The biological complexity of colorectal cancer: insights into biomarkers for early detection and personalized care

Abstract

Colorectal cancer has been ranked the third and second most prevalent of all cancers in men and women, respectively, and it represents the fourth most common cause of cancer deaths. In 2012, there were 1.4 million estimated cases of colorectal cancer worldwide, and 700,000 estimated deaths, which implies significant impact on public health, especially in economically-developed countries. In recent years, there has been an increase in the number of tumors, although this has been accompanied by decreased mortality, due to more appropriate and available information, earlier diagnosis, and improvements in treatment. Colorectal cancers are characterized by great genotypic and phenotypic heterogeneity, including tumor microenvironment and interactions between healthy and cancer cells. All of these traits confer a unique peculiarity to each tumor, which can thus be considered as an individual disease. Well conducted molecular and clinical characterization of each colorectal cancer is essential with a view to the implementation of precision oncology, and thus personalized care. This last aims at standardization of therapeutic plans chosen according to the genetic background of each specific neoplasm, to increase overall survival and reduce treatment side effects. Thus, prognostic and predictive molecular biomarkers assume a critical role in the characterization of colorectal cancer and in the determination of the most appropriate therapy.

Keywords

colorectal cancer epidemiology colorectal cancer molecular heterogeneity colorectal cancer tumorigenesis predictive biomarkers prognostic biomarkers

Colorectal cancer: epidemiology and therapy

Colorectal cancer (CRC) is one of the main causes of cancer deaths and the third most frequent cancer worldwide, with about 30% of cases being hereditary or familial [Ferlay et al. 2010, 2015; Karsa et al. 2010]. There are several different causes involved in CRC incidence. Sporadic CRCs arise following somatic mutations, and these account for around 75% of all CRCs. Germline-inactivating mutations in oncogenes or tumor suppressor genes cause hereditary CRC, while minor variant or single nucleotide polymorphisms in the same genes are responsible for familial CRCs. First-degree relatives of patients with CRC have a three-fold greater risk of CRC than that of individuals without familial predisposition [Jasperson et al. 2010; Duraturo et al. 2011; Rustgi 2007; De Rosa et al. 2015].

Hereditary nonpolyposis CRC (HNPCC), or Lynch syndrome, is an autosomal dominant disease that is caused by a germline mutation in one of the DNA mismatch repair (MMR) genes, such as MLH1 and MSH2, or also MSH6 and PMS2. HNPCC is the most common form of hereditary CRC, and it accounts for 1–6% of all colorectal malignancies [Aaltonen et al. 1998; Samowitz et al. 2001]. Patients with HNPCC should undergo colectomy with ileorectal anastomosis, while carriers of a mutation in these MMR genes should undergo colonoscopy every 1–2 years from the age of 20–25 years, and for woman, pelvic evaluation should be followed from the age of 25–35 years. Also, relatives of these patients should be tested for MLH1 and MSH2 mutations [Akiyama et al. 1997; Bronner et al. 1994; Fishel et al. 1993; Miyaki et al. 1997; Nicolaides et al. 1994; Niessen et al. 2009].

Adenomatous polyposis and hamartomatous polyposis are inherited syndromes that account for 2–5% of all colon cancers. Adenomatous polyposis syndromes include familial adenomatous polyposis (FAP), attenuated FAP (AFAP), and MUTYH-associated polyposis (MAP). Hamartomatous polyposis syndromes include Peutz–Jeghers syndrome, juvenile polyposis syndrome, and ‘PTEN hamartoma tumor syndrome’. AFAP and MAP, and also Lynch syndrome, can all be associated with only a few adenomas (i.e. 3–10 polyps); nevertheless, each syndrome has distinct cancer risks, characteristic clinical features, and separate genetic etiology. Thus, differential diagnosis is essential for correct management of the specific disease [Lucci-Cordisco et al. 2013]

Familial adenomatous polyposis is a hereditary autosomal dominant syndrome that is mainly associated with mutation of the tumor suppressor APC (adenomatous polyposis coli) gene, which is located on chromosome 5q21-q22 [Groden et al. 1991; Kinzler et al. 1991]. FAP is characterized by the onset of hundreds to thousands of adenomas at the level of the large intestine, which usually occurs at a young age. If these are untreated, progress is almost invariably (90%) toward the development of CRC at an average age of 40 years [Vasen et al. 2008]. Patients who carry an APC mutation should undergo colonoscopy every 1–2 years from the age of 10–12 years, and in the presence of adenomatous polyps, they need, instead, annual endoscopic follow up. Prophylactic colectomy is recommended if the polyps are endoscopically untreatable, and proctoscopy should be performed every 6 months, with monitoring for ileum–anal anastomosis every 2 years if total proctocolectomy has been performed [Jasperson et al. 2010].

Patients with ulcerative colitis and Crohn’s disease are associated with high risk of developing CRC. According to a relatively recent meta-analysis, the risk here is estimated at 2%, 8%, and 18% after 10, 20, and 30 years, respectively. The duration and extent of the ulcerative colitis or Crohn’s disease are the two most important factors that determine the patient cancer risk. Patients with irritable bowel disease who have a family history of CRC have double the cumulative risk of developing dysplasia, compared with those with a negative history [Eaden et al. 2001].

The trend in the incidence of CRC by age follows the typical curve of the majority of cancers, increasing with age. Indeed, CRC diagnoses are sporadic in the younger age groups, while the frequencies are of the order of 50 cases per 100,000 inhabitants/year between 30–50 years, which increases to 100 cases between 50–70 years, and reaching about 250 cases at ages >70 years. The incidence rates are similar between the two sexes up to the age of 45 years; however, the curve for males then shows a steeper increase and reaches higher values.

Surgery is the mainstay of treatment for early colorectal tumors. Colon cancers are treated by surgical excision and with complete mesocolic excision [Hohenberger et al. 2009; Sehgal and Coffey 2014] which appears to be as appropriate as total mesorectal excision for rectal cancer. These allow excellent oncological outcome, with a 5-year cancer-specific survival rate of 91.4% in stage II, and 70.2% in stage III [Siani and Pulica, 2014]. In the case of obstruction, perforation, and bleeding, emergency segmental colectomy for resection of the tumor is indicated [Sagar, 2011]. Several international multicenter randomized trials have shown that minimally invasive surgery is a safe and valid alternative to open surgery, with less postoperative pain, shorter duration of hospital stay, and similar oncological outcomes. Robotic surgery has not been shown to be better for either colon or rectal cancer, compared with an adequate laparoscopic approach [Tyler et al. 2013; Baek et al. 2013].

The staging of rectal cancer is crucial to define the most appropriate treatment strategy, with some important factors defined here, such as distance from the anal verge, circumferential margin, nodal stage, and vascular and nerve invasion. Local excision can be considered for very early tumors (i.e. stage 1) and for malignant polyps with an invasion level of ‘0–3’; this should also be supported by accurate preoperative diagnosis with endorectal ultrasound, magnetic resonance imaging, and computed tomography scanning. A transanal procedure as a video-assisted technique can be performed, with either transanal endoscopic microsurgery [Sgourakis et al. 2011; Guerrieri et al. 2014] or transanal minimally invasive surgery [Matz and Matz, 2012; Atallah et al. 2010]. In the more advanced disease stages, surgery with total mesorectal excision is the appropriate procedure, either performed laparoscopically or as open surgery. Neoadjuvant chemoradiotherapy is indicated in locally-advanced rectal cancer, to reduce local recurrence rates. This became standard practice after the publication of the results of the German CAO/ARO/AIO-94 randomized phase III trial [Sauer et al. 2012]. Long-course chemoradiation and short-course radiation can both be considered here [Sauer et al. 2012; Schmoll et al. 2012]. Infusional 5-fluorouracil has been commonly used in neoadjuvant treatment, although the oral capecitabine appears to be equally effective and more manageable [Hofheinz et al. 2012].

Stage IV disease (i.e. CRC with metastatic sites, mCRC) is commonly treated with chemotherapy, with several drugs used in combination with 5-fluorouracil and leucovorin (e.g. oxaliplatin–FOLFOX, irinotecan–FOLFIRI). There are two monoclonal antibodies against the epidermal growth factor receptor (EGFR) that are now commercially available (i.e. cetuximab and panitumumab) [Downward 2003], and these are used in combination with commonly used treatment schedules [Tveit et al. 2012; Peeters et al. 2010; van Cutsem et al. 2009]. The National Cancer Institute of Canada Clinical Trials Group CO.17 trial demonstrated that the anti-EGFR monoclonal antibody cetuximab improves overall survival (OS) and progression-free survival (PFS) in patients with advanced, chemotherapy-refractory CRC, particularly in patients with wild-type KRAS tumors [Au et al. 2009].

