Molecular Pathways and Biomarkers in Colorectal Cancer: From Carcinogenesis to Targeted Therapies Volume 61- Issue 1

Koumoundourou Dimitra*

• Department of Pathology, University Hospital of Patras, Greece

Received: March 12, 2025; Published: March 19, 2025

*Corresponding author: Koumoundourou Dimitra, Department of Pathology, University Hospital of Patras, Rio, Greece

DOI: 10.26717/BJSTR.2025.61.009537

Abstract PDF

Colorectal carcinogenesis is a multistep process that is driven by genetic mutations, epigenetic abnormalities, chronic inflammation, and interactions within the tumor microenvironment. The main molecular pathways involved in CRC progression are chromosomal instability (CIN), microsatellite instability (MSI), and CIMPCpG Island Methylator Phenotype. According to other investigators these pathways are interwoven in the two major clinicopathologic sequences: the adenoma -carcinoma sequence and the serrated neoplasia pathway. These pathways specify CRC subgroups with varying prognostic and therapeutic consequences. On the other hand, pathologists play an important role since they must identify crucial molecular biomarkers with prognostic and predictive role in patient’s outcome. In this review, we will discuss some of the critical molecular processes, the biomarkers that are evaluated in the tumors specimen and are strongly associated with targeted therapies, as well as future prospects, laying the framework for AI innovation of precision medicine.

Epidemiology: Colorectal cancer (CRC) is the third most frequently diagnosed cancer after lung and breast cancer and the second leading cause of cancer-related deaths worldwide according to the International Agency for Research on Cancer (IARC), with more than 900,000 deaths per year worldwide[1].The rising incidence among younger population underscores the importance of early identification and risk factor modification [2], highlighting the need for developing preventive strategies and treatment options (IARC, 2020). Significant disparities exist, since lower screening rates among certain populations lead to later stage diagnoses and thus worse outcomes [3] Incidence and mortality of CRC varies significantly across geographic regions, reflecting differences in a variety of socio-economic parameters, diet, lifestyle and access to health care [4]. Incidence rates are higher in developed countries, where screening programs have led to better survival [2]. In contrast, low- and middle-income countries have an increasing incidence of CRC due to the adoption of Westernized diet and to the decrease of physical activity [5]. CRC is most frequently encountered in North America, Europe, and Australia [4]. Strategies to reduce CRC-related mortality include early detection, adoption of healthy lifestyles and the application of precision therapies based on the unique genetic profile of a particular tumor [6]. Nevertheless, a large proportion of cases are still detected at an advanced stage so that treatment becomes more complicated, and survival decreases. While a 5-year survival rate for early-stage CRC is almost over 90%, it decreases to less than 15% in metastatic CRC. In these cases, the need for more effective therapies and targeted treatment approaches is obvious and necessary [7].

CRC typically evolves over decades, usually beginning with the formation of benign polyps that can be detected and removed during colonoscopy screenings, offering a chance for prevention [8]. Adenomatous polyps are the predominant precursors, accounting for 85– 90% of sporadic CRCs. However, only a small proportion of them, less than 10%, progress to malignancy. Adenomas that are characterized by size ≥1 cm, villous histology, or high-grade dysplasia, share a significantly higher risk, which increases with the patient’s age [9,10]. On the other hand, serrated polyps, which account for 10–15% of CRC cases, include hyperplastic polyps, sessile serrated adenomas, and traditional serrated adenomas, and were initially considered non-malignant. Nevertheless, large hyperplastic polyps, particularly in the proximal colon, are now recognized as potential precursors of CRC. Recent studies [11,12] found odds ratios for the development of CRC to be 1.79 for traditional serrated adenomas, 3.40 for sessile serrated adenomas, and 2.50 for conventional adenomas, compared to individuals without a polyp history. CRC is correlated with multiple risk factors, such as dietary patterns, history of abdominal radiation, smoking, alcohol use, lifestyle functions, inflammatory bowel disease, genetic predispositions, and dysbiosis of gut microbiota.

Preventive measures can be better linearized into regular screenings (colonoscopy, fecal occult blood tests), dietary therapies, increased physical exercise, and chemoprevention including use of non-steroidal anti-inflammatory drugs (NSAIDs). Lynch syndrome (also referred to as Hereditary Nonpolyposis Colorectal Cancer (HNPCC) is an autosomal dominant genetic disease with a markedly elevated lifetime risk of colorectal cancer (up to 80%), often at an earlier age (median age at diagnosis ~45 years) compared to sporadic cases. Lynch syndrome accounts for about 2–4% of all colorectal cancers and is linked with other malignancies such as endometrial, ovarian, gastric, urinary tract, small bowel, pancreatic and hepatobiliary cancers [1]. Another rare hereditary disease correlating with CRC isfamilial adenomatous polyposis (FAP) syndrome, which is due to a germline mutation in the APC gene.

The recent Consensus Molecular Subtypes (CMS) Consortium analyzing CRC expression profiling data from multiple studies described four CMS groups [13] The first subtype consists ofCMS1 (MSI-immune subtype 14% of all tumors) hypermutated tumorswith high microsatellite instability (MSI), high immune infiltration, and frequent BRAF mutation. The remaining MSS cancers are subcategorized into three groups: CMS2 (canonical, 37 %)that is characterized by chromosomal instability (CIN), activation of Wnt/MYC signaling and strong epithelial differentiation. CMS3 (the metabolic subtype13%) is characterized by KRAS mutations and metabolic dysregulation with changes in glucose metabolism and CMS4 (the mesenchymal subtype, 23 %) exhibits epithelial-to-mesenchymal transition (EMT), stromal invasion, and poor prognosis while a residual unclassified group is also recognized (mixed features, 13 %). This classification of CRC revolutionized treatment options. CMS1 tumors display high levels of microsatellite instability (MSI) and immune infiltration and correspond to immune checkpoint inhibitors therapies, while CMS4 tumors correlate with epithelial-to-mesenchymal transition (EMT) and poor prognosis, thus, are not responding to conventional therapies. To understand this classification elucidation of CRC molecular pathogenesis is necessary.

Three primary molecular mechanisms contribute to CRC carcinogenesis: chromosomal instability (CIN), microsatellite instability (MSI) and the CpG island methylator phenotype (CIMP). The chromosomal instability (CIN)is the most common molecular pathway responsible for about 65%–70% of sporadic CRC cases. It is characterized by chromosomal changes that include somatic copy number alterations (SCNAs) caused by aneuploidy, deletions, insertions, amplifications, or loss of heterozygosity (LOH). These karyotypic abnormalities are associated with mutations of tumor suppressor genes such as APC and TP53 and activating mutations of KRAS and PIK3CA (phosphatidylinositol4,5-bisphosphonate 3-kinase catalytic subunit alpha). Inactivation of APC is likely the earliest genetic alteration detected in colorectal tumorigenesis. APC (adenomatous polyposis coli) gene mutation is thought to be one of the earliest and most pivotal genetic changes in colorectal tumorigenesis and often acts as the initiating mutation in the adenoma-carcinoma spectrum. APC is a tumor suppressor that regulates cellular proliferation and differentiation mainly by controlling the levels of the key signaling protein β-catenin. APC inactivation, due to point mutations or deletions, leads to nuclear translocation of β-catenin which binds to T-cell factor/lymphoid enhancer factor (TCF/LEF) transcription factors and activates Wnt signaling pathway. This activation results in the transcription of numerous oncogenic target genes, including MYC and CCND1, that enhance cellular proliferation, thus mediating tumorigenesis and tumor progression. This dysregulation is important in the early stages of colorectal neoplasia contributing to the formation of an adenoma and predisposing for malignant transformation [14].

