Selecting breast cancer patients with T1-T2 tumors and one to three positive axillary nodes at high postmastectomy locoregional recurrence risk for adjuvant radiotherapy.

PURPOSE: To define the individual factors and combinations of factors associated with increased risk of locoregional recurrence (LRR) that may justify postmastectomy radiotherapy (PMRT) in patients with T1-T2 breast cancer and one to three positive nodes. METHODS AND MATERIALS: The study cohort comprised 821 women referred to the British Columbia Cancer Agency between 1989 and 1997 with pathologic T1-T2 breast cancer and one to three positive nodes treated with mastectomy without adjuvant RT. The 10-year Kaplan-Meier estimates of isolated LRR and LRR with or without simultaneous distant recurrence (LRR +/- SDR) were analyzed according to age, histologic findings, tumor location, size, and grade, lymphovascular invasion status, estrogen receptor (ER) status, margin status, number of positive nodes, number of nodes removed, percentage of positive nodes, and systemic therapy use. Multivariate analyses were performed using Cox proportional hazards modeling. A risk classification model was developed using combinations of the statistically significant factors identified on multivariate analysis.
RESULTS: The median follow-up was 7.7 years. Systemic therapy was used in 94% of patients. Overall, the 10-year Kaplan-Meier isolated LRR and LRR +/- SDR rate was 12.7% and 15.9%, respectively. Without PMRT, a 10-year LRR risk of >20% was identified in women with one to three positive nodes plus at least one of the following factors: age <45 years, Stage T2, histologic Grade 3, ER-negative disease, medial location, more than one positive node, or >25% of nodes positive (all p < 0.05 on univariate analysis). On multivariate analysis, age <45 years, >25% of nodes positive, medial tumor location, and ER-negative status were statistically significant predictors of isolated LRR and LRR +/- SDR. In the classification model, the first split was according to age (<45 years vs. >/=45 years), with 29.3% vs. 13.7% developing LRR +/- SDR (p < 0.0001). Of 123 women <45 years, the presence of >25% of nodes positive was associated with a risk of LRR +/- SDR of 58.0% compared with 23.8% for those with </=25% of nodes positive (p = 0.01). Of 698 women >45 years, the presence of >25% of nodes positive also conferred a greater LRR +/- SDR risk (26.7%) compared with women with </=25% of nodes positive (10.8%; p < 0.0001). In women >45 years with </=25% of nodes positive, tumor location and ER status were factors that could be used to further distinguish low-risk from higher risk subsets.
CONCLUSION: Clinical and pathologic factors can identify women with T1-T2 breast cancer and one to three positive nodes at high LRR risk after mastectomy. Age <45 years, >25% of nodes positive, a medial tumor location, and ER-negative status were statistically significant independent factors associated with greater LRR, meriting consideration and discussion of PMRT. Combinations of these factors further augmented the LRR risk, warranting recommendation of PMRT to optimize locoregional control and potentially improve survival. The absence of high-risk factors identifies women who may reasonably be spared the morbidity of PMRT.