70 Years of DON and Beyond: Glutaminase Inhibition as a Synergistic Tool to Struggle Cancer in Combination Therapy

1. Introduction

After glucose, glutamine (Gln) is the second most avidly consumed molecule by cancer cells, with most tumors showing increased Gln metabolism. Thus, approaches aimed at blocking nitrogen metabolism arose as essential antitumor strategies against cancerous growth and proliferation (Altman et al., 2016; Yang et al., 2017). In 1956, the discovery and biological studies of 6-diazo-5-oxo-L-norleucine (DON) were published for the very first time [3], and shortly after its isolation and characterization were also reported [4], all of which proved DON to be a new useful drug in anticancer therapy. In 1957, the first clinical assay using DON as monotherapy for patients with serious malignancies (mainly lung and breast cancers) was published, but very significant side effects were observed, including ulceration of the tongue, mouth, and lips, diarrhea, nausea, and vomiting [5]. DON was later identified as a structural analog of Gln and its chemotherapeutic usefulness was fully established in the 1970s [6], with later experiments proving the synergistic effects of combining Gln depletion strategies and DON (and another Gln analog, acivicin, chemically (2S)-2-amino-2-[(5S)-3-chloro-4,5-dihydro-1,2-oxazol-5-yl]acetic acid) as powerful molecules inhibiting glutaminase (GA) and cancer growth [7]. In the 1980s, several studies reported promising findings in both in vitro and in vivo models, but subsequent clinical trials produced unpromising results due to low specificity (inhibition of other enzymes whose substrate is Gln, amino acid transporters, and transglutaminase), likely leading to the observed generalized toxicity [8]. Nowadays, less toxic agents have been described and characterized as valuable drugs to inhibit GA [9,10,11], including natural products that have shown its interaction by molecular dynamics simulations but still need to be tested in more in-depth studies [12], as well as derivatives of a natural molecule, withangulatin A, that have already shown antitumoral properties in vitro and in mouse xenografts in the highly Gln-dependent triple-negative breast cancer (TNBC) subtype [13]. However, this review will be focused on fully characterized GA inhibitors, including compound 968, BPTES, and CB-839 (telaglenastat), as well as novel prodrugs that release the inhibitor at the tumor site with maximum efficacy while minimizing toxicity. On the other hand, we emphasize the importance of GA in the metabolic network of tumor cells and its crosstalk with other key molecular pathways, as well as its essential role in the regulatory signals that govern cancer metabolism and modulate the energetic and biosynthetic pathways that need to be targeted in anticancer pharmacological therapy [1,9]. In this review, we focus on recent findings and examine how the historical evolution of DON and GA inhibitors can inform current therapeutic development in oncology.

2. Glutaminase Is a Metabolic Target Against Cancer

Cancer metabolism might be described as a large city underground map, where, when one pathway is blocked, another can be activated to meet the energetic and biosynthetic requirements of cancer cells. Metabolic plasticity is a hallmark of tumors and contributes to the emergence of resistance mechanisms that limit therapeutic efficacy. However, although broad, metabolic plasticity is limited, and blocking several related pathways may circumvent resistance, aiming to synergistically arrest cancer development [14,15]. In this context, GA has been fully characterized as a key target in the metabolic therapy of cancer [16,17,18]. GA is the enzyme responsible for catalyzing the conversion of Gln to glutamate, releasing the amide group from Gln as free ammonia, and represents the first step in glutaminolysis [10]. Glutaminolysis is a commonly upregulated pathway in cancer cells, as it can provide both energy (via generation of α-ketoglutarate and incorporation into the tricarboxylic acid cycle) and biosynthetic precursors for the synthesis of proteins, lipids, and nucleic acids, positioning Gln as a crucial metabolite boosting survival, accelerated growth, and proliferation [19]. Given that the reaction catalyzed by GA is the gate to downstream glutaminolytic metabolism, intense focus has been placed on inhibiting GA as an alternative or parallel target for potential cancer treatment regimens, as this strategy might help tackle the heterogeneity among cancer cells [9,20,21,22]. Several human GA proteins have been identified, encoded by two paralogous genes, GLS and GLS2 [9], each of which gives rise to two different isoforms [23]. The transcripts known as KGA (kidney-type glutaminase) and GAC (glutaminase C) arise by alternative splicing of the GLS gene, being collectively named GLS; whereas two GLS2 transcripts have also been identified: the canonical long transcript termed GAB (glutaminase B) and the short transcript LGA (liver-type glutaminase), which was originally identified in rat liver. Both GAB and LGA can also be indistinctly referred to as GLS2 isozymes [19].