Furthermore, CIN tumorigenesis is often correlated with KRAS, NRAS, and BRAF mutations.KRAS (Kirsten rat sarcoma viral oncogene homolog) genes are mutated in about 40% of CRCs and play an oncogenic role by activating the MAPK and PI3K-AKT pathways [15]. KRAS activation leads to constitutive activation of the Raf-MEK–extracellular signal–regulated kinase (ERK) pathway and phosphoinositide 3 kinase (PI3K) signaling via MTOR (mammalian target of rapamycin) and the transcription Nuclear Factor κB (NF-kB). Analyses of gene expression patterns in colorectal tumors have identified a relationship between KRAS mutations and CMS-3 (also called the metabolic phenotype).

NRAS (neuroblastoma RAS viral oncogene homolog) and RAF family proteins are serine/threonine kinases that act on MEK1 (Mitogen- Activated Protein Kinase MAP2K1) and MEK2 (MAP2K2) (dual- specificity kinases)with a crucial role in cell proliferation, differentiation, and survival. MEK1 and MEK2function as upstream regulators of ERK1/2, which transduce signals from growth factors to promote tumor progression. The phosphorylation of ERK1 and ERK2 leads to the consequent phosphorylation of certain promoters of cell-cycle progression. [16]. Many colorectal tumors with mutations in KRAS also carry mutations in gene PIK3CA (the gene encoding the catalytic subunit-a of PI3K).PIK3CA is a coding gene for p110α catalytic subunit of phosphatidylinositol-3-kinase (PI3K), an important activator for PI3K-AKT-mTOR pathway that regulates cell proliferation, survival, and metabolism.PIK3CA mutations (found in about 10-20% of colorectal cancers) are associated with worse prognosis and may be targeted with therapeutic agents especially in metastatic colorectal tumors. PTEN is a tumor suppressor, acting as a negative regulator of the PI3K-AKT-mTOR signaling pathway by dephosphorylating PIP3 to PIP2, thus antagonizing PI3K activation. PTEN loss-of-function mutations lead to aberrant signaling through the PI3K pathway. This pathway promotes both tumor proliferation and therapy resistance. PTEN alterations (comprised of point mutations, deletions and promoter hypermethylation) occur in ∼20–30% of CRCs.

Approximately 60% of CIN tumors harbor inactivating mutations in TP53 [17]. TP53, a well-known tumor suppressor gene and the most frequently mutated gene in cancer, is located on human chromosome arm 17p. P53 protein mediates expression of genes regulating DNA repair and cellular responses to oxidative stress. P53 lossof- function mutations are more frequent in colorectal tumors than in adenomas with microscopic foci of invasive cancer, and they are more frequently found in invasive carcinomas than in benign adenomas [18]. The correlation between TP53 mutation and tumor stage suggests that mutant P53 acts at a late stage in tumorigenesis. Germline mutations in TP53, the underlying cause of Li–Fraumeni syndrome, are related to a modestly increased risk of CRC compared with the general population. An analysis of data from patient registries identified a correlation between germline alterations in TP53 and early-onset CRC (younger than 50 years) [19].

This observation is clinically important because young-onset CRC cases have markedly increased in the United States and some parts of Europe and Asia, while CRC incidence has declined in other age groups. A proportion of those cases may be due to hereditary genetic factors [20]. Another important change in CIN pathway is loss of heterozygosity (LOH) at chromosome 18qwhich occurs in more than 70% of advanced CRC cases and has been implicated in mutations of tumor suppressor genes of the TGF-beta pathway (e.g., SMAD2, SMAD4), as well as DCC (Deleted in Colorectal Cancer), contributing to tumor progression [21] LOH correlates with poor prognoses and greater metastatic potential [22]. In CIN tumors, aberrant activation can also result from growth factor receptor signaling, such as EGFR (epidermal growth factor receptor). The ErbB family of tyrosine kinases receptor (RTKs) plays a prominent role in the development of colorectal cancer mainly via the activation of distinct signaling cascades including EGFR (ErbB1), HER2 (ErbB2), HER3 (ErbB3), and HER4 (ErbB4). EGFR is overexpressed in 60–80% of colorectal cancers (CRCs) and activates MAPK and PI3K/Akt pathways promoting tumor growth and cell survival. 3–5% of colorectal tumors harbor human epidermal growth factor receptor-2 (HER2) amplification which is associated with a higher incidence of metastasis and overall worse outcomes [23].

The underlying mechanisms involved in chromosomal instability (CIN) as already mentioned include errors in chromosomal segregation, defective DNA damage response (DDR), and telomere dysfunction. Aneuploidy and chromosomal missegregation result from defective mitotic spindle assembly and centrosome duplication errors [15]. Telomere dysfunction causes chromosomal rearrangements and instability, which, combined with other alterations, promote tumor progression. DDR deficiency, such as ATM, ATR, and BRCA1 mutations, drives the accumulation of genomic alterations in a variety of CIN tumors (Santini et al.). Clinically, CIN tumors are usually microsatellite stable (MSS), though they display high levels of genomic instability [23]. These tumors are often clinically aggressive, with increased metastatic potential. They also have a poor immunogenic microenvironment, with a subsequent poor response to immunotherapy.

Microsatellite Instability (MSI) Pathway is the second prevalent molecular pathway of CRC that accounts for about 15% of CRC cases and is associated with defects in the DNA mismatch repair (MMR) system. MMR proteins recognize and fix errors such as mismatched bases or small insertion–deletion loops, during DNA replication and recombination. The heterodimer MSH2–MSH6 (known as MutSα) bind to single-base pair mismatches while the MSH2–MSH3 (or MutSβ) scan the newly formed DNA detecting base mismatches. When these heterodimers encounter a mismatch, they bind to the error site recruiting a mediator protein (MLH1–PMS2 heterodimer, in humans predominantly known as MutLα). and recruit supporting proteins like Replication Factor C (RFC) and Proliferating Cell Nuclear Antigen (PCNA) that stabilizeMMR proteins at the mismatch site [24]. The heterodimersalsoactivateExonuclease1 (Exo1) an enzyme that digests themismatched bases, leaving a gap in the DNA strand. During this excision process, the exposed DNA is stabilized by Replication Protein A (RPA). This interaction is ATP-dependent and helps coordinate the subsequent steps of repair. In the last stage, strand re-synthesis, the template strand is subsequently used to fill the gap by DNA Polymerase δ (Pol δ), synthesizing DNA that perfectly complements the template strand. The repaired segment is then sealed by the DNA Ligase I, restoring the integrity of the DNA molecule. All these mechanisms repair dsDNA breaks in order to avoid mutations and protect the genome from instability and oncogenic transformation [25].

Overall, MMR mechanism increases the fidelity of DNA replication up to 1,000-fold thus ensuring genomic stability. Defects of DNA mismatch repair (MMR) genes (including MLH1, MSH2, MSH6, and PMS2) or epigenetic abnormalities, such as hypermethylation-induced silencing of MLH1 lead to the MSI phenotype. Mutations in EPCAM (Epithelial Cell Adhesion Molecule), a gene that regulates MSH2 expression can also cause MSI (EPCAM is also known as TACSTD1 and encodes a transmembrane glycoprotein expressed in the intestinal mucosa). Microsatellites are short repeated tandem nucleotide sequences dispersed throughout the genome that are prone to errors during the DNA replication process (exacerbated by DNA polymerase lacking sufficient binding affinity). Defective MMR proteins are not capable to repair such errors, leading to a spectrum of mutations accumulating in microsatellite regions. MSI is frequently linked to CIMP (see section below), as CpG island hypermethylation can inactivate mismatch repair genes. Defects in this pathway can lead to microsatellite instability (MSI). MSI is correlated with various cancers, including CRC and other Lynch syndrome-correlated cancers. Lynch syndrome-associated CRC is caused by germline mutations in MMR genes [25]. Sporadic MSI-high tumors have CpG island methylation, leading to epigenetic silencing of MLH1. A distinctive feature of tumors with sporadic MSI is the increased incidence of BRAF V600E mutations, which are uncommon in CRC associated with Lynch syndrome. In colonic tumors with microsatellite instability (MSI), mutations frequently occur in genes, such as the transforming growth factor-beta receptor type 2(TGFBR2) [26]. The TGFBR2 gene encodes a receptor protein with tumor suppressor role. Mutations occur in more than 90% of MSI colorectal cancers and lead to inactivation of the receptor’s signaling functionality with subsequent loss of its suppression. TGF-β pathway displays context-dependent functions in CRC. Although this pathway acts as a tumor suppressor in early neoplastic events, it causatively switches to promote epithelial-to-mesenchymal transition (EMT) and metastasis in later disease stages [27]. A lot of other genes that encode proteins controlling cell proliferation are frequently mutated in MSI colorectal tumors (GRB1, TCF4, WISP3, ACVR2, IGF2R, AXIN2, CDX), or cell cycle arrest and apoptosis (CASP5, PRDM2, BCL10, PTEN, PA2G4, FAS), or proteins that play a role in DNA repair (MBD4, BLM, CHK1, MLH3, RAD50, MSH3, MSH6). Subsequently, a widespread disruption of molecular pathways is present in MSI colorectal tumors significantly promoting tumor progression and altering clinical outcomes.

