It mainly includes the intake of high fat, high protein, low dietary fiber and nitrite compounds; vitamin deficiency; intestinal flora imbalance and other factors. Obesity induced by high fat diet has been a high risk factor for CRC.

High fat diet not only can stimulate the increase of bile secretion, but also promotes the growth of some anaerobic bacteria in the intestine. Bile alcohol and bile salt are decomposed by anaerobic bacteria to form unsaturated cholesterol, such as deoxycholic acid and lithocholic acid, which is carcinogens or co-carcinogens (6). Studies have shown that high fat diet can reduce the expression of the major histocompatibility complex class II (MHC class II) genes in intestinal epithelial cells (including intestinal stem cells), thereby disrupting the intestinal flora and promoting intestinal tumorigenesis (7).

High protein diet can increase harmful metabolites in the gut, such as N-nitroso compounds (NOCs) and H₂S. NOCs can lead to DNA alkylation, which can induce genetic mutations that ultimately lead to cancer (8, 9). After ingestion, protein is decomposed into various absorbable amino acids including methionine and cysteine that can be further decomposed by sulfate-reducing bacteria to produce H₂S, which can promote the occurrence of CRC by inhibiting butyrate oxidation, destroying the intestinal barrier, and interacting with ROS to induce DNA damage (10, 11).

Dietary fiber can increase intestinal peristalsis, promote stool excretion, and reduce the food transit time in the colon, thus reducing the contact time of potential carcinogens with intestinal mucosa (12, 13). Moreover, the residues of dietary fiber can be fermented by intestinal microbiota to produce short chain fatty acids (SCFA), especially butyrate, which has important physiological effects. Butyrate is not only the main energy source of intestinal microbiota and intestinal epithelial cells, but also plays an important role in colon health by inhibiting cell proliferation, inducing cell differentiation, promoting cell apoptosis, and reducing tumor cell invasiveness (14, 15). Therefore, lack of dietary fiber will increase the risk of intestinal diseases.

Chronic inflammatory stimulation can lead to the development of CRC. Inflammatory bowel disease (IBD) mainly includes ulcerative colitis and Crohn’s disease, which is one of the high risk factors for the development of CRC (25). Compared with the general population, patients with IBD have an approximately 2-6 folds increased risk of developing CRC (26).

Population aging has a significant impact on the incidence of CRC. Surveys show that the incidence of CRC worldwide increases with age. In the United States, the incidence of CRC is three times higher in people over 65 years old than in 50-64 years old, and approximately 30 times higher than in 25-49 years old (27).

The CIN is the most classical pathway, accounting for 80-85% of all CRC cases. Aneuploidy or structural chromosomal abnormalities, frequent loss of heterozygosity at tumor suppressor loci, and chromosomal rearrangement are the main features (32). These changes will affect important genes related to the maintenance of cell function, such as adenomatous polyposis coli (APC), KRAS and TP53. About 80% of CRC patients have APC mutations, which can active Wnt signal pathway. There are three characteristic genes related to human tumors in RAS gene family including HRAS, NRAS and KRAS. Among them, KRAS is the most mutated in 40% of sporadic CRC (33). KRAS mutations can active MAPK and PI3K signaling pathway, which increases cell proliferation (34). It is well known that p53 is a tumor suppressor protein, which is encoded by the TP53 gene. TP53 mutations occurs in 40-50% of sporadic CRC, which is a key step driving the development of CRC (35).

Microsatellite is short nucleotide tandem repeat in the DNA sequence, which is easy to get a replication error due to repeated structure. Mismatch repair (MMR) system mainly including MLH1, MSH2, MSH6, PMS1 and PMS2 proteins can repair these errors. MSI is driven by the loss of function of MMR system, and promoter hypermethylation is considered to be the main cause of gene silencing. MSI accounts for about 15% of sporadic CRC. Moreover, germ-line mutations in these genes causes Lynch syndrome, which is the most common inherited type of CRC (36).