GLS is consistently associated with Gln addiction in tumors and has well-established oncogenic properties; however, the role for GLS2 appears to be more complex, as it has been described as a context-dependent tumor suppressor factor [9,16,17,18]. The heterogeneity in the expression of key metabolic genes—like GLS and GLS2—suggests that different tumors might have differential requirements for glutaminolysis [24]. Accordingly, increased Gln catabolism in mouse liver tumors was associated with decreased levels of Gln synthetase (GS) and a shift from GLS2 to GLS. In sharp contrast, MYC-induced lung tumors display increased expression of both GS and GLS and accumulate Gln. These and other findings make it clear that the tumor metabolic profiles depend on both the genotype and tissue of origin, with critical implications for the design of therapies targeting tumor metabolism [25]. Anyhow, specific GLS inhibition has triggered crucial effects in the therapy of different cancers, as will be reviewed in this article. Importantly, GLS not only regulates the availability of Gln to the tumor but also acts as a tumor supportive protein in many types of cancer, as it has an important impact in pro-tumorigenic signaling [26]. The metabolic shift toward glutaminolysis —and the implications that it has for cancer progression— is driven by oncogene c-Myc-mediated GLS overexpression in many cancers [27,28,29]. Notably, c-Myc coordinates key metabolic pathways, including glycolysis and glutaminolysis. Interestingly, in prostate cancer cells, the crosstalk between these two pathways is mediated by GLS. Specifically, c-Myc-induced GLS indirectly repressed thioredoxin-interacting protein (TXNIP), thereby increasing glucose uptake and glycolysis. Inhibition of GLS restored TXNIP expression and reduced glucose uptake in PC3 cells. [30]. Metabolic reprogramming in cancer cells relies on the overexpression of key enzymes in order that upregulate determinant pathways such as glycolysis, glutaminolysis, and fatty acid synthesis [1]. Combined inhibition of three enzymes essential in these pathways —hexokinase-2, GLS and fatty acid synthase— with lonidamine, DON, and orlistat, respectively, results in a significant reduction in cell viability in the human colon cancer SW480 cell line, as well as in vivo in mice. This approach shows good tolerance, overcomes resistance mechanisms, and produces synergistic effects [31]. This example illustrates a widely adopted strategy in cancer research: combination therapy.

In Figure 1, we depict the main circuits involved in Gln homeostasis in cancer cells, along with major targets susceptible to pharmacological intervention within a multi-target chemotherapeutic approach. On the one hand, and crucially, glutaminolysis is inhibited with a specific inhibitor, among which compound 968, BPTES, and CB-839 have shown the best results (sections 3, 4, and 5). In addition, DON prodrugs and next-generation compounds have emerged as key tools for combination therapies (sections 6, 7, and 8). On the other hand, other key proteins, metabolic pathways, cellular processes, and signaling pathways must also be targeted to effectively constrain tumor growth [10]. In the following sections, we describe GA inhibitors to better understand their effects in monotherapy and, especially, in combination therapies, looking for strategies that not only enhance the antitumor activity of the other drug(s) but also produce synergistic effects, as will be described in detail below.