MSI tumors are highly immunogenic as they carry a high mutational burden and retain production of MHC class I molecules (unlike other tumors that escape immune surveillance by downregulating major histocompatibility complex (MHC) class I). This fact ensures that the mutated peptides are efficiently presented on the tumor cell surface, allowing cytotoxic CD8+ T cells to recognize and attack the cancer cells. The proteomic overload caused by the abundance of mutation-derived neoantigens further promotes T cell infiltration, enhancing the anti-tumor immune response [28]. This brisk immune activation is the underlying reason for the excellence response of MSI-high tumors (MSI-H) to immune checkpoint inhibitors, such asPD-1/PD-L1 blockade (pembrolizumab, nivolumab [29]. PD-1 (Programmed Cell Death Protein 1) is an immune checkpoint receptor expressed on T, B, and natural killer (NK) cells that prevent excessive immune responses (maintaining self-tolerance). Many tumors overexpress PD-L1 (ligand) which binds to PD-1 on T cells minimizing the immune attack so that the cancer cells can evade destruction. As a result, tumors are no longer subject to immune surveillance. PD-1 inhibitors (such as nivolumab or pembrolizumab). The third important molecular pathway of CRC is CpG Island Methylator Phenotype (CIMP).Colorectal tumors falling under the CpG island methylator phenotype (CIMP) occur via an epigenetic instability pathway and are defined by extensive hypermethylation of promoter CpG island sites, leading to the silencing of numerous tumor suppressor genes or other tumor-associated genes [30].

CpG(Cytosine followed by a Guanine) islands are DNA regions commonly found in promoter regions rich in CpG dinucleotide, responsible for gene expression regulation in most human genes. DNA methylation is an enzymatic reaction of adding a methyl group to the 5-position of cytosine by DNA methyltransferases (DNMT), resulting in 5-methylcytosine. The term CpG has thus been used for a CG dinucleotide sequence, the favored DNMT substrate. Cancer cells have several unmethylated CpG dinucleotides (which are methylated in normal cells) or show aberrant hypermethylation of CpG islands [31]. The consequent result is silencing of gene expression, predisposing to carcinogenesis. However reduced gene expression is seen in a minority of genes methylated in colorectal cancer. Gene body methylation provides a mechanism for transcriptional activation occurring in methylation in other than promoter site, CpG sites. CIMP phenotype has been reported in a variety of tumor types, including gastric, lung, liver, ovarian, glioblastomas, endometrial and breast tumors, leukemias and CRC [23].

CIMP tumors are commonly designated as CIMP-high (≥3) and CIMP-low (<2) based on the number of positive methylation markers found at locations of 5 genes (MINT1, MINT2, MINT31, CDKN2A, and hMLH1). CIMP-high status is manifested in about 20% of colorectal tumors and is frequently expressed together with a BRAF mutation and hypermethylation of MLH1, hallmarks that are present in a large subset of MSI-H tumors (While CIMP is present in most sporadic MSI-H tumors, half of all CIMP tumors is not correlated with either MSI or MLH1 hypermethylation [15]. The exact underlying molecular mechanism responsible for the hypermethylation phenotype is not fully defined. An important mechanism relies on the transcriptional repressor MAFG (Maf Avian Musculoaponeurotic Fibrosarcoma oncogene homolog G) which recruits other proteins that bind and hypermethylate certain gene promoters, such as MLH1. During this process, it is important to note that mutant BRAF increases the expression levels of MAFG, enhancing its ability to bind to promoters CIMP status has been associated with clinical and pathological characteristics that affect therapeutic response, including chemosensitivity. Interestingly, the presence of CIMP has often been associated with better clinical outcomes [23].

CIMP can be detected during the early phase of CRC tumorigenesis. For example, in recent studies, microvesicular hyperplastic polyps were found to be more frequently CIMP-high than sessile serrated adenomas, traditional serrated adenomas (TSAs), and other more advanced lesions. Furthermore, a link exists between CIMP in normal tissue and serrated polyp development, showing a field effect. Meta- analyses revealed an association between BRAF-mutated tumors with mucinous or poorly differentiated histology and with CIMP [32], MSI [33] predominantly in the right colon, and older female patients [34]. However, the prognostic role of CIMP remains unclear and difficult to reproduce. In this point we have to emphasize that in CRC pathogenesis, the above mentioned molecular pathways-chromosomal instability (CIN), CpG island methylator phenotype (CIMP) and microsatellite instability (MSI)—co-occur within individual tumors. CIMP and MSI frequently overlap because hypermethylation of CpG islands silences mismatch repair genes, directly resulting in MSI [23]. These phenotypes are variously found in sporadic CRC, with approximately 85% of colorectal tumors harboringCIN, 20% of tumors exhibiting CIMP positivity, while 15% of tumors are MSIhigh. From another point of view, colorectal carcinogenesis based on clinicopathologic criteria and not strict, distinctor isolated molecular events, follows two major clinical pathways, the adenoma-carcinoma sequence and the serrated neoplasia pathway. The traditional adenoma- carcinoma sequence was first described by Fearon and Vogelstein (1990) and is the classical pathway of colorectal cancer development.

A defined series of genetic alterations and initiating mutations targeting the APC tumor suppressor gene, activating mutations in KRAS, chromosomal instability, and finally, loss of function mutations of TP53 are associated with this pathway. These tumors usually arise from traditional adenomas with tubular or villous architecture with presence of epithelial dysplasia. This pathway is responsible for about 60–70% of colorectal cancer cases and tends to evolve over years, providing early detection chances via screening colonoscopies. In contrast, the serrated neoplasia pathway, more recently recognized, is characterized by different morphological and genetic features [35]. This pathway encompasses serrated lesions, which include hyperplastic polyps, sessile serrated adenomas/polyps (SSA/Ps), and traditional serrated adenomas (TSAs) morphologically distinct. Common genetic changes in serrated lesions involve mutations in BRAF (especially V600E) and a frequent CpG island methylator phenotype (CIMPhigh) with broad epigenetic silencing of tumor suppressor genes like MLH1, eventually giving rise to microsatellite instability-high (MSI-H) colorectal carcinomas [36,37].

Most micro vesicular hyperplastic polyps have BRAF mutations, and this mutation is thought to be an early event in the serrated pathway, leading to constitutive activation of the MAPK–ERK pathway and unregulated cell division [38]. Following mutation of BRAF, serrated tumors arise by two different pathways and are classified into two major subtypes: MSI-high serrated tumors that acquire MSI belong to a subset with a relatively rapid transition from premalignant lesion to carcinoma. Like all MSI-high tumors, they share a brisk immune infiltration and a favorable prognosis. The second subtype consists of MSS tumors, which are clinically aggressive and have poor patient survival [13]. In detail, the first pathway overlaps with the MSI pathway, where mutations in an MMR gene lead to the MSI-high phenotype. These tumors commonly arise from sessile serrated adenomas and share clinical characteristics with CMS1 tumors. Alternatively, BRAF-mutated tumors can develop TP53 mutations along with the activation of oncogenic pathways, such as Wnt signaling, TGFB signaling, and features of epithelial-to-mesenchymal transition (EMT) [39]. The Wnt pathway is indirectly activated through missense APC mutations or RNF43 mutations [40,41]. RNF43 is an E3 ubiquitin ligase which inhibits Wnt signaling (mediated by R-spondin). Somatic mutation in RNF43 occurs in up to 85% of both MLH1-methylated and MSI-high tumors [42]. These tumor types generally exhibit features of CMS4 (mesenchymal type) tumors. Unlike CMS1, CMS4 tumors are MSS with CIN, low levels of hypermutation, and high SCNA (Somatic Copy Number Alterations). The CMS4 subtype is driven by pathways that promote an immunosuppressive microenvironment, enabling tumor invasion by activating the angiogenic pathway. These features might explain whyCMS4 tumors can escape immune response, leading to the worst survival rates of all CMS subtypes [43,44].

