The translation products of cancer-selective or -specific coding sequences are naturally suited for development as targets for virtually any therapeutic modality because of their potential to be targeted without adversely affecting normal tissues. We hypothesize that the translation products of prevalent cancer-selective transcripts (CSTs) first identified in the TCGA may be enriched for neo-antigens that are selectively presented by MHC class I or II molecules on the surface of malignant cells, and therefor represent attractive targets for immunotherapy. The most highly promising targets can be identified by a computational approach that integrates the TCGA RNA sequence data with well curated data resources profiling somatic tissues and with valuable local data resources that allow us to infer the translational activity of the CSTs and predict which CSTs are that are not derived from tumor infiltrating cells. Our preliminary work shows that we can identify, in each of several cancers we have evaluated, large numbers of CSTs that are (i) expressed in cancer tissues from multiple different patients, (ii) apparently absent in untransformed adult somatic tissues, and (iii) are derived from malignant cell component of the solid tumor, and (iv) are translationally active. CSTs we identify include families of alterations that have previously been shown to generate immunogenic peptides that can elicit CD8+ T cell immune responses, and they have been found in genes of central importance to cancers of those organ sites. While the identification of an apparently cancer-selective transcript in cancer tissue samples does not guarantee the presence of a neoantigen, it does provide a cost-effective and logical starting point for further analysis given the availability of proven high throughput verification approaches Our project will:
Aim 1) for each TCGA cancer site estimate prevalent CSTs and also predict the translation state for CSTs identified in ovarian, colon, and lung cancer patients.
Aim 2) for ovarian, colon, lung cancer identify which predicted cancer-selective polypeptides harbor epitopes that are recognized by CD4+ or CD8+ T Cells.
Aim 3) prospectively validate lung cancer antigens by determining whether patient immune-response is enhanced following disruption of T cell inhibitory pathways in samples collected from Phase III trials.
The potential for immunotherapy to treat the most challenging human cancers has been demonstrated in preclinical models and in clinical trials but progress has been hindered because the most common solid tumors tumor specific antigens (targets) are not yet well defined. Our project intends to identify large numbers of new targets that may be suitable for a variety of therapeutic approaches, and for a number of different tumor sites.
|Cancer Target Discovery and Development Network (2016) Transforming Big Data into Cancer-Relevant Insight: An Initial, Multi-Tier Approach to Assess Reproducibility and Relevance. Mol Cancer Res 14:675-82|
|Walter, Roland B; Laszlo, George S; Alonzo, Todd A et al. (2013) Significance of expression of ITGA5 and its splice variants in acute myeloid leukemia: a report from the Children's Oncology Group. Am J Hematol 88:694-702|
|Chou, Jeffrey; Fitzgibbon, Matthew P; Mortales, Christie-Lynn L et al. (2013) Phenotypic and transcriptional fidelity of patient-derived colon cancer xenografts in immune-deficient mice. PLoS One 8:e79874|