Abstract A28: Immune responses to mutated Ras - development of a yeast-based immunotherapeutic

Abstract

Abstract Background: While mutations in the ras gene and corresponding product are known to be etiologic in a number of human cancers, efforts to block the mutated protein have been largely unsuccessful leading to its characterization as “undruggable.” We have taken an alternative approach involving generation of T cell responses to the mutated protein. GI-4000 is a series of proprietary immunotherapeutics designed to target cells with activating ras mutations using heat-killed Saccharomyces cerevisiae yeast (named Tarmogens: Targeted Molecular Immunogens) genetically engineered to express Ras G12 and Q61 mutations. Tarmogens activate antigen-specific T cell-mediated immune responses that kill target cells expressing a number of cancer antigens including mutated Ras. Activating mutations in ras occur in > 90% of pancreas cancer cases and >20% cases of NSCLC. GI-4000 has been evaluated in phase 2 trials in both these indications. GI-4000 has demonstrated: i) protection in a murine model of lung cancer (1), ii) in the pancreas trial, improvements in median RFS and OS vs placebo in subjects with a favorable proteomic signature (2), iii) also in the pancreas trial, 3 month improvement in median OS (p=NS) in R1 subjects, with 5 month improvement for immune responders (2), iv) improved OS in subjects with NSCLC treated with GI-4000 compared to case matched controls: HR=0.577, p=NS. In addition, Tarmogens specific for other oncologic targets decreased human regulatory T cells (Tregs) in vitro with a reciprocal increase in effector T cells (3). Here we discuss immunologic outcomes in R0 subjects enrolled in the pancreas cancer trial. Methods: In the pancreas cancer study 176 subjects with Ras mutant+ adenocarcinoma of the pancreas post resection were randomized 1:1 to GI-4000/gemcitabine or placebo/gemcitabine (stratified by resection status: R0/R1). Three weekly injections of GI-4000 or placebo were followed by 6 cycles of gemcitabine 1000 mg/m2 iv (day 1, 8, 15, then every 28 days). Monthly GI-4000 or placebo were administered on gemcitabine off-weeks and continued monthly until disease recurrence, intolerable toxicity or death. R0 subjects (n=102) with adequate blood samples available from timepoints throughout the study were assayed for immune response by interferon-γ (IFNγ) ELISpot assay using Ras peptide pools containing the G12 mutation present in the subject's tumor and a mismatched peptide set identical to the mutation-specific set except at G12. Frequencies of Tregs at baseline and pre-gemcitabine were measured by flow cytometry (n=76). Results: In contrast to the R1 group, there was no increase in Ras mutation-specific IFNγ responses in the R0 GI-4000 group compared to placebo: 16/52 (30.8%) vs 22/50 (44.0%) based on pre-specified criteria. However, naive Tregs (CD4+/CD45RA+/Foxp3low) in the GI-4000 treated group were significantly decreased compared to placebo: 11/42 (26.2%) vs 3/34 (8.8%) subjects had a >2-fold decrease in this fraction (p=0.048 Fisher's exact test). IFNγ responses and OS were strikingly influenced by G12 mutation and associations of specific Ras mutations with outcome will be discussed. Conclusions: GI-4000 decreases frequencies of Tregs, which could be an important attribute for an immunotherapeutic in the treatment of cancer. GI-4000 also generates mutation-specific immune responses and appears to have clinical activity in pancreas cancer and NSCLC. Immune targeting of the activating mutation may be a promising approach to Ras mutated cancers.