A pilot study utilizing molecular profiling to find potential targets and select individualized treatments for patients with metastatic breast cancer.

TPS11123 Background: The objective of this prospective pilot study was to determine if treatment selected by molecular profiling (MP) of metastatic breast cancer (MBC) tumors at time of disease progression provides greater clinical benefit than empiric treatment selection. Methods: Eligible pts had MBC, ≥ 3 prior lines of therapy, and > 6 wks - < 6 months time to progression on last treatment prior to study entry. Fresh core biopsy samples were taken from the metastatic site for MP. Analysis by IHC, FISH, DNA microarray and reverse phase protein microarray (RPMA) a quantitative protein pathway activation mapping technique using laser capture micro dissected tumor cells was done, with results in ≤14 days reviewed by a Treatment Selection Committee. The primary objective was to compare the progression-free survival (PFS) using a MP selected treatment regimen to the preceding PFS. Clinical benefit for the pt is defined by PFS ratio (PFS on MP selected therapy/PFS on prior therapy) is ≥ 1.3, (JCO,28:4877-83:2010). N= 25 was determined (exact single-stage design for phase II studies, type I error rate of 5% [one-sided], power of 90%). MP selected therapy would warrant further investigation if ≥ 35% of the pts demonstrate a PFS ratio of ≥ 1.3. Results: Three U.S. sites enrolled 25 evaluable pts who were treated based on MP results. Pts were heavily pretreated with 4 -11 prior therapies (median 7.24). Ten pts (40%) met or exceeded the PFS ratio of 1.3. The most common targets used in drug treatment selection were TOPO1, TS, ER/PR, TOP2A, HER2. Sixty percent of pts’ samples had activation of drug targets determined by RPMA in 3 major clusters: pan-HER-AKT; EGFR/SRC/ERK/mTOR; IGFR/RAF/MEK/PLK1. Days on MP selected treatments range from 9 - 816+. 3 pts continue on treatment for 199, 254 and 816 days. Conclusions: This study demonstrates the feasibility of a highly multiplexed genomic and proteomic MP study for treatment selection in a timely fashion. Patient-specific target driven treatment selection based on MP of a metastatic lesion provided clinical benefit for 10 of 25 heavily pretreated MBC pts. Thus, this approach merits further investigation in future studies. Funded by The Side-Out Foundation. Clinical trial information: NCT01074814.