Site specific DNA hypermethylation of CG dinucleotides (CpG islands) associated with promoters of tumor suppressor genes and DNA repair genes leads to transcriptional silencing that promotes cancer initiation and progression. CIMP is first identified in CRC, which is related to the hypermethylation of CpG islands in the promoter region of tumor suppressor genes such as MINT1, MINT2, MINT3 and MLH1 (37, 38). It is a form of epigenetic modification and can be divided into two type. CIMP-high subtype is BRAF mutations and MLH1 methylation. CIMP-low subtype is KRAS mutations. Studies have shown that BRAF and KRAS direct the assembly of distinct corepressor complexes on a common promoter through different pathways leading to promoter hypermethylation and transcriptional silencing of CIMP (39). BRAF increases phosphorylation of MAFG (a small MAF protein) via the MEK/ERK signaling pathway, thereby protecting MAFG from polyubiquitination and subsequent proteasome-mediated degradation. Subsequently, MAFG recruits corepressor complexes including BACH1, CHD8 and DNMT3B, resulting in promoter hypermethylation and transcriptional silencing of CIMP (40). KRAS increases the level of zinc finger protein 304 (ZNF304) by transcriptionally upregulating the serine/threonine kinase protein kinase D1 (PRKD1) and ubiquitin-specific peptidase 28 (USP28). Subsequently, ZNF304 recruits corepressor complexes including KAP1, SETDB1 and DNMT1, resulting in promoter hypermethylation and transcriptional silencing of CIMP (41). CIMP is highly associated with the CRC developing from serrated polyps pathway, but the underlying mechanism remains to be further explored.

The main features are decreased mitochondrial volume, decreased or disappearance crista, increased mitochondrial membrane density, disruption of mitochondrial membrane and without chromatin condensation. However, the nucleus structure is intact and morphological changes are not obvious (53).

It is mainly manifested in the increase of reactive oxygen species (ROS), the aggregation of iron ions, the increase of oxidation level of reduced coenzyme II (NADPH), the inhibition of glutathione peroxidase 4 (GPX4) and cystine/glutamic acid antiporter transporter (system Xc^-) (54).

Ferroptosis not only affects innate immunity by affecting the number and function of immune cells (e.g. macrophages and neutrophils) but also affects adaptive immunity through triggering inflammatory or specific responses after ferroptotic cells recognized by immune cells (e.g. T and B lymphocytes). The death of immune cells caused by ferroptosis may impair the immune response. In contrast, non-immune cell death caused by ferroptosis can activate damage-associated molecular patterns (DAMP), thereby activating immune responses (55).

The genes/proteins that regulate ferroptosis can be mainly divided into three categories, among which SLC7A11, SLC3A2, ATF3, p53, etc. mainly regulate cystine uptake; GPX4, ACSL4, LPCAT3, FSP1, NRF2, etc. mainly regulate lipid metabolism; TFR1, HSPB1, IREB2, NFS1 and others mainly regulate iron metabolism (56).

Lipid peroxidation is key to the occurrence of ferroptosis. Compared with monounsaturated fatty acids, polyunsaturated fatty acids (PUFA) have diallyl moiety. Therefore, phospholipids containing polyunsaturated fatty acids (PUFA-PL) are very susceptible to peroxidation (57). The peroxidation of PUFA-PL is mainly catalyzed by non-enzymatic autooxidation, which is actuated by Fenton reaction and catalyzed by iron. Lipid peroxides can be formed by capturing an unstable hydrogen atom in diallyl group on PUFA-PL (58).

Acyl-coa synthetase long-chain family member 4 (ACSL4) can catalyze the conversion of long-chain fatty acids to fatty acyl-coenzymes in an ATP-dependent manner, which plays a crucial role in the process of iron-dependent oxidative stress (59). ACSL4 tends to bind long-chain polyunsaturated fatty acids such as arachidonic acid (AA) and adrenic acid (ADA), and then specifically esterifies PUFA to PUFA-CoA such as AA-CoA and ADA-CoA under the action of acetyl coenzyme A (CoA). Then, under the catalysis of lysophosphatidylcholine acyltransferase 3 (LPCAT3), PUFA-CoA combines with phospholipids on the cell membrane to form PUFA-PL, which is peroxidized to PUFA-PL-OOH under the regulation of dioxygenase (ALOX) (60–62). On the one hand, the accumulation of phospholipid peroxides makes the cytomembrane thin and increases its curvature, and the accelerated influx of oxidants intensifies the oxidation, which eventually leads to membrane perforation, rupture and the release of cellular contents (63). On the other hand, lipid peroxides are further broken down into active substances that can deplete nucleic acids and proteins, or accelerate the destruction of membrane structural integrity, leading to cell ferroptosis (64, 65).

