Multi-omic profiling of metastatic lesions to guide treatment selection: The Side Out 2 trial experience.

1077 Background: The aim of this prospective pilot study, the Side Out 2 trial (NCT01919749), was to explore if treatment selection based on Multi-omic Profiling (MoP) provides clinical benefits superior to empiric treatment selection in progressive metastatic breast cancers (MBC). Methods: Fresh core biopsies were collected from the metastatic lesions of 32 MBC patients. Samples underwent genomic and proteomic profiling, including: exome sequencing, RNA-Seq, IHC, and quantitative phosphoprotein-based protein pathway activation mapping by Reverse Phase Protein Microarray (RPPA). Whole tissue lysates were used for the exome sequencing, RNA-Seq and IHC; the RPPA data were obtained from microdissected tumor epithelia. A sample size of 25 evaluable patients was determined a priori according to the exact single-stage design for phase II studies, using a type I error rate of 5% (one-sided) and a power of 90%. Clinical benefit was defined as Growth Modulation Index (GMI) ≥ 1.3 assuming that MoP selected therapy would warrant further investigation if ≥ 35% of the patients demonstrate a PFS ratio of ≥ 1.3. GMI was calculated as a ratio between PFS on MoP selected therapy versus PFS on prior therapy (PMID:25209003). Results: Between 2014 and 2016, four US sites enrolled 32 previously treated MBC patients. Of the 32 patients enrolled, 29 received treatment based on their MoP and 25 met the follow-up criteria established by the trial protocol. Of the 25 patients, 14 (56%) met or exceeded a GMI of 1.3. The most frequently selected treatments were irinotecan based on TOPO1 expression (n = 12; single agent n = 6) and capecitabine based on TS expression (n = 9; single agent n = 3). Of the 7 patients that received endocrine therapy, 3 were treated with exemestane+everolimus. Based on HER2 amplification/pathway activation, HER2 targeted agents were given to 4 patients. Conclusions: This study confirmed the unique role of MoP to select effective treatments for MBC. This approach provided clinical benefit for 56% previously treated MBC patients, which met the primary objective of the study. Thus, this approach merits further investigation. This study also suggests that irinotecan may be an under-developed drug for MBC patients. Clinical trial information: NCT01919749.