3. Compound 968 Alone or in Combination Therapy

Compound 968 is a dibenzophenanthridine, chemically 5-[3-bromo-4-(dimethylamino)phenyl]-2,3,5,6-tetrahydro-2,2-dimethyl-benzo[a]phenanthridin-4(1H)-one (Figure 2), which binds GLS and, unlike other inhibitors, also targets the GLS2 isoenzymes [23]. It is an allosteric modulator of GA, which induces apoptosis in cancer cells [26]. The pro-apoptotic effects of compound 968 have been demonstrated in both monotherapy and combination therapy settings [32]. For instance, the presence of compound 968 alone activates the release of the pro-apoptotic serine protease granzyme B, which is primarily secreted by activated T lymphocytes (CD3+, CD4+, and CD8+), boosting tumor infiltration and immune stimulation. This effect is further potentiated in vivo when combined with anti-PD-L1 antibodies. In another study, the competitive mTORC allosteric inhibitor PP242 induced an increase in Gln metabolism. A combination of PP242 and GA inhibition by compound 968 in glioblastoma (GBM) patient–derived xenograft (PDX) models effectively blocked tumor growth. Notably, no significant induction of cell death was observed in normal tissues, including the brain (cortex and hippocampus), liver, and kidney in mice treated with PP242, compound 968 alone, or the combination. These results demonstrate that GA inhibition can reverse mTORC-targeted therapy resistance in vivo by impairing tumor bioenergetics along with PP242 (Table 1), while maintaining low toxicity in normal tissues [33]. These findings may have important implications for combining mTORC kinase inhibitors with GA inhibition in patients with GBM and potentially other mTORC-activated cancers.

Another drug demonstrating greater efficacy in tumor cells than in normal ones, dihydroartemisinin—a semisynthetic derivative of artemisinin—has been shown to exert potent anticancer effects by increasing reactive oxygen species (ROS) levels (Table 1). Importantly, glutaminolysis and, therefore, GA activity, play a critical role in maintaining redox homeostasis, as GA-generated Glu serves for glutathione (GSH) synthesis both through direct incorporation and also by favoring cysteine (cystine) uptake via the xCT transporter (SLC7A11) as depicted in Figure 1. Cancer cells typically rely on glutaminolysis to boost antioxidant systems and support survival. Hence, the combination of dihydroartemisinin with compound 968 showed good efficacy, decreasing antioxidant capacity by disrupting redox homeostasis in cancer cells, while sparing normal cells [34]. Similarly, inhibition of proliferation was selectively enhanced in cancer cells by compound 968, which inhibited tumor growth through down-regulation of the epidermal growth factor receptor (EGFR)/extracellular signal-regulated kinase (ERK) pathway and induction of G1/G0-phase cell cycle arrest [35]. Interestingly, in the case of overcoming adriamycin chemoresistance, the effect appears not to be exclusively linked to GLS inhibition, since the same outcome is not achieved when compound 968 is replaced with another allosteric inhibitor, such as the drug CB-839, which specifically targets GLS isoforms [36].