Though clinical discrete, both the adenoma-carcinoma sequence and the serrated neoplasia pathway hold certain key features of colorectal carcinogenesis in common. A gradual evolution from precursor lesions to invasive carcinoma, facilitated by the accumulation of genetic and epigenetic events over long periods of time, drives both processes. These two pathways give major clinical opportunities for early detection and intervention with colonoscopy-based screening programs. Moreover, molecular characterization, including mutational status of KRAS, BRAF as well as microsatellite instability status, is integral to both pathways and clinically relevant. Although the prevalent molecular pathways of CRC carcinogenesis have already been described, many other molecules and factors also play a vital role in CRC development and progression. Due to the complexity and heterogeneity of CRC, it is worth mentioning briefly these molecules and pathways in order to highlight the complexity of the network behind colorectal cancer biology.

PTGS2 (prostaglandin-endoperoxide synthase 2, or cyclooxygenase 2 [COX-2]) is a prostacyclin located on the luminal surface of the nuclear membrane and the endoplasmic reticulum [45]. COX-2 can be induced by cytokines and other stimuli and is overexpressed in adenomas and malignant colorectal tumors [46], though no somatic mutations of COX-2 have been identified in colorectal carcinogenesis. COX-2 may produce prostanoids, including prostaglandins like PGE2, which are involved in controlling proliferation of colorectal cancer cells (by transforming free arachidonic acid) [47]. Additionally, COX-2 regulates angiogenesis and tumor vascularization [48]. Mice lacking Ptgs2 showed reduced small- and large-bowel neoplasia [49] while other studies demonstrated association between use of COX inhibitors such as nonsteroidal anti-inflammatory drugs (including aspirin and selective COX-2 inhibitors) and reduced risk of lesions in the adenoma– carcinoma sequence, including metastatic CRC [50]. The JAK/ STAT signaling pathway, which is usually upregulated in response to cytokines and growth factors, also plays a critical role in CRC and is associated with inflammation, influencing the tumor microenvironment (TME), new blood vessels formation and mechanisms of immune escape. Similarly, Notch signaling, when deregulated, contributes to cell fate positioning, affecting cell proliferation, stem cell maintenance, and apoptosis [16].

It has been increasingly acknowledged that CRC progression is influenced not only by tumor cells but also by the local tumor microenvironment that promotes tumor growth, immune evasion and, ultimately, metastatic potential. The tumor microenvironment, composed of cancer-associated fibroblasts (CAFs), immune cells, blood vessels, and extracellular matrix components, interacts with tumor cells to form a permissive niche for cancer progression [51]. Pro-tumorigenic cytokines (e.g., IL-6, TNF-α, and IL-1β) act on signaling pathways that further promote tumor growth and enhance treatment resistance [52]. The inflammation that is present in chronic inflammatory diseases, such as inflammatory bowel disease (IBD), has the potential to enhance tumorigenesis by promoting DNA damage, cell proliferation, and modifying immune responses. Epithelial-mesenchymal transition (EMT) is one of the central processes of metastasis that enables tumor cells to escape from their primary site and invade neighboring tissues [53]. During this process, expression of epithelial markers (E-cadherin) is downregulated and a switch to expression of mesenchymal markers (vimentin and N-cadherin) is taking place. Thus, the tumor cells are allowed to migrate and spread via blood and lymphatic systems. EMT is a hallmark of CMS-4 CRC tumors with extensive stromal invasion, and poor prognosis [54].

Circulating tumor cells (CTCs) are crucial for enabling metastasis in CRC, especially since the liver and lungs serve as common sites of metastasis owing to their high vascularization [54]. Another key feature of tumor progression is metabolic reprogramming. As seen in the Warburg effect, tumor cells develop adaptations of their metabolism to support their rapid proliferation through increased glucose consumption and aerobic glycolysis, enhanced lipid metabolism, and modifications in mitochondrial function. These metabolic adaptations enable cancer cells’ survival in suboptimal oxygen tension, prevent apoptosis, and evade immune recognition [55]. Hypoxia-inducible factors (HIFs) are transcription factors that regulate metabolic alterations in CRC cells and promote angiogenesis as well as cell survival in nutrient-deficient milieu. Therefore, given the intrinsic and acquired metabolic reprogramming present in CRC cells that underlies their resistance to chemotherapy and targeted therapies, the development of new strategies targeting metabolic pathways in latestage disease may be necessary. Besides metabolic adaptations, the interaction between CRC cells and the gut microbiota has gained a lot of attention in recent years. Some of the beneficial effects of gut microbiota in modulating inflammation, immune responses, and carcinogenic processes in the colon are well established by various studies. Specific microbial species (e.g., Fusobacterium nucleatum) were shown to induce CRC progression via inflammatory response activation, cell adhesion strengthening, and tumor microenvironment remodeling [56]. Dysbiosis, or perturbation of the gut microbiota, can drive CRC pathogenesis through the production of pathogenic metabolites that promote genomic instability and modulate epithelial integrity. A better understanding of the interaction between CRC and the gut microbiota may lead to novel microbiota-targeted therapeutics and preventive strategies [23]. All these pathways and mechanisms do not operate in isolation but instead form part of a complex, interwoven network of signaling events. Pathways’ crosstalk not only complicates cellular identity and intracellular signaling responses but also allows the design of successful therapeutics. Understanding this complex interrelationship is critical for the development of multi-targeted therapies for CRC treatment, which could enhance the existing therapies and unravel novel therapeutic agents [57].

Pathologic evaluation of a colorectal tumor is essential for diagnosis, staging, prognosis, and treatment planning. A complete pathology report should include macroscopic, histopathological, molecular, and immune histochemical findings. The gross examination defines tumor location, specifying whether it is in the right colon (cecum, ascending, transverse), left colon (descending, sigmoid), or rectum. Tumor size should be recorded in millimeters, along with the tumor configuration, which may be polypoid, ulcerative, infiltrative, annular constricting, or flat. The presence of serosal invasion must be assessed, and the status of the proximal, distal, and circumferential resection margins must be noted, particularly for rectal tumors, where a positive circumferential resection margin is associated with higher recurrence risk. Microscopic examination determines the histological type (based on World Health Organization classification), tumor grade and depth of invasion. The presence of lymphovascular (LVI) and perineural invasion and the degree of tumor budding Tumor budding should also be assessed, as high tumor budding is associated with a greater likelihood of metastasis. should also be noted. Tumor stroma ratio is an emerging prognostic factor, as tumors with high stromal content tend to have worse outcomes. Lymph node evaluation is critical for staging and prognosis as well as the presence of peritoneal dissemination. Based on these features final TNM staging should be reported based on the American Joint Committee on Cancer (AJCC) and Union for International Cancer Control (UICC) classification. Molecular and immunohistochemical analysis provides additional critical information. Pathologists also apply immunohistochemical (IHC) staining for mismatch repair (MMR) proteins (eg, MLH1, MSH2, MSH6, PMS2) to diagnose microsatellite instability-high (MSI-H) tumors. Molecular analyses (testing for mutations in KRAS, NRAS, BRAF, and PIK3CA genes, as well as loss of PTEN expression) are becoming integral, as these molecular characteristics play key roles in therapy decisions, particularly in relation to anti-EGFR therapies. Testing MLH1 promoter hypermethylation or BRAF V600E mutation aids in differentiating sporadic MSI-H colon cancer from hereditary forms being most often linked to the presence of Lynch Syndrome, and can thus assist further decisions regarding genetic counseling and testing.