These therapies show very low objective response rates because of the high Ras/Raf mutation rates in CRC [Bazan et al. 2002]. Biological chemotherapy also includes the vascular endothelial growth factor (VEGF)-A-targeted antibodies bevacizumab and aflibercept, recombinant proteins that target VEGF-A, VEGF-B, and placental growth factor (PlGF), and these have been approved by the US Food and Drug Administration for mCRC, in combination with standard cytotoxic chemotherapy [Ciombor et al. 2015; Hurwitz et al. 2004; van Cutsem et al. 2012; Tabernero et al. 2014]. New therapeutic agents in the targeting of the EGFR and VEGF receptor (VEGFR) pathways, and those targeting other signal-transduction pathways, such as the MET, IGF1R, MEK, phosphoinositide 3-kinase (PI3K), Wnt, Notch, Hedgehog, and death-receptor signaling pathways, are now under evaluation for the treatment of mCRC [Seow et al. 2016].

Some groups claim that regular use of nonsteroidal anti-inflammatory drugs (NSAIDs), including aspirin, has an important role in the prevention of CRC; according to some studies, these can reduce the risk by 40% [Cuzick et al. 2009]. Furthermore, randomized studies have evaluated the preventive effects of aspirin on the development of adenomatous polyps, with reports of an inverse association between use of NSAIDs and incidence of CRC [Bosetti et al. 2006, 2009; Chan et al. 2005; Dubé et al. 2007]. Although numerous studies have shown preventive effects of aspirin on incidence and size of adenomatous polyps [Gann et al. 1993; Baron et al. 2003; Benamouzig et al. 2003; Chan et al. 2004], the dose and duration of treatment have never been specified, and some studies have failed to demonstrate effects of aspirin on the incidence of any specific cancer [Cook et al. 2005]. More randomized trials are mandatory to clarify the real preventive role of aspirin in CRC.

Model for colorectal cancer tumorigenesis

The development of CRC is characterized by the ‘adenoma–carcinoma sequence’. Thus, the normal epithelium acquires sequential genetic and epigenetic mutations in specific oncogenes or tumor suppressor genes, becomes a hyperproliferative mucosa, and later gives rise to a benign adenoma that changes into a carcinoma and metastases over an average of 10 years (Figure 1) [Pancione et al. 2012; Ewing et al. 2014; Vogelstein et al. 1988].

Figure 1.

Colorectal cancer heterogeneity during the adenoma carcinoma sequence.

Normal gastrointestinal epithelium is organized along a crypt–villus axis. At the bottom of the crypts, there is a pool of colon stem cells and progenitor cells, which are the most undifferentiated cell types that can undergo self-renewal and show pluripotency. These cells can move along the crypt–villus axis while differentiating into all of the epithelial colon lineages, such as Paneth cells, goblet cells, enterocytes, and enteroendocrine cells. In about 14 days, they arrive at the top of the villus, and undergo programmed cell death by apoptosis [Peifer, 2002; Kosinski et al. 2007]. This process is orchestrated by gradients of proteins, such as Wnt, bone morphogenetic protein (BMP) and transforming growth factor (TGF)-β, together with the extracellular matrix and stromal cells that form the cell niches [Medema and Vermeulen, 2011].

The stem cell niches thus have a critical role in the maintenance of stem cell properties, by providing them with protection from differentiation and apoptosis, and through controlling their turnover [Moore and Lemischka, 2006; Medema and Vermeulen, 2011]. It has been suggested that these stem cells are involved in cancer onset and progression, through their acquiring tumorigenic potential, and thus transforming into cancer stem cells (CSCs). These CSCs represent the only tumor cells that can initiate cancer transformation, progression, and metastases, and they are probably responsible both for relapse and resistance to therapy [Gangemi et al. 2009; Fanali et al. 2014]. The source from which CSCs arise is still unclear. It remains a topic for discussion whether differentiated cancer cells can de-differentiate, and thus transform into CSCs, or alternatively, whether healthy stem cells accumulate genetic mutations that confer them a neoplastic phenotype [O’Brien et al. 2007; Greaves and Maley, 2012]. Moreover, in contrast to the clonal hypothesis that was formulated to explain the mechanisms responsible for cancer onset and progression, the CSC hypothesis is a hierarchical theory according to which only a few undifferentiated cells, the CSCs, can promote tumor onset, progression, and metastases [Puglisi et al. 2009]. In agreement with this hypothesis, tumors should be composed mainly of differentiated cancer cells and of a few undifferentiated CSCs, which although slowly duplicating, are the only cells that can undergo self-renewal and show multipotency [Vermeulen et al. 2008].

It has been demonstrated recently that during epithelial to mesenchymal transition (EMT), epithelial cancer cells acquire not only mesenchymal traits, but also a stem cell-like phenotype, with the ability for self-renewal, unlimited proliferation, and resistance to apoptosis. This observation suggests that EMT might represent a tumor source of undifferentiated cancer cells, as the CSCs or stem cell-like cells [Dallas et al. 2009; Fan et al. 2012]. EMT consists of an essential phenotypic conversion of epithelial cells into cells with a mesenchymal phenotype. It is a reversible process that often occurs during embryonic development and tissue remodeling, and it also has a critical role in the early events that occur in invasion and metastases of many types of cancer, including CRC [Batlle et al. 2000]. EMT regulation is orchestrated through a group of transcription factors, including Snail, Slug, ZEB1 and Twist, although the tumor microenvironment also has a role in this conversion through different signals, such as TGF-β, EGF, Wnt and Notch [Cano et al. 2000; Li et al. 2014].

During EMT, epithelial cells lose E-cadherin expression, which specifically guarantees the epithelial phenotype, with destruction of their intercellular adhesion, and acquisition of mesenchymal characteristics and increased migratory and invasive properties [Thiery et al. 2009]. Furthermore, the EMT program induces stem cell-specific gene expression, thus promoting the self-renewal capability [Mani et al. 2008].

Metastases are the first cause of death in CRC, as well as with many other solid tumors, and this defines stage IV CRC, which is characterized by relatively short OS [Singh and Settleman, 2010]. During metastases development, the cells acquire the ability to invade the basal membrane that divides the epithelial tissues from the extracellular matrix, and they migrate into surrounding tissues. Later, they join the endothelial cells from lymphatic and blood vessels and arrive in the lumen, in a process known as intravasation. Subsequently, the cells that survive in the lumen of the vessels come out from the vases (i.e. extravasation), disseminate into the adjacent organs, and finally colonize them to generate micrometastases,

Thus, as described above, EMT confers to the epithelial cells the traits that are needed to enable them to complete this process [Singh and Settleman, 2010; Thiery et al. 2009].

Molecular heterogeneity of colorectal cancer

At the molecular level, CRCs are a very heterogeneous group of diseases, and this appears to have a critical role in the generation of drug resistance, and ultimately, in the mechanism of relapse. Loss of genomic integrity facilitates accumulation of multiple mutations during the development of CRC. Chromosomal instability, microsatellite instability (MSI), aberrant DNA methylation (i.e. the CpG island methylator phenotype), and DNA repair defects are all mechanisms that are involved in colorectal epithelial cell transformation, and they all have a significant role in CRCs (Figure 1) [Goel et al. 2007; Sideris and Papagrigoriadis, 2014; Pino and Chung, 2010; Boland and Goel, 2010; Colussi et al. 2013]. It is well accepted that a tumor is classified as positive in terms of the CpG island methylator phenotype if it shows methylation of at least three of the following markers: CACNA1G, IGF2, NEUROG1, RUNX3 and SOCS1 [Cohen et al. 2016].

Chromosomal instability arises in about 60% of CRCs, and this is a phenomenon that is related to the presence of abnormal chromosome numbers or complements [Pino and Chung, 2010]. MSI is caused by alterations to DNA MMR mechanisms, and it consists of variations in the numbers of repetitive units in each microsatellite sequence. MSI is a characteristic of patients with Lynch syndrome, although it is also associated with about 15–22% of sporadic CRCs [Corso et al. 2013]. Inactivation of the MMR genes (i.e. MLH1, MSH2, MSH6, PMS1, PMS2) through deletions, point mutations, or epigenetic silencing [Aaltonen et al. 1993; Ionov et al. 1993; Thibodeau et al. 1993; Vasen et al. 1999] is responsible for MSI, which causes the failure of the DNA MMR system to repair errors that occur during replication. On the basis of the observed degree of instability, it has been possible to distinguish two well defined MSI phenotypes in particular: MSI-high (MSI-H) and MSI-low (MSI-L; or microsatellite stable, MSS) [Boland and Goel. 2010; Colussi et al. 2013].