4. BPTES as an Effective GLS Inhibitor Against Cancer

BPTES is an allosteric inhibitor of GLS isoforms, chemically identified as bis-2-(5-phenylacetamido-1,2,4-thiadiazol-2-yl)ethyl (Figure 2). The efficacy of BPTES as a single antitumor agent has been demonstrated both in vitro and in vivo across several cancer types [37], although its highest potential is typically observed when used in combination with other drugs in multiple malignancies (Table 2). Tumor heterogeneity is a key factor in treatment design and optimization [25]. Therefore, understanding the gene expression profile of each specific tumor type is essential to decide which therapeutic strategy —or combination thereof— can achieve the greatest efficacy [21]. A study conducted at Columbia University provides a compelling example. Mutated NOTCH1, a common alteration in T-cell acute lymphoblastic leukemia (T-ALL), drives cell proliferation by regulating multiple pathways, including glutaminolysis. NOTCH1 signaling channels Gln metabolism into the tricarboxylic acid (TCA) cycle, promoting its use as a carbon source in NOTCH1-induced T-ALL. Dibenzozepine (DBZ), a gamma-secretase inhibitor (GSI), effectively blocks NOTCH1 activity. However, DBZ-treated leukemia cells that overexpress GLS show increased Gln utilization, suggesting that GLS overexpression may represent a potential mechanism of resistance to NOTCH1 inhibition. Combined treatment with DBZ and BPTES exhibited pronounced synergistic effects, impairing cell proliferation and sensitizing cells to NOTCH1 inhibition, with glutaminolysis assuming a dominant metabolic role over glycolysis. In xenograft mouse models, administration of BPTES plus DBZ resulted in strong tumor growth suppression. Notably, the Pten phenotype also appears to be a critical predictive factor for the efficacy of combined DBZ-BPTES therapy, as Pten-deleted T-ALL cells failed to response to GSI alone or in combination with GLS inhibition. Consequently, both GLS expression and Pten status could serve as predictive elements to guide treatment strategies in T-ALL [38].

5. CB-839 Is the Most Successful GLS Inhibitor in Cancer Therapy

CB-839 (telaglenastat) is a derivative of BPTES, chemically 2-(pyridin-2-yl)-N-(5-(4-(6-(2-(3-(trifluoromethoxy)phenyl)acetamido)pyridazin-3-yl)butyl)-1,3,4-thiadiazol-2-yl)acetamide (Figure 2). CB-839 monotherapy has demonstrated significant antitumor activity both in vitro and in vivo [46,47]. However, it is important to note that dual or triple combination therapy has shown a synergistic increase in antiproliferative effects without causing greater damage to healthy cells, while also allowing dose reduction, for example with doxorubicin, which is associated with severe side effects in patients [48]. Although combination therapy is the preferred strategy in most antitumor approaches (Table 3), some studies report that a single nitrogen metabolism inhibitor, such as CB-839, can be even more effective than combining different inhibitors. For instance, in patient-derived xenografts (PDX) of thyroid squamous cell carcinoma (SCC), CB-839 reduced the levels of the antioxidant enzyme NAD(P)H quinone oxidoreductase 1 (NQO1), thereby targeting and diminishing the master regulator nuclear factor erythroid 2-related factor 2 (NRF2) activity, which maintains redox homeostasis and protects cells from oxidative stress and damage. As an anticancer monotherapy, CB-839 was even more effective than the combination of the multireceptor tyrosine kinase inhibitor cabozantinib and the PI3K inhibitor GDC-0326 [49].

Interestingly, a recent study found that colorectal cancer (CRC) models developed resistance to CDK4/6 inhibitor palbociclib by upregulating glutaminolysis, and that combination with CB-839 overcame this resistance, producing synergistic effects [55]. Similar results were reported in lung adenocarcinoma models [56]. These studies highlight the relevance of Gln metabolism as a resistant and survival mechanism in cancer cells, due to its capacity to sustain multiple essential pathways. Therefore, GA inhibition appears to be a relevant adjuvant strategy, even in previously unsuspected contexts such as the CDK4/6 axis. In this case, although mechanisms are not yet fully elucidated, they likely involve downregulation of the E2F/Myc axis following CDK4/6 inhibition, thereby creating a synthetic lethal relationship with glutaminolysis and, consequently, with GA.

