Breast Cancer Molecular Subtypes as a Predictor of Radiotherapy Fractionation Sensitivity

Abstract

Breast cancer molecular subtypes have been shown to be predictive for certain systemic therapies; yet, there is a paucity of data regarding their use as a predictor of radiotherapy fractionation sensitivity. The hypothesis of this analysis is that the impact of hypofractionation (HF) in comparison to conventional fractionation (CF), does not differ across breast cancer subtypes in a population of women with invasive breast cancer treated with modern systemic therapy. Patients diagnosed with stage I - III breast cancer between 2005-2009 and referred to a population-based cancer care program were identified. Descriptive statistics were used to report clinico-pathologic and treatment parameters. Molecular subtype was determined using the American Joint Committee on Cancer (AJCC) classification system: Luminal A (ER+ AND PR+ AND HER2- AND Grade 1) OR (ER2-3+ AND PR2-3+ AND HER2- AND Grade 2); Luminal B (All other (ER+ OR PR+) AND HER2-); HER2+ (All Her2+); Triple Negative (ER- AND PR- AND Her2-). Radiotherapy fractionation was categorized as HF (16 fractions) versus CF (25-30 fractions). Multivariable Cox Regression modelling was used to identify predictors of local recurrence (LR). LR-free survival (LRFS) was determined using the Kaplan-Meier method and compared using the log-rank test. A total of 5,868 cases were identified. The median age was 58 years (range: 23 – 91). Patients with Luminal A subtype comprised 45% of the cohort (N=2,628), compared to 30% Luminal B (N=1,734), 15% HER2+ (N=903), and 10% Triple Negative (N=603). There was no difference between those that received HF versus CF in terms of margin status (p=0.67), presence of extensive ductal carcinoma in situ (p=0.30) or receipt of radiotherapy boost (p=0.54). On multivariable analysis, tumor grade (p=0.003), presence of LVI (p=0.006), use of systemic therapy (p=0.01) and receipt of endocrine therapy (p=0.001) were associated with the risk of LR. The overall 10-year LRFS was 97.1% (95 CI: 96.6-97.6) for the whole cohort. The 10-year LRFS based on molecular subtypes was 98.3% (95 CI: 97.6 – 98.7) Luminal A, 96.6% (95 CI: 95.5 – 97.4) Luminal B, 97.0% (95 CI: 95.5 – 98.0) HER2+, and 93.5% (95 CI: 91.1 – 95.3) Triple Negative (p <0.001). There was no difference in the 10-year LRFS between patients treated with HF versus CF in those with Luminal A (98.2% versus 98.4%; p=.42); Luminal B (96.6% versus 96.8%; p=.90); HER2+ (97.5% versus 95.8%; p=.12) or Triple Negative subtype (93.9% versus 92.2%; p=0.47). There was no significant interaction between breast cancer subtype and fractionation regimen. Within the AJCC’s four commonly recognized clinical subtypes of breast cancer, local recurrence-free survival was not affected by breast radiotherapy fractionation. This data supports the routine use of hypofractionated regimens across all breast cancer subtypes.