Targeting cancer metabolic pathways for improving chemotherapy and immunotherapy

Abstract

Recent discoveries in cancer metabolism have revealed promising metabolic targets to modulate cancer progression, drug response, and anti-cancer immunity. Combination therapy, consisting of metabolic inhibitors and chemotherapeutic or immunotherapeutic agents, offers new opportunities for improved cancer therapy. However, it also presents challenges due to the complexity of cancer metabolic pathways and the metabolic interactions between tumor cells and immune cells. Many studies have been published demonstrating potential synergy between novel inhibitors of metabolism and chemo/immunotherapy, yet our understanding of the underlying mechanisms remains limited. Here, we review the current strategies of altering the metabolic pathways of cancer to improve the anti-cancer effects of chemo/immunotherapy. We also note the need to differentiate the effect of metabolic inhibition on cancer cells and immune cells and highlight nanotechnology as an emerging solution. Improving our understanding of the complexity of the metabolic pathways in different cell populations and the anti-cancer effects of chemo/immunotherapy will aid in the discovery of novel strategies that effectively restrict cancer growth and augment the anti-cancer effects of chemo/immunotherapy.

Introduction

Cancer remains one of the world’s deadliest diseases despite significant progress in the development of novel therapeutic strategies [1]. In addition to surgical options, chemotherapy is considered first-line treatment for several types of cancer. However, treatment failures are common due to the development of drug resistance. Although multi-drug chemotherapy regimens have led to improvements in overcoming chemoresistance, challenges remain for curing many types of cancer [2]. Recent breakthroughs in immunotherapy have greatly advanced cancer treatment but this therapeutic advancement has only benefitted a relatively small subset of cancer patients. In addition, some responders inevitably develop resistance to immunotherapy after a period of treatment. To achieve durable responses, identifying new targets to restore the response to chemo- and immunotherapy represents a novel and attractive strategy [3].

Over the past few decades, cancer metabolism has been extensively studied. It is widely accepted that the rewiring of metabolic networks occurs in a variety of cancer types at different stages of tumor development. Generally, cancer reprograms metabolism with altered consumption of glucose, amino acids, or lipids, and through the generation of non-physiological concentrations of intermediate metabolites or “new” metabolites, and as a result, produces an aerobic, acidic, and nutrient-insufficient microenvironment [4]. These metabolic changes not only directly facilitate tumor cell proliferation and survival but also protect them from immune surveillance by creating an immunosuppressive tumor microenvironment [4, 5]. In addition, these changes negatively influence chemo- and immunotherapy, suggesting that targeting metabolic pathways in cancer can improve therapeutic outcomes. This article reviews the current therapeutic strategies to alter cancer metabolism and discusses how these strategies may influence current chemo- and immunotherapy (see Table 1).

Table1.

Combinations of metabolic modulator and chemo/immunotherapy

Metabolism Pathway	Metabolic modulator	Chemo/immunotherapy agent	Mechanism(s)
Glycolysis	2-DG	Gemcitabine	Suppressing the glycolysis-ROS-DCLK1 axis [43]
		Doxorubicin	Upregulating protein expression of AMPKα, P53, and caspase 3; downregulating the MDR-related protein and P-glycoprotein (P-gp) [44]
		Cisplatin	Induction of the oxidative stress [45]
		Trametinib	Induction of the lethal ER stress [46]
		Sorafenib/Anti-PD1	Increased production of CXCL9/CXCL10 through the AMPK-EZH2-H3K27me3 axis [216]
N-glycans synthesis		4–1BB antibody	Inhibiting the glycosylation of PD-L1 and its immunosuppressive function [55]
		CAR T cells	Disrupting the N-glycan expression on tumor cells and interfering with the PD-1-PD-L1 axis [217]
N-glycans synthesis	FR054	Gemcitabine	Reducing protein glycosylation; inducing a sustained unfolded protein response (UPR); attenuation of the pro-tumorigenic epidermal growth factor receptor (EGFR)-Akt axis [218].
Oxidative phosphorylation	Metformin	Doxorubicin	Increasing ROS production and ATP depletion; downregulating drug-resistant genes such as P-glycoprotein (Pgp) [70].
AMPK-mTOR signaling		Cisplatin	Induction of G0/G1 cell cycle arrest; activation of AMPK and repression of mTOR signaling pathways [219].
Oxidative phosphorylation		Anti-PD-L1	Inhibition of RNF5-mediated K48-linked ubiquitination of STING, which is dependent on AXIN-1[220].
AMPK signaling		Anti-CTLA4	Decreasing the expression of PD-L1 in the cancer cells by altering PD-L1 glycan structure, which promotes PD-L1 degradation [221].
Lactate production	Oxamate	Paclitaxel	Inhibiting the paclitaxel-induced glycolysis and increasing the apoptosis [93].
	Catechin	5-fluorouracil	Inhibiting the glycolysis and inducing the ROS-mediated apoptosis [94].
	NCI-737	IL-21	Major transcriptomic changes, including the suppression of IL-21-induced exhaustion markers LAG3, PD1, 2B4, and TIM3 [222].
	Oxamate	Anti-PD-1	Alleviation of the acidification of TME [99].
Lactate exportation	VB124	Anti-PD-1	Alleviation of the acidification of TME and elevating the chemokine (C-X-C motif) ligand (CXCL) 9/CXCL10 secretion through reactive oxygen species/NF-KB signaling pathway [223].
Lactate exportation	Diclofenac	Anti-PD-1/Anti-CTLA-4	Reducing the lactate accumulation and activating T-cells and NK cells [224].
Glutaminolysis	CB-839	Carboplatin	Increasing the redox and replication stress [108].
		5-fluorouracil	Upregulating uridine phosphorylase 1 (UPP1), an enzyme for 5-FU activation, in an ROS-dependent manner [109].
		Anti-PD-L1	Reducing cellular GSH levels and upregulation of PD-L1 expression by impairing SERCA activity, which activates the calcium/NF-KB signaling cascade [225].
Tryptophankynurenine pathway	Navoximod	Anti-PD-L1	Reversing T-cell exhaustion and suppressing hyperactivated regulatory T cells [226].
Arginine metabolism	ADI-PEG	Gemcitabine	Inhibiting PI3K/Akt/NF-KB signaling [227].
Nitric Oxide production	L-NMMA	Anti-PD-1	Suppressing MDSCs and TAMs; upregulating PD-L1 [140].
Asparagine depletion	PEGASNase	Anti-PD-1+Gemcitabine+Oxaliplatin	Suppressing protein synthesis of NK/T cell lymphoma [145].
Methionine	BCH	Anti-PD-1	Restoration of H3K79me2 and STAT5 in T cells [149].
Lipid synthesis	FASN siRNA	Cisplatin/Doxorubicin	Downregulating the compensatory NF-kB/SP1-mediated PARP-1 and DNA repair [228].
Lipid droplet biogenesis	Triacsin C	5-fluorouracil+Cisplatin	Enhancing the caspase cascade activation and ER stress responses; increasing the immunogenic cell death and CD8+ T cell infiltration[229]
Fatty Acid β-Oxidation	Perhexiline	Paclitaxel/Oxaliplatin	Disrupting NADPH and redox homeostasis and increasing the ROS generation [159, 160]
Cholesterol esterification	Avasimibe	Anti-PD-1	Enhancing T-cell receptor clustering and signaling as well as more efficient formation of the immunological synapse [171].
Lipid synthesis	A939572	Anti-PD-1	Enhancing the production of CCL4 by cancer cells through reduction of Wnt/β-catenin signaling, and by CD8+ effector T cells through reduction of endoplasmic reticulum stress [172].
Fatty acid accumulation	Lipofermata	Anti-PD-L1	Reversing ROS-mediated immunosuppression by MDSCs; promoting MDSCs differentiation to an immune-stimulatory phenotype [174]
Fatty Acid β-Oxidation	Bezafibrate	Anti-PD-1	Activating the T-cell mitochondria respiratory capacity and upregulating oxidative phosphorylation as well as glycolysis [176].
Production of 2-HG	Ivosidenib	Azacytidine	Increasing the cycling of rare leukemia stem cells and triggering transcriptional upregulation of the pyrimidine salvage pathway [230].
Production of 2-HG	Ivosidenib	Anti-PD-1	Increasing the STAT1 and downstream CXCL10; recovering the ATP-dependent TCR signaling and polyamine biosynthesis [183, 184]

