Pathologic tumor size and lymph node status predict for different rates of locoregional recurrence after mastectomy for breast cancer patients treated with neoadjuvant versus adjuvant chemotherapy

Purpose: To compare the pathologic factors associated with postmastectomy locoregional recurrence (LRR) in breast cancer patients not receiving radiation who were treated with neoadjuvant chemotherapy (NEO) vs. adjuvant chemotherapy (ADJ). Methods and Materials: We retrospectively analyzed the rates of LRR of subsets of women treated in prospective trials who underwent mastectomy and received chemotherapy but not radiation. These trials were designed to answer chemotherapy questions. There were 150 patients in the NEO group and 1031 patients in the ADJ group. In the NEO group, 55% had clinical Stage IIIA or higher vs. 9% in the ADJ group (p <0.001, chi-square test). Results: Despite the more advanced clinical stage in the NEO group, the pathologic size of the primary tumor and the number of positive lymph nodes (+LNs) were significantly less in the NEO group than in the ADJ group (p <0.001 for both comparisons). However, the 5-year actuarial LRR rate was 27% for the NEO group vs. 15% for the ADJ group (p = 0.001, log - rank). The 5-year risk for LRR was higher in the NEO patients for all pathologic tumor sizes: 0-2 cm (18% vs. 8%, p = 0.011), 2.1-5 cm (36% vs. 15%, p <0.001), and >5 cm (46% vs. 28%, p = 0.028). The risk of LRR by the number of +LNs was similar in the NEO and ADJ groups, except for the subset of patients with ≥4 +LNs (53% vs. 23%, p <0.001). The rates of LRR in the patients with primary tumors measuring ≤2.0 cm and 1-3 +LNs were similar in both groups. However, for the patients with a pathologic tumor size of 2.1-5.0 cm and 1-3 +LNs, the LRR was higher in the NEO group than in the ADJ group (30% vs. 15%, p = 0.016). Most failures in this NEO subgroup had clinical Stage III disease. In a subset of NEO and ADJ patients matched for clinical stage, no significant differences were found in the rates of LRR according to primary tumor size and number of +LNs when these variables were analyzed independently. Again, however, differences were found in the subgroup of patients with tumors pathologically measuring 2.1-5.0 cm with 1-3 +LNs (32% NEO vs. 8% ADJ, p = 0.030). Conclusion: The rates of postmastectomy LRR for any pathologic tumor size are higher for patients treated with initial chemotherapy than for patients treated with initial surgery. Radiotherapy should be offered to all patients with ≥4 +LNs, tumor size >5 cm, or clinical Stage IIIA or greater disease, regardless of whether they receive neoadjuvant or postoperative chemotherapy. The information assessing LRR rates in patients with clinical Stage II disease who receive neoadjuvant chemotherapy, particularly if 1-3 lymph nodes remain pathologically involved, is insufficient to determine whether these patients should receive radiotherapy.