Locoregional Recurrence After Mastectomy In Breast Cancer Patients With 1-3 Positive Lymph Nodes Treated With Modern Systemic Therapy

Post-mastectomy radiation therapy (PMRT) is indicated for breast cancer patients with 4 or more positive lymph nodes (LN+), but remains controversial for patients with 1-3 LN+. We attempt to identify prognostic factors for locoregional recurrence (LRR) in a cohort of 1-3 LN+ patients treated with mastectomy and modern adjuvant systemic therapy. After IRB approval, we conducted a retrospective review of all 369 breast cancer patients with 1-3 LN+ who underwent a mastectomy without neoadjuvant systemic therapy between 2000 and 2007 at Cleveland Clinic. Cox proportional hazards regression analysis was conducted to identify risk factors (RF) for LRR in the absence of PMRT. We identified 271 patients with 1-3 LN+ who did not receive PMRT and 98 who did receive PMRT. The median follow-up was 5.5 years, and the median number of LN dissected was 11. Of those not treated with PMRT, 60% were post-menopausal, 54% were stage T1, 79% received adjuvant chemotherapy (of whom 70% received a taxane), 79% received hormonal therapy, and 5% had no systemic therapy. Of the Her2/neu amplified tumors, 42% received adjuvant trastuzumab. The 5-year LRR rate was 8.9% without PMRT vs. 0% with PMRT (p = 0.004). On univariate analysis, five RF significantly (p < 0.05) correlated with LRR in the absence of PMRT: ER/PR negative (HR 2.6), lymphovascular invasion (HR 2.4), 2-3 LN+ (HR 2.6), extracapsular extension (HR 3.7), and Bloom-Richardson Grade III (HR 3.1), while tumor size >2 cm, close/positive surgical margins, age, and menopausal status did not. The 5-year LRR rate was 3.8% (95% CI, 0.5-7.2%) for patients with 0-1 RF vs. 15.2% (95% CI, 8.6-21.9%) for patients with ≥2 RF (p = 0.0005), respectively. In patients with ≥2 RF, the 5-year disease-free survival was 82.3% with PMRT vs. 71.9% without PMRT (p = 0.07), whereas it was 90.9% vs. 89.6% (p = 0.74) for patients with 0-1 RF, with and without PMRT, respectively. On multivariate analysis, extracapsular extension (HR 4.3, p = 0.0006) and Bloom-Richardson Grade III (HR 3.6, p = 0.004) remained statistically significant RF for LRR. PMRT offers excellent locoregional control for patients with 1-3 LN+, with no local failures to date. In patients treated without PMRT, those with 2 or more RF (ER/PR negative, lymphovascular invasion, 2-3 LN+, extracapsular extension, Grade III) had a 5-year LRR rate of 15.2%, and would likely benefit from PMRT.