Overview of metabolic changes in cancer

Altered metabolic activity plays an important role in supporting the proliferation and survival of cancer cells. One of the most well-known examples of a reprogrammed metabolic pathway in cancer is the Warburg effect [6]. In 1920, Otto Warburg discovered that cancer cells exhibit an “aerobic glycolysis” phenotype. Glycolysis is a well-regulated response to hypoxia in normal tissues under physiological conditions, but cancer cells constitutively consume glucose and produce lactate through glycolysis regardless of oxygen availability. Though glycolysis is a less efficient way to produce energy compared to oxidative metabolism, the increased glycolytic influx provides important biosynthetic intermediates to subsidiary pathways to fulfill the anabolic needs of proliferating cells such as synthesis of lipids, nucleosides or proteins. However, because of the Warburg effect, most glucose-derived carbons are secreted as lactate instead of entering the tricarboxylic acid (TCA) cycle as pyruvate. To maintain the carbon pool of the TCA cycle in cancer cells, anaplerotic pathways are utilized to allow TCA cycle intermediates other than acetyl-CoA to enter the cycle and replenish the source of carbon. One major metabolic pathway that provides anaplerotic flux in cancer cells is glutaminolysis, which utilizes glutamine-derived α-ketoglutarate (α-KG) to replenish TCA intermediaries, thereby guarantying the survival and proliferation of cancer cells [7]. The alteration of glutamine demand is also known as “glutamine addicted”— that the cancer cells are highly glutamine-consuming to maintain survival or proliferation compared to normal cells. Oxidation of the branched-chain amino acids (BCAAS) such as leucine and valine also provides an anaplerotic flux in some tumor tissues [8]. In addition, lipid metabolism also undergoes significant reprogramming in cancer cells at different stages [9]. Highly proliferative cancer cells show a strong lipid requirement, which they satisfy by either increasing the uptake of exogenous lipids and lipoproteins or upregulating their endogenous synthesis. Excessive lipid uptake leads to lipid droplets (LD) accumulation in cancer cells, causing them to be more resistant to chemotherapy. More importantly, abundant fatty acids support energy consumption in cancer cells through enhancement of the fatty acid β-oxidation (FAO) pathway, especially in nutrient- and oxygen-depleted environmental conditions [10]. Lastly, lipid composition is altered due to deregulation of de novo synthesis. As part of signal transduction mediators, altered lipid composition affects a variety of carcinogenic processes, including cell growth, migration and metastasis [11].

To support distinctive core metabolic functions like anabolism and catabolism in cancer cells and sustain survival, a finite set of genetic pathways become dysregulated. For example, phosphatidylinositol 3-kinase (PI3K) and its downstream pathways, AKT and mammalian target of rapamycin (mTOR), are activated in normal cells upon stimulation by growth factor, thereby promoting a robust biosynthesis and energy production process involving increased glycolytic influx, oxygen consumption and protein/fatty acid synthesis [12]. Tumor cells very commonly harbor mutation(s) that lead to constitutive activation of the PI3K-AKT-mTOR axis with independence from extrinsic stimulation by growth factor.

The alteration and enhancement of growth-related pathways is accompanied by a phenotype that is resistant to conventional chemotherapy called a “chemoresistance signature.” Constitutive activation of the mutated PI3K/AKT pathway leads to multidrug resistance through overexpression of multidrug resistance-associated protein 1 (MRP1) or P-glycoprotein (P-gp) transporter [13]. Beside chemoresistance, the alteration in the PI3K pathway initiated by metabolic stress leads to immunoresistance through suppression of MHCI-dependent antigen presentation [14].

Perturbations in metabolism also occur through the upregulation of the MYC pathway. MYC, a transcriptional factor, is an oncoprotein and has a profound pro-tumor impact. MYC plays a key role in regulating cancer metabolism from nutrient sensing, catabolic intermediary metabolism, and anabolic macromolecular synthesis by affecting the expression of many key genes that support glycolysis, glutaminolysis, nucleotide synthesis, lipid synthesis, and metabolic rewiring [15]. Studies show that MYC promotes chemoresistance through multiple mechanisms [16, 17] and is also associated with immune evasion and decreased immunotherapy efficacy [18–20]

In addition, gene induction is commonly coordinated by hypoxia-inducible factor-1 (HIF-1). Solid tumors contain significant heterogeneity of perfusion, such that many tumor cells in the tumor core reside in oxygen-poor environments ranging from 0–2% O₂ because tumor cell proliferation often exceeds the rate of angiogenesis. Activation of HIF-1 by the hypoxic environment induces changes in the expression of metabolic genes to boost glycolytic flux. Moreover, activation of HIF-1 induces stress response gene expression that supports survival and chemoresistance [21].

Apart from common mutations, specific mutations in certain types of cancer also drive the reprogramming of metabolism. Melanomas harboring the BRAF^V600E mutation have an order of magnitude higher uptake of glucose compared to normal tissues with diminished oxidative phosphorylation through the suppression of PGC1α [22]. Conversely, treatment of BRAF^V600E melanomas with BRAF inhibitors renders them addicted to oxidative phosphorylation. Inhibition of mitochondrial metabolism together with a BRAF inhibitor may represent an attractive strategy to enhance the effect of BRAF inhibitors in these patients [22]. In another study, pancreatic tumors harboring a KRAS^G12D mutation exhibit abnormal glutamine dependence. More importantly, pancreatic cancers rely on a distinct pathway in which glutamine-derived aspartate is transported into the cytoplasm where it can be converted into oxaloacetate by aspartate transaminase (GOT1). This distinct reprogramming of glutamine metabolism is dictated by KRAS [23].

While reprogramming the tumor cells themselves can have a major impact on chemotherapy, reprogramming the tumor microenvironment (TME) can have a more profound impact on the efficacy of immunotherapy [5]. Due to the oncogenic transformation and reprogrammed metabolic phenotype, the TME is profoundly changed. The poorly developed vasculature leads to limitation of nutrients and oxygen, as well as waste removal. While tumor cells reprogram themselves to adapt to nutrient deprivation, hypoxia and acidic TME, other cell types such as infiltrated immune cells are forced to undergo metabolic adaptations. The adapted immune cell phenotype is usually characterized by immune tolerance and immune resistance, including increased immune checkpoint expression on effector T cells [24, 25], enhanced differentiation of M2-like macrophages [26], increased accumulation of myeloid-derived suppressor cells (MDSCs) [27] and regulatory T cells (Tregs) [28] and dendritic cells (DCs) tolerization [29]. Ultimately these changes in immune cell populations can undermine the effectiveness of the anti-tumor immune response.

The tumor stroma, another important component of TME, also supports tumor growth and hinders chemo/immunotherapy efficacy through tumor cell-stroma crosstalk. Cancer-associated fibroblasts (CAFs) are a major component of tumor stroma [30], especially in pancreatic cancer, a stroma-rich cancer type. Under physiological conditions, normal fibroblasts (NFs) regulate the formation and turnover of extracellular matrix (ECM), control tissue homeostasis and contribute to wound healing [31]. NFs show increased energy demand for proliferation and protein synthesis when activated by tissue injury, which is fulfilled by increased aerobic glycolysis and glutaminolysis [32]. Similarly, fibroblasts are activated and hijacked by cancer cells in TME to differentiate as CAFs. CAFs exhibit a similar metabolic phenotype as activated NFs, such as enhanced glycolysis and glutaminolysis [32]. However, CAFs also present a distinct metabolic reprogramming that facilitates cancer progression. For example, in addition to the enhanced glutamine consumption, CAFs also show upregulated amino acid synthesis and secretion to meet the nutrient requirements of ovarian cancer cells by upregulating the expression level of glutamine synthetase [33]. CAFs are also capable of producing aspartate/asparagine as nitrogen sources for prostate tumor cells [34]. Additionally, CAFs demonstrate reprogrammed lipid metabolism. For example, high levels of lysophosphatidylcholines (LPCs) are secreted by CAFs in pancreatic cancer [35]. LPCs are then metabolized to lysophosphatidic acids (LPAs) that cause activation of HIF1a signaling pathways, and further rewiring of cancer cell metabolism [36].

