Colorectal cancer (CRC) is a major disease and in 2020, there was an estimated 1.93 million incidences globally [1], with a worrying increase in incidence in the younger population [2–4]. As CRC is multistep process, early detection and effective prevention remain the most promising way to reduce CRC-associated mortality. Aspirin has a long-established association with a reduced risk of CRC [5–12]. However, the optimal dose and duration for effective prevention remains up for debate and more recent studies suggest risks may outweigh benefits for older individuals [6, 13–15]. Importantly, the potential therapeutic use of aspirin is also being explored and significant research is being undertaken to define a role for aspirin in CRC treatment. To better use this widely available, cost-effective drug for the management of CRC, it is important to understand its mechanism of action.

Aspirin (also known as acetylsalicylic acid) is a nonsteroidal anti-inflammatory drug (NSAID) used as an analgesic and anti-inflammatory agent, often given to patients at high risk of cardiovascular disease [16, 17]. Like other NSAIDs, aspirin inhibits cyclooxygenases (COXs, COX1 and COX2), enzymes responsible for the generation of prostanoids from arachidonic acid [18, 19]. COX1 is constitutively expressed and important for the production of the thromboxane (TX), TXA2, which is involved in platelet aggregation and vasoconstriction. By contrast, COX2 is inducible and is often expressed in inflammatory and hypoxic conditions where it is involved in prostaglandin E₂ (PGE₂) synthesis. Although the COX pathway is a well-established target of aspirin, recent evidence has highlighted additional COX-independent actions which have become increasingly important in explaining the efficacy of aspirin for cancer prevention and therapy [20–22].

CRC is the fourth most common cancer in the UK and accounts for 10% of all cancer mortalities, placing it the second most common cause of cancer death in the UK [4]. Evidence of the association between aspirin use and reduced CRC incidence is substantial [5–12]. Evidence first arose from observational studies and meta-analyses of randomised control trials (RCTs) for the prevention of vascular disease and is the subject of many excellent reviews [23, 24]. A meta-analysis of all observational studies up to 2019 indicated a dose-dependent risk reduction of 10–35% (75–100 mg/day conveys 10% risk reduction and 325 mg/day conveys 35% risk reduction) and concludes that aspirin use is associated with a ≥ 25% CRC risk reduction in forty-five observational studies [25]. Based on this evidence, the United States Preventative Task Force (USPTF) recommends low-dose aspirin use for the prevention of CRC in 50–59 years old adults [26]. In the UK, aspirin is recommended for patients with Lynch syndrome, a familial condition characterised by germline mutations in DNA mismatch repair genes, most commonly MutS homolog 2 (MSH2) and MutL homolog 1 (MLH1) [27].

Although less well established, there is also evidence for the use of aspirin as an adjuvant therapy. Rothwell et al. [28] have shown that aspirin ranging from 75–1,200 mg/day is associated with a 21% reduction in mortality from any cancer. In the context of CRC, a cohort study identified doses of 75–300 mg/day of aspirin post-diagnosis to be protective against CRC-associated mortality, this was significant at the lowest dose (75 mg/day) [29]. Several other studies also associate aspirin use with decreased CRC mortality [30–34]. A commonality between these studies is that to observe reduced mortality, aspirin must be administered post-diagnosis rather than pre-diagnosis. The effects of aspirin on mortality are seen after a period of ~5 years [35], which is thought to be due to aspirin reducing the risk of metastasis. A review of five RCTs found a reduced risk of metastasis with daily aspirin use but this only included patients who started taking aspirin post-diagnosis [36]. Research into the use of aspirin as an adjuvant therapy is ongoing; the add-aspirin trial is designed to investigate aspirin use and colorectal survival (alongside breast, prostate, stomach, and oesophageal cancer) and includes doses ranging between 100–300 mg/day (tertiary prevention) [37].

One disadvantage of aspirin is that it carries a risk of gastrointestinal (GI) bleeding, with a low dose of aspirin use associated with a risk of 2–4 GI bleeds per 1,000 patients. However, this risk depends on the patient’s age, gender, and history of GI ulcers [38]. The ASPirin in Reducing Events in the Elderly (ASPREE) trial concluded that aspirin increased the risk of major hemorrhage in adults > 70 years old although this risk was age-dependent [14]. The absolute risk of serious GI bleeding was 0.25% for a 70-year-old with no additional contributing factors. However, this increased to a 5% risk for an 80-year-old with multiple risk factors (smoking, hypertension, and previous NSAID use) [14]. If aspirin is to be used routinely in a cancer setting, it is crucial to identify optimal treatment regimens and patient subgroups who would benefit from aspirin use.

