The Effect of Bolus on Local Control for Patients Treated with Mastectomy and Adjuvant Radiotherapy

Abstract

Patterns of practice surveys have shown that about half of radiation oncologists routinely use bolus for postmastectomy radiotherapy (RT). Bolus use causes well recognized acute and late toxicities and leads to non-completion of adjuvant radiotherapy in a small fraction of patients. Whether bolus use improves local control is unknown. The null hypothesis of this study was that the use of bolus during radiotherapy would not improve local control. This study was conducted for a defined population of patients, in whom bolus was used routinely for chest-wall RT and was omitted routinely for RT with a breast reconstruction. As the decision to have a reconstruction was usually made by the patient in consultation with their surgeon(s) and not with their oncologists, this practice offered a pseudo-randomized opportunity to evaluate local control outcomes in patients treated with and without bolus. The study included newly diagnosed invasive breast cancers (pT1-4a, any-N, M0) treated with mastectomy and adjuvant RT. Patients with skin involvement (T-stage cT4b, c, d or pT4b, c, d) were excluded. In this population, local recurrence events are actively sought until 15 years after diagnosis with annual queries to the primary practitioners of all breast cancer survivors. The primary endpoint was the cumulative incidence of local recurrence using regional recurrence, distant recurrence and death as competing risks in patients treated with and without bolus. A multivariable analysis of local recurrence was conducted with T-stage, N-stage, lymphovascular invasion (LVI), margin status, triple-negative subtype, as well as use of bolus, hormone therapy and chemotherapy. The entire cohort consisted of 1,891 women diagnosed from 2007 to 2011, 789 treated with bolus and 1,102 treated without bolus. The median follow-up was 8.7 years. The median age in the reconstruction cohort was lower (54 versus 56 years, p=0.03, median test); however, there were no significant differences in the distribution of tumor characteristics (T-stage, N-stage, LVI, margin status) or treatment characteristics (hormone therapy, chemotherapy). The 10-year cumulative incidence of local recurrence was 1.8% (95%CI: 1.1 – 2.9%) with bolus and 1.7% (95%CI: 0.9 – 2.8%) without bolus. The multivariable analysis demonstrated that local recurrence was not significantly related to use of bolus (HR = 1.08, 95%CI: 0.5 – 2.2) or the other variables. In this population-based study of patients who received contemporary systemic therapy for breast cancer without skin involvement, the absolute risk of local recurrence was low. The use of bolus did not impact long-term local control after mastectomy and RT. This retrospective study provides the best current evidence that routine use of bolus during radiotherapy should be questioned in patients treated with mastectomy.