Similar to CAFs, the crosstalk between tumor cells and cancer-associated adipocytes (CAAs) also remodels the metabolism and phenotype of adipocytes, leading to cancer progression and drug resistance in those cancer subtypes with enriched anatomical distribution of adipose tissue, such as breast cancer and pancreatic cancer. Unlike normal mature adipocytes, CAAs adopt a dedifferentiation phenotype with lipolysis and alteration of fat storage capacity, which is the major metabolic feature of CAAs. In cancer-related lipolysis, CAAs liberate free fatty acids (FFAs) under stimulation of cancer cells [37]. Released FFAs can act as a fuel for cancer cells through mitochondrial FAO. This process of altered lipid metabolism by CAAs has been reported to occur in breast cancer [38]. In addition to lipid metabolism, CAAs also support glutamine metabolism of cancer cells through glutamine transfer. Pancreatic cancer cells have been shown to decrease glutaminase expression in adipocytes, which would be expected to inhibit adipocyte glutamine catabolism and increase glutamine secretion [39]. Moreover, CAAs-related glutamine transfer also contributes to resistance to L-asparaginase (ASNase) therapy in leukemia cells [40].

Targeting glucose metabolism to improve chemo/immunotherapy

High rates of glycolysis brought by the Warburg effect and increased production of lactate result in a TME that is hypoglycemic and acidic. Interestingly, T-cells undergo similar metabolic reprogramming during activation. Quiescent T cells produce energy mainly through oxidative phosphorylation (OXPHOS) of glucose. However, effector T cells have increased energy demands to support rapid replication and production of signaling molecules such as cytokines. To meet this need, activated T cells utilize glycolysis and increase glucose uptake [41]. The similarity of metabolic reprogramming between activated T cells and tumor cells results in a competition for glucose between tumor and immune cells.

Inhibiting glycolysis and thereby removing the energy supply of cancer cells represents a straightforward anti-tumor mechanism. 2-Deoxyglucose (2DG), a widely used glycolysis inhibitor, has been shown to significantly improve the tumor inhibition effect of chemotherapy through multiple mechanisms such as induction of apoptosis and autophagy. 2DG has been shown to have synergistic tumor killing effects when combined with numerous chemotherapy drugs including gemcitabine [42, 43], doxorubicin [44], and cisplatin [45] in various cancer models. For example, targeted therapy such as a MEK inhibitor can be improved with the combination of low dose 2DG in a Kras^G12D-driven pancreatic cancer mouse model [46]. Moreover, 2DG can enhance CD8⁺ T-cell infiltration by augmenting CXCL9/CXCL10 production in the tumor, thereby increasing the sensitivity to anti-PD-1 antibody treatment [47].

Another consequence of 2DG treatment is de-glycosylation [48]. Glycosylation is a vital pathway for post-translational protein modification. Abnormal glycosylation of proteins will cause malfunctions in protein folding, trafficking, and secretion, ultimately disrupting the immune response [49] and promoting cancer cell survival and metastasis [50]. Inhibition of glycolysis can affect the glycosylation of proteins within and on the surface of cells and this effect largely benefits the anti-cancer effects of chemo- and immunotherapy [51, 52]. For example, 2DG can interfere with the N-linked glycosylation process and further potentiate the endoplasmic reticulum stress that is caused by chemotherapy drugs, therefore accelerating the apoptosis process [53, 54]. More importantly, most membrane proteins require glycosylation to maintain their structure and stability, including PD-1/PD-L1. Studies have shown that 2DG can further augment anti-PD-1 treatment through de-glycosylation of both PD-1 and PD-L1, revealing another important mechanism for improving immunotherapy through inhibition of glycolysis [55–57].

As mentioned above, inhibition of c-Myc will arrest multiple metabolic pathways, including through downregulation of several genes involved in glucose metabolism. Interestingly, our recent work shows that treatment with Myc inhibitors leads to significant induction of glutamine fructose-6-phosphate amidotransferase-1 (GFAT1) [58]. GFAT-1 is a rate-limiting enzyme in the hexosamine biosynthetic pathway (HBP), which utilizes glucose, glutamine, acetyl-coenzyme-A, and nucleotide UTP to synthesize UDP-N-acetyl-D-glucosamine (UDP-GlcNAc) [59]. UDP-GlcNAc serves as the substrate for O-GlcNAc transferase (OGT), which catalyzes the transfer of the GlcNAc moiety onto the free hydroxyl of serine and threonine residues [60, 61]. GFAT-1 has been reported to be upregulated in several types of cancer, and the level of GFAT-1 is negatively correlated with prognosis [62–64]. Simultaneous targeting of both Myc and GFAT-1 shows enhanced antitumor efficacy and improved tumor infiltration by lymphocytes [58].

Despite the promising anti-tumor effect brought by metabolic inhibition and de-glycosylation, clinical translation of 2DG is limited by the toxicity at doses that demonstrate therapeutic benefit. Clinical trials combining 2DG with chemotherapy have failed to improve the overall response rate (ORR) due to severe adverse effects [65, 66]. There are also concerns over the potential effect on immune cells. Like cancer cells, effector T cells utilize glycolysis to meet their high energy demands. Studies have demonstrated that potentiating the glycolytic process in CD4⁺ and CD8⁺ T-cells can result in enhanced antitumor T cell responses [67, 68]. Therefore, the use of glycolysis inhibitors in cancer treatment is further complicated by their effects on anti-tumor immunity. How to rebalance the glycolytic competition between immune cells and tumor cells to favor both chemotherapy and immunotherapy remains a critical challenge.

Metformin is another commonly used anti-glycolytic drug that can sensitize chemo-resistant tumors to various chemotherapeutics [69–71]. Interestingly, metformin treatment may represent a feasible strategy to alter the metabolism of effector T cells to avoid glucose competition with tumor cells. Metformin appears to facilitate the compensatory shift from a glucose-dependent anabolic state to a catabolic state of metabolism by blocking mTOR signaling downstream of AMP-activated protein kinase (AMPK) and restoring mitochondrial FAO [72, 73]. This shift rescues exhausted CD8⁺ T-cells and restores their effector function [74]. Additionally, metformin can improve PD-1 blockade therapy via other mechanisms including activation of Stimulator of Interferon Genes (STING) and reducing PD-L1 expression [75, 76]. The clinical benefits of combination therapy using metformin and PD-1 blockade are currently being investigated in several clinical trials in patients with colorectal cancer, non-small cell lung cancer, and melanoma (NCT03800602, NCT04114136, NCT03874000, NCT03048500, NCT03311308). Notably, a diagnosis of diabetes commonly occurs prior to or concurrent with a pancreatic cancer diagnosis. [77]. Clinical retrospective analysis has demonstrated that metformin use is associated with improved outcomes for patients with diabetes and pancreatic cancer [78], suggesting that metformin combined with chemo/immunotherapy may bring extra benefits for pancreatic cancer patients.

Another strategy to mitigate the consequence of glycolysis is to reduce the production of lactate. As both cancer and immune cells rely on glycolysis to meet energy demands, there is a buildup of lactate as a by-product, leading to the acidification of the TME. Additionally, lactate has emerged as a critical regulator of cancer chemoresistance and immunosuppression [79–81]. Numerous mechanisms are involved in lactate-mediated chemoresistance of tumor cells, including enhanced DNA repair through inhibition of histone deacetylase and activation of hydroxycarboxylic acid receptor 1 [82], induction of Snail expression to upregulate the expression of multidrug resistance-associated protein 1 [83], and the promotion of CD44 expression to support the epithelial to mesenchymal transition (EMT) and metastasis [84]. In addition to promoting the survival and proliferation of cancer cells, lactate demonstrates strong inhibitory effects on immune cells and immunotherapy. It is associated with suppressive effects on the proliferation of effector T-cells through the induction of NAD(H) redox stress in T cells [85]. Lactate also inhibits natural killer cell function, increases MDSC recruitment [86], and promotes M2-polarization through HIF1a stabilization [87], further enhancing the suppressive immune microenvironment. Moreover, lactate can directly impair PD-1 antibody-based immunotherapy by upregulating PD-1 expression on Treg cells but repressing the PD-1 expression on CD8⁺ T cells. Consequently, PD-1 invigorates Treg cells instead of effector T cells, resulting in immunotherapy failure [88].