6. DON Returns Like a Trojan Horse

Although some recent studies have successfully used DON in combination therapy, for example, with MDiVi-1, a mitochondrial fission inhibitor that demonstrated antitumor efficacy in vitro and in vivo in ovarian cancer models [78], toxicity concerns have prevented DON from becoming the anti-Gln drug of choice. In 2016, a breakthrough opened a new avenue for DON use in the form of a prodrug, ethyl 2-(2-amino-4-methylpentanamido)-DON, named JHU083, with the critical advantage of inactive systemic administration of the compound and preferential accumulation and activation in the tumor microenvironment, allowing adequate concentrations at the site of action while minimizing side effects [79]. In 2019, another DON prodrug, (S)-isopropyl 2-((S)-2-acetamido-3-(1H-indol-3-yl)propanamido)-6-diazo-5-oxohexanoate, known as DRP-104 (sirpiglenastat), was characterized, retaining the benefits of DON but avoiding side toxicity [80]. Notably, DRP-104 is bioactivated by serine proteases in the tumor, where it releases the antitumor drug DON, while it is bioinactivated by carboxylesterases in healthy gastrointestinal tissue to a non-toxic molecule [81]. Since then, several studies have confirmed the utility of both compounds in different tumor models (Table 4 and Table 5).

And, as if completing a circle, we return to the beginning with the chemical structure of DON (Figure 3). After decades of neglect, the inhibitory potential of DON has reemerged in the form of chimeric prodrugs that release the active compound—the Gln analog itself—directly at the tumor site to perform its precise function, while avoiding the drawbacks of its intrinsic toxicity. DON forms an irreversible covalent complex with GA by binding to the active-site serine residue and releasing the diazo group [96]. Thus, masked as precursors of the active agent, these prodrugs have once again highlighted the strong antitumor activity of DON. Interestingly, monotherapy with JHU083 for 14 days successfully reprogrammed metabolism in mouse models of melanoma, lymphoma, and colon cancer [82]. Notably, clinical trials are already underway for DRP-104 (NCT04471415, NCT06027086) and trials for JHU083 are also expected.

7. Targeting GA in Clinical Trails

As described above, the antitumor activity of CB-839 has been extensively characterized in multiple cell-based cancer models and in vivo (section 5). Highly significant effects have been observed with both monotherapy [68,69] and combination therapy (Table 1). One of the most frequently reported antitumor effects is increased oxidative stress, often combined with radiotherapy for enhanced efficacy [68,69,97]. However, the options are very diverse, as demonstrated across various models confirming its value as an antitumor drug (Table 1). CB-839 has been the most studied GA inhibitor to date. Although its solubility and pharmacokinetic properties are suboptimal, its high specificity and potency have driven its inclusion in numerous clinical trials. These include: as monotherapy in solid tumors, specifically metastatic or unresectable malignant peripheral nerve sheath tumors (NCT03872427) [98]; in combination therapy with the monoclonal antibody panitumumab (anti-EGFR) in CRC (NCT03263429), with the antimetabolite azacytidine in myelodysplastic syndromes (NCT03047993), using a 5-fluorouracil (5-FU) prodrug as capecitabine in CRC (NCT02861300) [98], with a taxane like paclitaxel in triple negative breast cancer (TNBC) (NCT03057600), with the tyrosine kinase inhibitor (TKI) cabozantinib in renal cell carcinoma (RCC) (NCT03428217) [99], together with the TKI palbociclib in metastatic solid tumors (NCT03965845), with the TKI osimertinib in non-small cell lung cancer (NSCLC) (NCT03831932), with the mTOR inhibitor everolimus in RCC (NCT03163667), together with the mTORC inhibitor sapanisertib in NSCLC (NCT04250545) [100], with the proteasome inhibitor carfilzomib in multiple myeloma (NCT03798678), and with radiation therapy and temozolomide in treating patients with isocitrate dehydrogenase (IDH)-mutated diffuse astrocytoma or anaplastic astrocytoma (NCT03528642) [101]. Additionally, it was used in a phase II clinical trial with radiotherapy and cisplatin in cervical cancer (NCT05521997), a phase I/II trial using the antibody nivolumab for targeting programmed cell death protein 1 (PD-1) in advanced cancers, i.e., melanoma, NSCLC and RCC [97], and in a phase II clinical trial with pembrolizumab to equally target PD-1 in NSCLC (NCT04265534) [73].

