2916 - Omission of Radiation Therapy in Patients with cT1-T2N0 HER2+ Breast Cancer with Pathologic Complete Response after Preoperative Systemic Therapy and Breast Conserving Surgery

Purpose/Objective(s): Patients with HER2+ disease that receive preoperative systemic therapy (PST) and achieve pathologic complete response (pCR) have excellent oncologic outcomes suggesting that pCR may be used to de-escalate radiation (RT) in the setting of breast conserving surgery (BCS). Omission of RT is currently being evaluated in cT1-2N0 HER2+ patients with pCR on NRG-BR008. The trial was recently amended to include tumors up to 5cm in size and to allow for residual DCIS (ypTis). Since RT omission is not the current standard of care (SOC), there are few data evaluating the safety of RT omission. We used the NCDB to test the hypothesis that RT omission results in equivalent OS in cT1–T2(=5cm) cN0 HER2+ patients treated with PST+BCS with pCR(ypT0/Tis ypN0).

Materials/Methods: We used the NCDB to identify patients with cT1-2 (=5cm) cN0 HER2+ breast cancer treated with PST and BCS with a pCR and identified patients that received RT and those that had RT omission. We used logistic regression to determine which factors were associated with RT omission. We then evaluated OS in a subset of patients, stratified by receipt of RT, using optimal 1:1 matching utilizing the logistic regression model to calculate propensity score. Optimal matching was chosen due to the rarity of death events. This propensity score-matched (PSM) analysis balanced clinical and demographic factors between groups to compare OS. All Kaplan–Meier survival estimates were based on the matched cohort. The effect of RT omission was assessed with a univariable Cox proportional hazards model.

Results: We identified 3,528 patients (3,161 RT; 367 no RT). On multivariate logistic regression analysis, omission of RT was strongly associated with older age: age 60-70 y vs. <50 yr: OR=1.7 (95% CI 1.2-2.4, p=0.002) and age 70+ yr vs. age <50 yr: OR=4.7 (3.3-6.7, p<0.001). Other factors associated with RT omission included cT1 tumors with OR=1.4 (1.1-1.8; p=0.007), HR+ with OR=2.3 (1.8-2.9; p<0.001), Hispanic patients with OR=2.1 (1.4-3.0; p<0.001) and Asian patients with OR=1.5 (1.0-2.3; p=0.05) ethno-racial categories. Survival data was available in 2,902 patients (2,633 RT; 269 no RT). The PSM cohort consisted of 269 matched pairs with median follow-up 42 months (95% CI, 39.4-46.1 months) and 28 deaths (11 RT arm; 17 no RT arm). RT omission was not significantly associated with an increased risk of death with HR=1.89 (95% CI, 0.89-4.1, p=0.1) and corresponding 3-year OS 94.1% with RT omission vs. 97.0% with RT.

Conclusion: We found no association between RT omission and increased risk of death in cT1–T2N0 HER2+ BC patients who achieve pCR after PST+BCS. Patients who did not receive RT tended to have cT1 tumors, to be older, to have HR-positive disease, and to identify as Hispanic or Asian. The numerically lower rates of 3-year OS in the RT omission arm should be interpreted with caution given that this is not a current SOC treatment strategy. Overall, these data support the continued prospective evaluation of RT omission in this patient population on NRG BR008.