In addition to these direct effects, accumulated lactate indirectly affects chemotherapy and immunotherapy through acidification of the TME. The inverted pH gradient between the inside and the outside of tumor cells can impair the distribution of weak base chemotherapy drugs, described as “ion trapping” [89]. These drugs will be ionized in acidic TME, decreasing the ability to permeate the lipophilic cell membrane. The low intracellular drug concentration can further lead to chemoresistance [90]. Low pH has also been shown to suppress T-cell function though V-domain Ig suppressor of T cells activation (VISTA) and further impair efficacy of PD-1 blockade [91]. Moreover, immune suppressive cell populations such as MDSCs are preferentially recruited to the acidic tumor area due to enhanced binding of P-selectin glycoprotein ligand 1 (PSGL-1) [92].

Several critical enzymes in the glycolysis pathway have been targeted to reduce the production of lactate. One target of particular interest is the lactate-producing enzyme lactate dehydrogenase (LDH), as it is one of the rate-limiting enzymes during lactate production. Targeting LDH has demonstrated enhanced efficacy of chemo- and immunotherapy. LDH inhibitors such as Oxamate or Galloflavin can re-sensitize chemotherapy-resistant tumors to Taxol, 5-fluorouracil (5-FU) or cisplatin through the induction of reactive oxygen species (ROS) and mitochondrial apoptosis in tumor cells, as well as alleviating TME acidosis [93–95].

The impact of lactate production inhibitors has also been examined in the context of immunotherapy. Immunotherapy, especially immune checkpoint blockade, has revolutionized the treatment of melanoma. However, it has been reported that the anti-CTLA-4 monoclonal antibody is not effective in melanoma with high glycolytic flux and lactate production (e.g., BRAF^V600E) [96, 97]. Therefore, combining the inhibition of lactate production and immune checkpoint blockade may advance immunotherapy efficacy. A combination of LDH inhibition with IL-21 and anti-PD-1 immunotherapy has shown a cooperative anti-tumor effect in melanoma and non-small cell lung cancer models [98, 99]. Another approach to reducing the effects of lactate is to inhibit lactate transporters, monocarboxylate transporters (MCT) 1 or 4. Interestingly, MCT4 inhibition also increases the production of chemokine CXCL9/10, which improves the therapeutic benefit of PD-1 immunotherapy through enhanced recruitment of cytotoxic T cells [100]. Notably, MCT1 and MCT4 are dramatically upregulated in metastatic melanomas [101], suggesting MCT1/4 inhibition may be an effective strategy that may benefit melanoma patients, in particular. Indeed, the non-steroidal anti-inflammatory drug (NSAID) diclofenac can lower lactate secretion by inhibiting MCT1 and MCT4, resulting in simultaneous activation of NK cells and T cells. This combination of diclofenac with anti-PD-1 or anti-CTLA4 further improves the immunotherapy efficacy in a melanoma model [102]. Another approach to directly increase the pH of the TME is to orally administer bicarbonate, which has been shown to arrest tumor growth when combined with anti-PD-1 immunotherapy, and improve survival when combined with an adoptive T cell transfer [103, 104]. However, these preclinical results require additional validation.

Targeting amino acid metabolism to improve chemo/immunotherapy

Due to a limited glucose source and increased demand due to glycolysis-dependent metabolism, amino acids become compensatory nutrients as 1) an energy supply and 2) a source of carbon and nitrogen to support tumor growth and survival. The elevated acquisition of certain amino acids by numerous cancer types has drawn attention to amino acid metabolism as a potential therapeutic target. Dependency on certain amino acids is driven by the activation of major oncogenes, including Ras and c-Myc, or loss of tumor suppressor genes such as PTEN and P53. Consequently, the expression of several key enzymes that are involved in amino acid metabolism becomes dysregulated. Ultimately, changes in major oncogenes lead to a reprogrammed requirement for both non-essential and essential amino acids, affecting tumor survival and immune cell activities profoundly. Targeting the metabolic pathway of amino acids represents a promising strategy to improve both chemo- and immunotherapy.

Glutamine

Glutamine is the most critical amino acid in cancer cells. Excessive consumption of glutamine has been observed in glycolysis-dependent cancer cells. Glutamine actively participates in the TCA cycle as an anaplerotic fuel source to maintain TCA flux, biosynthesis of nucleotides and non-essential amino acids, and maintenance of redox homeostasis [105]. Glutamine deprivation suppresses cancer growth and induces cancer cell apoptosis in neuroblastoma, leukemia and triple negative breast cancer [106, 107]. “Glutamine-addiction” suggests that the inhibition of glutamine metabolism represents a novel strategy to improve cancer chemotherapy.

Numerous studies have shown the synergy between inhibition of glutamine metabolism and broad chemotherapy agents. CB-839, a well-known inhibitor of glutamine metabolism via inhibition of glutaminase (GLS), enhances the apoptotic effects of platinum-based chemotherapy through an increase in redox and replication stress in a triple negative breast cancer model [108]. Similarly, enhanced ROS accounts for the synergistic antitumor effect of the combination of CB839 and 5-FU in PIK3CA-mutant colorectal cancer [109]. CB-839 also shows a synergistic effect with pomalidomide in myeloma through further dampening the mTOR pathway. Several combination therapies of CB-839 with chemotherapeutic agents have advanced to clinical trials, including combinations with paclitaxel (NCT02071862), Talazoparib (NCT03875313), carboplatin (NCT04265534) and Palbociclib (NCT03965845). Results from these clinical studies may help establish the inhibition of glutamine metabolism as a versatile method to improve chemotherapy.

Interestingly, unlike glucose, which is required by both cancer and immune cells for anabolic growth, glutamine is utilized differently by cancer and immune cells. In effector T cells, glutamine antagonism leads to compensatory upregulation of oxidative metabolism and the cells adopt a long-lived, highly activated phenotype, while glutamine dependence in cancer cells lacks plasticity [110]. Differentiation of macrophage sub-populations is also subjected to the regulation of glutamine metabolism. Enhanced glutamine metabolism favors M2 macrophage activation, whereas pro-inflammatory M1 macrophages can be suppressed by glutaminolysis. This differentiation is mediated by the core metabolite in the glutamine metabolism pathway, α-KG. Inhibition of glutamine metabolism skews M2 macrophages toward an M1-like phenotype [111]. In addition, reducing intracellular glutamine in macrophages through pharmacologic inhibition or genetic ablation of glutamine synthetase (GLUL) promotes an M1-like phenotype via activation of HIF-1α [112]. These studies suggest that inhibitors of glutamine metabolism repolarize the macrophage towards a M1-phenotype. In addition to its impact on T-cells and macrophages, glutamine inhibition leads to decreases in colony stimulating factor 3 (CSF-3) in tumor cells and hence alleviates the immune-suppressive microenvironment by reducing MDSC recruitment and infiltration [113].

Glutamine’s broad effects in modulating immune cell function make the inhibition of glutamine metabolism a promising strategy to potentiate immunotherapy. Indeed, broad blockade of glutamine metabolism, combined with anti-PD-1 or anti-CTLA4, improves anti-tumor effects by suppressing tumor cell metabolism while reprogramming T cell glucose metabolism, epigenetic alterations, and cytotoxic function [110, 114]. Inhibition of glutamine utilization can also upregulate the expression of PD-L1 on tumor cells by impairing Sarco/ER Ca²⁺-ATPase (SERCA) function, further synergizing with anti-PD-1 therapies [115]. Moreover, inhibition of glutamine metabolism by CB-839 can potentiate anti-tumor CAR-T cells but not auto-immune T cells [116]. However, a recent study reports that glutamine dictates tumor cell-DCs crosstalk and licenses DCs role in activating cytotoxic T cells through the transcription factor TFEB. Inhibition of glutamine metabolism in DCs causes therapeutic resistance to checkpoint blockade and T-cell mediated immunotherapy [117]. In conclusion, inhibition of glutamine metabolism represents a potential strategy to improve immunotherapy due to the differential glutamine dependency between tumor cells and immune cells. However, more studies are needed to understand the impact of inhibition of this pathway on different immune cell subpopulations to avoid undermining the overall therapeutic efficacy.