Many articles highlight the need for a combined therapeutic approach that includes GA targeting, as Gln metabolism affects multiple cellular processes and metabolic pathways, including oxidative stress [102], DNA damage [103], response to radiotherapy [97], alterations in the harsh tumor microenvironment (TME) [104], and changes in the immune response [90,105]. A significant step toward clinical translation occurred in 2020 with the development of IPN60090 (or IACS-6274), a BPTES-derived molecule with the chemical structure 1-[(2R)-4-[6-[[2-[4-(3,3-difluorocyclobutyl)oxy-6-methylpyridin-2-yl]acetyl]amino]pyridazin-3-yl]-2-fluorobutyl]-N-methyltriazole-4-carboxamide. IPN60090 showed greater potency and selectivity than CB-839, with optimal pharmacokinetic properties characterized in mouse, rat, and dog [106]. This drug is currently being evaluated in a phase I clinical trial (NCT03894540). In epithelial ovarian cancer, IPN60090 has been shown to inhibit Zinc finger SWIM-type containing 4 (ZSWIM4), a transcription factor that induces drug resistance. Its effects have been characterized in vitro and in mouse xenografts, where it sensitized tumors to a CBP inhibitor (targeting CREB-binding protein and its paralog p300), and reduced tumor volume through an apoptosis/glutathione-dependent mechanism. These antitumor effects were synergistically augmented when combined with paclitaxel in patient-derived organoids (PDO) models [107]. In acute myeloid leukemia (AML) cells, IPN60090 induced apoptosis, inhibited tumor growth, and suppressed NADPH and ATP synthesis. Accordingly, it exhibited synergistic antiproliferative effects when combined with with venetoclax, a BCL-2 inhibitor [67]. In a novel nanostrategy, IPN60090 was encapsulated in nanospheres targeting mitochondria, together with copper and IR780. While IRP60090 blocks GSH-mediated antioxidant defense, copper accentuates oxidative stress through cuproptosis, and IRP780 enhances the immune response and immunogenic cell death, achieving substantial synergistic antitumor effects in breast cancer cells both in vitro and in vivo [108].

8. Future Perspectives

Targeting Gln metabolism has been extensively validated as a powerful anticancer strategy [9,10,11,101,102,103,104,105]. In this context, targeting GA appears particularly promising, as it represents the main rate-limiting step of glutaminolysis, a pathway that provides bioenergetics, biosynthetic intermediates, and antioxidant protection. In fact, many cancers overexpress GA, particularly GLS, making its inhibition a prime opportunity to reverse cancer cell metabolic reprogramming, limit proliferation and render cells vulnerable to death induction, thereby enhancing sensitivity to chemoradiotherapy and improving therapeutic responses [10]. For immunotherapy-nonresponsive cancers, the prodrug JHU083 emerges as a promising novel treatment, even in the presence of immune-suppressing macrophages and scarce T-cells, positioning it as a potent option for tumors unresponsive to checkpoint inhibitors. Besides, it has demonstrated substantial therapeutic benefits in urological cancers, particularly by reducing angiogenesis in the TME [90].