MMR proteins: According to the College of American Pathologists screening impaired mismatch repair (MMR) in colorectal carcinomas (CRCs) is essential for diagnosing Lynch syndrome (HNPCC, ~2–3% of cases) because its presence influence treatment and imposes the screening of the family. Patients with high-MSI tumors may have a germline mutation in one of four DNA mismatch repair (MMR) genes (MLH1, MSH2, MSH6, or PMS2) or a mutated EPCAM (Epithelial Cell Adhesion Molecule) gene.MSI testing should be performed using a minimum of five microsatellite markers and recently used panels focus on mononucleotide repeats rather than dinucleotide markers to improve sensitivity and specificity. MSI testing is usually complemented by an immunohistochemical (IHC) assessment of mismatch repair (MMR) proteins (MLH1, MSH2, MSH6, PMS2). Loss of expression of MSH2 and MSH6, with intact MLH1 and PMS2, is suggestive of an MSH2 germline mutation, while discordant results (MSI-H with normal IHC or the reverse) should prompt a review of the sample. Retained protein expression (presence of immunohistochemical expression of all proteins) does not rule out Lynch syndrome, since missense mutations (particularly in MLH1) can give rise to a nonfunctional but intact protein that preserves its antigenicity [58]. Loss of MLH1 can occur in Lynch syndrome or in cases with promoter hypermethylation, the latter being also associated with a BRAF V600E mutation. MSI-high is a good prognostic marker and has a predictive role, as T3 MSI-high tumors have a poor response to 5-FU chemotherapy and may also predict a better response to oxaliplatin [59].

KRAS, NRAS and BRAF mutations: Detection of KRAS, NRAS and BRAF mutations in CRC tumor specimen have important implications for treatment selection. In CRC, almost 40-50% of patients have KRAS or NRAS mutations which have a great prognostic value since they are strongly correlated to anti-EGFR resistance (to monoclonal antibodies like cetuximab and panitumumab which are used as therapies in other cases of colon cancer). On the other hand, BRAF mutations, especially the V600E variant, are present in approximately 10% of CRC cases, are associated with a more aggressive tumor phenotypes and have been associated with a lack of clinical response to epidermal growth factor receptor (EGFR)-and VEGF-targeted therapies [60]. These mutations play an important predictive role as they indicate a need for alternative treatment options, such as BRAF inhibitors, MEK inhibitors, and chemotherapy, often in combination and have prognostic significance as they are strongly associated with an unfavorable patients’ survival.[61]. Tumors with microsatellite instability- high (MSI-H) frequently harbor somatic mutations at the BRAF V600E hotspot. This mutation is characteristic of sporadic MSI-H colorectal cancers but is absent in hereditary Lynch syndrome-associated cancers (with germline MLH1 or MSH2mutations). The presence of BRAF V600E is a poor prognostic factor, as mentioned above, but in combination with MSI-high share a better prognosis than MSS tumors without BRAF mutations. Studies have shown that patients with MSS and mutant BRAF tumors have shorter overall survival (hazard ratio, 2.16) and disease-specific survival (hazard ratio, 2.59) compared with classical CIN tumors (MSS, no mutations in BRAF) [62]. On the contrary, overall survival and disease-specific survival are longer in patients with MSI and mutant BRAF tumors [62]. PIK3CA Mutational Analysis: Mutations in PIK3CA activate the PI3K-PTEN-AKT pathway downstream of EGFR and the RAS-RAF-MAPK pathway. Additionally, PIK3CA mutations have been linked with poor survival among patients with stage I to III colon cancer, particularly in KRAS wild-type tumors [63].

Novel guidelines on PIK3CA mutational analysis establish that sequencing the tumor tissue is required to select targeted therapy. Mutations of PIK3CA are seen in 10–20% of CRC cases, mainly in exon 9 (helical domain) and exon 20 (kinase domain). A recent guideline recommends next-generation sequencing (NGS) to determine PIK- 3CA mutational status, given its ability to capture hotspot mutations. Alternative methods include single-nucleotide extension (Sanger sequencing, or allele-specific PCR), but NGS provides higher sensitivity and allows the concurrent evaluation of KRAS, NRAS and BRAF co-occurring mutations a fact that facilitates the choice of therapy. Functional classification of PIK3CA mutations is important because kinase domain (exon 20) mutations are generally more oncogenic than helical domain (exon 9) mutations. Studies have shown that PIK3CA mutation plays a role in mediating resistance to anti-EGFR therapies (cetuximab and panitumumab), especially when co-mutated with KRAS or NRAS. Routine testing for the PIK3CA mutation has been recommended for metastatic colorectal cancer (mCRC) patients by the 2023 guidelines of ESMO and NCCN, specifically in patients with RAS and BRAF wild-type tumors in order to intensify treatment strategies.

PTEN mutational analysis: PTEN has also been recognized as an important biomarker in colorectal cancer treatment especially in the context of targeted therapies. Recently released new guidelines on screening PTEN mutational analysis with protein expression by immuno histo chemistry (IHC) have been suggested. Loss of PTEN, identified by IHC, has been associated with poor response to anti-EGFR treatment and enhanced PI3K pathway drive. NGS and IHC combination becomes better as in some tumors PTEN loss is due to epigenetic silencing or post-transcriptional change. Emerging evidence indicates that PTEN-deficient CRCs are sensitive to PI3K or AKT inhibitors, especially in combination with MEK inhibitors. Assessment of PTEN status should be considered along with PIK3CA mutations in patients with PIK3CA mutations who are potentially eligible for EGFR-targeted therapy, consistent with the most recent recommendations from ASCO and ESMO for the evaluation of mutations in metastatic CRC. Molecular profiling for PIK3CA and PTEN mutations is considered now a standard part of precision oncology in CRC. New guidelines recommend the use of next-generation sequencing (NGS) for PIK- 3CA analysis and a joint NGS and IHC approach for PTEN evaluation. These biomarkers play a profound role in the development of personalized treatment strategies in colorectal cancer patients as well as in predicting response to therapy and in the search for novel targeted therapy. Increasing evidence of their efficacy has prompted the clinical evaluation of PI3K pathway inhibitors in treatment regimens, especially in PIK3CA-mutant or PTEN-deficient malignancies, although rigorous trials are required to determine best practices regarding the approved agents and their use in combination [64].

Other treatment options: PD-1 inhibitors have proven to be highly effective for MSI-H CRC by promoting immune system recognition of tumor cells. On the contrary, microsatellite stable (MSS) tumors (accounting for around 85% of CRC cases) display minimal response to immunotherapy and often necessitate alternative therapeutic strategies [65]. Consequently, MSI-high tumors show a robust response to immune checkpoint inhibitors like PD-1 inhibitors (e.g., pembrolizumab and nivolumab), radically changing treatment strategies for CRC patients [66]. MSI-high tumors have a better prognosis than MSS tumors, notwithstanding their limited response to fluoropyrimidine- based chemotherapy, [63]. This has resulted in the recommendation of routine MSI testing for CRC staging and treatment decision-making. Unlike MSI-high tumors, CIN tumors are typically microsatellite stable (MSS), though they display high levels of genomic instability and as they have a poorly immunogenic microenvironment they do not respond to immunotherapy [67]. These tumors are often aggressive, with significant metastatic potential. It is also known that EGFR-targeted therapies have no effect on KRAS- and BRAF-mutant CIN tumors [62]. BRAF-mutant CRC is less frequent; however, these tumors are characterized by aggressive clinical disease and poor responses to conventional chemotherapy. Although the BEACON CRC trial showed that encorafenib (a BRAF inhibitor) combined with binimetinib (a MEK inhibitor) and cetuximab improved survival outcomes in patients with BRAF V600E-mutant CRC, the overall response rates were modest, highlighting the need for further therapeutic development [68].

