Over the last decade, immunotherapeutic options have led to impressive clinical responses in patients with various cancer types and this has increased expectations for successful treatment of the disease. Despite immunotherapy results leading to clinical trials in melanoma, renal cell carcinoma and non–small cell lung cancer, the percentage of patients who respond to immunotherapy remains low; this highlights the need to identify the patient population that will best respond to these approaches.
Jedd D. Wolchok, MD, PhD
Over the last decade, immunotherapeutic options have led to impressive clinical responses in patients with various cancer types and this has increased expectations for successful treatment of the disease. Despite immunotherapy results leading to clinical trials in melanoma, renal cell carcinoma and nonsmall cell lung cancer, the percentage of patients who respond to immunotherapy remains low; this highlights the need to identify the patient population that will best respond to these approaches.
One of the main immunotherapeutic strategies is immune checkpoint blockade, the inhibition of the immune system’s intrinsic regulatory mechanisms to drive the activation of an improved anti-cancer immunological response.1 An improved understanding of the mechanisms that cancer cells use to hide from the immune system has contributed to the development of new approaches against cancer. Particular success has been reported following treatment with inhibitors against the immune checkpoint proteins cytotoxic T lymphocyte antigen 4 (CTLA-4) and programmed cell death protein 1 (PD-1). CTLA-4 is a membrane receptor found on effector T cells that binds cluster of differentiation (CD)80/CD86 on antigen-presenting cells, which leads to T cell arrest.2 PD-1 is also a cell membrane receptor of activated T cells, and it interacts with its ligands, PD-L1 and PD-L2, to induce T-cell inhibition.3,4
Because only a fraction of patients experience effective responses to these immune checkpoint inhibitors, new approaches targeting additional immune-regulating mechanisms or combinations of existing agents are under investigation.
“Like many people, I’m very interested in learning what are any updates on new immune modulators that are, first of all, being evaluated in the preclinical space and also those that are beginning human clinical trials,” explainedJEDD D. WOLCHOK, MD, PHD, the Lloyd J. Old/Virginia and Daniel K Ludwig chair in clinical investigation, chief of immuno-oncology service, director of the Parker Institute for Cancer Immunotherapy, and associate director of the Ludwig Center for Cancer Immunotherapy at Memorial Sloan Kettering Cancer Center in New York.
“The major unanswered questions we have are what additional pathways should be explored in a rational, mechanism-based way to add to the activity of CTLA-4 and PD-1?” Wolchok, co-chair of the “Immune Checkpoints: Newer Targets and Update on Combinations” session at the 34th Annual Meeting & Pre-Conference Programs of the Society for Immunotherapy of Cancer (SITC 2019), explained.
Wolchok anticipates discussions on topics surrounding the early activity of newer-generation immune modulators, including the presence of efficacy signals and whether new toxicities are observed. “In addition, what are potential biomarkers that we might use to enhance development of these medicines and others so that they are introduced to the patient population that might benefit the most?” said Wolchok.
“In lung cancer, as in multiple other tumor types, immunotherapy obviously has made a big difference in what we do and how well patients do,” explainedDON L. GIBBONS, MD, PHD, an associate professor at The University of Texas MD Anderson Cancer Center. According to Gibbons, there are patients who experience long-term responses with immune checkpoint therapies, including CTLA-4 or PD-1 inhibitors, and there is a subset of patients who either do not have any response or have a response that is not durable there is clear evidence that resistance develops.5-7
Results from animal models of lung cancer suggest that CD38 expression on tumor cells may play a dominant role in the development of resistance to single-agent immune checkpoint therapy. “And this caught us by surprise because I had never heard of CD38 before, really, and didn’t know anything about it,” said Gibbons, who is presenting “Targeting CD38 to Optimize PD-1/PD-L1 and CTLA-4 Blockade Immunotherapy” on Friday, Nov. 8, at 2:05 p.m. as part of Session 204 in the Potomac Ballroom.
CD38 is normally expressed on the surface of immune cells and nonhematopoietic cells. CD38 is an ectozyme and functions to convert NAD+ to ADP-ribose (ADPR) and cADPR, which are required for regulating extracellular metabolites, cell adhesion, intracellular Ca+2, and signal transduction pathways.8Although CD38 has been shown to have activity in many immune cell types and functions that vary during development, activation and differentiation of lymphocytes, the potential function of CD38 on tumor cells is not well described.
