79 A comparison of five external beam 3D conformal partial breast radiotherapy (3D-PBRT) techniques with standard tangents whole breast radiotherapy (WBRT)

Purpose/Objective: To compare dosimetry for five 3D PBRT techniques and standard WBRT for adjuvant RT after lumpectomy. Materials/Methods: 30 patients treated with lumpectomy and WBRT with a range of laterality, quadrant location & seroma sizes had WBRT tangent 3D dosimetry calculated on a volumetric planning CT using Varian Eclipse software to deliver 50Gy in 25 fractions. Five 3D-PBRT techniques were tested: (i) partial breast tangents; (ii) 3fld photons; (iii) 3fld photon/electron; (iv) 4fld non-coplanar (lt & rt breast), (v) 5fld non-coplanar (lt breast only). Each PBRT technique was 3D planned with a dose of 36Gy in 10 fractions. Volumes of interest for dosimetric evaluation included: heart, ipsilateral lung (IL),contralateral lung (CL), contralateral breast (CB), ipsilateral breast (IL), clinical target volume (CTV=Seroma+1.5cm) & planning target volume (PTV=CTV+1cm). Results: For the 3D-PBRT techniques (i to v) & the standard WBRT(vi) the mean dose volume histogram (DVH) results are. Heart(V2): 2.3%; 50.20%; 13.3%; 2.7%; 4.9%; 4.1% IL(V11): 7.8%; 22.4%; 9.7%; 9.2%; 8.5%; 9.7% CL(V2): 0.1%; 34.9%; 1.4%; 2.2%; 1.8%; 0.1% CB(V5): 2.0%; 1.8%; 4.0%; 1.2%; 1.1%; 4.1% IB(V18): 66.6%; 62.3%; 63.6%; 60.3%; 62.2%; 98.5% IB(V34): 50.6%; 32.7%; 33.1%; 35.9%; 32.5%; 92.0% PTV homogeneity index: 0.95; 0.96; 0.86; 0.97; 0.97; 0.90 PTV conformity Index: 0.30; 0.60; 0.60; 0.60; 0.66; 0.20 Conclusions: The 3fld photon/electron technique and the 4- and 5-field non-coplanar techniques improved PTV conformity and sparing of IB tissue compared to standard WBRT tangents, while maintaining acceptable doses to the heart, lungs and CB. Partial breast tangents did not adequately spare uninvolved ipsilateral breast tissue. The 3-field photon technique delivered unacceptably high doses to the heart, lungs & CB. The time/resource implications of the 3fld photon/electron and non-coplanar techniques need to be considered to select an optimal class solution for 3D PBRT.