Clinical target volume localization using conventional methods (anatomy and palpation) and ultrasonography in early breast cancer post-operative external irradiation

Purpose: To evaluate the accuracy of three methods, anatomy (A), palpation (P) and ultrasounds (US) in localizing the clinical target volume (CTV) in patients (pts) with early breast cancer (EBC) undergoing breast external irradiation as part of conservation therapy. Material and methods: One hundred consecutive pts with EBC (Tis 1%, T1 78%, T2 21%, N- 68%, N+ 32%), treated with conservation surgery and breast irradiation with opposed tangential portals, were prospectively analyzed. Anatomically, palpatory or ultrasound defined field borders for CTV localizations were determined in the same position thanks to the utilization of a vacuum-formed cellulose acetate immobilization cast, removed during CTV definitions. Results: P and US CTV localizations have been found to coincide on the four margins (superior, inferior, medial, lateral) in only(1(100)) pts, while no pt showed identical A and US CTV localizations. Only (31(397)) (8%) field measurements with A, and(98(395)) (25%) with P corresponded to US border definition. If mean and median values of each field border were considered, the CTV was generally over-estimated with P appearing more accurate than A in a gross definition of the target (P < 0.01). However, a geographical miss of at least one field border of CTV occurred in 55% of pts with A and in 36% of pts with P. The most critical margin to be defined with conventional methods was the superior one: an underestimation of the cranial border of CTV with A was observed in 51% and with P in 22% of pts (22% and 8%, respectively, when an underestimation by more than 1.5 cm was considered). When pre-menopausal and peri/post-menopausal groups of pts were separately analyzed, conventional methods were highly inaccurate to define the superior border in younger pts, in which a geographical miss was noted with A in 62% and with P in 35% of cases (P < 0.05). When an underestimation of more than 1.5 cm was evaluated, these values were reduced to 33% and 12%, respectively (P = 0.02). Also, in the subgroup of pre-menopausal patients the CTV defined by palpatory reference lines appeared to be more accurate than A-method (P < 0.01). Conclusions: Conventional methods frequently appear inadequate to localize residual mammary gland in EBC post-operative external irradiation. This analysis evidentiates the inaccuracy of empirical CTV definition, and suggests that palpation or anatomical reference borders should be critically used and that state of art imaging methods should be included in treatment planning, particularly in pre-menopausal pts. If economical or individual variables have to be taken into account, palpation might be considered in post-menopausal women