Skip to main content

Harnessing tumor metabolism to overcome immunosupression

Periodic Reporting for period 2 - ImmunoFit (Harnessing tumor metabolism to overcome immunosupression)

Reporting period: 2020-01-01 to 2021-06-30

Cancer immunotherapy has provided patients with a promising treatment approach. Therapeutic regimens such as adoptive T cell transfer (ACT), cancer vaccines and immune checkpoint inhibitors (e.g. anti-PD-1 or anti-CTLA-4 antibodies), harness the ability of the immune system to recognize and reject the tumor. However, despite an improved prognosis for a subset of melanoma patients and some other tumor types, for several other tumors such as colorectal cancer (CRC) and pancreatic ductal adenocarcinoma (PDAC), immunotherapy fails to show any clinical benefit. The next challenge is therefore to anticipate the reasons why in certain patients and various tumors, immune intervention does not offer a durable response, or in the worst case, the tumor is completely refractory to this treatment.
Recent studies highlight the lack of correlation between T cell infiltration in solid tumors, response to immunotherapy (i.e. anti-PD-1) and density of immunogenic antigens, leading to the key questions of which antigen-independent factors limit anti-tumor responses. Growing evidence shows that, even in the presence of an immunotherapeutic intervention, specific metabolic features of the tumor microenvironment (TME) can compromise functions and fate of tumor-infiltrating immune cells in a way to favor immunological tolerance and reduce anti-tumor effector functions. Here we propose a holistic, integrative approach to study the metabolic roadmaps and cross-talk between different cell compartments in the TME that lead to immunosuppression and resistance to immunotherapy. In the immune landscape, CD8+ T cells are surely the most powerful, cytotoxic, anti-tumor cells. Therefore, the overarching goal of this project is to metabolically modulate the TME or CD8+ T cells in order to restore T cell fitness and anti-tumor effector functions.

My research team has a strong track record in investigating the unique properties of the TME (such as hypoxia, metabolite composition, cytokine surge, etc.) and their influence on several tumor compartments (including cancer cells, endothelial cells, monocytes, macrophages, neutrophils, T cells). To fully appraise the key role of tumor metabolism in mounting immunosuppression, and its significance as therapeutic target in the field of onco-immunology, we now want to address new important questions: (i) Which are the metabolic alterations that define the composition of an immunosuppressive TME? (ii) Which of the tumor cell compartments and their cross-talk drives these metabolic alterations? (iii) Can we “normalize” the TME to sustain T cell fitness and effector functions? (iv) Can we directly modify the metabolism of T cells to increase their fitness and effector function in the harsh TME? (v) Can this knowledge be exploited for therapeutic purposes in the treatment of cancer and to overcome resistance to immunotherapy?

The society impact of this research will be broad. It is our tradition to attempt to set-up new spinoffs based on academic findings taking care of valorizing our research at the industrial level. This will create new platforms, networks, and working opportunities. Most important, all these data will allow to set up new immunotherapeutic regimens (cell-based or drug-based) that per se trigger the immune system against the tumor and keep under control a possible metastatic relapse, or in the best-case scenario, impose the shrinkage of already established metastatic lesions. On one hand, we aim to have new treatments options that work in monotherapy. On the other hand, we are assessing the efficacy of our cancer metabolism-based approach in combination with current immunotherapeutic regimens such as immune checkpoint inhibitors. Through our findings, many more patients will likely respond to immunotherapy, thus enlarging the population of cancer patients that can benefit from this strategy.

Given the large armamentarium of metabolic drugs currently used in preclinical and clinical settings for the treatment of other diseases including diabetes, lung disorders, etc., we will evaluate the possibility to reposition these (approved) drugs in the field of onco-immunology therefore mitigating the hard path of the clinical trials in terms of costs and time.

We believe that overcoming the immune suppressive “metabolic checkpoints” of the TME by pharmacologic intervention represents a novel, though yet clinically unexplored, frontier of cancer immunotherapy. Because of the integrative approach chosen, we are convinced that our therapeutic findings will be applicable to a large set of tumor types. The use and combination of patient and mouse datasets is allowing us to have at the same time, a workable, experimental model but also reach clinically relevant conclusions that can be quickly translated to humans. Finally, since other conditions as infections (viral, bacterial, etc.) and autoimmune diseases aim respectively to boost and hinder the immune system, our findings will open up a new avenue for the treatment of several disorders.
1. We integrated the first publicly available pretreatment transcriptomic datasets from patients (mostly melanoma and renal cancer patients) responding and non-responding to immune checkpoint inhibitors (i.e. anti-CTLA-4 and anti-PD-1) with bulk tumor transcriptomic and metabolomic profiling of responding (MC38), scarcely responding (CT26) and non-responding (Panc02) murine tumor models, generated in our lab (by using BIOMEX, Biological Interpretation of Metabolic Experiments). Among the genes that were up-regulated in all the “resistant” vs. “responsive” datasets (both in mouse and patients), we identified CDA, SLC4A4, and HPGDS as potential targets involved in immune checkpoint blockade unresponsiveness. By using single cell RNA sequencing and histological protein evaluation on patient samples in melanoma and PDAC tumors, we disclosed the induction of CDA and SLC4A4 in PDACs where they were mostly expressed by the transformed ductal epithelium (compared to the normal epithelium). Unlikely, HPGDS was found to be specifically induced in tumor-associated macrophages (but not in normal macrophages). All was confirmed in workable mouse models carrying clinically relevant mutations. We have combined metabolomics, metabolic tracing, mouse genetics, small molecule inhibitors, histology, FACS, and in vitro assays to study the metabolic cross-talk and how the perturbation of these metabolic pathways can change the immune landscape and immune cells functions in mice and in vitro. In several tumor models, that are resistant to immune checkpoint blockade, we showed that genetic KO of each of these genes or their pharmacological inhibition, has an anti-tumor effect in monotherapy and sensitizes the tumor to immunotherapy. Going back to patient samples, we have shown the that the expression of these proteins directly correlates with the infiltration of immunosuppressive macrophages and inversely correlates with cytotoxic T cell infiltration. Besides delivering fundamental biology on tumor metabolism and metabolic cross-talks, and on how conventional metabolic pathways are hijacked by the tumor to sustain immunosuppression, we provided proof-of-concept towards innovative anti-cancer therapies that bypass the immunosuppressive constraints of the TME and overcome resistance to state-of-the-art immunotherapies.

