Association of Dosimetric Parameters and Boost Volume with Quantitative Cosmetic Outcomes in a Randomized Trial of Conventionally Fractionated vs. Hypofractionated Whole Breast Irradiation

Abstract

<h3>Purpose/Objective(s)</h3> Whole breast irradiation (WBI) is the most common treatment in radiation oncology, yet there is a relative paucity of evidence delineating the relationship between dosimetric parameters and long-term outcomes. We conducted a prospective randomized trial with longitudinal follow up to compare two different WBI dosing schedules and sought to evaluate the association of dosimetric parameters and boost volume with long term cosmetic outcomes using novel, previously validated, quantitative metrics of breast cosmesis derived from 2-dimensional photographs. <h3>Materials/Methods</h3> A prospective randomized trial of patients with early breast cancer who were randomly assigned to conventionally fractionated or hypofractionated WBI plus boost was conducted from 2011-14. Anteroposterior photographs from the low-neck to umbilicus were obtained at baseline, 6 months, and yearly through 5 years. Fiducial points marking each nipple and inframammary fold were delineated on each photograph and used to calculate a measure of nipple symmetry (M1) and breast vertical symmetry (M2), as previously described (PMID 28333010). These values range from 0 to 1 with values closer to 1 indicating better symmetry. Multivariable logistic regression longitudinal models with M1 and M2 as outcomes, dichotomized at the median, evaluated associations of dosimetric parameters and boost volume with M1 and M2 adjusting for relevant clinical, demographic and treatment variables. Candidate dosimetric variables included volume of tissue encompassed by the 90% isodose line (PTV), tumor bed volume, volume encompassed by the 105% isodose line, maximum point dose in the breast (Dmax), central axis separation, percent dose delivered with 18 MV photons, boost energy, boost position, boost modality, and cross-sectional area of the boost field. <h3>Results</h3> Of the 287 patients included in this trial, 149 patients received CF-WBI and 138 received HF-WBI, with median follow up 48 months. Median age was 60 years. Median PTV was 1340cc (IQR 904-1760), median tumor bed volume was 23cc (IQR 13-43), and median boost field size was 64cm² (IQR 54-80). 50% of patients had a Dmax of ≥107% of the prescription dose. A total of 1600 photographs were annotated to calculate M1 and M2. In multivariable analysis, highest quartile of PTV (OR 0.40, P=0.002) and boost field size (OR 0.50, P=0.02) were associated with worse M1 outcome compared to the lowest quartile. Highest quartile of tumor bed volume was associated with worse M2 compared to the lowest quartile (OR 0.52, P=0.03). <h3>Conclusion</h3> Utilizing previously validated quantitative measures of cosmetic outcome derived from 2D photographs, we found that boost fields with cross-sectional area <54 cm² and PTV <1760cc were associated with better nipple symmetry, and tumor bed volume >43cc was associated with worse vertical breast symmetry. Understanding how dosimetric parameters and boost volume affect long-term cosmetic outcomes can facilitate optimization of breast radiation treatment plans.