In 2018, Gibbons and colleagues used established nonsmall cell lung cancer and melanoma tumor models to investigate immunotherapy resistance.9 They reported that tumors treated with antibodies against PD-1/PD-L1 develop resistance by means of CD38 upregulation, which is induced by all-trans retinoic acid and interferon (IFN)-β. Results from in vitro and in vivo studies demonstrate that CD38 inhibits CD8+ T cell function by means of adenosine receptor signaling. Additionally, they suggested that CD38 or adenosine receptor blockade may be effective strategies to overcome resistance. Analysis of large patient tumor data sets revealed CD38 expression in a subset of tumors with high levels of basal or treatment-induced T cell infiltration into the interior of the tumor, which is where immune checkpoint therapies are thought to have the greatest effect. In animal models, combination treatment with CD38 and PD-L1 inhibitors led to therapeutic benefit. These results suggest that a combination strategy may be beneficial for many patients in whom PD-1/PD-L1 blockade is currently indicated.
“At least part of the mechanism of what appears to be going on is that the enzymatic activity of CD38, because it acts upstream of CD73, is driving adenosine formation in the tumor microenvironment,” explained Gibbons, “and adenosine can be profoundly immunosuppressive.”
Since that 2018 publication, Gibbons and col leagues have been investigating combination checkpoint therapy. Treatment with combination immune checkpoint therapy that blocks both the PD-1 and CTLA-4 axes led to pronounced upregulation of CD38. There seems to be some level of effect from the adenosine mechanism; however, there is also a differential response to single-agent CD38 blockade. “There’s a much better response to single-agent CD38 blockade after the resistance to combination than resistance to the single agent, suggesting that there’s something totally different about the immune microenvironment in those tumors, even though they’re also resistant,” noted Gibbons.
Gibbons and his colleagues think they have found a mechanism by which CD38 is marking a subset of monocytes, normally present in tumors, that should become mature dendritic cells and will act as strong antigen-presenting cells for CD4+ and CD8+ T cells. This will drive a robust anti-tumor response. However, when CD38 expression is very high, both the immunosupeffect of adenosine as well as a lack of dendritic cell maturation result in resistance that develops over time.
“There are many more ways that tumors acquire resistance to immune therapies than what we currently know,” said Gibbons. “Obviously, this is an important area for continued exploration in lung cancer but also in many other cancer types.”
Interleukin (IL)-2 combination therapies are also under investigation. An advantage of IL-2 therapies is that they have the ability to integrate into most other forms of immune or nonimmune therapy.10
Therefore, IL-2 therapy may contribute to improved responses in any situation in which T or natural killer (NK) cells are involved in the mediation of immune responses. There is suggested value for combination therapy with immune checkpoint therapy because of their distinct mechanisms of action. Preclinical and clinical studies have demonstrated synergy of both IL-2- and IL-15-based therapies with monoclonal antibodies against PD-1 and CTLA-4.11-14
The first clinical trial designed to investigate this combination included 36 patients with metastatic melanoma who were treated with ipilimumab (Yervoy) and high-dose IL-2, and resulted in three complete responses during the initial trial period and six during follow-up.15,16 These findings provided support to the investigation of additional combinations involving IL-15/IL-15Rα complex- es, pegylated IL-2, and high- and low-dose IL-2 with anti-PD-1 antibodies pembrolizumab (Keytruda) and nivolumab (Opdivo) and the anti-PD-L1 monoclonal antibody atezolizumab (Tecentriq).
A case series of nine patients with metastatic melanoma who had progressive disease while receiving monotherapy with a PD-1 inhibitor were treated with intra-tumoral IL-2 in addition to PD-1 inhibitor therapy at the time of progressive disease.17Of these, three patients had complete responses, three patients had partial responses, and three patients had progressive disease. These data support the use of additional intralesional IL-2 therapy in patients with initial resistance to PD-1 inhibitor therapy and injectable lesions.
In patients who have immunologically “cold” tumors, activation of tumor-resident innate immune cells may be needed to prime an adaptive immune response to promote the efficacy of immune checkpoint blockade. Toll-like receptor ( TLR) agonists function to initiate immune activation within the tumor microenvironment and to break immunosuppression and tolerance (FIGURE).18,19TLR7 and TLR9 primarily are expressed by NK cells, B cells, macrophages and plasmacytoid dendritic cells. TLR7 and TLR9 are thought to target stromal cells but not tumor cells.20Because TLR agonists act on innate immune cells, whereas PD-1 blockade targets adaptive immune cells, it has been hypothesized that antigen-presenting cell activation in the tumor microenvironment through intratumoral injection of TLR agonists may improve the efficacy of the PD-1 blockade.