It is important to understand the mechanism by which aspirin is having anti-tumour effects, potentially opening the door to further “aspirin-related drugs” for CRC prevention and therapy. There has been an increasing number of studies contributing to the mechanistic understanding of aspirin’s anti-tumour effects in CRC including regulation of wingless-related integration site (Wnt), nuclear factor kappa light chain enhancer of activated B cells (NF-κB), hypoxia-inducible factor-1α (HIF-1α), mammalian target of rapamycin (mTOR), and phosphatidylinositol 3-kinase (PI3K) pathways which is the subject of excellent reviews (Figure 1) [20, 22, 39, 40]. More recently a role for aspirin in metabolic reprogramming of cancer cells has begun to emerge, which is the focus of this review.

The reprogramming of cellular metabolism is an established hallmark of cancer and is essential to support the increased demand for biosynthetic intermediates required for sustained proliferation [41]. In non-malignant cells, nutrient-sourced carbons are predominantly used to generate adenosine triphosphate (ATP). However, tumour cells increase the import of nutrients from the tumour microenvironment and reprogramme metabolic pathways, primarily to fuel biosynthetic processes (Figure 2). Shifting to anabolic metabolism allows biomass accumulation through nucleotide, amino acid, and lipid synthesis. Oncogenic pathways alongside mutations in metabolic enzymes can rewire nutrient consumption and utilisation to meet the bioenergetic and biosynthetic needs of the tumour cell. For example, the receptor tyrosine kinase (RTK)-PI3K-Akt pathway commonly drives the expression of glucose transporter 1 (GLUT1) and subsequently increased glucose uptake for tumour biomass production [42]. CRC cells exhibit common features of aberrant metabolism such as increased glycolytic flux (the Warburg effect) and glutamine utilisation [43–45]. The reprogramming of CRC cell metabolism supports tumour development and the switch to an increased glycolytic rate has been shown to occur early in CRC progression [46–48]. Therefore, targeting cancer metabolism is an attractive therapeutic approach. Antimetabolites such as 5-fluorouracil (5-FU) are routinely used as chemotherapy; these antimetabolites target the increased demand for nucleotide synthesis [49]. However, targeting proliferative metabolism in general often leaves an inadequate therapeutic window as non-malignant rapidly proliferating cells rely on a similar metabolic programme which results in toxicity. To combat this, there is a focus on targeting specific metabolic dependencies adopted by malignant cells for novel therapeutic strategies. In recent years, specific metabolic inhibitors have gained momentum for cancer therapy such as ivosidenib and enasidenib for relapsed/refractory IDH-mutated acute myeloid leukaemia [50, 51].

Glutamine dependency of tumour cells is an attractive therapeutic target, glutamine analogues reached clinical trials many years ago but were discarded due to lack of efficacy [52]. This is commonly seen when using metabolic inhibitors as single agents due to the ability of cancer cells to overcome metabolic perturbation by inducing compensatory pathways to survive. For example, upon glutaminase (GLS) inhibition in pancreatic ductal adenocarcinoma (PDAC) cells, the upregulation of pyruvate carboxylase (PC), activation of lipid biosynthetic pathways, and alternative pathways involved in glutamate production allowed cells to adapt and regain proliferation [53]. It is extremely important to analyse and understand the compensatory mechanisms induced by metabolic inhibition, as this gives rise to the possibility of combination therapies. Targeting metabolic enzymes/pathways in combination therapies restricts the adaptive metabolic network and the ability to adopt alternative pathways which are needed to circumvent metabolic inhibition. Here we propose that, through its action on cancer cell metabolism, aspirin could provide a simple, relatively safe, and cost-effective way to target this important hallmark of cancer and potentially be used as adjuvant therapy to improve the efficacy of metabolic inhibitors already in clinical trials.

There have been a number of studies investigating the role of aspirin in metabolic rewiring in CRC. A metabolomics study on human colon tissue taken from patients given 81 mg aspirin a day found the main metabolic processes associated with aspirin treatment were energy, nucleotide, and amino acid metabolism [54]. Importantly, 81 mg of aspirin a day was associated with a decrease in adenoma risk suggesting the regulation of these processes is key to the chemopreventative effect of aspirin. Mechanistically, aspirin has been found to acetylate several metabolic enzymes, 14 lysine residues were found to be acetylated by aspirin on the glucose-6-phosphate dehydrogenase (G6PD) protein [55]. G6PD is a key enzyme in the pentose phosphate pathway, a biosynthetic branchpoint from glycolysis generating ribose-5-phosphate, which is essential for nucleotide synthesis and nicotinamide adenine dinucleotide phosphate (NADPH), an important reducing agent for protection against reactive oxygen species (ROS). The acetylation of G6PD by aspirin markedly decreased its enzymatic activity, suggested to contribute to reduced synthesis of ribose sugars and NADPH [55]. Aspirin also acetylates other metabolic enzymes such as pyruvate kinase M2 (PKM2) and lactate dehydrogenase (LDH), however, this was not found to affect their enzymatic activity [56]. G6PD is overexpressed in CRC patient specimens and predicts poor prognosis [57], and therefore, suppression of G6PD activity may be an important mechanism underlying the anti-tumour effects of aspirin.

