Celldex Offers Safety Profile, Immuno Response Data from Phase 1/2 Study of Varlilumab, Nivolumab at AACR Meeting

Celldex Therapeutics, Inc.

announced today new safety and immune response data from the Phase 1 portion of a Phase 1/2 dose escalation and cohort expansion study examining the investigational combination of varlilumab, Celldex's CD27 targeting investigational immune-activating antibody, and Bristol-Myers Squibb's anti-PD-1 immunotherapy Opdivo® (nivolumab). The data were presented today in a poster at the American Association for Cancer Research (AACR) Annual Meeting 2016 in New Orleans. The Phase 1 portion of the study, conducted in patients with solid tumors, has completed enrollment (n=36) and primarily enrolled patients with colorectal (n=20) and ovarian cancer (n=8). The primary objective of the Phase 1 portion of the study was to evaluate the safety and tolerability of the combination. The Phase 2 portion of the study is open to enrollment. Key Highlights: Combining the potent immune activator, varlilumab, with the PD-1 inhibitor, nivolumab, showed acceptable tolerability and safety across all dose levels without any evidence of increased autoimmunity or inappropriate immune activation. Combination therapy led to marked changes in the tumor microenvironment including increased infiltrating CD8+ T cells and increased PD-L1 expression, which have been shown to correlate with a greater magnitude of treatment effect from checkpoint inhibitors in other clinical studies. Additional favorable immune biomarkers, such as increase in inflammatory chemokines and decrease in T regulatory cells, were also noted. In a subset of patients (n=17) on study who had both pre- and post-tumor biopsies available, preliminary evidence suggest a correlation between biomarker data and stable disease or better in seven of these patients (4 ovarian cancer, 2 colorectal cancer, 1 squamous cell carcinoma of the head and neck). "The combination of varlilumab and nivolumab demonstrated acceptable tolerability across all dose levels of varlilumab, showing that immune stimulation through CD27 was safely combined with PD-1 blockade," said Tibor Keler, Ph.D., Executive Vice President and Chief Scientific Officer of Celldex Therapeutics. "In addition, we observed favorable changes in intratumoral immune biomarkers, most notably an increase in tumor infiltrating lymphocytes, which is recognized to correlate with improved clinical outcome. Based on the strong preclinical data, scientific rationale and these recent results, we are very excited for the Phase 2 portion of the trial, which is now open to enrollment across six different indications." The Phase 2 portion of the study includes cohorts in advanced non-small cell lung cancer (n=35), colorectal cancer (n=18), ovarian cancer (n=18), head and neck squamous cell carcinoma (n=18), renal cell carcinoma (n=25) and glioblastoma (n=20). The primary objective of the Phase 2 study is overall response rate for all cohorts except glioblastoma, where the primary objective is the rate of 12-month overall survival. Secondary objectives include pharmacokinetics assessments, determining the immunogenicity of varlilumab when given in combination with nivolumab and further assessing the anti-tumor activity of combination treatment, including duration of response, time to response, progression-free survival and overall survival. The study is being conducted by Celldex under a clinical trial collaboration with Bristol-Myers Squibb Company. The companies are sharing development costs. Celldex and its collaborating investigators are presenting seven posters at the AACR Annual Meeting. As of Monday, April 18, four of these posters have been presented and summaries of these, including the Phase 1/2 varlilumab/nivolumab combination study, can be found below. Title: Phase 1 results from the combination of an immune activating anti-CD27 antibody (varlilumab) in combination with PD-1 blockade (nivolumab): activation across multiple immune pathways without untoward immune-related adverse events The Phase 1, dose-escalation portion of the study assessed the safety and tolerability of varlilumab at doses ranging from 0.1 to 10 mg/kg when administered with nivolumab (3 mg/kg). Enrollment to the Phase 1 study portion is complete with a total of 36 patients treated. Data for 35 patients are included in the poster: colorectal cancer (n=20), ovarian cancer (n=8), metastatic melanoma (n=4) and head and neck squamous cell carcinoma (n=3). 69% of patients had three or more prior therapies. All dose levels of the combination therapy showed acceptable tolerability and safety, without identification of a maximum tolerated dose. In the Phase 2 portion of the study, varlilumab will be administered at 3 mg/kg, which is based upon cumulative data across multiple studies. The safety profile of the varlilumab and nivolumab combination has been consistent with that of each agent individually, and no unexpected toxicities have been observed. The most frequent treatment related adverse events, occurring in more than 10% of patients, were fatigue (25.7%), lymphopenia (20%), nausea (20%), chills (17.1%), arthralgia (14.3%), pruritus (14.3%) and rash (11.4%), the majority of which were grade 1 or 2. Two patients experienced drug-related serious adverse events. In the 10 mg/kg cohort, grade 4 hepatitis and grade 3 renal insufficiency was observed in a patient with ovarian cancer. Also in the 10 mg/kg cohort, grade 2 paresthesia (tingling/numbness) was observed in a patient with colorectal cancer. Biomarker data from all varlilumab dose levels indicate increases in inflammatory chemokines and decreases in circulating T regulatory cells, which is generally consistent with varlilumab monotherapy. Importantly, in tissue biopsies from patients, the authors noted, where pre-treatment and on-study specimens were available (n=17), a marked increase of tumor infiltrating lymphocytes and an increase in PD-L1 expression. Although the Phase 1 portion of the study was focused on immune response and safety, a correlation between this biomarker readout and stable disease or better (n=7) was observed in this preliminary dataset.