In a study of mouse models of head and neck squamous cell carcinoma, intratumoral treatment with 1V270 (a TLR7 agonist) or SD-101 (a TLR9 agonist), in combination with an anti-PD-1 antibody led to suppression of tumor growth at the injected site and at distant body sites.21TLR7 agonist treatment led to an increased M1/M2 tumor-associated macrophage ratio and recruitment of activated CD8+ T cells. Combination therapy with the TLR7 agonist in conjunction with anti-PD-1 was shown to increase CD8+ T cell clonality locally and systemically. These data demonstrate a potential therapeutic mechanism for improving PD-1 blockade efficacy in head and neck squamous cell carcinoma.
The TLR7/8 agonist, MEDI9197, has been shown to activate human immune cells resulting in the secretion of IFN-α, IL-12,22and IFN-γ. It should be noted that other agonists, for TLR9 and STING, only induce the release of IFN-α. When considered together, the combination of all three agonists provides a broader cytokine profile that may enhance CD8+ T cell responses. Furthermore, MEDI9197 led to tumor growth inhibition in syngeneic models. When combined with T cell-targeted immunotherapies in preclinical models, the anti-tumor activity of MEDI9197 was enhanced.
T cell immunoglobulin and mucin-domain containing protein-3 ( TIM-3) is constitutively activated on innate immune cells; however, the expression of TIM-3 on T cells has been associated with activated and terminally differentiated states.23-26Phospha- tidylserine, galectin-9, carcinoembryonic antigen-related cell adhesion molecules (CEACAM)-1 and high-mobility group box 1 (HMGB1) are proposed ligands of TIM-3.
Studies have shown that the cytoplasmic domain of TIM-3 is able to mediate intracellular signaling in myeloid cells and in T cells despite a lack of classical inhibitory or activating signaling molecules. The association of human leukocyte antigen (HLA)-B-associated transcript 3 with the cytoplasmic domain of TIM-3 has been shown to prevent T cell exhaustion and dysfunction.27Unlike PD-1 and CTLA-4, there are no reports of antigen-presenting cell-expressed membrane-bound ligands for TIM-3. In a study of breast cancer samples, TIM-3-positive cells were of myeloid origin, which implies that therapeutic approaches that target TIM-3 may strongly impact dendritic cells, macrophages and other antigen-presenting cells.28
Distinct premalignant lesions with unique genetic aberrations can result from cell injury, and a subset of these lesions may escape immune surveillance and develop into invasive cancer. To date, the mutational landscape that may facilitate the prediction of progression is not well understood. To more effectively develop preventive and interception strategies, improved understanding of tumorigenesis, premalignant lesion composition, and the associated microenvironment is needed.
Kostyantyn Krysan, PhD, et al, assessed lymphocyte infiltration in premalignant lesions to identify somatic mutations. This resulted in the identification of sequenced exomes from 89 premalignant atypical adenomatous hyperplasia lesions.29The presence of an adaptive immune response was suggested by the high presence of CD8+ and CD4+ T cells. The predicted neoantigens resulting from these mutations were highly correlated with CD8+ and CD4+ T cell infiltration and PD-L1 upregulation in premalignant lesions. Based on unique patterns of somatic mutations and neoantigens for each patient, these data demonstrate evidence of mutational heterogeneity, pathway dysregulation and immune recognition in adenomatous premalignancy.
To date, immune checkpoint inhibitor therapy has caused large alterations in the oncology therapy landscape. Studies investigating combinations and additional targets may contribute to even more beneficial results for patients by revealing strategies to overcome resistance to immune checkpoint inhibitor therapy. “There is an assumption that the type of resistance that one gets to single-agent therapy is probably the same as it is to combination immune checkpoint therapy, and I think that that’s probably a fallacy,” explained Gibbons. “It’s important to think about the fact that the types of resistance we see in one setting may not be the same as in another.”
Recent studies have shed light on new combinations and targets for immuno-therapy, including targeting CD38 to op- timize PD-1/PD-L1 and CTLA-4 blockade immunotherapy, investigating combinations of immune checkpoint inhibitors with IL-2 agents or TLR agonists, targeting TIM-3 and investigating the immune landscape of premalignant lesions to better predict progression to cancer.