Tryptophan

Unlike glutamine, tryptophan (Trp) is an essential amino acid that is catabolized in local microenvironments of tumors, immune-privileged sites, or sites of inflammation. In these tissues, cancer cells, immune cells, or specialized epithelial cells (e.g., syncytiotrophoblasts in the placenta) create an environment that suppresses antigen-specific T-cell responses both by depletion of Trp and by accumulation of immunosuppressive Trp catabolites [118]. Trp is metabolized via the kynurenine (Kyn) pathway, which is regulated by crucial rate limiting enzymes indoleamine-2,3-dioxygenase 1 (IDO1), IDO2 or tryptophan-2,3-dioxygenase (TDO2). These enzymes catalyze the oxidative cleavage of the indole moiety of Trp that leads to the formation of N-formyl-L-kynurenine, which is then degraded to Kyn and further downstream metabolites. In cancer, dysregulated IDO and TDO activity results in suppressed anti-tumor immunity. High IDO1 expression has been shown to correlate with reduced populations/frequencies of infiltrating CD3⁺ T cells, CD8⁺ T cells, CD57⁺ NK cells, and B cells but increased frequencies of MDSCs and Tregs [119–122]. TDO activity also synchronizes with IDO activation, leading to CD8⁺ T cell exhaustion and death [123]. Clinically, the accumulation of Kyn is correlated to a poor prognosis [120, 124]. Studies also show that chemotherapy often induces IDO1 expression [125], potentially undermining the efficacy and causing immunoresistance. Therefore, blocking the utilization of Trp by cancer cells and the subsequent reduction of kynurenine may improve the efficacy of cancer therapy.

IDO1 inhibition is the most thoroughly studied strategy to arrest Trp metabolism. Currently, IDO1 inhibition provides a major benefit in cancer immunotherapy and shows promise as a potential adjuvant to boost immunotherapy. Multiple IDO1 inhibitors are undergoing evaluation in clinical trials in combination with anti-PD1 or anti-CTLA4 immunotherapy for a variety of oncology indications [126]. Completed studies suggest that combination of IDO1 inhibition and immune checkpoint blockade shows more favorable ORR than immune checkpoint blockade treatment alone in advanced melanoma (NCT02073123), urothelial carcinoma (NCT03374488), and endometrial cancer (NCT04106414) with more cancer types under recruitment (NCT03358472, NCT03459222, NCT03854032). Inhibition of other rate-limiting enzymes are also being investigated in combination with chemotherapy and immunotherapy. For example, TDO inhibition shows the capability to reverse chemoresistance in prostate cancer by interrupting the Trp/TDO2/Kyn/AhR/c-Myc loop [127]. Also, a TDO inhibitor is being evaluated in combination with IDO inhibitors in clinical trials to improve immunotherapy [128].

Arginine

Arginine is a non-essential amino acid that plays important roles in a variety of biological functions such as protein synthesis, cell proliferation, and cell survival. It is also a precursor associated with production of nitric oxide, polyamines, proline, creatinine and glutamate. Arginine and its metabolic by-product, nitric oxide (NO), support the survival of both cancer cells and immune cells.

Depletion of arginine is generally considered an anti-tumor strategy in arginine-dependent tumor types. Usually, these tumors are deficient in arginosuccinate synthetase (ASS), an enzyme used to synthesize arginine as compensation for arginine starvation. The arginine deprivation agent, arginine deiminase (ADI), enhances the therapeutic efficacy of gemcitabine by inhibiting NF-κB signaling [129]. Activation of the eIF2α/ATF4/CHOP autophagy-related pathway is another potential mechanism as demonstrated in a study combining several chemotherapy agents, including oxaliplatin, 5-fluorouracil, gemcitabine, and cisplatin [130]. Several clinical trials have been completed evaluating the therapeutic potential of combinations of chemotherapy drugs with the arginine deprivation agent, PEG-arginase (ADI-PEG), in metastatic breast cancer (NCT01948843), advanced pancreatic cancer (NCT02101580), and acute myeloid leukemia (NCT05001828), respectively.

The impact of arginine depletion on the TME, especially immune cells, is complicated and conflicting results have been reported. Arginine depletion has been reported to suppress T-cell responses by blocking proliferation and cell-cycle progression in normal activated T cells [131], and at the same time, induce the accumulation of MDSC through serine/threonine-protein kinase, a key mediator of the effects induced by amino acid starvation [132]. However, another study has shown that depletion of arginine levels by ADI-PEG improves T-cell infiltration and inhibits the induction of Treg cells [133]. Nonetheless, targeting NO, one of the by-products of arginine metabolism through nitric oxide synthase (NOS), appears to be a viable strategy to improve immunotherapy outcomes in all reported studies. NO activates cyclooxygenase-2 (COX-2) and other inflammatory mediators, thereby creating oxidative stress in the microenvironment that supports cancer cell growth and suppresses antitumor immunity [134–136]. For example, inhibition of NO production reverses MDSC-mediated immunosuppression by blocking MDSC recruitment to the tumor [137, 138]. Moreover, high levels of expression of NOS correlate with poor prognosis but the inhibition of NOS improves the efficacy of PD-1 antibody immunotherapy by upregulating the PD-L1 levels on breast cancer cells [139, 140]. Taken together, balancing the role of arginine metabolism in both cancer cells and immune cells might be critical in developing an arginine inhibition strategy to improve the overall anti-tumor efficacy. Also, the roles of arginine and NOS in both tumor cells and immune cells need further investigation.

Asparagine

Asparagine is a non-essential amino acid that can be produced through de novo synthesis from glutamine or taken up from the environment. Recently, studies have shown that protein recycling becomes another unique pathway to acquire asparagine when facing asparagine deprivation [141]. Asparagine is actively involved in cell proliferation and survival through the activation of the mTORC1 pathway, which is responsible for coordinating protein and nucleotide synthesis. The importance of asparagine for tumor growth has been demonstrated by ASNase treatment of acute lymphoblastic leukemia (ALL) that has low levels of asparagine synthetase (ASNS). However, thus far, ASNase has only had clinical success in ALL [142]. One major reason is the high expression of ASNS in most other tumor types. Despite the failure of an asparagine deprivation strategy alone in clinical trials targeting other tumor types, a combination of ASNase with chemotherapy drugs has shown promising results. For instance, a phase II clinical study has shown that ASNase combined with gemcitabine or a modified FOLFOX-6 regimen in pancreatic cancer is associated with improvements in overall and progression-free survival [143]. Notably, this improvement is independent of ASNS expression. Although ASNase represents the most successful strategy to target asparagine, no other inhibitors of asparagine metabolism have been developed. A better understanding of asparagine’s unique role in cancer development is required to develop novel inhibitors of asparagine metabolism.

As a growth-signaling support molecule, asparagine favors CD8⁺ T cell activation and anti-tumor immune responses through activation of T cell receptor (TCR) signaling, which is mediated by enhancement of the phosphorylation of lymphocyte-specific protein tyrosine kinase (LCK) in CD8⁺ T cells. Removal of asparagine or treatment with ASNase reverses the activation of CD8⁺ T cells [144]. However, a clinical study shows that the combination of anti-PD-1 antibody and peg-ASNase improves the therapeutic efficacy of chemoradiotherapy in NK/T cell lymphoma [145]. More in-depth mechanistic studies are needed to better understand the impact of asparagine deprivation on immunotherapy. Moreover, the differential impact of asparagine deprivation on tumor growth and TME also needs further evaluation.

Methionine

Methionine (Met) is an essential amino acid for protein synthesis and is involved in biochemical reactions required for cell viability and growth. Cancer shows specific alterations in Met metabolism to fulfill its high energy demand for rapid proliferation. Depletion of Met decreases invasion and metastasis while promoting apoptosis in tumor cells. The most important metabolite of Met is S-adenosylmethionine (SAM). SAM is the universal methyl donor for RNA, DNA, and chromatin methylation, and directly links nutrient availability and cellular metabolism with epigenetic regulation. Various studies have shown that depletion of Met can improve the efficacy of multiple chemotherapy drugs including doxorubicin, 5-FU, and cisplatin [146]. One of the mechanisms is that Met depletion can decrease the ATP pool and glutathione, thus alleviating the drug resistance mediated by the ATP-dependent transporters such as Pgp and/or MDR [146]. Restriction of dietary Met can overcome the chemotherapy resistance in RAS-driven colorectal cancer through different mechanisms. Met restriction can disrupt the flux of one-carbon metabolism and create vulnerabilities in redox and nucleotide metabolism that can be exploited by administration of other chemotherapies [147]. Moreover, by influencing methylation, Met restriction shows enhanced anti-tumorigenesis and growth inhibition in lung cancer and glioma [147, 148].

