Biweekly neoadjuvant gemcitabine, epirubicin, and nano-particle albumin bound (nab) paclitaxel (GEA) with tumor SPARC analysis correlated with pathologic responses: Results of a multicenter phase II trial

Abstract

11060 Background: The triplet combination of gemcitabine (G), anthracyclines and taxanes have demonstrated significant activity in the metastatic as well as neoadjuvant breast cancer setting with objective response rates of 90% and pathologic CRs of 20–25% with a variety of schedules. SPARC, a poor prognostic indicator, is a glycoprotein expressed in many solid tumors, including breast cancer, and is thought to mediate active tumor transport of nab paclitaxel resulting in high intratumoral concentrations. The present tumor analysis correlates SPARC expression with pathologic responses to neoadjuvant GEA. Methods: Patients with clinical T1c-T4d and/or N0–3, M0 (T1N0M0 excluded) breast cancer, chemo naive, ECOG PS 0–2, normal LVEF, adequate organ function were treated with neoadjuvant G 2000 mg/m2, epirubicin 50 mg/m2, nab paclitaxel (A)175 mg/m2 IV q14 x 6 cycles followed by surgery. Postoperative therapy: G 2000 mg/m2 and A 220 mg/m2 q14 x 4 cycles. Neulasta 6 mg sc D2 was given with all cycles although alternative granulocyte stimulating factors was permitted. Enrollment planned for 120 pts. Results: 72 pts are enrolled, 48 patients evaluable for toxicityand pathologic responses available for 35 pts. Median age: 48 years (29–73), ECOG PS 0 - 45 pts (96%), 1 - 3 (6%), median tumor size 4.5 cm, 79% ductal, 4% lobular, 13% inflammatory. 54% ER-/PR-, 81% HER2 negative. 21 (60%) were triple negative, basal subtype. Heme toxicity was minimal: G3/4 neutropenia: 4 (8%) pts, thrombocytopenia: 3 (6%) pts. There were no episodes of febrile neutropenia. Nonheme toxicity significant for G3/4 arthraglias: 5 (10%) pts & fatigue in 3 (6%) pts. 3 pts hospitalized (infection, PE, abd pain). 1 pt developed PD during neoadjuvant therapy; 5 pts did not complete post op chemo (pt compliance: 2, PD: 2, MD discretion: 1). pCR was noted in both breast and LNs in 7 pts (20%) and pPR in 26 (74%). Conclusions: Neoadjuvant dose dense GEA is feasible and extremely well tolerated. Significant activity with a pCR of 20% is evident with the basal subtype of breast cancer. SPARC tumor analysis correlations with observed pathologic responses will be presented and updated with the continuing accrual to this trial. No significant financial relationships to disclose.