Identifying cell surface targets for antibody treatment of cancer

Abstract

Whilst cancer immunotherapies have seen progress in their ability to treat patients with several different cancer types, the long-term response rates of even the most promising treatments such as ipilimumab and nivolumab (CTLA-4 and PD-1 checkpoint inhibitors, respectively) remain low for most cancers at around 20%. A strong contributing factor to this is the immunosuppressive nature of the tumour microenvironment (TME). T-regulatory cells (Tregs) are a subset of CD4+ T cells that can be an important component of this TME. Under normal physiological conditions, they dampen the body’s immune responses to self-antigens and thus play an important part in preventing autoimmunity. However, within the TME, they can reduce the efficacy of the host’s immune response against the tumour, so limiting the effectiveness of many current immunotherapies. Selectively targeting and depleting these tumour Tregs is thus an appealing strategy to boost the response rate of current treatments. With this aim, novel cell surface targets expressed on tumour Tregs were sought using bioinformatic approaches. This was done by filtering mouse Treg RNA-seq data for genes with suitable characteristics for a therapeutic target, which includes differential expression in Tregs, presence on the cell surface and homology to human genes. 130 murine genes were found to match these criteria, which were then ranked based on their enrichment in Tregs versus other immune cell populations. To test the accuracy of the bioinformatic filtering process, mRNA and cell surface protein expression levels of already established Treg-expressed genes (TNFRSF members) were compared between RNA-seq and flow cytometry, showing high concordance. Following these analyses, 7 potential targets were identified for further laboratory analysis: Zan, Itgb8, Ccr8, Gpr83, Tigit, Klrg1 and Lrrc32. Based upon reagent availability, Ccr8, Gpr83, Tigit, Klrg1, Itgb8 (αvβ8) and Lrrc32 (GARP) were evaluated for cell surface protein expression on mouse immune cells within the tumour. Target expression was further assessed using RNA-seq data from human and mouse tumour infiltrating leukocytes (TILs). Of the 7 characterised targets, ITGB8/αvβ8 showed promising expression patterns and target biology and as a result was selected to generate novel antibodies against. Antibodies were generated against αvβ8 by phage display selection using BioInvent’s nCoDeR library, with 460 single chain variable fragments (scFvs) being screened for binding. Of these, 192 showed effective binding and were sequenced, resulting in 47 unique clones. 43 of these were then converted to full IgGs and their binding titrations and functional abilities tested. One showed functional activity against the target. In summary, this project identified 130 tumour Treg targets, shortlisted and characterised the expression of 7 of these and culminated in the generation and in vitro characterisation of 43 novel antibody clones against the most promising target, αvβ8.

More information

Identifiers

ORCID for Timo Leopold Kuerten: orcid.org/0000-0002-8541-838X

ORCID for Kirstie Cleary: orcid.org/0000-0001-6200-4945

ORCID for Mark Cragg: orcid.org/0000-0003-2077-089X

Catalogue record

Export record