Tumor cells with an MSI-positive phenotype accumulate mutations consequent to the defective DNA repair mechanism, and this might activate key oncogenes or inactivate tumor suppressor genes. The TGF-β tumor suppressor gene [Markowitz et al. 1995], the TGF-β type II receptor (TGF-βR2), BAX, caspase 5 apoptotic regulators [Duval and Hamelin, 2002; Vousden and Prives, 2009], and the tumor suppressor gene TCF4 (which has been implicated in deregulation of the Wnt/β-catenin/TCF signaling pathway) are the most frequently altered genes in these tumors [Cuilliere-Dartigues et al. 2006].

The mechanism of MMR gene inactivation is different between sporadic and hereditary cancers. The most frequent MMR gene alteration in sporadic CRCs consists of promoter hypermethylation of the hMLH1 gene, instead of point mutations of the hMLH1 or hMSH2 genes, as described for HNPCC [Markowitz and Bertagnolli, 2009]. The MSI sporadic CRCs are frequently associated with the BRAF V600E mutation, which mediates extensive epigenetic silencing in CRC [Fang et al. 2016]. This therefore represents a diagnostic marker to distinguish sporadic from hereditary CRC [Capper et al. 2013].

Several altered molecular signaling pathways are involved in CRC onset, such as the Wnt, RAS/MAPK, PI3K, TGF-β, NF-κB pathways or DNA MMR genes. These alterations are responsible for responsiveness or resistance to antitumor agents (Figure 2), and they can confer individual susceptibility to cancers when they are germlines [Markowitz and Bertagnolli, 2009; Najdi et al. 2011; De Rosa et al. 2015].

Figure 2.

Functional cross-talk between the signaling pathways involved in colorectal cancer.

Activating mutations of the Ras gene and pathway, including alterations of many of its regulators or effectors, probably represent the most frequent genetic alterations in human tumors [Bos, 1989; Rojas and Santos, 2002; Parsons et al. 2005; Mandal et al. 2015; Moura et al. 2015; Ratner and Miller, 2015]. Ras is a multitasking protein that cooperates with several effectors in the regulation of many cellular processes, such as cell proliferation, differentiation, apoptosis, and senescence [Moodie et al. 1993; Warne et al. 1993; Hallberg et al. 1994;]. The Ras genes (i.e. K-, H- and NRas) are localized on chromosomes 12, 11 and 1, respectively, and encode small proteins with GTPase activity that associate with the plasma membrane.

Ras mutations usually activate Ras signaling by enhancing GTP levels, and in this way mediating phosphorylation and activation of the Raf proteins (A, B and C) (Figure 2) [Vojtek et al. 1993; Zhang et al. 1993; Malumbres and Barbacid, 2003]. It is now believed that oncogenic Ras can promote tumor onset and progression [Chin et al. 1999], although multiple Ras-interacting proteins, such as the PI3Ks, appear to be required to sustain tumor growth [Lim and Counter, 2005]. For example, oncogenic Ras and PI3K cooperate to promote the loss of anoikis, so taking part in the progression of an advanced tumor into metastases [Chiarugi and Giannoni, 2008; Simpson et al. 2008].

The PI3K pathway is often dysfunctional in both sporadic and hereditary CRC, and it activates cell growth, inhibits apoptosis, and regulates cellular transformation, adhesion, motility, and survival [Cantley, 2002; Fruman et al. 1998] in response to several extracellular signals, such as growth factors, cytokines, hormones, heat and oxidative stress, hypoxia, and hypoglycemia (Figure 2) [Testa and Bellacosa, 2001; Martini et al. 2014; Vivanco and Sawyers, 2002; Castellano and Downward, 2011].

Several genes and alternative splicing mechanisms are responsible for the synthesis of the PI3K heterodimeric lipid kinases complex, which consists of the catalytic, adaptor and regulatory subunits. This protein family has three main classes (class I, II, and III), with class I as the best characterized and the most clearly known to be associated with human cancers [Yuan and Cantley, 2008; Castellano and Downward, 2011]. The main reported alterations are recurrent mutations in the PIK3CA gene [Engelman et al. 2006; Samuels et al. 2004], such as gene amplifications, deletions, and somatic missense mutations, which probably increase the kinase activity of PIK3CA, to contribute to cellular transformation [Karakas et al. 2006].

PTEN is a negative regulator of the PI3K/AKT pathway (Figure 2), and is a tumor suppressor gene, alterations to which are involved in several sporadic and hereditary CRCs. As with small quantitative alterations of PTEN expression, somatic mutations in the PTEN gene have been shown to be associated with different kinds of cancers [Eng 2003; Manfredi 2010].

The hepatocyte growth factor (HGF) receptor is known as mesenchymal–epithelial transition factor (cMET), and it belongs to the tyrosine kinase receptor super-family. The cMET binds its ligand, HGF, which leads to its dimerization, phosphorylation, and finally, activation (Figure 2). The cMET proto-oncogene, which is encoded by 21 exons, and is localized on chromosome 7 at position q21–q31 [Duh et al. 1997; Liu, 1998]. This activation of the HGF/cMET pathway induces EMT, with the regulation of specific mechanisms, such as E/N-cadherin switching and extracellular matrix degradation processes. These, in turn, promote cell motility, survival, and proliferation. It is now well known that EMT is a physiological cellular mechanism that tumor cells borrow during the metastatic process. In this way, in many human cancers, the HGF/cMET pathway participates in malignant progression and cell scattering, invasion, and protection from apoptosis, and in metastasis formation [Fujita and Sugano, 1997; Samamé Pérez-Vargas et al. 2013; Smith and Bhowmick, 2016].

Several mechanisms are involved in this cMET activation, and many human cancers have been described with point mutations, gene amplification, mRNA overexpression, or ligand autocrine loops [Samamé Pérez-Vargas et al. 2013]. Another way by which the HGF/cMET signaling pathway contributes to tumor progression and metastases is through stimulation of angiogenesis and lymphangiogenesis, thus acting in a synergistic manner with the VEGF/VEGFR pathway (Figure 2). Moreover, the HGF/cMET signaling pathway promotes the growth of endothelial cells and overexpression of pro-angiogenic factors, such as VEGF itself [Ding et al. 2003; Zhang et al. 2003; Matsumura et al. 2013].

The role of angiogenesis in tumor progression and metastases has been well investigated for a long time. The mammalian VEGF family consists of five glycoproteins (i.e. VEGFA, VEGFB, VEGFC, VEGFD, PlGF), among which VEGFA is the best characterized [Dvorak, 2002; Hicklin and Ellis, 2005; Fan et al. 2005]. In-vitro studies have shown that a monoclonal antibody against VEGFR1 blocks tumor cell migration and invasion in CRC cells that overexpress this receptor [Ellis and Hicklin, 2008]. The VEGFR is a tyrosine kinase protein that binds its ligand and activates downstream signaling pathways that promote pro-angiogenic processes (Figure 2) [Kowanetz and Ferrara, 2006]. A role for VEGF was also recently suggested in the escape of tumors from immune responses, through the block of dendritic cell differentiation [Gabrilovich et al. 1996; Ohm et al. 2003; Ellis and Hicklin, 2008]. Indeed, dendritic cells are antigen-presenting cells that have a critical role in presenting cancer antigens to several immune cell types [Steinman et al. 2003; Banchereau and Palucka, 2005; Colonna et al. 2006].

The microenvironment is mainly represented by the immune system and the extracellular matrix, and it also provides tumor heterogeneity [Vermeulen et al. 2010; Herrera et al. 2013; Berdiel-Acer et al. 2014]. In both localized tumors and the whole organism, the interactions between normal and tumor cells contribute to inter-individual genetic heterogeneity. These interactions are the reason for the different responses between individuals with apparently similar tumors, which are due to their different genetic backgrounds.

