Background: Clinical trials for HER2-positive metastatic BC excluded pts with prior AdjT and/or relapsing within 1 year (yr) from the end of AdjT. With this gap in mind, we aimed to address the presentation, the therapeutic options, and the clinical outcome of BC pts after failure to AdjT. Methods: Medical charts of BC pts relapsing on AdjT and admitted at the National Cancer Institute of Milan were reviewed. Fisher's exact test was used for contingency tables, long-rank test for survival distribution; p<0.05 was considered statistically significant. Data were matched with literature. Results: Forty pts were identified in the 2013 first quarter. Median age at relapse was 49 (26-80) yrs; 67.5% of pts had stage III; 93% neo-/adjuvant anthracycline+/-taxane; 55% hormonotherapy. Median TTR was 23.5 (5.2-108.5) months (mos). Relapses occurred during (early) or after (late) AdjT in 12 (30%) and 28 (70%) pts. Median AdjT cycles were 15 (2-18) and 18 (6-19) in pts with early and late relapses. Early and late relapses shared a similar distribution of hormone receptor positive and negative status (35% vs 25%, p = 0.73). Lung (8% vs 36%, p = 0.12), bone (25% vs 36%, p = 0.72) and multiple (8% vs 36%, p = 0.12) metastases were less likely to occur in pts with early than in late relapses; liver was affected by 42% and 29% of early and late relapses (p = 0.48). T was the 1stline tretament in 77.5% of pts. After a median follow-up of 21.3 (0.6-98.6) mos, 29 pts (72.5%) progressed. Median Time To Progression (TTP) was 12.7 (9.0-15.3) mos, with a response rate (RR) of 61%. Pts relapsing during AdjT had a shorter TTP than pts relapsing after AdjT, 9.2 (95%CI:3.9-13.3) vs 15 mos (95%CI:9.1-27.9), p = 0.048. T was the 2ndline treatment in 65.5% of pts; with a RR 51%. The literature review showed that pts with prior AdjT represent 0.9-11.7% of overall study-population in just 5 of 8 registrational trials in HER2-positive metastatic BC. No trials included pts relapsing during AdjT. Compared to pts from clinical practice, pts of the above trials were less likely to have locally-advanced BC at diagnosis and neo-/adj chemotherapy. Conclusions: T was the choice in both 1st and 2ndline treatment of pts relapsing on AdjT. Anti-tumor activity was consistent accross lines and with literature. These data should be interpreted with caution, as the cross-sectional design may have favored the inclusion of pts at good prognosis. Notwithstanding the above, pts with early relapse to AdjT showed a worse outcome as compared to the rest. Further investigation to tailor therapy for these pts is recommended. Patient data collection continues and an update will be presented at the meeting.

Citation Information: Cancer Res 2013;73(24 Suppl): Abstract nr P4-12-22.