System Xc^- is a sodium independent transmembrane cystine/glutamate antiporter located on the plasma membrane. It belongs to heteromeric amino acid transporters (HATs), which are composed of the light chain subunit SLC7A11 (xCT) and the heavy chain subunit SLC3A2 (4F2hc) linked by covalent disulfide bonds. SLC7A11 is a member of the SLC7 family, which mainly includes two subfamilies of cationic amino acid transporters (CATs, SLC7A1-4 and SLC7A14) and L-type amino acid transporters (LATs, SLC7A5-13 and SLC7A15) (66). LATs can bind specifically to two members of SLC3 (SLC3A1 and SLC3A2) to form HATs (67).

The activity of system Xc^- is mainly determined by SLC7A11, which is highly specific to cystine and glutamate. It can pump glutamate out of the cell in a 1:1 ratio while transferring extracellular cystine into the cell (68). Excessive release of glutamate will increase the concentration of extracellular glutamate, which in turn regulates the function of the system Xc^-, reducing the intake of cystine and the excretion of glutamate. At the same time, released glutamate is an excitatory neurotransmitter with dual effects of neurotoxicity and excitabilit (69).

The cystine that enters the cell is decomposed into cysteine in a highly reduced environment. Then cysteine is converted to GSH under the catalysis of glutamate cysteine ligase (GCL) and glutathione synthetase (GSS) (70). GPX4 is a kind of intracellular antioxidant enzymes, which can catalyze the conversion of GSH into its oxidized form GSSG, and convert harmful lipid peroxide (L-OOH) into non-toxic lipid alcohol (L-OH), so as to inhibit lipid peroxidation and prevent the occurrence of ferroptosis. With the help of glutathione reductase (GSR), excess GSSG is reduced to GSH by NADPH and enters the next cycle. On the contrary, when system Xc^- or GPX4 is inhibited or inactivated, the redox balance in the cell is dysregulated, which will promote the ferroptosis (71).

Fe²⁺ and Fe³⁺ are two oxidation states of iron. Iron bines with transferrin in blood circulation and exists in the form of Fe³⁺. After entering cells via the transferrin receptor 1 (TfR1) on the cell membrane, Fe³⁺ is deoxidized and converted to Fe²⁺ under the action of iron oxide reductase six transmembrane epithelial antigen of the prostate 3 (STEAP3) (72). After that, Fe²⁺ is transported to the labile iron pool (LIP) in the cytoplasm by divalent metal transporter 1 (DMT1) (73). Intracellular iron is mainly stored in ferritin, and the autophagic degradation of ferritin, which is mediated by nuclear receptor coactivator 4 (NCOA4) (74), can release iron into LIP. Thus, blocking NCOA4 can reduce the level of LIP and inhibit ferroptosis (75). In contrast, enhanced ferritin phagocytosis can increase the LIP and promote ferroptosis. Dynamic iron pools can maintain iron balance under normal physiological conditions while Fe²⁺ accumulates in cells under pathological conditions. Excessive Fe²⁺ can not only react with endogenous H₂O₂ in Fenton reaction to produce a large amount of ROS, but also enhance the activities of multiple metabolic enzymes (such as LOXs, PDH1, ALOX and POR) with iron as a co-factor to promote the generation of ROS (76). ROS can promote the peroxidation of PUFA-PL on the cell membrane to generate lipid peroxides and cause ferroptosis. Therefore, factors related to iron metabolism are potential target sites for inducing ferroptosis.

Several metabolic processes in mitochondria are also significant in triggering ferroptosis. Mitochondria are the main sites for the production of intracellular ROS, which is critical for lipid peroxidation and ferroptosis. Complexes I, II and III in mitochondria are mainly located in the respiratory chain and generate superoxide, which is then converted to H₂O₂ by superoxide dismutase (77). H₂O₂ reacts with labile iron to generate hydroxyl radicals (·OH) and drives PUFA-PL peroxidation through Fenton reaction. In addition, electron transport and proton pumps in mitochondria are vital to produce ATP, which also contributes to ferroptosis (78). When ATP is depleted, PUFA-PL and ferroptosis are inhibited by activation of AMP-activated protein kinase (AMPK) and inactivation of acetyl-CoA-carboxylaze (ACC). In contrast, in the condition of sufficient energy, PUFA-PL synthesis and ferroptosis are promoted (79).