Combinations aimed at boosting immune response are of particular interest here, since Gln depletion in the TME, caused by unrestrained incorporation and metabolism by Gln-addicted tumors, critically limits lymphocyte activation and growth, which critically rely on Gln for clonal expansion, thereby constituting an immune escape mechanism for cancer cells. Hence, limiting glutaminolysis by tumors increases Gln availability to immune cells in the TME, boosting immune response. Very recently, a new DON chimera has been designed by fusing DON with JQ1 to leverage JQ1’s well-established downregulation of PD-L1; chemically (6S)-4-(4-chlorophenyl)-2,3,9-trimethyl-6H-thieno [2-f]triazolo [3-a]diazepine-6-acetic acid 1,1-dimethylethyl ester, resulting in the new compound HB023 [109]. This new drug has shown great chemotherapeutic capacity in a murine model of colon cancer, overcoming tumor immune evasion and increasing anticancer efficacy by cooperatively activating both adaptive and innate antitumor immunity. Similar results have been observed in a murine model of clear cell renal cell carcinoma (ccRCC) with DRP-104, and with the GLS inhibitor CB-839 combined with PD-L1, achieving synergistic increases in antitumor immunity [110]. Indeed, anti-PD-1 and anti-PD-L1 therapies are revolutionizing cancer treatment, as drug resistance remains a major challenge; immune checkpoint inhibitors thus emerge as valuable tools to overcome multidrug resistance and collaborate with Gln metabolic reprogramming to achieve durable anticancer responses [32,33,72,73,82,85,86,90,93,105,110]. Collectively, the investigations described above emphasize the importance of combination therapy and the optimization of synergistic approaches, in which targeting Gln metabolism serves as a valuable tool.

Notably, Gln-based positron emission tomography (Gln-PET) imaging using (2S, 4R)-4-[¹⁸F]Gln has proven effective for tumor diagnosis [111], disease tracking [112], therapeutic monitoring [113], and pharmacodynamic of drugs such as CB-839 [78]. This technique exhibits the highest reproducibility and repeatability in preclinical cancer models [114], and has been validated in a phase I/II clinical trial of CB-839 (NCT03263429) in patients with metastatic CRC [75]. In personalized oncology, Gln-PET enables organ- and patient-specific monitoring based on the organ being studied and the patient [77], including analysis of the TME profiling, as demonstrated in a mouse model and a patient with brain metastases from invasive ductal breast carcinoma [115]. In addition to addressing some of the main challenges in antitumor therapy, such as cancer heterogeneity, drug resistance, and immune evasion, nano-delivery strategies will be essential for modulating the TME through combination therapy and immunometabolic crosstalk [116]. Of particular interest are chimeric compounds that exhibit immunometabolic bifunctionality. For instance, Zhang et al. combined the prodrug JHU083 (which blocks Gln metabolism) with MSA-2, an agonist of the interferon gene activation pathway that regulates innate immunity in the TME and elicits a robust immunogenic response [117]. This bifunctional construct has demonstrated very promising results in vitro and in vivo in colon cancer models., with synergistic enhancement when combined with 5-FU or anti-PD-L1 therapy [117]. Nevertheless, future research will must study how, for every single cancer, Gln blockage can support or impair antitumor immunity, depending on tumor type, immune context, dosage, compartment and treatment combination.

9. Conclusions

Many tumors exhibit significant intrinsic plasticity, triggering metabolic reprogramming to develop drug resistance and enhance biosynthetic and signaling pathways that sustain their proliferation program [118,119]. Therefore, combination therapy represents one of the few viable strategies for successful neoplasm treatment (Table 1, Table 2, Table 3, Table 4 and Table 5). Several reviews have described the activity of multiple GA inhibitors [9,10,11,101], as Gln metabolism constitutes a central network for tumor growth and survival, providing energy and essential substrates [23,105,120]. In this work we emphasize the need to design targeted therapeutic interventions based on blocking and/or activating different metabolic and/or cell signaling pathways looking to counteract the great adaptability of tumors to oxidative and/or metabolic stress conditions. Inhibition of GA, either through allosteric inhibitors (i.e.,: compound 968, BPTES, CB-839, IPN60090) or Gln analogs (i.e.,: DON, JHU083, DRP-104, HB023) represents a useful intervention targeting glutaminolysis, which as discussed constitutes an essential pathway for cancer survival and progression. This approach offers dual advantages: synergistic efficacy with a broad range of drugs, and reduced dosing requirements for certain agents, thereby minimizing toxicity to healthy cells. Combination therapy can integrate Gln metabolism inhibitors with other chemotherapeutic drugs, radiotherapy, or immunotherapy, optimizing overall efficacy by targeting multiple pathways while mitigating resistance development [121]. Recent research and ongoing clinical trials position CB-839 and DRP-104 as the most promising drugs targeting nitrogen metabolism, either by blocking GA (CB-839) or inhibiting Gln across multiple enzymes (DRP-104) [23]. Both demonstrate efficacy as monotherapy but excel in combination regimens, with CB-839 notably restricting TCA cycle flux, and DRP-104 broadly suppressing proliferative nodes like purine biosynthesis [122]. Moreover, DRP-104 uniquely modulates antitumor immune activity [81] by reshaping the TME [90,93]. A dysregulated, tumor-supportive TME represents a hallmark of cancer [123], and targeting Gln metabolism shows great potential for abrogating the tumor-supportive environment in both preclinical models [78] and patients [115]. Efforts to optimize inhibitors—including prodrug formulations—continue, exemplified by a tert-butyl ester derivative of DRP-104 that offers improved solubility and in vivo stability [124].

