BACKGROUND: The BCIRG006 study (Slamon, et al. N Engl J Med 2011) demonstrated similar efficacy and superior safety of adjuvant docetaxel, carboplatin, and trastuzumab (TCH) over an anthracycline regimen in HER2+ BC. Dual HER2-blockade with lapatinib (Ty) and H demonstrated synergism preclinically and significant activity in the metastatic and neoadjuvant settings (Blackwell, et al. J Clin Oncol 2012; Baselga, et al. Lancet 2012). This study evaluates the clinical and molecular effects of neoadjuvant TC plus H and/or Ty in HER2+ BC.

METHODS: This is an open-label, randomized phase II study in which pts with stage I-III, operable BC were assigned to 1 of 3 arms: Arm 1-TCH, Arm 2-TCTy, and Arm 3-TCHTy. To test the safety of TCHTy, the first 20 pts received TCHTy. The remaining pts were randomized (1:1:1) to the 3 arms. Pts received a run-in cycle of Ty (1000 mg/d days 1-21) and/or H (8 mg/kg iv), followed by 6 cycles of q3-wkly TC (T: 75 mg/m2, C: AUC 5 or 6) plus H (6 mg/kg) and/or Ty (1000 mg/d d1-21). The first 6 pts were part of a dose-escalation evaluation of C and received AUC5. The remaining pts received C at AUC6. Biopsies were taken at baseline, post run-in cycle, and at surgery. The primary endpoint was pCR rate in each arm (defined as the absence of viable tumor cells in both the breast and axillary lymph nodes). Safety evaluation, molecular effects, and cardiac events were secondary endpoints.

RESULTS: From October 2008 to December 2012, 130 pts were enrolled at 13 centers in the US. Two pts withdrew from study prior to starting any Tx (1 ineligible, 1 withdrew consent) and are excluded from analyses. As of May 2013, complete data was available for 106 pts: 32 in Arm 1, 27 in Arm 2 and 47 in Arm 3. Of these, 16 patients came off study tx prior to surgery (6 in Arm 2, 10 Arm 3) but are included in the ITT analyses. Median age was 48 years (range 27-76). Hormone receptors were both negative (HR-) in 43 pts (41%) and were ER and/or PR positive (HR+) in 63 (59%). At presentation, 5 (5%) pts had clinical stage I, 71 (67%) stage II and 30 (28%) stage III BC. The overall pCR was 42% (45/106), including 43% (13/30) in Arm 1, 25% (7/28) in Arm 2 and 52% (25/48) in Arm 3 (Chi-squared test P = 0.069). Using a pair-wise comparison, pCR is significantly lower in Arm 2 than in Arm 3 (p = 0.021) but no difference was detected between Arms 1 and 2 (p = 0.14) or 1 and 3 (p = 0.45) The pCR in HR+ and HR- tumors were 33% HR+ vs 58% HR- for Arm 1, 13% HR+ vs 42% HR- for Arm 2, and 41% HR+ vs 68% HR- for Arm 3. There were no deaths and no episodes of CHF reported. LVEF decreased >10% from baseline and below the lower limits of normal in 5 pts (1 pt in Arm 1, 3 in Arm 2, 1 in Arm 3). The most common gr >3 AEs in Arms 1/2/3 respectively were pain (10%, 25%, 17%), diarrhea (3%, 11%, 27%), neutropenia (7%, 11%, 13%), anemia (10%, 11%, 4%) hypokalemia (7%, 7%, 8%), and infection (7%, 11%, 6%).

CONCLUSIONS: To our knowledge, this is the first report of this 4–drug combination in the neoadjuvant setting. Final efficacy, safety and molecular analyses from all 128 pts will be presented.

Citation Information: Cancer Res 2013;73(24 Suppl): Abstract nr S1-02.