Colorectal cancer heterogeneity can be observed among antitumor immune surveillance systems, mainly in terms of the major histocompatibility complex, although this heterogeneity is also seen for other important molecules that are dedicated to immunological regulation, such as galectin-3 and programmed death ligand-1 [Bernal et al. 2011]. There is now strong evidence that subcellular populations showing CD3+, CD8+, and CD45RO+ antigens have a central role in antitumor immunological responses. Altered expression of major histocompatibility complex antigens represents an early event in CRC tumorigenesis. This induces the so-called phase of ‘escape’, in which clones of tumor cells can evade the immunological control and initiate metastatic spreading [Garrido et al. 1993, 1997]. Interestingly, the presence of peritumoral inflammatory infiltrate is a characteristic of MSI tumors that correlates with a good prognosis. This confirms that tumor immunological characteristics and responses in CRC are related to the genetic background of the tumor [Sinicrope and Sargent, 2012].

As described above, during EMT, the cancer cells can detach from the primary cancer, migrate into the surrounding tissues, and start the metastatic development that is characterized by intravasation, dissemination, extravasation, and colonization. Thus, once in the vessels, these cancer cells can circulate in the blood. CRCs show great heterogeneity not only in the primary tumor, but also in these circulating cancer cells. Indeed, they show different differentiation grades and epithelial, mesenchymal, or stem cell biomarker expression [Yu et al. 2013]. Investigations into the molecular biomarkers of these circulating cancer cells represent an interesting field in CRC clinical research. The cancer cells that show mesenchymal and stem cell-like phenotypes are more able to undergo extravasation and colonization of distant organs, compared with those with an epithelial phenotype [Tsai and Yang, 2013; Kalikaki et al. 2014].

As described above, CRCs are characterized by this great genotypic and phenotypic heterogeneity that includes the tumor microenvironment and the interactions between healthy and cancer cells. All of these traits confer unique peculiarities to each tumor, and thus each tumor can be considered as a single disease. Here, well conducted molecular and clinical characterization is essential, with the view to implement precision oncology, and thus personalized care.

Prognostic and predictive biomarkers in colorectal cancer

Prognostic and predictive biomarkers for cancer predisposition

Molecular characterization of inherited CRCs allows presymptomatic diagnosis to identify at-risk individuals, and thus improves cancer surveillance. Differential diagnosis can be very difficult between syndromes because of their phenotypic variability. Attenuated FAP, MAP and Lynch syndrome can all be associated with only a few adenomas (e.g. 3–10 polyps); nevertheless, each syndrome has distinct cancer risks, characteristic clinical features, and separate genetic etiology. Differential diagnosis is also essential for suitable patient management of hamartomatous polyposis syndromes, because each of these syndromes has its own characteristic organ-specific manifestations, and each requires a precise surveillance approach. According to the literature, we suggest that next-generation sequencing (NGS) is at present the best and most efficient technique for molecular diagnosis of hereditary colorectal polyposis syndrome, hereditary colorectal cancer, and familial colorectal cancer. Indeed, NGS allows a number of different genes associated with colon cancer to be considered for differential diagnosis, which include APC, MUTYH, PTEN, STK11, BMPR1A, SMAD4, MLH1, MSH2, MSH6, PMS2, CDH1, CHEK2, EPCAM, TP53, and other genes related to these molecular signaling pathways. This approach allows the detection of previously unidentified low-frequency allelic variants, including novel candidate loci.

Prognostic and predictive pathological biomarkers

Personalized patient care is one of the main goals of research in oncology. The aim here is the standardization of therapeutic plans designed and chosen according to the genetic background of each specific neoplasm, in order to increase OS and reduce treatment side effects. Thus, prognostic and predictive molecular biomarkers assume a critical role in characterization of the disease and in determination of the most appropriate therapy. Prognostic biomarkers improve the characterization of the disease, while predictive biomarkers allow the prediction of the likely response to therapy, and also the potential disease outcome.

To date, tumor node metastases (TNM) staging still represents the ‘gold standard’ for tumor classification, and it offers valuable prognostic information and a guide to the necessary therapy decisions. However, the relationship between the TNM stage and patient prognosis is very complex, because each cancer stage is also a heterogeneous group [Gunderson et al. 2010], and indeed, stage II patients who do not receive postoperative adjuvant chemotherapy sometimes show the same disease outcome as stage III patients. Taking into account these considerations, other biomarkers will be useful for combination with the TNM classification for the prediction of responses to therapy and for the planning of personalized care.

Tumor budding is defined as the presence of a single tumor cell or a group of up to five such cells that have detached at the invasive tumor front and have broken off from the main tumor body. These then ‘blend’ into the tumor microenvironment, as the classical peritumoral budding, or remain inside the tumor, known as intratumoral budding [Prall, 2007; Komori et al. 2010; Zlobec et al. 2014; Ueno et al. 2002]. These cells represent tumor cells that have undergone EMT [Dawson and Lugli, 2015; Attramadal et al. 2015], and according to their mesenchymal characteristics, they also show hypoproliferative status and resistance to apoptosis and anoikis [Dawson et al. 2014; Guadamillas et al. 2011]. As expected, high grades of tumor buds are associated with high grades of neo-angiogenesis, lymph-node infiltration, and distant metastases, and therefore, poor patient prognosis [Ishikawa et al. 2008; Nakamura et al. 2005; Wang et al. 2009]. Tumor budding is now considered to be a useful predictive marker of node positivity in patients who are pTN0 [Park et al. 2005; Choi et al. 2009; Tateishi et al. 2010] or have early pT1-2 disease; no difference in survival has been reported between patients who are pTN0 and show high-grade tumor budding and patients who are pTN positive [Lugli et al. 2012; Okuyama et al. 2003]. However, tumor budding has also been suggested as a marker to select patients with adenoma or early T1 disease who will need complete resection after endoscopic resection, including the lymph nodes [Yasuda et al. 2007].

Given the feature of these tumor cells invading the tumor microenvironment, it can often be difficult to distinguish these from peritumoral cells, and thus on this basis, pan cytokeratin detection might represent a useful marker to allow precise cytological characterization [Horcic et al. 2013]. At the molecular level, tumor budding is more represented in APC-mutated tumors than in microsatellite-unstable tumors. Nevertheless, in both of these, tumor budding is a marker for poor patient prognosis [Jass et al. 2003; Zlobec et al. 2012]. However, there is cross-talk between invading tumor cells, as seen by peritumoral budding, and immune defender cells; indeed, the negative effects of tumor budding are mitigated by the presence of immune cells [Zlobec et al. 2011], which suggests that each prognostic biomarker needs to be interpreted in the context of the tumor heterogeneity. Thus, as discussed above, tumor budding represents a strong marker that is useful for precise tumor classification and that needs to be improved in the near future, initially with the definition of an internationally accepted scoring system.

As cancer outcome is the result of several factors that particularly involve the immune system, a novel prognostic index has been developed, which has been defined as the ‘immunoscore’. This is based on the lymphocyte populations at the tumor invasive front. Specifically, the CD3+/CD45RO+, CD3+/CD8+, or CD8+/CD45RO+ cell concentrations can be converted into a score that lies between I–0 and I–4. There is strong correlation between the immunoscore and tumor outcome, such that a higher immunoscore is associated with improved prognosis. As with tumor budding, the immunoscore has also been shown to be more helpful toward better tumor classification than the TNM alone, and thus it has been suggested that the immunoscore should be included as part of the TNM staging system, to generate the TNM-Immunoscore (TNM-I) system [Galon et al. 2014; Mlecnik et al. 2011; Broussard and Disis, 2011; Blanco-Calvo et al. 2015].

Prognostic and predictive molecular biomarkers

As indicated above, MSI status has been classically classified into two or three groups: MSI-H, as tumors that show instability in ⩾30% of the microsatellites analyzed; MSI-L, which defines tumors that show 10–30% as unstable microsatellites; and MSS tumors, with stable microsatellites [Lee and Chan, 2011; Jass 2007]. More rarely, MSI has been divided into positive or negative, with MSI-L and MSS considered as a unique group [Domingo et al. 2013; Simons et al. 2013; Hagland et al. 2013]. A panel of five mononucleotide markers (i.e. Bat-25, Bat-26, NR-21, NR-24, MONO-27) is currently being used by most clinical laboratories to detect MSI.