The above findings underscore the value of identifying possible targeted therapy-induced resistance mechanisms to design novel combination approaches that achieves the previously discussed advantages. Recent strategies have proposed integrating Gln metabolism into cancer diagnosis, classification, treatment, and monitoring [14]. In animal models, many tumors are especially dependent on Gln metabolism [21]. This observation suggests that imaging, quantifying, or blocking Gln metabolism in human cancers could be incorporated into the diagnosis and management of the disease [125], including PET-based technologies [77,110,126]. In most cases, it would be beneficial to use in vivo perioperative administration of isotope-labeled biomarkers (glucose and Gln) to cancer patients to differentiate metabolic pathways between tumors and benign tissue [127]. Additionally, in vivo Gln metabolic studies may help predict which tumors will respond to metabolism-targeted therapies [2].

In contrast, it should not be overlooked that several studies have identified important limitations and challenges associated with glutaminase inhibitors, including: (i) the need for isoenzyme-specific GLS inhibition when GLS2 acts as a tumor suppressor [17]; (ii) intestinal side effects caused by mucosal damage and cell death following GA inhibition [128]; (iii) metabolic reprogramming in some tumors through increased lipid oxidation to compensate for reduced glutaminolysis, thereby limiting the efficacy of GA inhibition [23]; (iv) compensatory increases in glucose metabolism through pyruvate carboxylase [14]; (v) the ability of some KRAS-mutant cancers to acquire Gln through alternative mechanisms, reducing the effectiveness of GA as a therapeutic target [75,121]; (vi) the incompatibility of GA inhibition with KEAP1 and STK11/Lkb1 co-mutations in KRAS-mutant lung adenocarcinoma because glutamate is required for CD8 T-cell activation [129]; and (vii) the high dependence of immune-system T cells on Gln metabolism for proliferation, requiring careful evaluation of GA blockade to achieve a balanced therapeutic response [130]. Hence, further research is needed to elucidate the metabolic consequences of GLS inhibition in combination with immunotherapy, as well as potential resistance mechanisms and side effects. However, there are also highly promising therapeutic findings, for example in ovarian cancer [32] as well as in lung and colon cancer [45]. For possible personalized use in the future a strong and in deep study must be done, including tumor type, selected biomarkers, immune context, glutamine dependency, GLS versus GLS2 biology, metabolic plasticity, resistance mechanisms, and adequate patient selection. In addition, understanding the genetic and epigenetic circuits regulating GLS will facilitate the development of improved combination therapies that exploits Gln metabolism reprogramming for immune evasion, ultimately leading to novel anticancer drugs and precision oncology approaches [82,105,118,131,132]. In this field, the most promising research strategies appear to integrate anti-PD-1/PD-L1 therapies with targeting of metabolic interactions within the TME that promote tumorigenesis (Figure 4) [133]. Indeed, several very recent analyses have identified compelling combination therapy approaches showing clinical efficacy [134].