Many cancers become desensitized to therapy by upregulating autocrine or paracrine growth factors that activate antiapoptotic signaling pathways.7-9 Notably, heregulin (HRG) -driven ErbB3 signaling mediates insensitivity to a broad range of therapies by activating the phosphoinositide 3-kinase (PI3K)/Akt signaling pathway.10-15 In ovarian cancer, ErbB3 promotes cell proliferation in preclinical models, and approximately 30% of patients show evidence of an HRG/ErbB3 autocrine loop in tumor cells derived from malignant ascites.16 Together, these data suggest that blocking HRG/ErbB3 may increase sensitivity to therapy when this pathway is active. Seribantumab (MM-121; Merrimack Pharmaceuticals, Cambridge, MA) is a fully human immunoglobulin G2 antibody that targets ErbB3, blocks HRG from binding, and downregulates the receptor.17 Here, we describe the results of a phase II study in which women with advanced platinum-resistant or -refractory ovarian cancer were randomly assigned to receive once-per-week paclitaxel or paclitaxel in combination with seribantumab. plus paclitaxel compared with 3.68 months with paclitaxel alone (hazard ratio [HR], 1.027; 95% Goat polyclonal to IgG (H+L)(Biotin) CI, 0.741 to 1 1.425; = .864). Among patients whose tumors had detectable HRG mRNA and low HER2 (n = 57 [38%] of 151 with available biomarker data), increased treatment benefit was AG-494 observed in those receiving seribantumab plus paclitaxel compared with paclitaxel alone (PFS HR, 0.37; 95% CI, 0.18 to 0.76; = .007). The HR in patients not meeting these criteria was 1.80 (95% CI, 1.08 to 2.98; = .023). Conclusion The addition of seribantumab to paclitaxel did not result in improved PFS in unselected patients. Exploratory analyses suggest that detectable HRG and low HER2, biomarkers that link directly to the mechanism of action of seribantumab, identified patients who might benefit from this combination. Future clinical trials are needed to validate this finding and should preselect for HRG expression and focus on cancers with low HER2 levels. INTRODUCTION Ovarian cancer is a leading cause of cancer-related death among women, with approximately 240, 000 cases diagnosed annually.1 Although most patients respond to initial treatment, most eventually develop resistance to platinum-based therapy. At this stage, single-agent therapies such as once-per-week paclitaxel confer a median progression-free survival (PFS) benefit of approximately 3 to 6 months.2-6 Patient outcomes could potentially be improved by identifying mechanisms of drug resistance and developing drugs that effectively combat such mechanisms. Many cancers become desensitized to therapy by upregulating autocrine or paracrine growth factors that activate antiapoptotic AG-494 signaling pathways.7-9 Notably, heregulin (HRG) -driven ErbB3 signaling mediates insensitivity to a broad range of therapies by activating the phosphoinositide 3-kinase (PI3K)/Akt signaling pathway.10-15 In ovarian cancer, ErbB3 promotes cell proliferation in preclinical models, and approximately 30% of patients show evidence of an HRG/ErbB3 autocrine loop in tumor cells derived from malignant ascites.16 Together, these data suggest that blocking HRG/ErbB3 may increase sensitivity to therapy when this pathway is active. Seribantumab (MM-121; Merrimack Pharmaceuticals, Cambridge, MA) is a fully human immunoglobulin G2 antibody that targets ErbB3, blocks HRG from binding, and downregulates the receptor.17 Here, we describe the results of a phase II study in which women with advanced platinum-resistant or -refractory ovarian cancer were randomly assigned to receive once-per-week paclitaxel or paclitaxel in combination with seribantumab. Additionally, this study was designed to examine five potential biomarkers of ErbB3 activation that had been previously highlighted by computational modeling: HRG, ErbB3, human epidermal growth factor receptor 2 (HER2), epidermal growth factor receptor (EGFR), and betacellulin (BTC).15,17 In preclinical studies, the levels of these proteins predicted response to seribantumab in mouse xenograft models.15 The hypotheses underlying these biomarkers are summarized in Appendix Table A1 (online only). Because it was not known how these biomarkers are altered by disease progression, they were measured in both archived tissue and mandatory pretreatment biopsies. PATIENTS AND METHODS Study Design This was a multicenter, open-label, randomized phase II study of seribantumab combined with paclitaxel versus paclitaxel alone in patients with advanced platinum-resistant or -refractory ovarian cancer. Patients were randomly assigned at a ratio of two to one to receive seribantumab with paclitaxel or paclitaxel alone, with two stratification factors: Eastern Cooperative Oncology Group (ECOG) performance status18 (0 to 1 1 2) and number of prior therapies (one to two three or more). The primary end point was PFS as assessed by RECIST (version 1.1).19 Secondary objectives included correlation of PFS with each of five biomarkers, efficacy of combination treatment with regard to overall survival (OS) and objective response rate (ORR), assessment AG-494 of treatment according to health-related quality-of-life (HRQOL) scores, further characterization of the safety profile of seribantumab, and examination of predictive biomarkers. Patients Eligibility criteria included confirmed advanced or recurrent epithelial ovarian, fallopian tube, or primary peritoneal cancer that was platinum resistant or refractory per Gynecologic Oncology Group criteria.20,21 Patients underwent mandatory pretreatment core needle biopsy and also submitted archived tumor samples, as available. All patients supplied written informed consent, and local institutional review boards and/or ethics committees approved the study protocol. Study Procedures Patients were randomly assigned at a ratio of two to one to either the experimental (seribantumab plus paclitaxel) or AG-494 control arm (paclitaxel alone). Seribantumab was administered intravenously (40 mg/kg loading dose, then 20 mg/kg once per week). Paclitaxel was administered intravenously (80 mg/m2 during cycle one, with optional modification in subsequent cycles to 80 mg/m2 once per week for 3 weeks followed by 1 week of.