Abstract Purpose: To evaluate the early therapeutic outcome of accelerated partial breast irradiation(APBI) in low-risk breast cancer patients treated with breast conservative surgery(BCS) in a Chinese population, to analyze the intra-fractional and residual error with image-guidance.Material and methods: Eligible patients were age ≥18 years, unicentric Stage I invasive carcinoma or ductal carcinoma in situ(DCIS), treated with BCS with margin>2mm. Sixteen patients were enrolled between June 2008 and April 2009, with median age 53 ys (46-71). Clinical target volume(CTV) is defined by surgical cavities plus 1.5cm margin, another 1cm is expanded to form the planning target volume(PTV), with edition to no closer than 5 mm to the skin surface and no deeper than the anterior chest wall. Three-dimensional treatment plannings with field in field technique were generated with dose of 38.5Gy/10F, bid with interval≥6 hours. Two online cone beam CTs (CBCT) was acquired for each fraction, using Elekta Synergy, X- Ray Imaging system. One is after setup with skin-markers and the other at the end of fraction. Three different registrations: bony structure, grey and grey with manual adjustment(gmanual) were compared. The clip box included: whole breast, ipsilateral lung and part of the vertebras. The registration results of each fraction were recorded. The mean shifts and standard deviation (SD) were calculated on the CBCT pre and post-treatment.Results: The ratio of APBI PTV/whole breast PTV was 25%-40% (median 29.3%). 6MV photon with median fields of 4(3-6) was used. Grade I skin toxicity was found in all the 16 patients after 15-20(median 17) days of treatment, grade II skin reaction occured in 2 patients.The setup errors detected by the initial CBCT are 7mm in lateral(X), 10mm in longitudinal(Y) and 7mm in vertical directions(Z).Table 1 Setup errors detected by the differences between pre-treatment CBCT with skin marker and planning CT grey gmanual x (cm)y(cm)z(cm)x (cm)y(cm)z(cm)Mean-0.010.14-0.1 0.15-0.13SD0.260.370.290.280.370.28system error0.210.320.230.210.310.23random error0.180.190.180.180.230.2margin0.650.930.700.650.940.72 The bony algorithm showed statistical difference with grey or gmanual algorithms (p<0.05). Shift was applied when the registration result by gmanual was >3mm in any directions. The residual error and intra-fraction error was then analyzed after on-line shift and the post-fraction CBCT, which resulted in 6.9mm(X)l, 6.4mm(Y)and 7.3mm (Y) directions, see Table 2.Table 2 The summary of residual errors and intra-fraction errors after shift alignment x (cm)y (cm)z (cm)mean0.010.00-0.07SD0.260.290.29system error0.210.200.23random error0.230.200.22margin0.690.640.73 Conclusion: APBI with 3-D conformal external beam and image guidance is a feasible technique. Grey registration with manual adjustment is recommended for clinical application. The initial set-up error data confirmed the safety of 1cm margin from CTV to PTV. With online shift of any error above 3mm, the combination of residual error and intra-fraction error is 6.9mm, 6.4mm and 7.3mm in X, Y and Z directions, which might provide potential for margin reduction with image guidance. Citation Information: Cancer Res 2009;69(24 Suppl):Abstract nr 4111.
Read full abstract