Recently the competition between T cells and tumor cells over Met has attracted attention. The importance of Met to T cells is underscored by its requirement for remodeling the histone methylation landscape during T cell differentiation. T cells undergo rapid proliferation and differentiation upon recognition of antigens. Exogenous Met is required during this process to support protein synthesis and generate SAM for methylation reactions required for T cell activation. Studies show that cancer cells outcompete T-cells in Met consumption with overexpression of SLC43A2, the major transporter responsible for Met uptake, on cancer cells, resulting in impaired T cell immunity. Met supplementation or treatment with a SLC43A2 inhibitor, can boost T-cell function and synergize with PD-1 blockade treatment [149]. Another study shows that Met restriction also enhances antitumor immunity by regulating the N6-methyladenosine (m6A) methylation and translation of immune checkpoint molecules. Met restriction also synergizes with PD-1 blockade for better tumor control [150]. The consequences of Met restriction are complex, which may impede further clinical application in combination with immunotherapy. With advancement in spatial epigenomics, the role of Met in regulating cancer and cancer immunity will be further elucidated.

Other amino acids

Increasing evidence shows that a variety of amino acids are involved in cancer survival, chemotherapy resistance and TME modulation. Most of them are novel topics that require more research. Also, most of these amino acids lack well-defined and effective inhibitors to modulate their metabolism. Recently, serine/glycine metabolism has been reported to be reprogrammed in ovarian cancer cells resistant to platinum-based chemotherapy, indicating that intratumor serine levels can predict the development of platinum resistance in a subgroup of patients [151]. Also, serine in TME supports effector T cell expansion, suggesting supplementation of serine may improve T-cell mediated anti-tumor efficacy [152]. Histidine is another emerging target for chemotherapy sensitivity. Studies have shown that histidine supplementation sensitizes tumors to methotrexate treatment [153]. Also, high serum histidine levels predict better responses to PD-1 blockade therapy in non-small cell lung cancer, though the mechanism is unclear [154].

Targeting lipid metabolism to improve chemo/immunotherapy

Lipid metabolism influences cell function in various ways. Different lipids are involved in energy production by FAO, synthesis of lipid building blocks for cell membranes, signaling, and posttranslational protein modification. Highly proliferating cancer cells show distinctive features of lipid metabolism. These are caused by increased uptake of exogenous lipids, upregulation of lipid-metabolic enzymes and altered metabolic routes of lipids to both support energy needs and facilitate mitogenic and/or oncogenic signaling.

The response and resistance of tumor cells to chemotherapeutic agents have long been linked to altered lipid metabolism through a number of mechanisms. Chemotherapy-resistant cells often exhibit enhanced FAO and de novo lipid synthesis catalyzed by acetyl-CoA carboxylase/fatty acid synthase, all of which are associated with the abnormal, increased enzymatic activity involved in lipid and cholesterol biosynthesis. Also, chemo-resistant cells have demonstrated altered lipid composition of their cell membranes, leading to reduced membrane fluidity, endocytosis and passive diffusion of anticancer drugs. Finally, altered lipid molecules and their intermediates are involved in enhanced cell survival, reduced apoptosis, and multidrug resistance (MDR). Therefore, targeting lipid metabolism in cancer cells has received attention as a potential chemo-sensitization strategy.

Inhibitors have been developed to target lipid synthesis, lipid oxidation or modulation of lipid composition. For lipid synthesis, pharmacological inhibition of fatty acid synthase (FASN) has been widely studied. FASN inhibition sensitizes doxorubicin-induced DNA damage by inhibiting the compensatory DNA repair in breast cancer [155]. Also, many chemotherapy drugs are capable of inducing ceramide production, facilitating caspase 8-mediated apoptosis. However, FASN is negatively correlated with TNF-α production to counter the apoptosis effect. Thus, inhibition of FASN recovers the TNF-α production and synergizes with chemotherapy drugs to promote tumor cell apoptosis in breast cancer [156]. In another example that targets lipid synthesis, treatment with an inhibitor of long-chain fatty acyl-CoA synthetase leads to restoration of sensitivity to 5-FU and oxaliplatin through the prevention of lipid droplet accumulation in colon cancer [157]. Such treatment similarly enhances the efficacy of paclitaxel by downregulating chemoresistance-related ABC transporter in breast cancer [158]. Targeting FAO to limit energy generation and break the redox balance is a more straightforward strategy than targeting lipid synthesis. Carnitine palmitoyl transferase 1 (CPT1) is a key FAO enzyme and inhibition of CPT1 is a promising chemo-sensitizing approach for a range of cancers [159–161]. Finally, to modulate the aberrant glycosphingolipid composition on the membrane of cancer cells, PPMP (1-phenyl-2-palmitoylamino-3-morpholino-1- propanol), an inhibitor of glucosylceramide synthase, has shown synergistic anticancer effects in combination with a variety of chemotherapy drugs [162–164].

Many studies have been published examining the impact of altered lipid metabolism on the TME, especially on the function of immune cells. Lipid accumulation is a major consequence of altered lipid metabolism in TME. Individual immune cells in the TME are affected by lipid accumulation differently but excess lipid accumulation generally leads to an immunosuppressive microenvironment. Lipid droplets in the TME polarize tumor associated macrophages (TAMs) into an immunosuppressive M2-subtype [165]. Also, Treg cells can modulate the lipid metabolism in M2-macrophages to enhance T-cell suppression. Treg cells suppress CD8⁺ T cell secretion of interferon-γ (IFNγ), which would otherwise block the activation of sterol regulatory element-binding protein 1 (SREBP1)-mediated fatty acid synthesis in immunosuppressive (M2-like) TAMs [166]. Tumor-infiltrating DCs show impaired antigen presentation and low T cell priming efficiency due to a tolerogenic phenotype resulting from abnormal FAO and excessive lipid accumulation in the TME [167, 168]. While T cells partially depend on FAO as a compensatory energy source, lipid accumulation in the TME induces T-cell exhaustion [169, 170].

Inhibition of lipid metabolism in TME has been studied as an approach to improve immunotherapy. Avasimibe, an inhibitor of acetyl-CoA acetyltransferase 1 (ACAT1), a key cholesterol esterification enzyme, has been shown to improve anti-PD-1 therapy by improving TCR clustering [171]. Inhibition of stearoyl-CoA desaturase 1 (SCD1), another rate-limiting enzyme involved in the synthesis of monounsaturated fatty acids, can enhance anti-PD-1 therapy by alleviating ER stress in T-cells and increasing the CCL4-mediated recruitment of DCs [172]. Another strategy is to block lipid accumulation by inhibiting fatty acid transport protein 2 (FATP2) in MDSCs. In the TME, MDSCs accumulate lipids via uptake of exogenous fatty acid, leading to enhanced mitochondrial function and activation of ROS [173], and a more immunosuppressive phenotype of MDSCs. Inhibition of FATP2 blocks ROS-mediated immunosuppressive functions of MDSCs and promotes MDSCs differentiation to an immune-stimulatory phenotype, thereby enhancing anti-PD-L1 cancer immunotherapy [174]. In addition, inhibition of sterol regulatory element-binding protein 1 (SREBP1) can sensitize anti-PD1 immunotherapy by suppressing M2-like TAMs [175]. Interestingly, increasing T-cell fatty acid oxidation through the administration of bezafibrate, an agonist of peroxisome proliferator–activated receptor γ (PPARγ) coactivator 1-α (PGC-1α)/transcription factor complexes, enhances the efficacy of PD-1 blockade [176]. There is an increased interest in targeting the abnormal lipid metabolism in DCs with the advancement of a mRNA-based cancer vaccine in melanoma treatment [177]. Inhibition of CPT1A, a rate-limiting enzyme in the FAO pathway can potentiate anti-PD-1 therapy in BRAF^V600E melanoma [29]. Moreover, pronounced induction of antigen specific CD8⁺ T-cell is shown after CPT1A inhibition by a small molecule ETO, suggesting an enhancement in DCs-mediated T cell priming [29]. Therefore, combination of a lipid metabolism modulator with anti-PD-1 therapy and an mRNA cancer vaccine may represent a novel and effective regimen for improved melanoma treatment. Clearly, the role of lipid metabolism in immune responses is complex but provides a great opportunity to improve the TME to generate more robust anti-cancer immunity.

