How does knowledge of three-dimensional excision margins following breast conservation surgery impact upon clinical target volume definition for partial-breast radiotherapy?

Abstract

Background and purpose To compare partial-breast clinical target volumes generated using a standard 15 mm margin (CTV standard) with those generated using three-dimensional surgical excision margins (CTV tailored30) in women who have undergone wide local excision (WLE) for breast cancer. Material and methods Thirty-five women underwent WLE with placement of clips in the anterior, deep and coronal excision cavity walls. Distances from tumour to each of six margins were measured microscopically. Tumour bed was defined on kV-CT images using clips. CTV standard was generated by adding a uniform three-dimensional 15 mm margin, and CTV tailored30 was generated by adding 30 mm minus the excision margin in three-dimensions. Concordance between CTV standard and CTV tailored30 was quantified using conformity (CoI), geographical-miss (GMI) and normal-tissue (NTI) indices. An external-beam partial-breast irradiation (PBI) plan was generated to cover 95% of CTV standard with the 95% isodose. Percentage-volume coverage of CTV tailored30 by the 95% isodose was measured. Results Median (range) coronal, superficial and deep excision margins were 15.0 (0.5–76.0) mm, 4.0 (0.0–60.0) mm and 4.0 (0.5–35.0) mm, respectively. Median CoI, GMI and NTI were 0.62, 0.16 and 0.20, respectively. Median coverage of CTV tailored30 by the PBI-plan was 97.7% (range 84.9–100.0%). CTV tailored30 was inadequately covered by the 95% isodose in 4/29 cases. In three cases, the excision margin in the direction of inadequate coverage was ⩽2 mm. Conclusions CTVs based on 3D excision margin data are discordant with those defined using a standard uniform 15 mm TB–CTV margin. In women with narrow excision margins, the standard TB–CTV margin could result in a geographical miss. Therefore, wider TB–CTV margins should be considered where re-excision does not occur.