Colorectal cancer (CRC) is the third diagnosed malignant tumor worldwide, accounting for approximately 9% of annual cancer death (1). More than 90% of CRCs are adenocarcinoma, resulting from normal glandular colonic and rectum epithelium. Other rare types include carcinoid tumors, gastrointestinal stromal tumors, colorectal lymphoma, squamous cell carcinomas, leiomyosarcomas, and melanomas (2). Approximately 65% of CRC cases developed sporadically, without a family history or inherited genetic mutations predisposition, through multiple acquired somatic genomic and epigenetic alterations (3, 4). Other cases are associated with heritable components such as family history (25%), hereditary cancer syndrome (5%), some known CRCs low-penetrance genetic variations (<1%), and other unknown inherited genomic alterations (4, 5).

At the molecular level, CRC, like other solid tumors, is a heterogeneous disease. This can be attributed to at least three major molecular pathways (Figure 1). The most common pathway is the chromosomal instability (CIN), occurring in 85% of sporadic CRC (sCRC), characterized by chromosome structure and number abnormalities, frequent loss of heterozygosity (LOH) at tumor suppressor gene loci, gain or loss of chromosomal segments and chromosomal rearrangements, further resulting in gene copy number variations (6). These alterations typically are associated with mutations in specific oncogenes or tumor suppressors genes such as adenomatous polyposis coli (APC), Kirsten rat sarcoma virus (KRAS), phosphatidylinositol 4,5-bisphosphate 3-kinase catalytic subunit-α (PIK3CA), b-raf proto-oncogene (BRAF), SMAD family member 4 (SMAD4) or p53 (7), which regulate cell proliferation and cell cycle and play pivotal roles in CRC initiation and progression pathways. Another important pathway to CRC is the microsatellite instability (MSI), caused by dysfunction of DNA mismatch repair (MMR) genes during DNA recombination, DNA replication and DNA damage, which encodes MutL Homolog (MLH) proteins or MutS homolog (MSH) proteins. Therefore, it is often associated with genetic hypermutability (8). CpG island methylator phenotype (CIMP) comprises the third major pathway to CRC. CIMP positive tumors can be divided into CIMP^high tumors, with BRAF mutations, MLH1 methylation, and silencing of O-6-methylguanine-DNA methyltransferase (MGMT) or cyclin-dependent kinase inhibitor 2A (CDKN2A) and CIMP^low tumors, with KRAS mutations. In fact, CRC pathogenesis is normally associated with multiple pathways. Specifically, approximately one third CRC cases develop via a serrated pathway that is associated with either KRAS or BRAF mutations in addition to CIMP mutations (CIMP^low or CIMP^high).

Generally, CRC tumorigenesis initiates with the transformation of normal colorectal epithelial cells by a spontaneous mutation, environmental mutagens, genetic or epigenetic alterations. Then, these initiated cells rapidly expanded to form aberrant crypt foci and early adenoma, driven by mutations that cause hyperproliferation, such as APC mutations, or other signaling pathways such as WNT-β-catenin, cytokines, chemokines, and growth factors from the tumor microenvironment (TME) (9). Thus, mutations such as SMAD4, cell division control protein 4 (CDC4) and transforming growth beta factor 2 (TGFBR2) as well as chromosomal aberration such as LOH 18q further lead to the outgrowth of these clones into late adenoma and malignant tumors, known as tumor promotion. Further mutations (p53 and Bcl-2 Associated X-protein (BAX)), pro-angiogenic factors, extracellular matrix-degrading factors and other factors which promote CRC cells invasive motility facilitate these tumors efficiently to metastasize to distant organs and tissues, known as tumor progression.

Although genetic understanding of these CRC tumors and paired metastases has led to major advances in elucidating early driver genes responsible for carcinogenesis and metastasis, the pathophysiological contribution of transcriptional and epigenetic alterations in this malignancy influences many central signaling pathways have attracted attention recently. Therefore, the present review summarized the current knowledge on the molecular network of CRC, including cellular signaling pathways, CRC microenvironment modulation, epigenetic changes, CRC and inflammatory bowel disease (IBD), and CRC biomarkers for early diagnosis and prognostic/predictive use and provide an overview of opportunities for CRC treatment.