Some recent studies have suggested that the growth-inhibitory effects of aspirin may be dependent on targeting glutamine metabolism. Glutamine deprivation was found to limit the anti-growth effect of aspirin in HCT-15 and HCT116 cells [CRC cell lines harbouring phosphatidylinositol-4,5-bisphosphate 3-kinase catalytic subunit alpha (PIK3CA) mutations] [58]. Interestingly, a number of studies have linked the efficacy of aspirin to PIK3CA mutational status [59, 60]; evidence has shown that aspirin inhibits the phosphorylation and therefore activation of PI3K and Akt [59]. Aspirin treatment appears to mimic the effects of glutamine depletion on the cell cycle and biosynthetic pathways, by both inducing G1 arrest and causing mTOR complex 1 (mTORC1) inhibition [58]. Consistent with this, aspirin was shown to upregulate glutaminolysis-associated genes such as GLS 1, alanine-serine-cystine transporter 2 (ASCT2), large neutral amino acid transporter 1 (LAT1), and glutamic-pyruvic transaminase 2 (GPT2) in PIK3CA mutant CRC cells [58]. This upregulation by aspirin is proposed to be via the induction of activating transcription factor 4 (ATF4) signalling, a major metabolic regulator often upregulated in response to glutamine deprivation [61].

In support of this work, aspirin was shown to regulate cellular metabolism in CRC cells but this was not dependent on PIK3CA mutational status [62]. Aspirin was found that aspirin downregulates a number of key glutaminolysis enzymes including GPT2, glutamate dehydrogenase 1 (GLUD1), glutamic-oxaloacetic transaminase 2 (GOT2), and the glutamine transporters SLC7A11 and SLC7A5 leading to reduced levels of glutaminolysis [62]. This work was conducted in cells treated with aspirin long-term (across 52 weeks), which may account for the differences with other published studies. As a likely compensatory mechanism (and consistent with previous work), it was found that CRC cells upregulate GLS1 upon aspirin exposure [62]. This upregulation of GLS1 exposed a metabolic vulnerability which was exploited in a treatment combination with telaglenastat (CB-839), a potent and selective inhibitor of GLS1 [63]. CB-839 is in phase I clinical trials in patients with solid tumours (ClinicalTrials.gov identifier: NCT02071862) and advanced CRC (ClinicalTrials.gov identifier: NCT02861300) but has shown limited efficacy when used as a monotherapy. It was demonstrated that both long-term (52 weeks) and short-term (72 h) aspirin treatment sensitised CRC cell lines to CB-839. Importantly, the combination of aspirin and CB-839 was shown to reduce colon crypt proliferation in vivo [62]. Excitingly, these findings add weight to the growing evidence that aspirin has therapeutic potential, in addition to chemoprevention, as an adjuvant to metabolic inhibitors currently under clinical investigation.

As mentioned above, oncogenic pathways frequently dysregulated in CRC such as Wnt, NF-κB, mTOR, and HIF-1α are reported to be targeted by aspirin. These key signalling nodes also drive metabolic reprogramming in tumour cells suggesting that the regulation of these pathways by aspirin may be key to its metabolic effect (summarised in Figure 3).

This review focuses on the emerging role of aspirin as a regulator of metabolic reprogramming. Cancer cells frequently undergo metabolic rewiring, driven by oncogenic pathways such as HIF, Wnt, mTOR, and NF-κB, to support the increased proliferative rate as tumours develop and progress. Although specific metabolic inhibitors have gained momentum for cancer therapy, their use has often proved ineffective due to the metabolic plasticity of cancer cells. Cellular metabolic reprogramming has been identified as a key mechanism of action of aspirin and includes the regulation of key metabolic drivers, glycolytic and glutaminolysis enzymes, and altered nutrient utilisation upon aspirin exposure. The targeting of metabolic plasticity by aspirin exposes metabolic vulnerabilities and provides an exciting opportunity for the use of aspirin in combination with specific metabolic inhibitors (including GLS inhibitors such as CB839 and IACS-6274, currently in clinical trials) for the treatment of CRC and potentially other cancers. Although there remains debate around the optimal dose and duration of aspirin treatment to minimise risk in the context of prevention [6], research also highlights a new role for aspirin in improving the efficacy of a new generation of metabolic inhibitors currently undergoing clinical investigation.