Colorectal cancer (CRC) is a heterogeneous disease that is being triaged by the gradual application of precision medicine, including liquid biopsy, epigenetic therapies, artificial intelligence, molecular subtyping, microbiome modulation, and tumor metabolism to newly identified actionable targets. Liquid biopsy is an emerging, non-invasive method to detect CRC early, assess the treatment response, and analyze minimal residual disease. Biomarkers like ctDNA and CTCs offer an on-time view of tumor evolution and therapy resistance, allowing therapy refinements. With the addition of artificial intelligence (AI) and machine learning algorithms, the predictive potential of liquid biopsy has been elevated even further to enable precise therapeutic intervention [69]). In recent times, AI-based deep learning models have started to be employed on histopathological, imaging, and genetic data, aiding in CRC classification and guiding the treatment process. Extracellular vesicles (EVs) have been increasingly recognized as critical mediators in CRC progression, metastasis, and therapy resistance. Tumor-derived EVs transport oncogenic cargo, including microRNAs, proteins, and lipids, modify the extracellular space and promote pre-metastatic niche formation. As EVs are recently appreciated as mediators of intercellular communication and immune modulation, their targeting represents an attractive therapeutic strategy to CRC. Fusobacterium nucleatum, a bacterium associated with CRC tumorigenesis, has also emphasized a new area of research focusing on the tumor microenvironment. Fusobacterium nucleatum facilitates immune evasion and therapy resistance, which makes it an appealing target for microbiome-based interventions [70]. Epigenetic therapy targeting reversal of CIMP-related gene silencing is an active field of CRC therapeutic development. DNA methyltransferase inhibitors (DNMTi) and histone deacetylase inhibitors (HDACi) are being explored to restore normal gene expression and enhance the sensitivity of tumor to immunotherapy. Particularly, MLH1 hypermethylated MMR-deficient tumors have demonstrated good responses to immune checkpoint inhibitors combined with epigenetic regulators. While clinical trials are still underway to confirm the efficacy of these combination strategies, initial results have indicated that they improve treatment outcome and decrease resistance [71]. Molecular subtyping is transforming CRC classification and therapeutic stratification. The concurrent analysis of genomics, transcriptomics, proteomics, and metabolomics, dubbed multi-omics, has improved CRC subtypes, allowing for subtype-targeting therapies. Therefore, high-throughput sequencing technologies and personalized analysis of molecular signatures are being developed to better stratify patients thus bolstering targeted therapeutic decisions [13].

Conclusively, the development of CRC regimens that include specific microbiome-targeted therapies, metabolic inhibitors, and immune- modulating agents is leading to a more personalized cancer treatment paradigm. Tumor metabolism has recently gained increasing attention in CRC biology and unique metabolic vulnerabilities have recently shown promise as therapeutic targets. The dysregulated metabolic pathways such as glycolysis, glutaminolysis, and lipid metabolism play a critical role in CRC progression. Preclinical models have identified metabolic inhibitors of these pathways as potent therapeutic agents, particularly in combination with conventional therapies. This phenomenon is being studied in early-phase clinical trials, which currently assess whether metabolic-targeted therapies can be leveraged to enhance overall treatment efficacy [72]. Artificial intelligence (AI) is increasingly utilized in the diagnosis and treatment of CRC. AI-based tools also expand predictive modeling, radiomics, and pathomics, enabling clinicians to interpret complicated datasets for more precise decisions. The research and development of new drugs, which traditionally takes years, is being accelerated through AIbased drug discovery, identifying new candidate molecules in a much shorter time frame. AI’s analysis results in personalized treatment algorithms that are now being integrated into clinical workflows to increase treatment efficacy and optimize therapeutic outcomes [14].

Immunotherapy is not static and personalized immunotherapeutic approaches using T cell therapy, cancer vaccines and microbiota immune modulation are emerging strategies. These promising results have led to the investigation of novel immunotherapeutic strategies against CRC-specific mutations targeting neoantigens, such as personalized vaccines that are engineered to induce robust anti-tumor immune responses [73]. Moreover, the composition of gut microbiome has been attributed as a key to the efficacy of immunotherapy. Modulation of the microbiome to potentiate immune checkpoint blockade response is an exciting area of exploration in CRC therapy. The potential for personalized therapy in CRC is emerging where the completeness of next generation sequencing (NGS) and multi-omics analysis facilitates the high-definition classification of CRC tumors. Tailoring treatment regimens based on genomic, proteomic, and metabolomic data leads to increased values in patient outcomes and reduced unnecessary toxicity. AI-based decision support systems (DSSs) are suggested to improve treatment adjustments in real time based on genomic and clinical data, taking personalized CRC management to an entirely new level [74-75].

Efforts to develop novel therapeutics are targeting major signaling pathways implicated in CRC progression including WNT/β-catenin and PI3K/AKT pathways. Inhibitors to these pathways are under investigation in preclinical and early-phase clinical trials, with evidence suggesting efficacy at potentiating standard treatments. These targeted agents may eventually move into the CRC therapeutic arsenal as research evolves towards precision-guided therapeutic solutions for aggressive CRC cases. Technological and molecular advances are shaping the future of CRC research and treatment. Liquid biopsy, epigenetic therapies, and microbiome-based strategies, combined with AI and target-driven metabolic approaches make a novel foray towards high precision semi-personalized CRC therapies. High-throughput sequencing and AI-powered analytics will continue to inform molecular stratification, allowing for earlier detection, more accurate treatment strategies, and ultimately improved patient survival. As CRC research enters the precision medicine era, a significant challenge is to convert these novel findings into clinically actionable strategies, which can be adopted across-the-board to enhance patient outcomes.

The study of colorectal cancer (CRC) has significantly evolved, uncovering the complex interplay of genetic mutations, epigenetic alterations, and interactions with the tumor microenvironment that drive carcinogenesis. The identification of key molecular pathways, including chromosomal instability (CIN), microsatellite instability (MSI), and the CpG island methylator phenotype (CIMP), has enhanced our understanding of tumor development and progression. These pathways, in conjunction with clinicopathologic sequences such as the adenoma- carcinoma pathway and the serrated neoplasia pathway, have refined CRC classification and prognostication, ultimately guiding targeted therapeutic interventions. A comprehensive pathology report is critical for ensuring accurate staging, guiding treatment decisions, and predicting prognosis. Molecular profiling, including MSI status, MMR protein expression, KRAS, BRAF, PTEN and PIK3CA mutations, as well as PD-L1 expression, plays a key role in selecting patients for appropriate targeted therapies and immunotherapy. Molecular profiling has become an indispensable tool in the clinical management of CRC, enabling precision medicine approaches tailored to individual patient profiles. The integration of next-generation sequencing (NGS) and artificial intelligence (AI) further enhances predictive modeling and real-time decision-making, optimizing patient outcomes.

Despite advances in targeted therapies and immunotherapy, challenges persist in improving response rates. Tumor heterogeneity, dynamic genetic alterations, and the influence of the tumor microenvironment remain significant barriers to treatment success. However, emerging strategies such as liquid biopsy, epigenetic therapy, microbiome modulation, and metabolic reprogramming offer promising avenues for early detection, treatment monitoring, and novel therapeutic development. The future of CRC treatment is increasingly focused on integrating multi-omics data, AI-driven diagnostics, and personalized immunotherapeutic approaches. The application of artificial intelligence in drug discovery, histopathological analysis, and clinical decision-making holds the potential to further refine CRC management. Additionally, microbiome-based interventions and metabolic- targeted therapies present exciting opportunities to enhance treatment efficacy and overcome drug resistance. As research advances into the era of precision medicine, the challenge lies in translating scientific discoveries into widely accessible clinical applications. The continued development of high-throughput sequencing, molecular stratification, and AI-driven analytics will further refine treatment paradigms, ensuring earlier detection, more effective treatment strategies, and improved survival rates. The ultimate goal remains the implementation of personalized, evidence-based therapies that optimize patient outcomes and reduce the burden of colorectal cancer on global health.