Other factors affecting ferroptosis include coenzyme Q10 (CoQ10), NADPH, p53, nuclear factor E2 related factor 2 (NRF2), etc. Ferroptosis suppressor protein 1 (FSP1) is localized to the plasma membrane and acts as an NADPH dependent oxidoreductase that reduces ubiquinone CoQ10 to ubiquinol (CoQH2) to prevent lipid oxidation and inhibits ferroptosis (80, 81). p53 can inhibit cystine uptake and ferroptosis by down-regulating the expression of SLC7A11 (82). NRF2 plays a significant role in maintaining intracellular redox equilibrium. It up-regulates the level of many genes (NQO1, HO1 and FTH1, etc.) involving in the metabolism of iron and ROS through the p62-Keap1-NRF2 pathway, then inhibits ferroptosis (83).

Liver metastasis is the most common cause of CRC, and the liver is a favorable site for ferroptosis. Liver disease is currently the most studied type of ferroptosis. On the one hand, programmed cell death triggered by lipid accumulation in hepatocytes is considered as a possible cause of liver tissue damage and inflammation (84). On the other hand, the liver is the main organ of iron deposition, and the dysregulation of iron metabolism leads to the production of a large amount of free iron, which can significantly increase the sensitivity of hepatic cells to ferroptosis (85). We know that nonalcoholic steatohepatitis (NASH) is a severe chronic liver disease characterized by lipid droplet accumulation, hepatocyte death, and inflammatory cell infiltration. It is an important risk factor for liver cirrhosis and carcinogenesis. Studies have shown that ferroptosis is the preferential form of cell death in NASH. Liver tissue inflammation is significantly inhibited by specific inhibition of ferroptosis (86), thereby reducing the possibility of CRC.

In fact, dietary fat is decomposed into fatty acids and monoacylglycerols by pancreatic lipase in the intestinal cavity. Fatty acids penetrate the mucus layer on the surface of microvilli and are absorbed by villous intestinal epithelial cells (CD36, FABP, etc.) then transport to the endoplasmic reticulum (87). With the action of multiple enzymes such as fatty acid synthase, long-chain fatty acids and triglycerides are resynthesized, which participates in the formation of cell membranes or reserve energy for the body (88). Based on studies of ferroptosis in liver disease, we can make reasonable assumptions. Since the gastrointestinal tract is the primary site for the metabolism and absorption of nutrients, including fat and iron, it provides favorable conditions for ferroptosis of gastrointestinal cells (89).

In addition, in sporadic CRC, p53 is mutated in the late stage of carcinogenesis, which promotes the further progression of adenoma to adenocarcinoma (90). In colitis associated CRC, p53 is more important and the mutating early can initiate tumorigenesis (91). As the most important tumor suppressor gene, p53 can not only cause apoptosis, cell cycle arrest and senescence, but also play a tumor suppressor function by inducing ferroptosis as described above.

There may be a correlation between ferroptosis and prognosis of CRC. Biomarkers of ferroptosis, including GPX4, NOX1 and ACSL4, are important prognostic markers in CRC. Moreover, IFN- γ may be involved in tumor death, suggesting that patients with CRC have a good prognosis (92). Obesity is closely related to poor prognosis in patients with advanced CRC, because fat-derived exosomes reduce the susceptibility to ferroptosis in CRC and promote chemotherapy resistance to oxaliplatin (93).

Are ROS-promoting compounds effective in inducing ferroptosis? We knew that endogenous ROS were mainly formed in the mitochondrial respiratory chain, and low levels of ROS played an important role in regulating the biological functions of cells. However, excessive ROS could oxidize proteins, lipids and DNA, break redox homeostasis, and lead to cell death. ROS-induced lipid peroxidation could induce programmed cell death in the form of apoptosis, autophagy and ferroptosis.

Firstly, the formation of lipid peroxides can be achieved through different pathways, mainly including non-enzymatic and enzymatic pathways. The non-enzymatic pathway was an iron-dependent process and an important link in the formation of ferroptosis. As the name implied, the formation of lipid peroxides by lipoxygenase catalysis was the enzymatic pathway. Secondly, lipid peroxides could initiate different forms of programmed cell death through different signaling pathways. They could not only initiate apoptosis through pathways such as NF-κB, MAPK and PKC, but also initiate autophagy through AMPK/mTORC and JNK-Bcl-2/Beclin 1 (94). For example, knockdown of protocadherin 7 (PCDH7) affects autophagy and induces ferroptosis, which enhances the sensitivity of colon cancer cells to chemotherapy by inhibiting the MEK1/2/ERK/c-FOS axis (95). In addition, they could achieve ferroptosis by interacting directly with the cell membrane and disrupting membrane integrity. Therefore, ferroptosis was different from other forms of cell death in cell morphology and mechanism.