The role of MSI as a prognostic biomarker has been well demonstrated. A retrospective finding confirmed by the Pan-European Trials in Adjuvant Colon Cancer III (PETACC III) trial showed better relapse-free survival and OS, and decreased risk of metastases, for MSI-H tumors, compared with MSI-L and MSS, independent of the disease grade [Marisa et al. 2013; Sideris and Papagrigoriadis, 2014]. Moreover MSI tumors show particular clinicopathological features, including their location in the proximal colon, and an increased cancer onset age with poor or mucinous differentiation [Church et al. 2012; Domingo et al. 2013; Hagland et al. 2013], lymphocytic infiltration, and an inflammatory reaction [De Smedt et al. 2015]. Thus, the advanced stage of MSI-H tumors resembles the early stage of MSI-L and MSS tumors [Erstad et al. 2015].

The prognostic role of MSI for the prediction of response to therapy is still controversial. Several reports have described the relationship between MSI-H status and worse response to 5-fluorouracil, compared with that obtained for MSI-L and MSS tumors [Webber et al. 2015; Kawakami et al. 2015]. However, there are also several findings that are not in agreement, and a definitive answer is awaited [Erstad et al. 2015].

As indicated above, the Ras proto-oncogene acts downstream of EGFR and has as its major targets the RAF-MEK-ERK and PI3K pathways. A mutation in the KRAS gene represents an early event in colorectal tumorigenesis, which is associated with more than 40% of patients with CRC. Hot-spot mutations located at codons 12, 13, 61, and 143 account for about 97% of the known CRC-associated mutations [http://www.sanger.ac.uk/genetics/CGP/cosmic].

At present, EGFR and VEGFR inhibitors represent the only approved biological drugs for mCRC, which can be combined with traditional chemotherapy [Esin and Yalcin, 2016; Goldstein et al. 2015]. Thus, studies concerning alterations in all of the genes and pathways related to EGFR, such as K-, H-, and NRAS, BRAF, PI3KCA and PTEN, have become of great interest in clinical cancer research. Moreover, only KRAS mutations have been validated and are accepted in clinical practice as predictive biomarkers of response to EGFR inhibitors.

Several studies over the last decade, starting with the CO.17 (KRAS Mutations and Benefit From Cetuximab in Advanced Colorectal Cancer) trial, have provided evidence that not all KRAS mutations interfere equally with patient responses to EGFR monoclonal antibodies. However, tumors carrying a G13D mutation have better PFS and OS when classical chemotherapy is combined with cetuximab, compared with tumors with all of the other KRAS mutations or with wild-type KRAS treated without cetuximab [Karapetis et al. 2008; De Roock et al. 2010a, 2010b; Tejpar et al. 2011, 2012; Osumi et al. 2015]. Then, the two clinically available agents of cetuximab and panitumumab became the standard third-line treatment as monotherapies for chemotherapy-refractory patients with mCRC. Furthermore, mutations in codons 12 and 13 of the KRAS gene confer the same resistance when patients are treated with panitumumab in addition to chemotherapy, as FOLOFOX4 or FOLFIRI [Amado et al. 2008; Peeters et al. 2010; Peeters et al. 2013; Taieb et al. 2016].

Recently, Rowland and colleagues conducted the first systematic review and meta-analysis of eight randomized controlled trials that evaluated the impact of KRAS G13D on survival benefit of anti-EGFR monoclonal antibody therapy. This study highlighted that the efficacy of anti-EGFR monoclonal antibody therapy does not differ significantly between KRAS G13D and other KRAS mutations in mCRC tumors, whereas it confirmed that significant PFS and OS benefits were observed for patients with KRAS wild-type tumors [Rowland et al. 2016].

The ICECREAM (Irinotecan Cetuximab Evaluation and Cetuximab Response Evaluation Among Patients with a G13D Mutation) study was the first prospective study to evaluate the efficacy of cetuximab alone or in combination with irinotecan in patients with quadruple wild-type KRAS, NRAS, BRAF, and PI3KCA mCRC and with tumors with KRAS G13D mutations. The results of this study were recently reported by Segelov and colleagues [Segelov et al. 2016] and confirmed the findings of Rowland and colleagues [Rowland et al. 2016] by demonstrating that in patients with KRAS G13D-mutated chemotherapy-refractory mCRC there was no statistically significant improvement in disease with either cetuximab monotherapy or cetuximab plus irinotecan. No responses were seen with single-agent cetuximab, although a modest response was seen with the combination of cetuximab and irinotecan, which might reflect drug synergy or persistent irinotecan sensitivity [Segelov et al. 2016]. KRAS codon 61 and 146 mutations have been described as having roles in reducing PFS, but not OS, in patients with mCRC treated with cetuximab in addition to classical chemotherapy [Loupakis et al. 2009; De Roock et al. 2010a].

In a retrospective analysis, De Roock and colleagues suggested that patients with G13D-mutated tumors had worse OS than those with wild-type KRAS tumors and those with tumors bearing other KRAS mutations, however this poor prognosis was mitigated by cetuximab treatment. [De Roock et al. 2010b]. The potential prognostic role of KRAS mutations remains unclear. In mCRC, it is difficult to distinguish KRAS effects on patient survival from its effect on patient response to therapy [Garassino et al. 2008]. However, ambiguous data have also been obtained from patients with nonmetastatic CRC [Roth et al. 2010; Hutchins et al. 2011].

BRAF is a serine-threonine kinase that is phosphorylated and activated by KRAS. BRAF mutations, which are mutually exclusive to KRAS mutations, are frequently found in solid tumors, and specifically in sporadic CRC, with a frequency of around 8–10%. The V600E mutation represents a hotspot mutation that when present, locks the cell in a constitutively active status of the Ras/RAF/MEK/ERK pathway, in a manner similar to KRAS mutations [Di Nicolantonio et al. 2008]. BRAF represents a better prognostic than predictive biomarker. Colorectal tumors with BRAF mutations benefit less from anti-EGFR therapy than do wild-type BRAF tumors. However, the resistance of BRAF-mutated tumors to EGFR inhibitors is still controversial [De Stefano and Carlomagno, 2014; Benvenuti et al. 2007; Haraldsdottir and Bekaii-Saab, 2013]. Moreover, BRAF mutation is often associated with sporadic MSI-H CRC and a sessile/ serrated adenoma sequence, and so it can be considered a diagnostic biomarker that can be used to distinguish hereditary from sporadic MSI-H tumors [van Lier et al. 2010; Sideris and Papagrigoriadis, 2014]. The role of BRAF mutation as a prognostic biomarker depends on the combination with other genetic alterations. Indeed, BRAF mutations are associated with poor prognosis in MSS and MSI-L CRC, but their known effect is reduced for MSI-H status [Roth et al. 2010; Pai et al. 2012; Ogino et al. 2012; de Cuba et al. 2016].

Neuroblastoma-RAS (NRAS) is a member of the RAS family that encodes proteins with GTPase activities. The NRAS gene is located on chromosome 1, and its mutations that occur in about 3–5% of sporadic CRC [De Roock et al. 2010b] are mutually exclusive to KRAS and BRAF mutations [Hawkes and Cunningham, 2010]. It is already clear that NRAS mutations also confer resistance to EGFR inhibitors [De Roock et al. 2010b]. Thus, according to all of the issues discussed here, mutational analysis of the RAS genes is recommended to better identify patients who can benefit from therapies with EGFR inhibitors.

PIK3CA encodes the catalytic subunit of the PI3K enzyme [Lai et al. 2015]. Mutations in this gene occur in about 10–20% of patients with sporadic CRC, whereby two hotspot mutations located in exons 9 and 20 account for about 80% of all mutations found to date. In a RAS wild-type tumor, PIK3CA mutations have been suggested to cause resistance to anti-EGFR treatment, although this finding is still under discussion [de la Rochefordiere et al. 2015]. Furthermore, the relevance of PIK3CA mutations as predictive biomarkers appears to be connected to the response to aspirin in patients with CRC. There have been several studies that have suggested that only when a tumor carries a PIK3CA mutation do the patients have a longer OS when aspirin is regularly used [Liao et al. 2012; Ogino et al. 2012].

The phosphatase and tensin homologue (PTEN) gene is frequently altered in human sporadic and inherited cancers, according to several mechanisms, including gene mutations, allelic losses, and promoter hypermethylation. This last epigenetic alteration is a characteristic of MSI-H CRC, as it occurs in about 19% of these tumors [Lin et al. 2015], whereas about 20% of unselected tumors show loss of the PTEN protein, as determined by immunohistochemistry assays [Sartore-Bianchi et al. 2009]. In contrast to RAS and BRAF mutations, PTEN and RAS mutations are not mutually exclusive events. Recent studies have shown that PTEN mutations can have roles in resistance to cetuximab or panitumumab, and that PTEN promoter methylation and mutation status might be useful as a predictive biomarker for responses to anti-EGFR therapies [Therkildsen et al. 2014; Lupini et al. 2015; Sood et al. 2012].

