1147 - Local Recurrence Risk in Patients with Microinvasive vs. Early-stage Invasive Breast Cancer: Long Term Outcomes from a Multi-institutional Series
Presenter(s)
C. Hardy Abeloos1, I. J. Choi2, C. White3, Z. Zhang3, A. J. Khan4, J. Xiao5, L. Z. Braunstein6, S. N. Powell4, O. Cahlon7, J. Gurewitz8, and N. K. Gerber7; 1NYU Grossman School of Medicine, New York, NY, 2New York Proton Center, New York, NY, 3Department of Epidemiology and Biostatistics, Memorial Sloan Kettering Cancer Center, New York, NY, 4Department of Radiation Oncology, Memorial Sloan Kettering Cancer Center, New York, NY, 5Department of Radiation Oncology, NYU Grossman School of Medicine, New York, NY, 6Memorial Sloan Kettering Cancer Center, New York, NY, 7New York University Grossman School of Medicine, Department of Radiation Oncology, New York, NY, 8Department of Radiation Oncology, NYU Langone Health, New York, NY
Purpose/Objective(s): A secondary analysis from the Ontario Clinical Oncology Group hypofractionation trial found a significantly higher 10-year local recurrence (LR) rate in patients with T1mi disease compared with T1a-2 disease, suggesting that microinvasive disease may be more locally aggressive than early-stage invasive cancer. The goal of our study was to compare long-term outcomes between patients with T1mi disease and T1a-2 disease after hypofractionated whole breast irradiation (WBI) and to identify prognostic factors.
Materials/Methods:
We included patients with T1mi-2 N0 breast cancer who received hypofractionated WBI between 2013-2019 at two institutions. Univariable (UVA) and multivariable analysis (MVA) of time to LR was performed using Fine-Gray competing risks regression. The cumulative incidence of local (LR), regional (RR) and distant recurrence (DR) was compared between patients with T1mi disease vs. T1a-2 disease using Gray’s test, and overall survival (OS) was compared using log rank test.Results:
Out of 4,115 patients, 170 (4.1%) had T1mi disease. Of those with T1mi disease, 48 patients (28%) had positive margins defined as <2 mm, and 2 patients (1.2%) had positive margins defined as tumor on ink. Among patients with T1a-2 disease, 64 patients (1.6%) had positive margins defined as tumor on ink. Patients with T1mi disease were more likely to have HER2 positive disease, 27.4% vs. 9%, respectively (p < 0.001). Out of those with hormone receptor status positive, patients with T1mi disease were less likely to receive endocrine therapy (ET), 70% vs. 95%, respectively (p < 0.001). At a median follow-up of 6 years, the rate of LR was 10% in patients with T1mi disease and 2% with T1-2 disease. Adjusting for T1mi positive margins defined as “tumor on ink” on MVA, T1mi was independently associated with increased risk of LR (HR 3.89; 95% CI: 1.17, 12.9; p=0.026). Adjusting for T1mi positive margins defined as “< 2 mm”, T1mi was no longer associated with LR (HR 1.91; 95% CI:0.42, 8.71; p=0.4), while margin status was significant (HR 2.96; 95% CI: 1.07-8.21; p=0.037). In the former MVA model, younger age (HR 2.37; 95% CI: 1.42, 3.96; p < 0.001), higher grade (HR 4.09; 95% CI: 1.64, 10.2; p=0.003), and lack of ET (HR 2.52; 95% CI: 1.09, 5.85; p=0.031) were independently associated with increased risk of LR. These associations remained regardless of which T1mi margin definition was included in MVA. There were no significant differences in RR, DR, and OS between those with T1mi vs. T1-2 disease.Conclusion:
In the largest series of patients with T1mi breast cancer, we noted a higher risk of LR among patients with T1mi disease compared to those with T1a-2 disease. This increased risk was no longer significant when adjusting for T1mi positive margin status <2 mm, perhaps highlighting the importance of margin width defined by SSO expert panel in 2016 as = 2 mm for optimal local control. Lower rates of ET were noted in the patients with microinvasive disease, and lack of ET treatment was also associated with increased LR.