The changes of ferrous ion concentration, lipid peroxides and ferroptosis-related proteins were the key indicators to identify the occurrence of ferroptosis (64). For example, Liu et al. demonstrated that oxaliplatin could promote ferroptosis by inhibiting the NRF2 signaling pathway. The results showed that after HT29 cells were treated with oxaliplatin, the Fe²⁺ concentration in the cells was significantly increased and the ferroptosis-related protein GPX was significantly decreased, thus contributing to the production of a large amount of ROS and lipid peroxides (95). Dichloroacetic acid (DCA) can cause iron death in colorectal cancer stem cell (CSC), manifested as increased iron concentrations, lipid peroxides, and glutathione levels (96).

Although many chemotherapeutic drugs could induce ROS production, there was no significant change in the ferroptosis-related characteristics during cell death. Therefore, it did not belong to ferroptosis. They may cause the destruction of proteins or DNA by ROS or the production of lipid peroxides by enzymatic pathways, thereby causing cell death through apoptosis and autophagy. For example, Liu et al. found that the small molecule compound VB1 (extracted from Vitex negundo) showed anti-tumor activity, which led to DNA damage and apoptosis by promoting the accumulation of ROS in cells (97).

Ferroptosis inducers (FIN) are compounds that can induce ferroptosis in cells. In tumor therapy, they can be mainly divided into two kinds according to their mechanism of action. 1) FIN associated with System Xc^-. These inducers act primarily by inhibiting SLC7A11-mediated cystine uptake in system Xc^-. It mainly includes erastin and its derivatives (such as imidazole ketone erastin, perazine erastin), sulfasalazine, sorafenib, etc. 2) FIN associated with GPX4. These inducers can be further subdivided into three kinds: a) directly blocking GPX4 enzyme activity, including RSL3, ML162, ML210, DPI7, etc.; b) depleting GPX4 protein, binding squalene synthase (SQS) and depleting antioxidant CoQ10, such as FIN56; c) directly oxidizing Fe²⁺ and indirect inactivating GPX4, such as FINO2 (90, 98).

According to reports, there are mainly 14 compounds (such as erastin, RSL3, talaroconvolutin-A, etc.) that can be used in CRC treatment ( Table 1 ). RSL3 is a classical inhibitor by directly inhibiting the catalytic activity of GPX4. It utilizes the electron-philic chloroacetamide fraction to covalently bind to the selenocysteine residue of GPX4, which promotes the accumulation of intracellular ROS and lipid peroxidation, thereby triggering ferroptosis (99). Cetuximab inhibits the Nrf2/HO-1 axis by activating p38 MAPK and enhances RSL3-induced ferroptosis (111). Erastin was the first FIN to be discovered. By binding with SLC7A11, erastin can inhibit its activity and affect cystine transport, thereby reducing GSH synthesis, which results in the failure of removing lipid peroxides in time, then causing cell membrane damage and triggering ferroptosis (112). Talaroconvolutin-A (TalaA) is a novel ferroptosis inducer, which is more effective than erastin according to report. TalaA can promote ferroptosis through not only promoting ROS production but also down-regulating SLC7A11 and up-regulating arachidonate lipoxygenase 3 (ALOXE3) (103). Lysionotin (Lys) is a flavonoid compound that promotes the accumulation of ROS in CRC cells and increases the degradation rate of Nrf2 protein, resulting in ferroptosis (113). Petunidin 3-O-[rhamnopyranosyl-(trans-p-coumaroyl)]-5-O-(β-D-glucopyranoside) (Pt3R5G) further inhibits the proliferation of human colonic adenocarcinoma cells (RKO) primarily by down-regulating SLC7A11 to inhibit ferroptosis (114). Auriculasin promotes CRC apoptosis, ferroptosis, and oxidative apoptosis by inducing ROS production, thereby inhibiting cell infiltration (114). Double-targeted PI3K and HDAC inhibitor BEBT-908 developed by Fan et al. can induce cancer cell ferroptosis and effectively inhibit tumor cell growth by up-regulating MHC class I molecule of tumor cell and activating endogenous IFN- γ signal through STAT1 signaling pathway (115). Tagitinin C induces ferroptosis through endoplasmic reticulum stress-mediated activation of the PERK-Nrf2-HO-1 signaling pathway (110).

Since the concept of ferroptosis was proposed in 2012, the initial design ideas of some drugs like those listed in Table 1 were not based on the mechanism of ferroptosis. For example, oxaliplatin exerted anti-tumor effect mainly by inhibiting DNA replication and transcription. It produced biologically active hydrated derivatives in body fluids that form intra- and inter-strand crosslinks with DNA (116). However, its ferroptosis-inducing effect has been found in recent studies. At present, there is no clear evidence that oxaliplatin has a more obvious effect on inhibiting DNA synthesis or inducing ferroptosis. However, according to the existing research results, ferroptosis is an effective adjuvant treatment method at least, which can enhance the overall treatment effect (117).

In addition to the drugs mentioned above, some other proteins and enzymes have extensive effects, including miR-19a, OTUD1, miR-15a-3p, HSPA5 and N-acetyltransferase 10 (NAT10), etc., which can also be involved in the regulation of iron death and thus affect the occurrence and development of colorectal cancer. For example, oncogenic miR-19a negatively regulates ferroptosis inducer iron responsive element binding protein 2 (IREB2), inhibiting the growth of CRC cells and reducing the risk of ferroptosis (118). OTUD1 is a deubiquitinating enzyme of IREB2 that stabilizes IREB2-mediated iron transport, leading to increased ROS production and ferroptosis (119). Overexpression of miR-15a-3p can inhibit GPX4, resulting in increased levels of ROS, intracellular Fe²⁺ and malondialdehyde (120). HSPA5 protects cells against ferroptosis, which promotes the occurrence and development of CRC by maintaining the stability of GPX4 and inhibiting ferroptosis (121). N-acetyltransferase 10 (NAT10) can affect the mRNA stability and expression of ferroptosis suppressor protein 1 (FSP1). The overexpressed of NAT10 can enhance the cells proliferation, migration, invasion, tumorigenesis and metastasis, and reduce the patient survival time (122).

Nanotechnology has been extensively studied to develop advanced nanoparticle drug delivery systems for anticancer drug delivery, which has significant advantages in improving drug availability and targeted delivery properties. The disadvantages of low solubility and poor membrane permeability can be well overcome by preparing anticancer drugs into nanoparticles (123). More importantly, nanomedicine has unique advantages in cancer treatment due to the enhanced permeability and retention (EPR) effect and easy surface modification of nanoparticles (124). Recently, there has been increasing interest in ferroptosis driven nanomedicine due to the potent antitumor activity offered by the combination of ferroptosis and nanotechnology (125). Pan et al. reported that zinc oxide nanoparticles could inhibit GSH synthesis by scavenging H₂S from CRC to induce ferroptosis (126). Li et al. designed a nanoplatform (GCMNPs) based on glycyrrhetinic acid (GA)/poly (lactic-co-glycolic acid), which could induce ferroptosis by increasing intracellular H₂O₂, Fe²⁺ and lipid peroxidation levels and suppressing GPX4 expression. Moreover, they confirmed that combination of GCMNPs and ferrotherapy could enhanced Fenton reaction and immune response in CRC mouse models (127). Chen et al. constructed nanoelicitors, which included two immune-elicitable polyphenols (Chlorogenic acid and Mitoxantrone) and Fe³⁺ ions. The results showed that nanoelicitors could activate tumoricidal immunityto enhance ferroptosis in CRC treatment (128). Han et al. reported core-shell structure nanoparticles of ZnP@DHA/Pyro-Fe, which was used to co-deliver dihydroartemisinin (DHA) and pyropheophorbide-iron (Pyro-Fe). This strategy could remarkably enhance ferroptosis and therapeutic effect of DHA in CRC mouse models (128). Carbon dots (CDs) are zero-dimensional nanomaterials. Tian et al. prepared pentacyclic triterpenes (PTs), a natural anticancer product, into PTs-CDs, which can induce and selectively kill cancer cells by targeting apoptosis, autophagy and ferroptosis of tumor mitochondria (129).

As mentioned above ( Table 2 ), there are few reports on the nanoparticles by promoting ferroptosis for CRC treatment. Moreover, due to poor effect of monotherapy in tumor treatment, the combination of ferroptosis with tumor targeted imaging, phototherapy, chemotherapy, autophagy or immune regulation is going to be a new trend. Therefore, the nanomedicine based on ferroptosis for CRC treatment has great space for development.