Recently, the evaluation of the ‘Quadruple index’ has been proposed, which associates the probability of EGFR monoclonal antibody responses with changes in one of the following genes: KRAS, BRAF, PIK3CA and PTEN. This study showed that 70% of CRCs show one alteration at most. Furthermore, the presence of only one mutation gives a response probability of about 5%, which then becomes 0% in a tumor carrier with two changes. Tumors without mutations in these genes have about 51% possibility of responding to these therapies [Sartore-Bianchi et al. 2009, Loupakis et al. 2009; Therkildsen et al. 2014].

Angiogenesis represents another critical mechanism in tumor formation and it is thus a target for biological therapy. Although the use of anti-angiogenic drugs is growing in clinical practice for patients with advanced CRC, there are no validated biomarkers that can be used to predict the response to anti-angiogenic drugs [Jubb and Harris, 2010; Jayson et al. 2011]. In the AVF107 pivotal study, the relationships between all of the VEGF proteins and receptors and their responses to bevacizumab were investigated. This suggested that VEGF-D expression might represent a good predictive biomarker for anti-angiogenic response, with high expression of the VEGF-D protein associated with longer PFS and OS than seen for tumors with low expression of VEGF-D. Another trial demonstrated an association between high expression of CD31, VEGF-A, and EGFR-2 and benefit from bevacizumab-including regimens [Luo and Xu, 2014]. Moreover, hypertension is a common side effect of anti-angiogenic therapy, and thus has been suggested as a predictive biomarker to reveal a favorable response; however, this issue deserves further investigation [Tahover et al. 2013]. Although there are no validated predictive biomarkers relating to the use of anti-angiogenic drugs, VEGF has a prognostic value: high VEGF expression is associated to poor prognosis for patients with CRC, low response to preoperative radiotherapy, and relapses. Instead, VEGF-C has been suggested to be a prognostic biomarker in rectal cancer [Yin et al. 2013].

The hepatocyte growth factor receptor is a threonine kinase receptor that has been suggested to be useful in the prediction of responses to cetuximab combined therapies. Some studies have reported recently that overexpression of cMET is associated with shorter median PFS and OS, compared with tumors that express normal or low levels of cMET [Sadanandam et al. 2013]. Another threonine kinase receptor that appears to have a role as a prognostic and predictive biomarker is insulin-like receptor 1 (IGF1R). This is a transmembrane receptor that is overexpressed in about 50–90% of all CRCs [Koda et al. 2004; Takahari et al. 2009], and preclinical studies have shown that its overexpression is correlated with poor prognosis and bad responses to EGFR inhibitors in several tumor types [Ma et al. 2015; Choi et al. 2015].

Prognostic and predictive stemness biomarkers

According to the hierarchical model of cancer onset, CSCs are the only cancer cells that can initiate neoplastic transformation, progression and metastases, and they are probably responsible for resistance to therapy and for relapse [Doherty et al. 2016]. The EMT mechanism also has a role in these processes, as it appears to represent a source of CSCs or to confer motility to epithelial cancer cells through their interconversion into a mesenchymal subtype [Pietilä et al. 2016]. One of the first molecular biomarkers associated with cancer stemness was CD133. It later became clear that, instead, the CD133 protein is expressed both in differentiated and undifferentiated cancer cells, and that there is no exclusive relationship between its expression and cancer cell stemness, self-renewal, and tumorigenicity, nor tumor progression and patient survival [Lugli et al. 2010]. Other biomarkers, such as CD29, CD44, EpCAM, CD166, ALDH1A1 and ALDH1B1, are now considered to be linked to stemness and also to cancer stage, differentiation, invasiveness and metastases [Lugli et al. 2010; Zeuner et al. 2014].

CD44 is a transmembrane glycoprotein that interacts with components of the extracellular matrix, such as hyaluronic acid, and it is present in cells in several isoforms that arise from alternative splicing mechanisms. CD44 is a target of the Wnt/ β-catenin signaling pathway, and it has been considered for a long time as a marker of cancer stemness [Ozawa et al. 2014; Lugli et al. 2010]. Although CD44 knockdown reduces cancer aggressiveness in colon cancer in vitro and in vivo, it is probable that only the CD44v6 alternative isoforms represent a specific colon cancer stem cell marker that is useful for the recognition of cells that can invade the surrounding tissues and initiate the cancer metastasis process [Subramaniam et al. 2010; Kemper et al. 2010; Todaro et al. 2014].

Epithelial cell adhesion molecule (EpCAM) is a transmembrane glycoprotein that has a role in epithelial cell–cell adhesion, and also takes part in cell migration, proliferation and differentiation [Patriarca et al. 2012]. It has been suggested that EpCAM is a diagnostic marker because its overexpression has been associated with advanced cancer stages and shortened median OS for patients with colon and rectal cancer [Kim et al. 2015; Goossens-Beumer et al. 2014].

Activated leukocyte cell adhesion molecule (ALCAM), which is also known as CD166 (cluster of differentiation 166), is implicated in cell adhesion and migration. Multiple alternatively-spliced transcript variants encode different isoforms. ALCAM expression is very variable in CRC tissues; moreover, its overexpression has been associated with shortened OS and poor patient prognosis. The use of ALCAM to identify a CD44+/EpCAM+ subpopulation of cancer cells has also been suggested [Lugli et al. 2010; Tachezy et al. 2012; Langan et al. 2012].

CD24 is a sialoglycoprotein that modulates cell growth and differentiation signals. As such, it has been described as being associated with cell differentiation, lymph node metastases, and shortened OS [Lim and Oh, 2005; Seo et al. 2015].

Aldehyde dehydrogenase is the subsequent enzyme after alcohol dehydrogenase in the major pathway of alcohol metabolism, the overexpression of which has been associated with the stemness of several epithelial cell types, such as breast, stomach, and colon [Deng et al. 2010]. The literature data show that high ALDH1 expression is associated with high tumor grade and bad prognosis of disease [Langan et al. 2012; Chen et al 2015].

The leucine-rich-repeat-containing G-protein-coupled receptor 5 (Lgr5) gene encodes the R-spondin receptor, which is an agonist of the canonical Wnt pathway, and has been described to be responsible for maintenance of stemness in adult intestinal epithelial cells. It has been suggested that Lgr5-positive cancer cells might be adult stem cells that become CSCs in tumors [Barker et al. 2007]. Recently, high LGR5 expression has been reported in CRC-invading cells, which was characterized by high CD44 and low KRT20 expression, combined with a remarkable increase in the stem cell population [Baker et al. 2015]. In CRC, high LGR5 protein expression has also been associated with high risk of relapse, decrease of disease-free survival, and high rate of hepatic and lymph node metastases [Jiang et al. 2016].

As indicated by these details above, a statistically validated profile of CSCs and other prognostic markers would be expected to provide a better and more comprehensive assessment of the patient susceptibility to metastatic disease than any single molecular marker.

Schölch and colleagues developed an orthotopic mouse model of CRC that reproduces the process of CRC dissemination. Circulating tumor cells (CTCs) derived from this model showed stem cell-like characteristics and formed colonies in vitro and tumors in vivo. This study has provided new insight into the biology of CRC-derived CTCs, and it might provide new therapeutic targets in mCRC [Schölch et al. 2016]. Also recently, Ong and colleagues showed that expression of the stem-like group of markers (i.e. CD44, LGR5, SOX2, OCT4) was associated with significantly worse patient prognosis compared with cases with their lower expression. Furthermore, patients with high levels of these stem-like markers showed greater benefit from adjuvant treatments than other patients. The biologically diagnostic/ prognostic relevance of these stem-like markers was also observed in early stage I cancers, which suggests a potential opportunity for identification of aggressive tumors at a very early stage of the disease, as well as the identification of the small subgroup of patients who will derive benefit from adjuvant chemotherapy following surgery [Ong et al. 2015].

Taking these findings together, we suggest that stem cell marker expression should be investigated further, together with EMT cell biomarkers (e.g. vimentin, Twist-1, Snail, Slug, Zeb-1), with a view to the definition of their role in therapeutic decision-making. Furthermore, in agreement with the opinion of the scientific community, we believe that the CSC hypothesis opens interesting perspectives for cancer diagnosis and care, because specific factors associated with CSC and EMT appear to represent the best diagnostic, prognostic and therapeutic targets for patients with CRC.