1. (2020) International Agency for Research on Cancer. Global cancer statistics.
2. Siegel RL, Miller KD, Fuchs HE, Jemal A (2021) Cancer statistics, 2021. CA Cancer J Clin 71: 7-33.
3. Levin B, Lieberman DA, McFarland B, Robert A Smith, Durado Brooks, et al. (2018) Screening and prevention of colorectal cancer: A consensus statement. J Natl Cancer Inst 110: 1196-1209.
4. Melina Arnold, Mónica S Sierra, Mathieu Laversanne, Isabelle Soerjomataram, Ahmedin Jemal, et al. (2017) Global patterns and trends in colorectal cancer incidence and mortality. Gut 66: 683-691.
5. Freddie Bray, Jacques Ferlay, Isabelle Soerjomataram, Rebecca L Siegel, Lindsey A Torre, et al. (2018) Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide. CA Cancer J Clin 68: 394-424.
6. Jasperson KW, Tuohy TM, Neklason DW, Burt RW (2010) Hereditary and familial colon cancer. Gastroenterology 138: 2044-2058.
7. Wood LD, Parsons DW, Jones S, Tobias Sjöblom, Rebecca J Leary, et al. (2007) The genomic landscapes of human breast and colorectal cancers. Science 318: 1108-1113.
8. Dolan RD, McSorley ST, Horgan PG, Barry Laird, Donald C McMillan, et al. (2022) The role of the systemic inflammatory response in predicting outcomes in patients with colorectal cancer. Ann Oncol 33: 174-183.
9. Conteduca V, Sansonno D, Russi S, Franco Dammacco (2013) Precancerous colorectal lesions. Int J Oncol 43: 973-984.
10. Hermann Brenner, Michael Hoffmeister, Christa Stegmaier, Gerhard Brenner, Lutz Altenhofen,et al. (2007) Risk of progression of advanced adenomas to colorectal cancer by age. Int J Cancer 121: 1066-1071.
11. Crockett SD, Nagtegaal ID (2019) Terminology, molecular features, epidemiology, and management of serrated colorectal neoplasia. Gastroenterology 157: 949-966.
12. Charles Muller, Akihiro Yamada, Sachie Ikegami, Haider Haider, Yuga Komaki(2022) Risk of colorectal cancer in serrated polyposis syndrome: a systematic review and meta-analysis. Clin Gastroenterol Hepatol 20: 622-630.
13. Vogelstein B, Kinzler KW (1993) The multistep nature of cancer. Trends Genet 9: 138-141.
14. Justin Guinney, Rodrigo Dienstmann, Xin Wang, Aurélien de Reyniès, Andreas Schlicker, et al. (2015) The consensus molecular subtypes of colorectal cancer. Nat Med 21: 1350-1356.
15. Laetitia Marisa, Aurélien de Reyniès, Alex Duval, Janick Selves, Marie Pierre Gaub, et al. (2013) Gene expression classification of colon cancer into molecular subtypes: characterization, validation, and prognostic value. PLoS Med 10: e1001453.
16. Nguyen LH, Goel A, Chung DC (2020) Pathways of colorectal carcinogenesis. Gastroenterology 158: 291-302.
17. Ranganathan S, Doucette MM, Hayward SW (2011) Comparative biology of prostate carcinoma and benign prostatic hyperplasia. Front Biosci 16: 251-268.
18. Bert Vogelstein, Nickolas Papadopoulos, Victor E Velculescu, Shibin Zhou, Luis A Diaz Jr, et al. (2013) Cancer genome landscapes. Science 339: 1546-1558.
19. Russo A, Bazan V, Agnese V, et al. (2005) DNA mismatch repair and colorectal cancer: state of the art and perspectives of treatment. J Surg Oncol 92: 124-133.
20. Yurgelun MB, Kulke MH, Fuchs CS, Allen BA (2015) Diagnosis and management of hereditary gastrointestinal cancer syndromes: Lynch syndrome as a model. J Clin Oncol 33: 353-364.
21. Siegel RL, Miller KD, Jemal A (2017) Cancer statistics, 2017. CA Cancer J Clin 67: 7-30.
22. Mehlen P, Fearon ER (2004) Role of the dependence receptor DCC in colorectal cancer pathogenesis. J Clin Oncol 22: 3420-3428.
23. Michal Sheffer, Manny D Bacolod, Or Zuk, Sarah F Giardina, Hanna Pincas, et al. (2009) Association of survival and disease progression with chromosomal instability: a genomic exploration of colorectal cancer. Proc Natl Acad Sci USA 106: 7131-7136.
24. Qing Li, Shan Geng, Hao Luo, Wei Wang, Ya-Qi M,etal. (2024) Signaling pathways involved in colorectal cancer: pathogenesis and targeted therapy. Sig Transduct Target Ther 9: 266.
25. Konstantinos Venetis, Chiara Frascarelli, Luca Boscolo Bielo, Giulia Cursano, Riccardo Adorisio, et al. (2024) Mismatch repair (MMR) and microsatellite instability (MSI) phenotypes across solid tumors: A comprehensive cBioPortal study on prevalence and prognostic impact.
26. Bessa X, Garcia M, Andreu M (2023) Genetic and epigenetic instability in colorectal cancer: clinical implications. Cancer Genet Cytogenet 195: 102-109.
27. R Parsons, L LMyeroff, B Liu, J K Willson, S D Markowitz, et al. (1995) Microsatellite instability and mutations of the transforming growth factor-beta type II receptor gene in colorectal cancer. Cancer Res 55: 5548-5550.
28. Massagué J (2008) TGFβ in cancer. Cell 134: 215-230.
29. Fearon ER, Vogelstein B (1990) A genetic model for colorectal tumorigenesis. Cell 61: 759-767.
30. Zhao P, Li L, Jiang X, et al. (2019) Prognostic and predictive value of PD-L1 expression in MSI-H colorectal cancer. Cancer Immunol Immunother 68: 841-851.
31. Ehsan NazemalhosseiniMojarad, Peter JkKuppen, Hamid Asadzadeh Aghdaei, Mohammad Reza Zali (2013) The CpG island methylator phenotype (CIMP) in colorectal cancer. Gastroenterol Hepatol Bed Bench 6: 120-128.
32. Lao VV, Grady WM (2011) Epigenetics and colorectal cancer. Nat Rev Gastroenterol Hepatol 8: 686-700.
33. Zong L, Setia N, Mody K, et al. (2016) CpG island methylator phenotype in colorectal cancer: recent advances. Lab Invest 96: 1063-1072.
34. Liu C, Karam R, Zhou Y, et al. (2018) MSI and BRAF mutation in colorectal cancer and their implications for survival. J Clin Oncol 36: 325-333.
35. Issa JP (2004) CpG island methylator phenotype in cancer. Nat Rev Cancer 4: 988-993.
36. Toyota M, Ohe-Toyota M, Ahuja N, et al. (1999) Distinct methylation patterns of CpG island loci in colorectal cancer with or without microsatellite instability. ProcNatlAcadSci USA 96: 8681-8686.
37. Snover DC, Ahnen DJ, Burt RW, Odze RD (2010) Serrated polyps of the colon and rectum and serrated polyposis. WHO Classification of Tumours of the Digestive System (4^th)., pp. 160-165.
38. Domingo E, Niessen RC, Oliveira C, Alhopuro P, Moutinho C, et al. (2005) BRAF-V600E is not involved in the colorectal tumorigenesis of HNPCC in patients with functional MLH1 and MSH2 genes. Oncogene 24(24): 3995–3998.
39. Kambara T, Simms LA, Whitehall VL, et al. (2004) BRAF mutation is associated with DNA methylation in serrated polyps and cancers of the colorectum. Gut 53(8): 1137-1144.
40. Borowsky J, Dumenil T, Bettington M, et al. (2018) The role of BRAF mutations in colorectal serrated pathway cancers. J Pathol 245(2): 139-151.