Targeting oncometabolites to improve chemo/immunotherapy

Oncometabolites are metabolites that accumulate in tumor cells as a result of reprogrammed metabolic pathways, most significantly the Krebs cycle. With advancements in studies on cancer genetics, multiple abnormalities in Krebs cycle enzymes have been reported in cancers, which are responsible for generating different oncometabolites. 2-Hydroxyglutarate (2-HG), succinate, and fumarate are the 3 well-established oncometabolites which are produced as a result of mutations in isocitrate dehydrogenase (IDH), succinate dehydrogenase (SDH), and fumarate hydratase (FH) genes, respectively. These metabolites are extensively involved in reprogramming patterns seen in cancer, including hypermethylation, redox homeostasis, and metabolic reprogramming.

2-HG

IDH1/2 mutations are neomorphic. Under normal conditions, isocitrate is converted to αKG. However, mutated IDH catalyzes conversion of isocitrate to the oncometabolite, 2-HG. IDH mutations are widely seen in many types of cancer but are most prominent in glioma and acute myeloid leukemia (AML). IDH mutation-derived 2-HG has been shown to exert pleiotropic effects on cell biology including chromatin methylation, cellular differentiation and ROS regulation. IDH1/2 inhibitors have been developed and tested in clinical trials in glioma and AML to block the production of 2-HG. IDH1/2 inhibition has shown clinical benefit in AML patients with IDH mutations and is now an FDA-approved therapeutic option for this disease subtype. In addition, inhibition of IDH1 by AG-120 synergizes with the chemotherapy drug azacytidine, leading to improved efficacy in an AML model [178]. However, the efficacy of these inhibitors in solid tumors remains unclear. Interestingly, results from recent clinical studies suggest that patients with IDH1/2 mutant glioma and cholangiocarcinoma have longer median survival times and better responses to conventional radiotherapy and chemotherapy than their wild-type (WT) counterparts [179, 180]. In addition, 2-HG has been reported to sensitize PARP inhibitor (PARPi) therapy in glioma [181, 182]. A study in our laboratory also shows that co-administration of exogenous 2-HG and PARPi leads to enhanced antitumor activity in an IDH WT non-small cell lung cancer model (Unpublished data).

In addition to its effect on tumor cells, tumor cell-derived 2-HG is released in the TME and taken up by T cells. Excessive 2-HG in T-cells suppresses their activity by inhibiting the nuclear factor of activated T cells (NFAT) and polyamine synthesis [183]. The presence of 2-HG in glioma cells can also suppress STAT1 activation, resulting in reduced chemokine production (CXCL9 and CXCL10) and decreased recruitment of cytotoxic T cells to the TME [184]. Inhibition of IDH1 has also been shown to enhance anti-PD-1 immunotherapy. Currently, the combination of the IDH1 Inhibitor, Ivosidenib, and the anti-PD-1 antibody, Nivolumab, is under investigation in a phase II clinical trial in patients with IDH1 mutant glioma and other advanced solid tumors (NCT04056910).

Succinate and fumarate

Succinate and fumarate are critical metabolic intermediates in the Krebs cycle, producing energy, sustaining cell respiratory process and enabling anabolic biosynthesis. In mitochondria, succinate is converted by succinate dehydrogenase (SDH) to fumarate, followed by fumarate hydratase (FH)-mediated reversible conversion between fumarate and malate. Loss-of-function mutations in SDH and FH in cancer results in cytoplasmic accumulation of succinate and fumarate. Elevated levels of succinate or fumarate lead to similar tumor-favoring consequences in cancer development. Studies have shown that both succinate and fumarate are responsible for the activation of the hypoxia pathway through HIF1a, thus activating multiple oncogenic survival pathways [185–187]. Moreover, overproduction of ROS is commonly seen in SDH or FH deficiency due to the perturbation in mitochondria function, leading to activation of signaling pathways such as PI3K, MAPKs, and NF-κB, which are required for tumorigenesis [188–191]. More recently, there is evidence of epigenetic modulatory effects as a consequence of succinate and fumarate accumulation. For example, both succinate and fumarate are considered epigenetic modifiers that elicit epithelial-to-mesenchymal transition, which is an important biomarker for cancer metastasis and drug resistance [192, 193]. In addition, similar to 2-HG, succinate in the TME leads to T-cell exhaustion after uptake by T cells [194]. However, a recent study shows that succinate and fumarate suppress DNA-repair in cancer cells by inducing hypermethylation and cells with a high level of succinate and fumarate are sensitive to synthetic lethality therapy [195]. More studies are needed to explore the therapeutic potential of treatments targeting the metabolism of succinate and fumarate.

Nanotechnology for synergistically targeting cancer metabolism

What is clear from the information outlined thus far is that cancer metabolism and its effects on TME and cancer cells themselves are complex but, altering or inhibiting aspects of cancer metabolism has the potential for dramatic therapeutic benefit. However, obstacles still exist for advancement to clinical application. One challenge is to separate the metabolic inhibition of cancer cells from that of immune cells. As discussed above, cancer cells and immune cells compete for nutrients. Metabolic inhibitors that affect both tumor cells and immune cells may suppress immune cell activity, which may be attributed to the mixed responses seen with inhibition of either of the 3 major metabolic pathways discussed.

Similar to other therapies, metabolic inhibitors must be efficiently delivered to the tumors. Due to the non-discriminative nature of the inhibitors in tissue distribution, it is challenging to deliver sufficient amounts of drugs to tumors without affecting the normal tissues, especially those that are highly active in metabolism and are in high demand of energy and nutrients. Thus, off-target toxicity is a major barrier that needs to be overcome. For example, 2-DG administration has been reported to cause cardiotoxicity [196]. A more challenging task is effective delivery of several drugs to the tumor without causing significant toxicity. As discussed before, metabolism inhibitors are often combined with other therapies such as chemotherapy to maximize the outcome of treatment. It is quite common that drugs of different physicochemical properties are combined. Great efforts are needed to ensure that all drugs can be delivered to the tumors at effective concentrations and at optimal ratios.

Nanoparticles (NPs) represent an attractive approach to improve the delivery of therapeutic agents to tumors while minimizing their distribution to normal tissues, especially the vital organs. Another advantage of NPs is the feasibility of codelivery of several drugs to tumors simultaneously. Many types of NPs have been reported and some are designed to co-formulate hydrophobic drugs and hydrophilic drugs [48, 197, 198]. NPs can also be designed for codelivery of small molecule agents and nucleic acids-based therapeutics [199]. NPs have been used for various combination therapies via codelivery of a metabolism inhibitor and another therapeutic agent or codelivery of different metabolism inhibitors [200]. For example, a liposomal nanoparticle formulation has been employed to co-deliver doxorubicin (DOX) and 2-DG, resulting in improved antitumor activity and decreased off-target toxicity [201]. Similarly, DOX and V9302, an inhibitor of a glutamine uptake transporter, have been co-loaded into carbon quantum dots with an albumin coating and are shown to have enhanced antitumor activity [202]. In another study, a NP formulation is designed to enable the delivery of siRNA to inhibit key glycolysis enzymes, leading to improved antitumor activity in combination with docetaxel [203].