Circulating biomarkers

Despite progress in the knowledge of the molecular basis of CRC onset and progression, and also the efforts to identify new prognostic/ predictive peripheral biomarkers, the only two blood biomarkers available to monitor patients with CRC are carcinoembryonic antigen (CEA) and carbohydrate antigen 19-9 (CA19-9). CEA is a high-molecular-weight glycoprotein that is specific to embryonic tissue and is also found in colorectal malignancies. At the moment, it represents the antigen that is used in clinical practice as a prognostic marker to follow disease progression after diagnosis, as the CA19-9 antigen is less sensitive and specific for CRC compared with CEA [Locker et al. 2006].

There is evidence that the presence and percentage of CTCs in the blood correlate with disease status and are associated with progression of cancer and metastatic disease. It would thus be useful to monitor advanced disease in cases that do not show overexpression of other circulating cancer biomarkers, such as CEA or CA19-9 [Cohen et al. 2008]. Detection of cancer-specific mutations in CTCs will fulfill a controversial role as a diagnostic/ prognostic factor, because the cancer mutation panel needs to be known at diagnosis. However, methylation analysis of specific genes, such as Vimentin, NGFR, SEPT9, and TMEFF2, and also p16, APC, hMLH1, HLTF, and DAPK, has been under evaluation [Li et al. 2009]. It has been reported that SEPT9 methylation detects CRC with 70% sensitivity and 90% specificity, and a PCR-based assay to identify SEPT9 methylation is now commercially available as a kit (Epi proColon Early Detection Assay, Epigenomics, Germany) [deVos et al. 2009].

To date, CSC panels of multiplexed predictive molecular biomarkers should be used in the new age of liquid biopsies for solid tumors. However, all candidate circulating CRC-specific markers need to be further validated in large, randomized trials to determine their clinical utility in screening for CRC.

Conclusion

As described so far, several alterations at the molecular level favor CRC onset and progression, and the formation of metastases. Thus, molecular characterization of cancer-associated mutations can provide valuable information on disease prognosis and patient response to therapy. The advent of NGS technology has allowed better, more rapid, and also cheaper, molecular characterization of cancers, and clinically-relevant biomarkers can be measured. In summary, germline mutational screening of genes responsible for hereditary and familial CRC represent essential molecular tests for cancer predisposition, while several somatic predictive/ prognostic molecular biomarkers are available today. As described above, KRAS mutations are the only molecular marker actually validated and accepted in clinical practice as a predictive biomarker in response to EGFR inhibition. However, the ‘Quadruple index’, which associates EGFR monoclonal antibody responses with changes in one or more of four genes, as KRAS, BRAF, PIK3CA and PTEN, might provide further information. The MSI status has a role as a prognostic biomarker, although its role as a predictive biomarker for response to therapy remains to be clarified. In our opinion, the TNM-I system coupled with tumor budding might represent the best pathological prognostic biomarkers, which need to be validated for clinical practice. Furthermore, stemness and EMT molecular biomarkers might not only improve cancer screening and early detection, but might also help to define the best postsurgery follow up, not only for patients with advanced and mCRC, but also for patients with early stage disease, such as node-negative patients. Only stem cells have the potential for unlimited proliferation, multi-lineage differentiation, and colonization of new sites, and thus these represent the most likely precursors for invasive and mCRC, and therefore these CSCs require much further attention.

As we have previously reported, a multidisciplinary approach that includes genetics, molecular biology, surgery, and clinical oncology is the only way to achieve the best results for patient outcomes [De Rosa et al. 2015]. On this issue, future studies in CRC need to be concentrated on the standardization of procedures to better determine the genetic profile of each cancer and to define a panel of predictive/ prognostic biomarkers that will allow surgeons and clinical oncologist to choose the best treatment strategy for each, and every single patient.

Footnotes

Acknowledgements

We are grateful to Christopher Berrie for editing the text. Ugo Pace and Paolo Delrio contributed equally to the manuscript.

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Conflict of interest statement

The authors declare that there is no conflict of interest.

References

Aaltonen

Peltomäki

Leach

Sistonen

Pylkkänen

Mecklin

. (1993) Clues to the pathogenesis of familial colorectal cancer. Science. 260: 812–816.

Aaltonen

Salovaara

Kristo

Canzian

Hemminki

Peltomäki

. (1998) Incidence of hereditary nonpolyposis colorectal cancer and the feasibility of molecular screening for the disease. N Eng J Med 21: 1481–1487.

Akiyama

Sato

Yamada

Nagasaki

Tsuchiya

Abe

. (1997) Germ-line mutation of the hMSH6/GTBP gene in an atypical hereditary nonpolyposis colorectal cancer kindred. Cancer Res 57: 3920–3923.

Amado

Wolf

Peeters

van Cutsem

Siena

Freeman

. (2008) Wild-type KRAS is required for panitumumab efficacy in patients with metastatic colorectal cancer. J Clin Oncol 26: 1626–1634.

Atallah

Albert

Larach

(2010) Transanal minimally invasive surgery: a giant leap forward. Surg Endosc 24: 2200–2205.

Attramadal

Kumar

Boysen

Dhakal

Nesland

Bryne

(2015) Tumor budding, EMT and cancer stem cells in T1–2/N0 oral squamous cell carcinomas. Anticancer Res 35: 6111–6120.

Karapetis

O’Callaghan

Moore

Zalcberg

. (2009) Health-related quality of life in patients with advanced colorectal cancer treated with cetuximab: overall and KRAS-specific results of the NCIC CTG and AGITG CO.17 Trial. J Clin Oncol 27: 1822–1828.

Baek

Carmichael

Pigazzi

(2013) Robotic surgery: colon and rectum. Cancer J 19: 140–146.

Baker

Graham

Elia

Wright

Rodriguez-Justo

(2015) Characterization of LGR5 stem cells in colorectal adenomas and carcinomas. Sci Rep 5: 8654.

10.

Banchereau

Palucka

(2005) Dendritic cells as therapeutic vaccines against cancer. Nature Rev Immunol 5: 296–306.

11.

Barker

van Es

Kuipers

Kujala

van den Born

Cozijnsen

. (2007) Identification of stem cells in small intestine and colon by marker gene Lgr5. Nature 449: 1003–1007.

12.

Baron

Cole

Sandler

Haile

Ahnen

Bresalier

. (2003) A randomized trial of aspirin to prevent colorectal adenomas. N Engl J Med 348: 891–899.

13.

Batlle

Sancho

Francí

Domínguez

Monfar

Baulida

. (2000) The transcription factor Snail is a repressor of E-cadherin gene expression in epithelial tumour cells. Nat Cell Biol 2: 84–89.

14.

Bazan

Migliavacca

Zanna

Tubiolo

Grassi

Latteri

. (2002) Specific codon 13 K-RAS mutations are predictive of clinical outcome in colorectal cancer patients, whereas codon 12 K-RAS mutations are associated with mucinous histotype. Ann Oncol 13: 1438–1446.

15.

Benamouzig

Deyra

Martin

Girard

Jullian

Piednoir

. (2003) Daily soluble aspirin and prevention of colorectal adenoma recurrence: one-year results of the APACC trial. Gastroenterology 125: 328–336.

16.

Benvenuti

Sartore-Bianchi

Di Nicolantonio

Zanon

Moroni

Veronese

. (2007) Oncogenic activation of the RAS/RAF signaling pathway impairs the response of metastatic colorectal cancers to anti-epidermal growth factor receptor antibody therapies. Cancer Res 67: 2643–2648.

17.

Berdiel-Acer

Sanz-Pamplona

Calon

Cuadras

Berenguer

Sanjuan

. (2014) Differences between CAFs and their paired NCF from adjacent colonic mucosa reveal functional heterogeneity of CAFs, providing prognostic information. Mol Oncol 8: 1290–1305.

18.

Bernal

Concha

Sáenz-López

Rodríguez

Cabrera

Garrido

. (2011) Leukocyte infiltrate in gastrointestinal adenocarcinomas is strongly associated with tumor microsatellite instability but not with tumor immunogenicity. Cancer Immunol Immunother 60: 869–888.

19.

Birchmeir

Birchmeier

Gherardi

Vande Woude

(2003) Met, metastasis, motility and more. Nat Rev Mol Cell Biol 4: 915–925.

