Immune checkpoint inhibitors are novel therapeutic agents directed against immune checkpoints- “the brakes of the immune system” that tumors frequently evade to protect themselves. Despite the impressive clinical activity of such drugs in patients with a variety of metastatic cancers, tumor relapse is inevitable.
At the ASCO 2021 virtual annual meeting, Dr. Antoni Ribas from UCLA, Dr. Patrick Alexander Ott from Dana Farber together with Dr. Kurt A. Schalper from Yale University teamed to discuss resistance mechanisms of checkpoint inhibitors, potential therapeutic combinations, neoantigen landscape, tumor mutational burden, and the role of tumor microenvironment in cancer immunotherapy.
Therapeutic resistance can be of two types, viz, primary resistance- wherein patients generally do not respond to therapy, and acquired resistance- where patients respond initially but still undergo disease progression.
Resistance Mutations in Cancer
Whole-exome sequencing in melanoma unveiled a novel truncating mutation in the beta-2-microglobulin (B2M) gene and homozygous loss-of-function mutations in Janus kinase 1 (JAK1) and Janus kinase 2 (JAK2). While B2M inactivating mutation leads to loss of surface expression of MHC class I, JAK1/JAK2 loss-of-function mutation results in defective IFN-γ–induced growth arrest in relapsed tumors.
Combination therapy with a CD122-preferential IL-2 pathway agonist, bempegaldesleukin (NKTR214), and nivolumab reverse B2M loss-of-function resistance by inducing activation and expansion of other effector cells independent of MHC I, such as CD4 T cells and NK cells.
Notably, co-treatment with TLR-9 agonist (CMP-001) and anti-PD-1 increases tumor infiltration of T cells and NK cells and overcomes resistance to JAK1/JAK2 loss-of-function mutation. Activation of dsRNA sensors such as TLR-3, protein kinase R (PKR) in combination with anti-PD-1 restores MHC I expression, thus vindicating hopes in JAK1/JAK2 loss-of-function relapsed tumors.
Multiple studies in different cancers have shown a correlation between the efficacy of checkpoint therapy and neoantigen load. Neoantigens are attractive therapeutic targets encoded by somatic alterations exclusively present in cancer cells.
Several factors such as DNA damage, neoantigen non-recognition, clonal distribution and types of mutations, overexpressed cell antigens, and other native antigens impact tumor mutational burden (TMB).
In silico neoantigen prediction, either alone or complemented with mass spectrometry analysis of MHC-associated neoantigens, or in vitro T-cell validation assay or a robust ex vivo ATLAS bioassay. serves as an alternative approach to identify neoantigens.
Neoantigen cancer vaccines and adoptive cell therapy (ACT) can induce endogenous neoantigen-specific T-cell responses. The key strengths of neoantigen vaccines lie in the breadth of the immune response, scalability, and feasibility for the development in early-stage disease, while cell-based therapy supposedly elicits a stronger immune response and overcomes immunosuppression.
Key findings from initial clinical trials of personalized neoantigen vaccine NeoVax together with TLR-3 agonist (poly IC-LC) in high-risk melanoma and glioblastoma demonstrated that the neoantigen vaccine is safe, feasible, and elicits a strong T cell immune response.
Recent advances in the field of multiplex tissue analysis, single cell sequencing have facilitated a better understanding of tumor microenvironment (TME), crucial in cancer therapeutics. In solid tumors, TME is composed of a tumor compartment and a stromal bed. Different cells in the TME exhibit variable expression of epitopes, and the underlying signaling cascade in the TME impacts disease prognosis and treatment.
In ~80% of NSCLC and metastatic melanoma, upregulation of other immune inhibitory molecules like PDL-1 such as B7-H3 and FGL1 is associated with worse overall survival. Bi-allelic loss of B2M gene is associated with immune evasion and acquired resistance to immunotherapy. Upregulation of immunosuppressive cytokine IL-8 in the plasma of NSCLC patients is related to poor outcomes.
Single cell analysis with CytoF facilitated the characterization of a sub-population of effector T-cells, called effector burned-out T-cells (Ebo) in TME of NSCLC. These cells are uniquely characterized by a high proliferation rate, upregulation of key immune inhibitory molecules, and overexpression of pro-apoptotic molecules. Essentially, Ebo cells take over the TME, and blocking the PD-1 axis inhibits the clonal expansion of these cells. However, further studies are warranted to understand the conversion of Ebo cells into effector cells in cancer immunotherapy.
TME analysis using single-cell technologies, advanced tissue imaging, and integrated computational analysis will further strengthen the understanding of TME, vulnerabilities of treatment resistance, and rational design of cancer immunotherapy.
Thoughts by Key Opinion Leaders
When asked to shed some light on the possibility of different resistance mechanisms across different tumor types, the panelists opined that there could be some dominant mechanism of resistance or some commonalities spanning all tumor types. However, the resistance mechanisms might differ based on the type of resistance viz, primary or acquired. Immunogenicity of cancer, the activity of the immune system, the role of effector T-cells, convergence in antigen-presenting machinery, IFN-γ signaling may play a role in defining resistance mechanism.
Regarding prioritizing the treatment strategies in the midst of immunotherapy resistance, all the panelists agreed that identification of the mechanism of resistance, characterizing the dominant mechanism would be ideal in prioritizing the treatment modality.
When asked about the mutational basis of the emergence of resistance pathways, Dr. Ribas emphasized that mutation is not the norm of the resistance pathways instead, epigenetic changes that result in those mutations may be more common. However, elucidation of epigenetic changes underlying the tumor resistance mechanism warrants further investigation. Rapid advancements in precision single-cell technologies may facilitate a detailed understanding of the non-mutational basis of tumor resistance.
Dr. Antoni Ribas, UCLA
Regarding the response of patients to immunotherapy strategies with very low tumor mutational burden and potentially low neoantigens, the panelists agreed that there is no linear correlation between TMB and response to immunotherapy instead, it largely depends on the functional state of the effector T-cells and the identification of the targetable epitope either native or neoantigen to generate an effective anti-tumor immune response.
Dr. Patrick also shed light on the stability of the neoantigens during the evolution of the disease or as a result of treatment. He emphasized that analysis of mutational landscape suggests the evolution of neoantigens. Yet, the mechanisms of evolution of neoantigens and their role in cancer immunotherapy are still elusive.
Dr. Patrick Alexander Ott, Dana Farber Cancer Institute
About mechanisms of dysfunction of CD4 and CD8 T-cells, Dr. Schalper mentioned that there are some similarities in the markers between CD4 and CD8 compartments however, the complexity in the structure and dynamics of the TME cannot be precluded in attributing differences to the mechanisms of dysfunction of CD4 and CD8 T-cells.
Dr. Kurt A. Schalper, Yale University
On the subject of the potential strengths and limitations in the usage of Class II antigens for diagnostic prediction and identification of therapeutics, all the panelists agreed that the HLA loci are the most polymorphic locus in the human genome. The peptide binding to HLA Class II is complex in structure, therefore, prediction based on Class II warrants further investigation. Studying neoantigens, variable peptide presentation, differentia expression of immunogenic proteins as a result of mutations poses challenges in studying neoantigens and prediction of the immune response.