Treatments of CRC include surgical resection, chemotherapy, targeted therapy, immunotherapy, gene therapy and combination therapies.

Chemotherapeutic intervention coupled with surgery is the traditional treatment for survival rate enhancement of metastatic CRC. For decades, commonly used CRC chemotherapeutic agents include 5-Fluorouracil, Irinotecan, Oxaliplatin, Calcium folinate, Capecitabine, S-1 (Tegafur/gimeracil/oteracil), and TAS-102 (Trifluridine/Tipiracil). CRC chemotherapies are usually combinations of several of those chemotherapeutic agents, such as FOLFOX (5-Fluorouracil, Calcium folinate, Oxaliplatin); FOLFIRI (5-Fluorouracil, Calcium folinate, Irinotecan); CAPEOX (Capecitabine, Oxaliplatin); FOLFOXIRI (5-Fluorouracil, Calcium folinate, Irinotecan, Oxaliplatin) (112).

Targeted therapies can be divided into two major categories: small molecule signal transduction inhibitors and monoclonal antibodies. Small molecule inhibitors can eliminate cancer cells by specifically blocking the signaling pathways necessary for tumor growth and proliferation and they can be administered orally, with high specificity and short half-life. Monoclonal antibodies, with high specificity and long half-life, recognize tumor cells through the specific antigen and antibody interaction, which are mostly administered intravenously, without hepatic metabolism. Genomic profiling for somatic mutations detection is very important because it identifies which treatments may be effective for CRC patients. Several molecular biomarkers such as KRAS mutations, BRAF mutations, and deficient MMR/MSI are commonly observed in CRC.

KRAS, a downstream pathway of EGFR, is the most observed mutation in CRC. Therefore, inhibition of EGFR can suppress the activation of KRAS and its downstream pathways. Cetuximab (Epitol), a monoclonal antibody specifically used for blocking EGFR, is one of the most representative drugs of targeted therapy and is currently approved for first-line monotherapy in CRC patients (113). However, CRC patients with KRAS mutation are very unlikely to benefit from anti-EGFR therapy. Eventually, its downstream pathways such as BRAF, MEK and ERK were continuously activated by KRAS, resulting in tumor cells proliferation and metastasis. Thus, CRC patients with KRAS mutations need to be treated with other effective therapies. Cetuximab can also be used in combination with Irinotecan, 5-Fluorouracil, and calcium folinate in the first-line treatment regimen for CRC. It was reported that the overall survival (OS) of CRC patients with wild type KRAS under Cetuximab and FOLFIRI combination treatment was improved 8.2 months compared with patients treated with FOLFIRI alone in the CRYSTAL study (114, 115). EGFR can initiate cell proliferation downstream pathway in addition to activation of KRAS. Particularly, most patients with advanced CRC are accompanied by EGFR overexpression and metastasis in the liver and lung, resulting in poor prognosis outcomes with first- and second-line therapies. Currently, clinical studies have confirmed the beneficial effects of third-line treatment of metastatic CRC with Cetuximab in Australia and Europe (116, 117).

Angiogenesis plays an important role in CRC progression and vascular endothelial growth factor (VEGF) is a major angiogenic factor in CRC (118). Increased VEGF levels were significantly correlated with CRC progression and metastasis (119). Bevacizumab, a humanized IgG monoclonal antibody, was approved by FDA as the first VEGF-targeted agent for metastatic CRC (120) and it is still the only FDA-approved VEGF-targeted agent used as a first- and second-line for CRC treatment (121). What’s more, various approved novel anti-VEGF receptor (VEGFR) agents such as Aflibercept (a VEGFR-1 and VEGFR-2 extracellular domain recombinant fusion protein) and Ramucirumab (a fully humanized monoclonal VEGFR-2-targeted IgG antibody) remained as second-line treatment of metastatic CRC. This is mainly because the combination of Aflibercept or Ramucirumab with FOLFOX regimen did not show beneficial results in progression-free survival (PFS) or response rate (122, 123). Other agents, such as various tyrosine kinase inhibitors, including Regorafenib (124) and Fruquintinib (125) are recommended for chemotherapy against refractory metastatic CRC.

