Post-mastectomy Radiation: Should Subtype Factor into the Decision?

Traditional indications do not factor molecular subtype into the decision making for post-mastectomy radiation (PMRT). We sought to determine whether constructed subtype was associated with receipt of PMRT in an academic cancer center and to assess differences in locoregional recurrence (LRR) by constructed subtype.

Patients treated with mastectomy as the primary surgical therapy were identified. Univariate and stepwise multivariate logistic regression analyses examined the association between covariates and PMRT. Kaplan-Meier estimates for the time to either the earlier of LRR or last follow-up were obtained for each subtype, and Cox proportional hazards regression examined the effect of covariates on time to LRR in both univariate analyses and stepwise multivariate analysis.

Overall, 884 patients with invasive breast cancer who underwent a primary mastectomy between January 2002 and May 2012 were included in the study. A total of 359 patients (41.6 %) received PMRT. Compared with other subtypes, triple negative (TN; HR-/HER2-) cancers were more likely to be smaller (95 % T1/T2; p = 0.02) and have a lower nodal burden (N0 65 %; p < 0.0001). On multivariate analysis, age < 50 years, lymphovascular invasion (LVI), T stage, N stage, and close or positive margins remained significantly associated with PMRT, while constructed subtype was not associated with PMRT. Compared with all other subtypes, TN had the highest rate of LRR [hazard ratio (HR) 5.70; p < 0.0001]. On multivariable analysis, TN status, LVI, and not receiving chemotherapy were significantly associated with LRR.

Despite significant differences in LRR by constructed subtype, receptor status does not appear to be associated with the receipt of PMRT. In our series, TN cancers had the highest risk of LRR despite their relatively smaller size and limited nodal disease.