Abstract CT105: Randomized trial of dendritic vs tumor cell patient-specific vaccines: 5-year analysis

Autor: Robert O. Dillman, Thomas Amatruda, Carol DePriest, Denysha Carbonell, Andrew N. Cornforth, Edward F. McClay
Rok vydání: 2016
Předmět:
Zdroj: Cancer Research. 76:CT105-CT105
ISSN: 1538-7445
0008-5472
DOI: 10.1158/1538-7445.am2016-ct105
Popis: Eltrapuldencel-T (CLBS20) consists of autologous dendritic cells loaded with antigens from irradiated, self-renewing, autologous tumor cells that potentially present the entire repertoire of unique patient-specific tumor-associated antigens resulting from nonsynonymous mutations in each patient's melanoma tumor cell line. In a single-arm phase 2 trial, metastatic melanoma patients treated with s.c. injections of CLBS20 had a 2-year overall survival (OS) of 73% (NCT00948480). In a randomized phase 2 trial (NCT00436930), 2-year OS was 72% compared to 31% for a tumor cell vaccine (TCV) consisting of irradiated cancer cells from an autologous tumor cell line. Mild local injection site reactions was the most common toxicity. This report focuses on 5-year follow up data from the randomized trial. During October 2007 to February 2011 42 patients were randomized 1:1 to receive CLBS20 or TCV. After resection of a metastatic tumor, if/when a cell line was established, patients were eligible for randomization if/when they were referred by their managing physician. Both products were mixed in 500 micrograms GM-CSF and injected s.c. weekly for three weeks, then monthly for up to 5 months. Cell lines were successfully established rapidly enough for possible clinical use for 78/183 (43%); 42/78 (54%) patients were referred for randomization. The median time from tumor submission to cell line success was 3.0 months, then 2.5 weeks for safety testing and documentation, then another 3.7 months to randomization. At the time of tumor harvest 24 patients were stage 4 and 18 were recurrent stage 3; at randomization 33 were stage 4 and 9 stage 3. Trends toward imbalances of baseline characteristics were biased against the CLBS20 arm (elevated LDH, detectable disease, brain metastases). There were no differences in tumor resection site, days in cell culture, or days to randomization. Because of leukapheresis and dendritic cell production for CLBS20, the median time from randomization to first dose was 43 days for CLBS20 vs 8 days for TCV. All patients were treated as randomized; survival was calculated from randomization date. At this analysis, 33 (79%) were dead and the 9 survivors had been followed 5 years. OS was higher in the CLBS20 arm: median 42.2 vs 19.9 months for all and 40.4 vs 16.9 for stage 4 patients, and in subsets defined by measurable disease, and serum LDH. The only variable associated with 3-year OS was randomization to CLBS20 (p = 0.018). A patient with refractory progressive measurable disease experienced a delayed complete response that was ongoing at 5 years. In an era before widespread use of anti-BRAF/anti-MEK and anti-checkpoint agents, CLBS20 monotherapy was associated with encouraging 5-year OS. Because of its unique mechanism of action, absence of toxicity, and apparent benefit regardless of tumor burden, CLBS20 is worthy of further evaluation as a monotherapy or in combination with other anti-melanoma therapies. Citation Format: Robert O. Dillman, Edward F. McClay, Thomas T. Amatruda, Carol DePriest, Denysha J. Carbonell, Andrew N. Cornforth. Randomized trial of dendritic vs tumor cell patient-specific vaccines: 5-year analysis. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr CT105.
Databáze: OpenAIRE
Externí odkaz: