The cumulative incidence of CAC varies across studies. A large Swedish population-based cohort study reported an overall cumulative incidence of CAC of 1.0%, 1.5%, and 2.7% after 10, 20, and 30 years of disease, respectively[1]. The cumulative risk of UC-associated CRC appeared to be higher in Asia and was reported to be 0.02%, 4.81%, and 13.91% in Asian patients[10] as well as 1.15%, 3.56%, and 14.36% in Chinese[11] patients at 10, 20, and 30 years of disease diagnosis. In addition, a meta-analysis of population-based cohort studies reported the cumulative risks of CAC as 1% after 10 years, 2% after 20 years, and 5% after over 20 years of disease duration[6].

In addition to disease duration, disease classification and extent are also considered to be important parameters influencing an individual’s risk of CAC. Compared with the general population, patients with UC and CD are associated with 1.47- to 2.70-fold and 1.51- to 2.10-fold increased risks of CRC, respectively[1,4-7]. Compared to those with proctitis UC, patients with extensive UC and left-sided UC are at higher risk of CRC[12]. Specially, certain populations with autoimmune diseases are known to have a higher risk of developing anal cancer than average, such as UC patients with an incidence rate of 276167 py and CD patients with an incidence rate of 614830 py[13].

It is noteworthy that the incidence of UC-associated CRC has steadily declined over the last six decades. The cumulative incidence declined from 33.1 per 1000 patients in studies published in the 1950s to 9.1 per 1000 patients in studies published in the last decade, while the incidence rates declined from 4.29 per 1000 py to 1.21 per 1000 py, respectively[14]. The prognosis for CAC has dramatically improved. Compared to the general population, cancer risk in IBD patients decreased from a 5-fold increase in the 1960s to a 2-fold increase in the 2000–2004 follow-up period[1].

The worldwide cancer incidence rates of UC-associated CRC show considerable geographical variation. The overall incidence rate varies from 5/1000 person-years duration (pyd) in the USA, 4/1000 pyd in the UK, and 2/1000 pyd in Scandinavia[9]. Geography also plays an underlying role in CAC prognosis. The time of malignant transformation started after 10-20 years of CAC duration in Asian patients, whereas it significantly increased to more than 30 years in North American patients[12].

Recent data have shown an association between family history and CAC. Patients with a family history of CRC in a first-degree relative have an almost 8-fold increase in the risk of CAC, while those without a family history have a 4-fold increase in risk[15]. Additional risk factors include hyperlipidemia, obesity, and alcohol consumption for early-onset colorectal cancer[16].

Wijnands et al[17] summarized the data that describe the prognostic factors for advanced colorectal neoplasia in IBD, identifying risk factors (including extensive disease, low-grade dysplasia, colonic strictures, post-inflammatory polyps, primary sclerosing cholangitis, and family history of CRC) and protective factors (including colonoscopic surveillance, 5-aminosalicylic acid, thiopurines, and smoking).

DNA is the basis of genetic information, and maintaining its integrity is essential for life and health. In the early 1900s, scientists developed the idea that “at the molecular level, tumors are the result of damage to cellular DNA”. DNA damage, such as genetic mutations, double-strand breaks (DSBs), and oxidative stress, can be caused by endogenous (e.g., reactive oxygen species) and exogenous factors (e.g., UV light). According to the widely accepted theory, DSBs and oxidative stress are closely related to cancers. DSB is one of the most critical and dangerous types of DNA damage that, if not repaired, can lead to cell death. Oxidative stress to DNA and DSBs might drive colitis-associated colorectal carcinogenesis in IBD patients[18]. In contrast to UC, CAC has a unique mutational profile. It is notable that mutations in NFKBIZ are of high frequency in UC but are rarely found in CAC, which suggests a discrete mechanism in colorectal carcinogenesis[19]. A study investigating somatic mutations in CAC found high frequencies of RNF43 mutations in CAC somatic cells. RNA-Seq analysis revealed elevated c-Myc and target gene expression in RNF43-mutated tumors, suggesting that RNF43 is a driver of colorectal tumorigenesis[20]. In addition, the proto-oncogene BMI1 and its target anti-oncogene Reg3b were identified as being closely associated with the development of CAC. A separate series of in vitro and in vivo studies demonstrated that BMI1 expression was elevated in CAC patients, that high BMI1 expression was associated with a lower response rate to antitumor necrosis factor α (TNF-α) therapy[21] and that BMI1 and its homologue MEL18 promoted cancer by inhibiting Reg3b expression[22].

Epigenetics is important for tumor initiation. The hypothesis of “epigenetic triggers in cancer initiation” is that once endogenous or environmental stimuli trigger epigenetic initiation in cancer-initiating cells, this leads to the development and progression of tumors[23]. ELF4, a member of the E-Twenty-Six domain transcription factor family, is involved in the regulation of a variety of DNA damage repair mechanisms. ELF4 suppression caused by methylation of the promoter region is prevalent in UC and CAC, supporting the “epigenetic triggers in cancer initiation” hypothesis[24]. One study by Emmett and colleagues evaluated DNA methylation patterns in CAC and sporadic CRC and found that several genes were highly methylated in CAC, such as MINT1, MYOD, and the promoter regions of EYA4 and ESR[25]. In addition to DNA methylation, defects in DNA glycosylation have also been implicated in the pathogenesis of CAC. A breakdown of the colonic mucus barrier mediated by impaired O-glycosylation expression was identified to result in spontaneous CAC in mice[26]. Based on these findings, genetics and epigenetics are involved in the development of CAC.

The most comprehensively studied proinflammatory and protumor pathways in CAC are the nuclear factor κ-light-chain-enhancer of activated B cells (NF-κB), β-catenin-mediated Wnt signaling, signal transducer and activator of transcription 3 (STAT3)/interleukin-6 (IL-6), and IL-23/T helper 17 cell (Th17) pathways. It has been well established that activation of intestinal NF-κB signaling promotes colitis-associated carcinogenesis[27]. DHX9 was overexpressed in CRC tissue and CAC mouse models. DHX9 was found to promote NF-κB-mediated transcriptional activity and enhance the expression of downstream targets of NF-κB, thereby promoting colorectal carcinogenesis[28]. In the early stages of colitis-associated carcinogenesis, the Wnt pathway is activated in 50% of cases[29]. Recent studies support the involvement of the WNT signaling pathway in the pathogenesis of CAC. The Wnt/β-catenin signaling pathway is activated through the deletion of NHE8, a multifunctional protein expressed in colon stem cells, leading to increased expression of Lgr5 in colon tissue, which is a target gene for Wnt signaling and ultimately exhibits a favorable outcome of colitis-associated tumorigenesis[30]. Excess yes-associated protein 1 triggers the Wnt/β-catenin signaling pathway and promotes colitis-associated tumorigenesis[31]. Furthermore, upregulation of IL-23 and IL-17 has been demonstrated in clinical CRC specimens and CRC mouse models. It has been shown that barrier disruption due to genetic damage leads to the invasion of microbial products, triggering IL-23- and IL-17-driven inflammatory infiltration of tumors, which in turn drives tumor growth[32].

Both the innate and adaptive immune systems play important roles in the development and progression of CAC. Macrophages are the most well-known member of the intrinsic immune system. Owing to the known potential of Pou3f1 in regulating immune responses and the immune system, investigators have investigated the role of Pou3f1 in colitis-associated colorectal tumorigenesis. Reduced secretion of inflammatory mediators in macrophages by knocking out Pou3f1 inhibited CAC development[33]. Ornithine decarboxylase in macrophages promotes colon carcinogenesis by impairing the M1 immune response[34]. Tumor-infiltrating T cells have been demonstrated to contribute to CAC immunosurveillance. Interestingly, the incidence of CAC was found to increase after an appendectomy, which may accelerate tumor development by inhibiting T-cell initiation and subsequently reducing cancer immune surveillance, as well as by dysbiosis of intestinal microbes and impairment of the intestinal barrier[35,36].

There are 3 main microbial hypotheses associated with the pathogenesis of CAC, including the “α-bug” hypothesis[37], the “driver-passenger” hypothesis[38], and the “common ground” hypothesis. According to the “common ground” hypothesis, exogenous and endogenous factors (unhealthy diet, exogenous contaminants, and chronic inflammation) create a “leaky gut” that allows pathogens to become highly permeable across cells and internalize bacteria, leading to chronic inflammation and morphological changes[39]. Accumulated evidence supports a significant role for the microbiota in CRC development and progression, focusing on Escherichia coli (E. coli), Bacteroides fragilis (B. fragilis), and Fusobacterium nucleatum (F. nucleatum).

E. coli can produce colibactin, which is encoded by the pathogenicity island polyketide synthesis (pks) gene. E. coli carrying the pks island causes DSBs and activation of the DNA damage checkpoint pathway, which may be a predisposing factor for CRC development[40]. A novel area of research confirmed a distinct mutational signature with unique single-base substitution, insertion, and deletion mutations in CRC caused by colibactin from pks⁺E. coli[41]. However, there appeared to be no significant differences between pks⁺E. coli and the prevalence of CRC or colorectal adenoma lesion patients in a Japanese prospective cohort study[42] and a Canadian cohort study[43]. Adherent invasive E. coli (AIEC)-associated genes, including chitinase 3-like1, carcinoembryonic antigen-related cell adhesion molecule 6, and claudin-2, as well as the protein expression of these genes, were found to be upregulated in CAC samples, suggesting an underlying link between AIEC and CAC[44].

Unlike pks+ E. coli, no unique mutational signature was produced by enterotoxigenic B. fragilis (ETBF). Allen et al[45] performed whole-exome sequencing and whole-genome sequencing and found that errors in DNA mismatch repair and homologous recombination DNA damage repair were involved in the pathogenesis of ETBF-induced CRC. Notably, a unique ETBF-associated colonic immune infiltrate was observed in a CRC murine model. The combined effect of IL-17 and B. fragilis toxin, specifically secreted by ETBF on colon epithelial cells, ultimately led to the suppression of T-cell proliferation[46].

In addition, F. nucleatum may contribute to the development and progression of CRC in several ways. It promotes epithelial-mesenchymal transition through the epidermal growth factor receptor signaling pathway, thereby accelerating the progression of CAC[47]; it recruits tumor-infiltrating immune cells, leading to a proinflammatory microenvironment conducive to colorectal carcinogenesis[48]; and it activates the NF-κB signaling pathway by activating Toll-like receptor 4 signaling, thereby upregulating microRNA-21 expression and leading to accelerated tumor progression[49]. Further efforts must be made to unravel the complexity between the microbiota and CAC.

According to the widely accepted concepts, aminosalicylic acid (ASA) has been an essential drug in the treatment of mild to moderate UC. Several studies have investigated its potential value in the chemoprevention of CAC. Long-term use of 5-ASA compounds has been shown to reduce the risk of UC-associated CRC in both human[74] and animal models[75]. Three meta-analyses[76-78] and a large epidemiological study[79] confirmed a protective association between the application of 5-aminosalicylate compounds and CRC. However, in another meta-analysis of nonreferral populations, there appeared to be no protective effect of 5-ASA on CAC[80].

Despite some conflicting results, most studies have confirmed the chemopreventive effect of 5-ASA in preventing CAC. The chemopreventive effects of 5-ASA vary for different disease classifications and drug types: The risk of CAC is greatly reduced in UC patients but not significantly in CD patients; mesalazine significantly reduces the risk of CAC, but salazosulfapyridine does not show any protective effect[76,77]. The dose-effect relationship on the protective effect of mesalazine is apparent[77]. A daily dosage of mesalazine ≥ 1.2 g is more protective against colorectal neoplasia than < 1.2 g. A case-control study demonstrated that a cumulative mesalamine dose of over 4500 g led to a CRC risk reduction of 97.6%[81]. Given this evidence, the 2017 ECCO Consensus recommended the use of mesalazine for the chemoprevention of CRC in all patients with UC[55].

The antineoplastic effect of thiopurines is debatable due to inconsistent results across studies. A cohort study in the Netherlands found a significant protective effect of thiopurines on the risk of advanced neoplasia[82]. However, data from a meta-analysis[83] and a case–control study[84] have demonstrated that thiopurine use was not associated with a significantly lower risk of colorectal neoplasia. In addition, the effectiveness of thiopurines varies in different disease classifications and geographical regions. Thiopurine treatment was associated with a reduced risk of colorectal neoplasia in European studies but not in African and Asian studies; thiopurine treatment also reduces the risk of CAC, which is significant in patients with UC but not in patients with CD[85].

Although thiopurines may reduce the risk of CAC, they are also associated with an increased risk of cancer[86]. There was no excess risk of developing any cancers in individuals with older-onset IBD in a population-based study[87]. Nevertheless, a nested case-control study has identified that exposure to thiopurines for over 5 years is related to a significantly higher risk of nonmelanoma skin cancer and lymphoproliferative disorders[86] but unrelated to a higher risk of melanoma or colorectal cancer[88]. Therefore, given this evidence, the American College of Gastroenterology (ACG) guidelines recommended that chemopreventive medical therapy alone is not an appropriate way to prevent UC-associated CRC and is not a substitute for colonoscopy[51].

In addition to thiopurines, given the therapeutic effect of thalidomide in moderate UC and moderate CD, its efficacy in CAC is under investigation. Lu et al[89] demonstrated that thalidomide suppressed macrophage polarization in the tumor microenvironment, which not only relieved colonic inflammation to promote mucosal healing but also inhibited the development of CAC. These findings may shed light on the potential use of thalidomide for the chemoprevention of CAC.

Biological agents commonly used in the guidelines for the treatment of IBD include antitumor necrosis factor-α (anti-TNFα) agent (Infliximab), α4β7 integrin antibody (Vedolizumab), Janus kinase (JAK) inhibitor (Tofacitinib), and IL-12/IL-23 antagonist (Ustekinumab)[70]. The guidelines recommend these drugs as an alternative treatment for moderate to severe UC and moderate to severe CD. However, biological agents appear to have no chemopreventive effect on CAC. Two population-based studies, one from France[87] and the other from Canada[88], found no association between anti-TNFα exposure and CRC. Concerns about the carcinogenic risk of antineoplastic drugs are of significant importance. Fortunately, a meta-analysis showed that antineoplastic drugs do not increase the risk of CAC[90]. The evidence-based medical evidence on chemoprevention in the last few years is presented in Table 3[91-97].

Recent attention has focused on the anticancer activity of probiotics, despite the lack of data from large clinical trials. Although the protective effect of probiotics on CAC has not yet been reported, the use of probiotic supplements has been indicated to significantly reduce the risk of postoperative complications in patients undergoing CRC surgery[98]. Importantly, some studies appear to form the basis of future explorations into probiotics for the treatment of CAC. As mentioned earlier, genomic instability has been linked to carcinogenesis in IBD. Bifidobacterium infantis has been demonstrated to alleviate colonic inflammation by activating DNA repair pathways and enhancing genomic stability[99]. Lactic acid bacteria, which have been reported to be successfully used in managing sporadic CRC[100], have also been shown to have a potential chemopreventive effect on CAC. One study by Silveira et al[101] established the CAC murine model and found that Lactobacillus bulgaricus downregulated certain cytokine levels in the intestine and tumors and inhibited tumor growth.

Traditional Chinese medicines (TCMs) are a unique but helpful health resource in China. Many classic TCMs have a long history of adjunct therapy in patients with mild to moderate IBD and are still in use today[102]. Qingchang Wenzhong decoction[103] and Sini decoction[104], effective TCM prescriptions, have been indicated to inhibit colitis-associated carcinogenesis, possibly through the improvement of intestinal flora dysbiosis and the intestinal barrier.

Improving microbial dysbiosis by modulating the gut microbiota is a novel strategy for the prevention and treatment of CRC. As mentioned above, probiotics have shown great potential in the prevention and treatment of CRC. Other approaches to gut microbiota modulation, such as prebiotics, postbiotics, and fecal microbiota transplantation (FMT), are theoretically promising for the prevention and treatment of CRC. Despite the lack of clinical evidence, two clinical trials of FMT in CRC are underway. It is anticipated that these drugs are effective in the prevention and treatment of CRC.