Whole Blood RNA Transcript Models Predict Clinical Response in Two Clinical Studies Advanced Melanoma Patients Treated with Tremelimumab

Background: Tremelimumab is an antibody that blocks CTLA-4 and demonstrates clinical efficacy in a subset of advanced melanoma patients. An unmet clinical need exists for blood-based response-predictive gene signatures to facilitate clinically effective and cost-efficient use of such immunotherapeutic interventions. Methods: Peripheral blood samples were collected in PAXgene® tubes from 210 treatment-naïve melanoma patients receiving tremelimumab in a worldwide, multicenter phase III study (discovery dataset). A central panel of radiologists determined objective response using RECIST criteria. Gene expression for 169 mRNA transcripts was measured using quantitative PCR. A 15-gene pre-treatment response-predictive classifier model was identified. An independent population (N=150) of refractory melanoma patients receiving tremelimumab after chemotherapy enrolled in a worldwide phase II study (validation dataset). The classifier model using the same genes, coefficients and constants for objective response and one-year survival after treatment was applied in the validation dataset. Results: A 15-gene pre-treatment classifier model (containing ADAM17, CDK2, CDKN2A, DDP4, ERBB2, HLA-DRA, ICOS, ITGA4, LARGE, MYC, NAB2, NRAS, RHOC, TGFB1, and TIMP1) achieved an area under the curve (AUC) of 0.86 (p< 0.0001) in the discovery set. This model was validated in the validation set with AUCs of 0.641 (p= 0.049) for objective response and 0.68 for one-year survival (p=0.0002). Conclusions: To our knowledge, this is one of the largest blood-based biomarker studies of a checkpoint inhibitor, tremelimumab, which demonstrates a validated pre-treatment mRNA classifier model that predicts clinical response. The data suggest that the models capture biological signatures representative of genes needed for a robust anti-cancer immune response. They also identify non-responders to tremelimumab significantly prior to radiological evidence of progression. Patient Population and Gene Selection: Both the discovery and validation datasets resulted from pre-treatment blood samples collected in multinational, open-label studies of tremelimumab administered to advanced melanoma patients. Only patients in which both a pre- and post-treatment blood sample were available were included in our analysis. The pre-treatment discovery dataset was a randomized phase III study which enrolled 210 patients, with N=28 responders and N=182 non-responders. The pre-treatment validation dataset was a non-randomized phase IIb study which enrolled 150 patients, with N=20 responders and N=130 non-responders. In both studies response was determined by RECIST criteria. The patients in the pre-treatment discovery dataset were treatment-naïve, while the patients in the pre-treatment validation dataset were chemotherapy-refractory. The selection process for the 169 genes tested included genes associated with inflammation, immunity, the CTLA4 pathway, oncogenes, and melanoma-specific genes. Whole-blood was collected in PAXgene® tubes and processed to RNA that met quality and integrity standards (RNA integrity number ≥6.3) per the Bioanalyzer 2100 in combination with the RNA 6000 Nano or Pico Series II LabChip. First-strand complementary DNA was synthesized from random hexamer-primed RNA templates using TaqMan® reverse-transcription reagents. Individual target-gene amplification was multiplexed with the 18S rRNA endogenous control and run in triplicate in 384-well format on the 7900HT fast real-time PCR system.