2. One of the main metabolic pathways characterizing the metabolic landscape of immunosuppressed/immunotherapy-resistant tumors was glutamine metabolism. Starting from our initial observation the glutamine synthetase in tumor-associated macrophages correlates with wound-healing, immunosuppressive macrophages, we applied models of cancer and tissue regeneration to assed how glutamine production or glutamine anaplerosis can impinge on these disorders. On one hand we have used genetic and pharmacological molecules (drug repurposing) to block GS in the control of tumor growth and synchronous or metachronous metastatic relapse in lung, breast, and melanoma mouse tumors. On the other hand, we proved that blocking glutamine anaplerosis tilts the balance towards glutamine production via glutamine synthetase upregulation in macrophages, which creates a favorable condition for muscle stem cell proliferation and differentiation, which in turn improves muscle regeneration in case of damage or ageing. The latter study is a clear example of how our approach and concept of metabolic cross-talk can be exploited for the treatment of disorders other than cancer.

3. In order to probe in a high throughput fashion, the complex and interconnected metabolic and epigenetic networks of T cells in vivo, we have generated tailored CRISPR/Cas9 pooled-gRNA libraries to perturb the metabolism (metabolic library) and the epigenome (acetylome and methylome libraries) of T cells. Our genetic perturbation platform coupled with well-validated adoptive T cells transfer approaches, sorting strategies, NGS, and downstream bioinformatic analyses will pinpoint all those genes that, when targeted, do not only increase CD8+ T cells infiltration/fitness within the hypoxic vs. normoxic tumor niches (i.e. in vivo hypoxic probe), but also their cytotoxicity and effector functions (i.e. in vivo CD107 degranulation) within these two metabolically distinct niches. Moreover, we will compare metabolic targets emerging at the primary tumor site vs. different metastatic sites. The targets identified in the screens will be validated and characterized in a clinically relevant set of tumor models, all corroborated by specific in vitro assays. Finally, our findings will be therapeutically exploited using ACT, small molecule inhibitors or novel biologics.
The metabolic and epigenetic coordination of T cell immune responses is becoming one of the most active areas with respect to drug development. However, nearly all studies investigating metabolic and epigenetic consequences on T cell biology have been performed in vitro. Our in vivo approach is forward looking and could discover clinically relevant therapeutic targets in primary and metastatic tumor sites, all characterized by different TMEs. The valorization analysis indicates that this study, beside high-impact publications, is expected to deliver several patents, partnerships with national and international companies and new startups to fuel innovation. This covers also the field of gene and cell therapy, where T cells could be metabolically or epigenetically rewired (i.e. CRISPR/Cas9 gene editing) or preconditioned (i.e. small molecules or biologics).
The novelty of this project lies in the combination of several cutting-edge approaches for in vivo selection and validation of functionally relevant metabolic targets in different tumor compartments, in casu cancer cells, TAMs, and T cells. The selection and validation steps are supported by unique and innovative (i) tailored metabolic gRNA libraries, (ii) clinically relevant metastatic tumor models as well as surrogate neoantigens, and (iii) transgenic mouse models. With this research, we will shed light on how metabolism and the metabolic cross-talk among cells contribute to the immunosuppressive TME, and it will build novel concepts into T cell metabolism to support T cell fitness in the adverse growth conditions of the TME. With our approach we move far beyond the current concept of targeting cancer metabolism to kill cancer cells directly. Here, we rather aim to hinder a metabolic cross-talk within the TME in order to trigger the action of the immune system.

In the short run, studying the link between metabolism and anti-tumor immunity will disclose a list of pathways and genes that are key regulators of the immunosuppressive properties of cancer cells (CDA, SLC4A4) and TAMs (GS, GLUD1, HPGDS), or that are causative for inefficient T cell fitness within the harsh TME, with particular attention to specific niches such as hypoxic regions, peri-vascular regions, different primary tumor versus metastatic sites (ad hoc designed screenings in CD8+ T cells). We will provide proof-of-concept of pharmacological targeting and initiate drug discovery campaigns within our spinoffs, with VIB Discovery Sciences and the Centre for Drug Design and Discovery, or with external industrial collaborators. Furthermore, studying how metabolism affects T cell fitness directly, will allow us to select for targets that can be safely modified ex vivo in T cells prior ACT. The ground-breaking nature is underscored by the recent green light to use the CRISPR/Cas9 platform to boost the therapeutic efficacy of patient’s T cells. Overall, our approach will not only open a brand-new way of tumor treatment but also offer the possibility to refine current tumor treatment options, as we are implementing immune checkpoint inhibitors and ACT in our experimental settings. Finally, all our effort and analysis of existing and newly generated mouse and, especially, patient samples (blood/plasma, tumor, metastasis, primary cells) will unseal the expression of the target, metabolic profiles, in situ or circulating metabolites, and new cellular neighbourhood descriptions as possible biomarkers to select the best subset of patients responding to current immunotherapies, or to one of our novel immunotherapeutic targets (i.e. companion diagnostic and predictive biomarkers), and to follow-up disease outcome (i.e. prognostic biomarkers).
Graphical Abstract