Abstract A024: Protein microarrays for the immunological profiling of melanoma

Abstract There is increasing evidence that the aberrant expression of cancer-testis (CT) antigens - a family of ca. 150 proteins that are both auto-immunogenic and mainly restricted to tumors in various types of human cancers - makes them attractive immunotherapy targets, as well as possible cancer diagnostic markers. The underlying hypothesis of our study was that there were measurable differences in auto-antibody repertoires between pre- and post-treatment cancer patient samples, which would correlate with the likelihood, nature and extent of response of these individual patients to a given therapeutic treatment. We carried out a retrospective serological study of primary and secondary autoimmune responses of selected malignant melanoma cancer patients prior to and following kinase inhibitor treatment (either dabrafenib, vemurafenib, trametinib, or dabrafenib and trametinib in combination), using archived human serum samples. Kinase inhibitor treatment has been seen to be associated with increased immune cell infiltrates, raising the possibility that clinical benefit in some of these patients may have, in part, an immune basis, thus justifying the use of an immune monitoring platform for evaluating a treatment approach that is not primarily intended to target immune mechanisms. Therefore, our goals were to explore the utility and general applicability of our recently developed and validated novel cancer antigen protein microarray platform in the cancer immunology field. In addition, we sought to cross-correlate this protein microarray data with both patient clinical information and in vitro T-cell re-stimulation assays for a selected subset of patients, as a means of determining the biological significance of this data. We observed a large number (81% of the cohort) of autoantibody responses – statistically significant changes in antibody titers – to treatment, and successfully identified well-established therapeutic targets – NY-ESO-1 and MAGEB1 – which were the two most abundantly expressed proteins amongst those tested, present across 61% and 56% patients, respectively. Furthermore, our CT antigen array data suggested preliminary evidence of direct correlation between our observed autoantibody responses and the clinically reported ones across a subset of patients (approximately 35% of the cohort). However, it appears plausible that other factors, which we have yet to determine, may be contributing towards a lack of correlation amongst the remaining cohort. Such factors may include patient-specific differences such as tumor-induced systemic immune suppression or evasion mechanisms, the absence or presence of tumor infiltrating lymphocytes and the favorability of the tumor microenvironment. In conclusion, we showed that our novel protein microarray platform represents a sensitive, high-throughput and readily customizable means to detect and quantify the presence of large panels of cancer-specific human autoantibodies in serum, obtaining consistently robust, high quality and reproducible data, and demonstrating its potential feasibility and inferred biological significance. Applications of this tool include prospective use in identifying novel diagnostic, disease progression, prognostic, treatment resistant and even predictive biomarkers, which could aid in the detection and management of cancer. Most importantly, these could lead to the discovery of novel cancer therapeutic targets, patient stratification biomarkers prior to treatment, and a means to monitor cancer patient responses to treatment. Citation Format: Jessica Duarte, Janique Peyper, Katherine Woods, Simon Tsao, Georgina Long, Richard Kefford, Richard Scolyer, Jonathan Cebon, Jonathan Blackburn. Protein microarrays for the immunological profiling of melanoma. [abstract]. In: Proceedings of the CRI-CIMT-EATI-AACR Inaugural International Cancer Immunotherapy Conference: Translating Science into Survival; September 16-19, 2015; New York, NY. Philadelphia (PA): AACR; Cancer Immunol Res 2016;4(1 Suppl):Abstract nr A024.