Growing evidence indicates that the hepatocyte growth factor (HGF) and its tyrosine kinase receptor, mesenchymal-epithelial transition factor (c-MET) pathway plays a crucial role in tumor invasive growth and metastasis of CRC (126, 127). MET amplification and HGF/c-MET overexpression have been reported as causing factors for CRC carcinogenesis. This aberrant signaling axis triggers a series of signaling cascades activation, including EGFR/MAPK, PI3K/AKT, and JAK/STAT pathways, and is associated with dysregulated cell proliferation, apoptosis, and poor prognosis (121). Thus, various newly developed monoclonal antibodies, small molecules or miRNAs targeting HGF/c-MET have emerged. Recently, a few monoclonal HGF neutralizing antibodies have been investigated under clinical trials. Rilotumumab, a fully humanized monoclonal antibody, has demonstrated median prolonged PFS and OS in patients with MET overexpression using Rilotumumab in combination with ECX (epirubicin, cisplatin, and capecitabine) compared with those in the placebo plus ECX arm (128). There are also several novel MET antibodies, like ABT-700 and Emibetuzumab, that have been used in clinical trials (121). Small molecule c-Met kinase inhibitors target activation sites of the receptor inside the cells via inhibiting phosphorylation and its downstream signaling pathway. PHA-665752, c-Met inhibitor, with the combination of cetuximab suppressed more efficiently the CRC cell growth in vitro and in vivo compared with either single agent treatment (129). In addition, miRNA, such as miR-206, has been identified to be able to affect the c-MET/HGF signaling pathway via inhibiting CRC cells proliferation and invasion (130).

In CRC, in addition to common mutations in KRAS and BRAF, mutations in human epidermal growth factor receptor (HER) 2 are also present. Indeed, HER2 amplification has been detected in approximately 2-3% of CRC patients. However, there was no approved targeted therapy in HER2-positive metastatic CRC patients. Most recently, an open-label, multicenter, phase 2 study confirms the role of Enhertu (DS-8201, an antibody-drug conjugate targeting HER2) in HER2 overexpressing CRC. It has demonstrated that Enhertu showed promising activity in terms of objective response rate, disease control rate, PFS and OS with long-term follow-up in CRC patients with HER2 expressing which brings new hope to the majority of these patients (DS8201-A-J203, NCT03384940).

Immunotherapy, as an innovative type of cancer treatment, fights tumors by activating the body’s immune system without direct attacking tumor cells. Immune checkpoint inhibitors exhibit an immense breakthrough in cancer therapeutics and the most representative ones are PD-1/PD-L1 inhibitors and CTLA-4 inhibitors. Some of these immune checkpoint inhibitors have been approved for CRC treatment. For example, nivolumab and pembrolizumab, the two most promising PD-1 inhibitors, have been approved by FDA for the second-line treatment of metastatic CRC patients with high levels of MSI and deficient MMR in 2017 (131). However, a recent clinical study reported limited response with antibodies blocking PD1 and CTLA4 in CRC patients with MMR-MSI mutations (132). Therefore, other alternative approaches of immune modulations are required to enhance the recruitment of immune cells to the tumor, such as combination therapy. In addition to pro-proliferative effects on tumors, activation of the RAS/MAPK/MEK pathway has been associated with decreased T cell infiltration into CRC tumors and MEK is a transducer of this signaling cascade and plays a crucial role in CRC development and progression. Thus, several clinical trials with the combination of MEK inhibitors with PD-1/PD-L1 and chemotherapies are ongoing.

