Patterns of locoregional failure in the CALOR (Chemotheraphy as Adjuvant for Locally Recurrent Breast Cancer) Trial.

Introduction: The risk of developing distant metastasis is elevated in patients (pts) who experience isolated locoregional recurrences (ILRR) after initial breast cancer treatment. Surgical resection of ILRR, radiation therapy (RT), and adjuvant hormonal therapy are common practice strategies. However, the role of chemotherapy (C) is unknown. Pts with complete excision of an ILRR were randomly assigned to C or no C. We report treatments, estrogen (ER) and progesterone receptor (PR) status in primary tumors and ILRR, incidence of second-ILRR failure events, and deaths. Methods: 162 pts were enrolled in IBCSG 27-02, BIG 1-02, NSABP B-37. RT was recommended but not required if a completion/salvage mastectomy was performed after an ipsilateral breast tumor recurrence. Hormonal therapy was required for ER+ and/or PR+ tumors and anti-HER2 therapy was given at investigator discretion. Stratification criteria were hormone receptor status of recurrence, site of recurrence (breast, mastectomy scar/chest, regional lymph nodes), and use of prior C. Results: Primary surgery was breast conserving: 98 (60%) and mastectomy: 64 (40%). Types of nodal surgery for primary cancer included sentinel node resection: 20 (12%), axillary node (AN) sampling: 37 (23%), AN dissection: 88 (54%), more extensive nodal surgery: 6 (4%), and no nodal surgery: 8 (5%). The sites of first ILRR were breast: 88 (54%), mastectomy scar/chest wall: 53 (33%), and regional lymph nodes: 21 (13%), and were similarly distributed by treatment arms. Of the 98 pts who originally had a lumpectomy, 92 (94%) received RT. Among these 92, ILRR were in-breast: 83 (90%), chest wall: 1 (1 %), and regional node: 8 (9%). Salvage mastectomy was performed in 76/98 (78%) with prior lumpectomy. Of the other 22 pts, 17 had repeat lumpectomies and 5 had AN resection. Of the 64 who originally had mastectomy, 52 (81%) recurred in the scar/chest wall, 12 (19%) in the regional nodes and 15 pts (23%) received post-mastectomy RT, 4 of whom had regional nodal recurrences. Of the 143 pts who had ER status reported for the primary tumor, 21 (15%) had a different ER status reported for the ILRR (table). PR expression was discordant in 35 (26%) of the patients with known values. Average time from surgery for ILRR to surgery for second ILRR was 1.6 yrs (range: 0.08 -4.8). The average pt age for second LRR was 58.9 yrs versus 54.6 yrs for those who developed distant failures. At median follow up of 4.9 yrs, 15 pts developed a second ILRR (10 local, 5 regional), 7 (47%) of whom have died. Of the 37 (23%) pts who developed a distant failure 19 (51%) have died. Conclusions Similar to the effect observed in distant failure rates, a second ILRR following multimodality therapy for an ILRR is a strong prognostic indicator of subsequent death in this population. [ABSTRACT FROM AUTHOR]