes in 76 excess of 105%. For patients whose breast contours vary slowly within the tangential fields, a three-slice CT scan appears to be adequate for clinical decision. However, for patients with large variation of contours within the tangential fields, a full CT scan gives more accurate dose distributions than the three-slice model.