Gene therapy has shown remarkable efficacy in recent clinical studies (133). Lack of efficient and reliable delivery methods is the most common challenge for gene therapy, although retrovirus delivery systems are the most used. Recently, nanotechnology-based gene therapy has been approved in the clinic because it can be well-controlled (134). A group found that cancer-targeted gene therapy pigment epithelium-derived factor DNA liposomes (R-LP/PEDF) demonstrated enhanced inhibitory effects on migration, invasion, and pro-apoptosis of CRC cells and without any toxic pathological changes in the vital organs of mice that received the R-LP/PEDF treatment (135). What’s more, a nano vector made from ginger-derived lipids was reported that can serve as a delivery platform for the therapeutic agent doxorubicin to treat CRC (136). This study expands our current understanding of drug-delivery systems and could be used as a foundation for a less toxic drug-delivery approach.

In recent years, proteolysis targeting chimeric (PROTAC) technology that degrades endogenous disease-related proteins through the ubiquitin-proteasome system has achieved remarkable efficacy in tumor growth inhibition and is also a promising strategy for the drug-resistant target. As discussed previously, KRAS is a well-known “undruggable” target in CRC due to its lack of deep pockets and is associated with poor prognosis and drug resistance. It has been reported that potent KRAS-specific degraders can target tumors with mutant KRAS (137). A research group led by Professor Craig M. Crews, a pioneer of PROTACs technology, firstly reported the development of endogenous KRASG12C degrader LC-2, where over half of KRAS mutant lung cancer and 3% of CRC express KRASG12C. It was demonstrated that LC-2 rapidly and consistently degraded KRASG12C, leading to inhibition of MAPK signaling. Suggesting that PROTACs-mediated degradation is a feasible strategy to attenuate oncogenic KRAS levels and downstream signaling in cancer cells (138). Therefore, PROTAC approaches can broaden drug target scope, overcome drug resistance, enhance target selectivity and this technology could be a key therapeutic modality that provides innovative treatment options for cancer patients.

It should be noted that patients with colon and rectal cancer are managed differently, and their epidemiology and clinical outcomes also exhibit different patterns (139). Zhang et al. recently reported that several genes can be used as biomarkers to identify colon cancer from rectal cancer based on the prediction of vector machine method algorithm and RNA-sequencing data, while additional research is required to support this finding (140). However, a study conducted by The Cancer Genome Atlas (TCGA) showed similar patterns of genetic alterations in the colon and rectal tissues from non-hypermutated tumors. There were almost indistinguishable differences in copy number, CIMP, mRNA and miRNA between the colon and rectal cancers from the above samples (141). A follow-up proteomics study of colon and rectal tumors analyzed previously by TCGA identified that CRC subtypes are similar to those detected by transcriptome analysis (142). The main difference between colon cancer and rectal cancer is the anatomical location, which determines the differences in their treatments modalities such as differences in surgical approaches, and preoperative and postoperative adjuvant therapies (radiotherapies). Drugs approved for colon and rectal cancer are identical according to their identical biology. Nevertheless, chemotherapies were used differently for different purposes and indications caused by their different anatomical location (143–145).

In summary, as discussed in this review, CRC represents a complex and heterogeneous group of disorders at molecular levels and signaling pathways, in which different patterns of genetic mutations and epigenetic alterations contribute to the onset and progression and are responsible for responding to the specific therapy. CRC results from the activation of multiple signaling pathways and cannot be targeted with a single treatment. Thus, a combination of conventional therapeutics with innovative inhibitors targeting different dysregulated pathways is urgently needed. A deeper understanding of CRC is required, and the efficiency of targeted therapies and the development of more efficient biomarkers provides an encouraging prospect for the future management of CRC. We believed that with the discovery of more novel targeted therapeutics, the disease burden of CRC can be decreased in the future.

ZH and MY wrote different sections of the manuscript. MY revised, wrote, and prepared the manuscript. All authors contributed to the article and approved the submitted version.

This work was supported by Natural Science Foundation of Liaoning Province (20170541001).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.