Prodrug-based nanocarriers represent another strategy to co-load different drugs and achieve controlled drug release in a temporal spatial manner. Our group has developed a series of strategies to target cancer metabolism based on a prodrug polymer. One such carrier is based on 2-DG-conjugated polymer (p-2DG). P-2DG retains the activity of 2DG to inhibit glycolysis. Importantly, p-2DG self-assembles to form micelles that are capable of loading another drug such as V9302. We hypothesize that V9302 is rapidly released, leading to the killing of tumor cells. Meanwhile, 2-DG is slowly released and helps to block increased glucose metabolism induced by the inhibition of glutamine uptake [48]. Indeed, codelivery of 2-DG and V9302 via a p-2DG-based nanocarrier has led to improved antitumor activity along with decreased toxicity [48]. In another study, PPMP, a novel glucosylceramide synthase inhibitor, has been used to create a prodrug nanocarrier (p-PPMP) to achieve codelivery of DOX and PPMP. DOX treatment leads to increased levels of pro-apoptotic lipid species such as ceramide. PPMP released from the prodrug polymer helps to sustain high levels of the pro-apoptotic lipids via inhibition of the glucosylceramide synthase-mediated metabolism of ceramide. Systemic administration of DOX-loaded p-PPMP NPs shows improved anti-tumor efficacy in comparison to Doxil, a liposomal formulation of DOX [204]. A novel immunochemotherapy has been similarly developed using an IDO1 inhibitor (NLG-919 or indoximod) prodrug-based nanocarrier. IDO1 is induced by many chemotherapy drugs such as DOX and PTX, resulting in limited efficacy due to the IDO1-mediated immunosuppression. Following codelivery of DOX with NLG-919 or indoximod using the prodrug carrier, NLG-919 or indoximod that is slowly released from the prodrug carrier effectively blocks the activity of the DOX-induced IDO1, resulting in an improvement in tumor immune profile and enhanced antitumor activity [125, 205].

NPs can also be designed to achieve targeted delivery to specific component in TME, such as distinctive populations of immune cells. For example, anti-CD3 antibody-conjugated NPs encapsulating a lipid metabolism-activating drug, fenofibrate, have been reported to relieve metabolic stress and selectively enhance the activity of effector T cells [206]. Another promising strategy is to target immunosuppressive cell populations. For example, scavenger receptor type B-1 (SCARB1), a high-affinity receptor for spherical high-density lipoprotein (HDL), is highly expressed on MDSCs. HDL-based NPs have been developed to achieve targeted delivery to MDSCs. HDL NPs show strong MDSC inhibition and a CD8 T cell potentiation effect, ultimately inhibiting tumor progression in melanoma and lung carcinoma mouse models [207]. Another example is to inhibit Tregs using a tLyp1 peptide ligand. Neurophilin-1 (Nrp-1) appears to be a Treg-selective marker and tLyp1 peptide binds to Nrp-1 with high affinity and specificity. Imatinib loaded into tLyp1-conjugated liposome effectively depletes Treg cells by blocking STAT3 and STAT5 signaling. Combination with anti-CTLA4 therapy further improves the anti-tumor efficacy [208]. Similarly, other membrane receptors such as mannose receptors and scavenger receptors have been reported to be overexpressed on immunosuppressive TAMs [209], and mannose-modified PLGA NPs are designed to target this subpopulation of immune cells [210]. Folate receptor-beta (FOLR2) is overexpressed by both tumor cells and TAMs. Dual targeting to tumor cells and TAMs by folic acid (FA)-conjugated NPs has been shown to improve breast cancer therapy [211]. NPs have also been engineered to target CAFs through different strategies. Fibroblast activation protein (FAP) is a type II integral membrane serine protease that is overexpressed by CAFs and is exclusively expressed in 90% of tumor tissues but not in the granulation tissue in normal, healthy adults during the wound repair [212]. Selective gene silencing of CXCL12 in CAFs using anti-FAP scFv-conjugated siRNA NPs led to enhancement of CD8⁺T cell infiltration in a prostate cancer model [213]. Further combination with anti-PD-L1 immunotherapy shows a synergistic effect in anti-tumor efficacy in pancreatic cancer [214]. Table 2 shows some of published works of targeting different cell populations in TME using NPs. However, nanoparticles targeting immunosuppressive cells have been rarely exploited to deliver metabolism modulation molecules to potentiate the anti-tumor immune response. The potential of these strategies in selective delivery of metabolic inhibitors to a subpopulation of cells in TME warrants future studies.

Table 2.

Representative targets for targeting various cell populations in TME

Cell population	Targets	Application
CD8+ T cells	CD90/CD45	Delivery of TGF-β inhibitor to enhance adoptive cell therapy [215]
CD4+ T cells	CD7	Delivery of siRNA to CD4+ T cells [216]
CD8+ T cells	CD3	Delivery of fenofibrate to CD8+ T cells to reshape fatty acid metabolism in T cells [206]
MDSC	SCARB1	SCARB1-targting high-density lipoprotein-like nanoparticles to inhibit MDSC activity [207]
Treg	Nrp-1	Delivery of imatinib to Tregs to suppress function of Treg cells [208]
TAMs	CD206	Delivery of DOX to TAMs to eliminate TAMs in TME [217]
TAMs	FOLR2	Delivery of gemcitabine and DOX by dual-targeting of TAMs and tumor cells [211]
DCs	CD209	Delivery of tumor antigen to DC as cancer vaccine [218, 219]
CAFs	FAP	Delivery of CXCL12 siRNA to CAFs to improve T cell infiltration [213]
CAFs	Sigma receptor	Overcoming fibroblast-mediated binding site barrier [220]

Summary and future direction

Aberrant cancer metabolism offers a promising opportunity to identify new targets to improve chemo/immunotherapies. Although many studies focus on a single pathway in cancer or immune cells, it is evident that due to the metabolic plasticity of tumors, inhibition of a single enzyme, transporter or specific pathway is unlikely to be a panacea. Instead, combination approaches that target multiple metabolic pathways coupled with chemo- or immunotherapeutics or targeted interventions offers the greatest potential to cure cancers. However, such approaches also add to the complexity and design of an effective combination therapy and necessitates our deep understanding of each pathway and the complex interplay among multiple pathways. Future studies of cancer metabolism will benefit from emerging and advanced technology as illustrated in several recent publications. For example, in one study that couples metabolomics and genomic sequencing, BRCA1 RNA-level was positively correlated with metabolites of fatty acid β-oxidation and inversely correlated with long-chain fatty acids, highlighting the role of BRCA1 in regulating the balance between fatty acid synthesis and oxidation [215]. In another study, coupling ChIP-seq and metabolomic analysis led to the identification of three major oncometabolites (2-HG, succinate and fumarate) that negatively impacted DNA repair through epigenetic regulation [195]. The continuous advancement of multi-omics technology in the future shall not only further improve our understanding of cancer metabolism but also facilitate the development of precision medicine for the optimal treatment of cancer patients. In addition, the advancement in nanotechnology will help to further improve treatment through improved delivery of versatile agents, especially selective delivery to a subpopulation of cells of interest.

Figure 1. Tumor environment and its impact on tumor cells, immune cells, stroma cells, and chemo/immunotherapy.

The tumor microenvironment (TME) is often characterized by low levels of nutrients and oxygen, acidic pH, and accumulation of oncometabolites. Low levels of nutrients and oxygen activate several growth and survival pathways such as Hif1a, MAPK, and mTOR. Additionally, a metabolic shift towards glycolysis and glutaminolysis leads to epigenetic reprogramming such as abnormal methylation and acetylation that support proliferation, metastasis and chemotherapy resistance. The TME supports immunosuppressive populations (MDSCs, Tregs, and M2-like macrophages), while anti-tumor immune cell populations (M1-like macrophages, NK cells, and effector T cells） develop an exhausted phenotype, resulting in immune evasion and poor response to immunotherapy. Metabolic reprogramming and secretion of immunosuppressive chemokines in CAFs and CAAs also contribute to both chemotherapy resistance and immunotherapy resistance.

Figure 2. Nanotechnology to facilitate targeting of cancer metabolism.

Anti-tumor immune cells compete for nutrients with tumor cells in the TME (①). Without proper targeting, immune cells would also be significantly affected by metabolic inhibition. By targeting tumor cells, nano-drug delivery systems localize metabolic inhibitors in tumor cells to inhibit metabolic pathways in cancer, reducing the undesired impact on immune cell populations (②). Restricting metabolic inhibition to tumor cells also helps to decrease the production of oncometabolites (③), and therefore minimize their impact on immune cells (④). Additionally, nano-drug delivery system allows maximizing the therapeutic dose and the synergistic effect between cytotoxicity agent and metabolic inhibitor (⑤).

Highlight.

• Cancer undergoes significant metabolic reprograming, which provides attractive targets to modulate cancer progression and responses to chemotherapy and/or immunotherapy.

• Success of combining metabolic inhibitors with immunotherapy necessities our improved understanding of the differential effects of metabolic inhibition on cancer cells and different immune cell subpopulations.

• Nanotechnology represents an attractive approach to facilitate cell-type selective delivery and improve the combination therapy of metabolic inhibition with chemotherapy of immunotherapy.