Modeling the tumor microenvironment

In this free on-demand SelectScience® webinar, join Dr. Agapitos Patakas, Chief Scientific Officer at Antibody Analytics as he shares the benefits and limitations of modeling the tumor microenvironment immune cell infiltrate with an in vitro cell-based assay. The assay is well suited to the interrogation of the therapeutic value of targeting checkpoint aces and is benchmarked with an approved checkpoint inhibitor (CPI) e.g. nivolumab.

During the webinar, a T cell exhaustion model will be presented, employing healthy human donor cells. The value of the model in the context of numerous T cell-directed therapies (modality agnostic) will also be discussed. Additionally, the model can be used in a translational sense to investigate combination approaches (e.g. with CPIs) or to determine any additive or synergistic effects of multiple therapeutic targets. The benefits of pre-screening donors to increase assay success and employing multiple, high-content readouts in parallel to maximize data output will also be explored.

Key learning objectives

• Discover how to model T cell exhaustion in vitro
• Understand the power of multi-parameter flow cytometry and kinetic cytotoxicity assays for phenotypic and functional characterization of T cell states
• Explore the range of drug modalities and targets capable of reversing T cell exhaustion
• Understand the power of orthogonal readouts for the determination of T cell effector cell function; MLR; cytotoxicity, and cytokine release

Read on for highlights from the live Q&A session, or watch the webinar on demand, at a time that suits you.

Have you comprehensively characterized gene expression profiles of exhausted T cells from your CD3/CD28 repeat STIM versus T cells exhausted by chronic exposure to tumor cells or peptide/MHC? And what data do you have, besides surface expression of canonical exhaustion markers, suggesting that exhausted T cells generated in your model are similar to T cells exhausted in a tumor microenvironment?

AP: We have characterized, by bulk RNA-sequencing, the cells, which confirms, at a transcript level, the expression of many exhaustion-related markers. There is a difficulty with using bulk RNA-seq in correlating results with existing data, such as from patient material, as most samples are single-cell RNA-seq which makes it more challenging. However, with the information generated from the bulk RNA-seq we can generate the hallmark characteristics of exhaustion in T cells. These are based on the metabolic basis and the expression of different transcription markers and inhibitor molecules, which could be observed in patients, for example. In response to the second half of the question, there will be differences with patient-derived retroviral TILs from patients with chronic infections. We don't expect them to be identical, however most of the hallmarks of exhaustion are encountered in our system, as well as cells displaying dysfunction. We believe that's a good substitute, especially for higher throughput screening of multiple molecules or medium throughput.

Do you prime T cells prior to the HeLa cell assay by co-culture with APCs loaded with HeLa extract?

AP: Specifically for this assay, results are demonstrated with the HeLa cells use a bystander killing approach. In order to stimulate targeted killing, we introduced CD3 antibody in the system. This differs to a targeted approach, as we observed using a bispecific T cell engager for a target expressed better HeLa cells, which is a limitation of this system. We are now moving away from that approach and are using either a bispecific T cell engager or we can generate CAR-T specific, like exhausted, T cells. This represents another specific system that allows us to increase physiological relevance to targeted killing assays.

What is the best antibody for characterization of exhausted T cells?

AP: The typical markers we use are expressions of PD13 and T3, and in some cases, we incorporate description factors, such as TOX or TCF-1 antibodies. This combination provides an element of information towards the cell phenotype. Further, we use functional characterization, an important factor of this model, as many of the listed markers are also expressed by activated T cells.

How do you think strong stimulation with CD3/CD28 Dynabeads™ reflects T cell stimulation in the tumor setting where antigen levels are usually low?

AP: The levels of antigen, such as when there are low levels of MHC Class I results in a different type of stimulation rather than what is observed when using Dynabeads. However, regarding the end results, if you compare chronic viral infections versus the type of stimulation you have in the tumor microenvironment, we don’t expect them to be 100% the same. What we see in the model is that we generate a lot of these characteristics with the strong stimulation that allows us to use it for testing therapeutics at a level one approach. We always suggest using orthogonal approaches, such as using tissue-based approaches or animal models, as a secondary approach. It’s important to note that we use a reductionist approach that allows assessment of molecules in a way that is prohibitive if we use tissues due to limitations in access and amount of material, at least in a human setting.

Learn more about Modeling the tumor microenvironment: Watch this webinar on demand here>>

SelectScience runs 10+ webinars a month across various scientific topics, discover more of our upcoming webinars>>