41. Bond CE, McKeone DM, Kalimutho M, et al. (2016) RNF43 and ZNRF3 in serrated neoplasia pathway. Carcinogenesis 37(6): 546-555.
42. Yan HH, Lai JC, Ho SL, et al. (2017) RNF43 is frequently mutated in colorectal and endometrial carcinomas. Nat Genet 49(5): 775-782.
43. Budinska E, Popovici V, Tejpar S, et al. (2013) Gene expression subtypes of colorectal cancer. Genome Med 5(4): 63.
44. Schlicker A, Beran G, Gyorffy B, et al. (2012) Subtypes of colorectal cancer. Genome Biol 13(12): R92.
45. Chandrasekharan NV, Simmons DL (2004) The cyclooxygenases. Genome Biol 5(9): 241.
46. C E Eberhart, R J Coffey, A Radhika, F M Giardiello, S Ferrenbach, et al. (1994) Up-regulation of cyclooxygenase 2 gene expression in human colorectal adenomas and adenocarcinomas. Gastroenterology 107(4): 1183–1188.
47. C S Williams, M Tsujii, J Reese, S K Dey, R N DuBois (2000) Host cyclooxygenase-2 modulates carcinoma growth. J Clin Invest 105(11): 1589–1594.
48. Ryo Fukuda, Kiichi Hirota, Fan Fan, Young Do Jung, Lee M Ellis, et al. (2003) Insulin-like growth factor-1 induces hypoxia-inducible factor 1-mediated vascular endothelial growth factor expression, which is dependent on MAP kinase and phosphatidylinositol 3-kinase signaling. J Biol Chem 278(12): 10337–10344.
49. Wang D, Fu L, Sun H, et al. (2015) Cyclooxygenase-2 deficiency reduces colorectal cancer risk in experimental models. Cancer Res 75(14): 3051–3058.
50. Catherine Dubé, Alaa Rostom, Gabriela Lewin, Alexander Tsertsvadze, Nicholas Barrowman, et al. (2007) The use of aspirin for primary prevention of colorectal cancer: a systematic review. Ann Intern Med 146(5): 365–375.
51. Quail DF, Joyce JA (2013) Microenvironmental regulation of tumor progression and metastasis. Nat Med 19(11): 1423–1437.
52. Grivennikov SI, Greten FR, Karin M (2010) Immunity, inflammation, and cancer. Cell 140(6): 883–899.
53. Jean Paul Thiery, HervéAcloque, Ruby Y J Huang, M Angela Nieto (2009) Epithelial-mesenchymal transitions in development and disease. Cell 139(5): 871–890.
54. Héctor Peinado, Haiying Zhang, Irina R Matei, Bruno Costa-Silva, Ayuko Hoshino, et al. (2017) Pre-metastatic niches: organ-specific homes for metastases. Nat Rev Cancer 17(5): 302–317.
55. Pavlova NN, Thompson CB (2016) The emerging hallmarks of cancer metabolism. Cell Metab 23(1): 27–47.
56. TaChung Yu, Fangfang Guo, Yanan Yu, Tiantian Sun, Dan Ma,etal. (2017) Fusobacterium nucleatum promotes chemoresistance to colorectal cancer by modulating autophagy. Cell 170(3): 548–563.e16.
57. Justin Guinney, Rodrigo Dienstmann, Xin Wang, Aurélien de Reyniès, Andreas Schlicker, et al. (2015) The consensus molecular subtypes of colorectal cancer. Nat Med 21(11): 1350-1356.
58. Jinru Shia, Zsofia Stadler, Martin R Weiser, Michael Rentz, Mithat Gonen,et al. (2013) Immunohistochemical staining for DNA mismatch repair proteins in colorectal carcinoma: how reliable is it in the presence of tumor-infiltrating lymphocytes? Am J Surg Pathol 37(4): 563–569.
59. Enric Domingo, Renée C Niessen, Carla Oliveira, Pia Alhopuro, Catia Moutinho, et al. (2005) BRAF-V600E is not involved in hereditary nonpolyposis colorectal cancer tumorigenesis. Oncogene 24(24): 3995–3998.
60. Astrid Lièvre, Jean-Baptiste Bachet, Delphine Le Corre, Valérie Boige, Bruno Landi, et al. (2006) KRAS mutation status: Predictive of response to cetuximab therapy in colorectal cancer. Cancer Res 66(8): 3992–3995.
61. Frank A Sinicrope, Qian Shi, Thomas C Smyrk, Stephen N Thibodeau, Rodrigo Dienstmann, et al. (2015) Molecular markers identify subtypes of stage III colon cancer associated with patient outcomes. Gastroenterology 148(1): 88-99.
62. Andreas Schlicker, Garry Beran, Christine M Chresta, Gael McWalter, Alison Pritchard, et al. (2012) Subtypes of primary colorectal tumors correlate with response to targeted treatment in colorectal cell lines. BMC Med Genomics 5: 66.
63. Shuji Ogino, Katsuhiko Nosho, Gregory J Kirkner, Kaori Shima, Natsumi Irahara, et al. (2009) PIK3CA mutation is associated with poor prognosis among patients with curatively resected colon cancer. J Clin Oncol 27(9): 1477–1485.
64. Xiaoyun Liao, Teppei Morikawa, Paul Lochhead, Yu Imamura, Aya Kuchiba, et al. (2012) Prognostic role of PIK3CA mutation in colorectal cancer: Cohort study and literature review. Clin Cancer Res 18(8): 2257-2268.
65. Rawla P, Sunkara T, Barsouk A (2019) Epidemiology of colorectal cancer: incidence, mortality, survival, and risk factors. Prz Gastroenterol 14(2): 89-103.
66. Dung T Le, Jennifer N Uram, Hao Wang, Bjarne R Bartlett, Holly Kemberling, et al. (2015) PD-1 blockade in tumors with mismatch-repair deficiency. N Engl J Med 372(26): 2509-2520.
67. Yvette Schwitalle, Matthias Kloor, Susanne Eiermann, Michael Linnebacher, Peter Kienle, et al. (2004) Immune response against frameshift-induced neopeptides in HNPCC patients and healthy HNPCC mutation carriers. Gastroenterology 126(3): 831-842.
68. Scott Kopetz, Axel Grothey, Rona Yaeger, Eric Van Cutsem, Jayesh Desai, et al. (2019) Encorafenib, binimetinib, and cetuximab in BRAF V600E–mutated colorectal cancer. N Engl J Med 381(17): 1632-1643.
69. Diaz LA, Bardelli A (2014) Liquid biopsies: genotyping circulating tumor DNA. J Clin Oncol 32(6): 579-586.
70. Shailesh M Advani, Pragati Advani, Stacia M DeSantis, Derek Brown, Helena M VonVille, et al. (2018) Clinical, pathological, and molecular characteristics of CpG island methylator phenotype in colorectal cancer: a systematic review and meta-analysis. Transl Oncol 11(5): 1188-1199.
71. Taieb J, Gallois C, Laurent-Puig P, et al. (2021) Methylation biomarkers in colorectal cancer: the challenge of translational application. BiochimBiophys Acta Rev Cancer 1875(1): 188467.
72. Ronald J Hause, Colin C Pritchard, Jay Shendure, Stephen J Salipante (2016) Classification and characterization of microsatellite instability across 18 cancer types. Nat Med 22(11): 1342-1350.
73. Leong KW, Karsan A (2021) Recent insights into the role of the immune system in colorectal cancer development and progression. Front Immunol 12: 729031.
74. C R Boland, S N Thibodeau, S R Hamilton, D Sidransky, J R Eshleman, et al. (1998) A National Cancer Institute Workshop on Microsatellite Instability for cancer detection and familial predisposition. Cancer Res 58: 5248-5257.
75. Liu D, Li L. (2020) Neoantigen load correlates with the immune response in colorectal cancer. J Immunother Cancer 8: e000940.