20.

Blanco-Calvo

Concha

Á.

Figueroa

Garrido

Valladares-Ayerbes

(2015) Colorectal cancer classification and cell heterogeneity: a systems oncology approach. Int J Mol Sci 16: 13610–13632.

21.

Boland

Goel

(2010) Microsatellite instability in colorectal cancer (review). Gastroenterology 138: 2073–2087.

22.

Bos

(1989) Ras oncogenes in human cancer: a review. Cancer Res 49: 4682–4689.

23.

Bosetti

Gallus

La Vecchia

(2006) Aspirin and cancer risk: an updated quantitative review to 2005. Cancer Causes Control 17: 871–888.

24.

Bosetti

Gallus

La Vecchia

(2009) Aspirin and cancer risk: a summary review to 2007. Recent Results Cancer Res 181: 231–251.

25.

Bronner

Baker

Morrison

Warren

Smith

Lescoe

. (1994) Mutation in the DNA mismatch repair gene homologue hMLH1 is associated with hereditary nonpolyposis colon cancer. Nature 368: 258–261

26.

Broussard

Disis

(2011) TNM staging in colorectal cancer: T is for T Cell and M is for memory. J Clin Oncol 29: 601–603

27.

Cano

Pérez-Moreno

Rodrigo

Locascio

Blanco

del Barrio

. (2000) The transcription factor Snail controls epithelial-mesenchymal transitions by repressing E-cadherin expression. Nat Cell Biol 2: 76–83.

28.

Cantley

(2002) The phosphoinositide 3-kinase pathway. Science 296: 1655–1657.

29.

Capper

Voigt

Bozukova

Ahadova

Kickingereder

von Deimling

. (2013) BRAF V600E-specific immunohistochemistry or the exclusion of Lynch syndrome in MSI-H colorectal cancer. Int J Cancer 133: 1624–1630.

30.

Castellano

Downward

(2011) Role of RAS in the regulation of PI 3-kinase. Curr Top Microbiol Immunol 346: 143–169.

31.

Chan

Giovannucci

Meyerhardt

Schernhammer

Curhan

Fuchs

(2005) Long-term use of aspirin and nonsteroidal anti-inflammatory drugs and risk of colorectal cancer. JAMA 294: 914–923.

32.

Chan

Giovannucci

Schernhammer

Colditz

Hunter

Willett

. (2004) A prospective study of aspirin use and the risk for colorectal adenoma. Ann Intern Med 140: 157–166.

33.

Chen

Xia

Jiang

Chang

Yuan

. (2015) Prognostic value of cancer stem cell marker ALDH1 expression in colorectal cancer: a systematic review and meta-analysis. PLoS ONE 10: e0145164.

34.

Chiarugi

Giannoni

(2008) Anoikis: a necessary death program for anchorage-dependent cells. Biochem Pharmacol 76: 1352–1364.

35.

Chin

Tam

Pomerantz

Wong

Holash

Bardeesy

. (1999) Essential role for oncogenic RAS in tumour maintenance. Nature 400: 468–472.

36.

Choi

Sohn

Chang

Lim

Choi

Jeong

(2009) Indications for subsequent surgery after endoscopic resection of submucosally-invasive colorectal carcinomas: a prospective cohort study. Dis Colon Rectum 52: 438–445.

37.

Choi

Kang

Nam

Lee

Kim

Ryu

. (2015) Bidirectional signaling between TM4SF5 and IGF1R promotes resistance to EGFR kinase inhibitors. Lung Cancer 90: 22–31.

38.

Church

Midgley

Kerr

(2012) Biomarkers in early-stage colorectal cancer: Ready for prime time? Dig Dis 30: 27–33.

39.

Ciombor

Goldberg

(2015) Recent therapeutic advances in the treatment of colorectal cancer. Annu Rev Med 66: 83–95.

40.

Cohen

Punt

Iannotti

Saidman

Sabbath

Gabrail

. (2008) Relationship of circulating tumor cells to tumor response, progression-free survival, and overall survival in patients with metastatic colorectal cancer. J Clin Oncol 26: 3213–3221.

41.

Cohen

Gourgioti

Wirtz

Raptou

. (2016) Evaluation of CpG island methylator phenotype as a biomarker in colorectal cancer treated with adjuvant oxaliplatin. Clin Colorectal Cancer 15: 164–169.

42.

Colonna

Pulendran

Iwasaki

(2006) Dendritic cells at the host-pathogen interface. Nature Immunol 7: 117–120.

43.

Colussi

Brandi

Bazzoli

Ricciardiello

(2013) Molecular pathways involved in colorectal cancer: implications for disease behavior and prevention. Int J Mol Sci 14: 16365–16385.

44.

Cook

Lee

Gaziano

Gordon

Ridker

Manson

. (2005) Low-dose aspirin in the primary prevention of cancer: the Women’s Health Study: a randomized controlled trial. JAMA 294: 47–55.

45.

Cooper

Park

Blair

Tainsky

Huebner

Croce

. (1984) Molecular cloning of a new transforming from a chemically transformed human cell line. Nature 311: 29–33.

46.

Corso

Pascale

Flauti

Ferrara

Marrelli

Roviello

(2013) Oncogenic mutations and microsatellite instability phenotype predict specific anatomical subsite in colorectal cancer patients. Eur J Hum Genet 21: 1383–1388.

47.

Cuilliere-Dartigues

El-Bchiri

Krimi

Buhard

Fontanges

Fléjou

. (2006) TCF-4 isoforms absent in TCF-4 mutated MSI-H colorectal cancer cells colocalize with nuclear CtBP and repress TCF-4-mediated transcription. Oncogene 25: 4441–4448.

48.

Cuzick

Otto

Baron

Brown

Burn

Greenwald

. (2009) Aspirin and non-steroidal anti-inflammatory drugs for cancer prevention: an international consensus statement. Lancet Oncol 10: 501–507.

49.

Dallas

Xia

Fan

Gray

Gaur

van Buren

. (2009) Chemoresistant colorectal cancer cells, the cancer stem cell phenotype, and increased sensitivity to insulin-like growth factor-I receptor inhibition. Cancer Res 69: 1951–1957.

50.

Dawson

Koelzer

Karamitopoulou

Economou

Hammer

Muller

. (2014) The apoptotic and proliferation rate of tumour budding cells in colorectal cancer outlines a heterogeneous population of cells with various impacts on clinical outcome. Histopathology 64: 577–584.

51.

Dawson

Lugli

(2015) Molecular and pathogenetic aspects of tumor budding in colorectal cancer. Front Med (Lausanne) 10: 11.

52.

De Cuba

Snaebjornsson

Heideman

van Grieken

Bosch

Fijneman

. (2016) Prognostic value of BRAF and KRAS mutation status in stage II and III microsatellite instable colon cancers. Int J Cancer 138: 1139–1145.

53.

De la Rochefordiere

Kamal

Floquet

Thomas

Petrow

Petit

. (2015) PIK3CA pathway mutations predictive of poor response following standard radiochemotherapy ± cetuximab in cervical cancer patients. Clin Cancer Res 21: 2530–2537.

54.

Deng

Yang

Lassus

Liang

Kaur

Q.laga

. (2010) Distinct expression levels and patterns of stem cell marker, aldehyde dehydrogenase isoform 1 (ALDH1), in human epithelial cancers. PloS ONE 5: e10277.

55.

De Roock

Claes

Bernasconi

De Schutter

Biesmans

Fountzilas

. (2010a) Effects of KRAS, BRAF, NRAS, and PIK3CA mutations on the efficacy of cetuximab plus chemotherapy in chemotherapy-refractory metastatic colorectal cancer: a retrospective consortium analysis. Lancet Oncol 11: 753–762.

56.

De Roock

Jonker

Di Nicolantonio

Sartore-Bianchi

Siena

. (2010b) Association of KRAS p.G13D mutation with outcome in patients with chemotherapy-refractory metastatic colorectal cancer treated with cetuximab. JAMA 304: 1812–1820.

57.

De Rosa

Pace

Rega

Costabile

Duraturo

Izzo

. (2015) Genetics, diagnosis and management of colorectal cancer (Review). Oncol Rep 34: 1087–1096.

58.

De Smedt

Lemahieu

Palmans

Govaere

Tousseyn

van Cutsem

. (2015) Microsatellite instable vs stable colon carcinomas: analysis of tumour heterogeneity, inflammation and angiogenesis. Br J Cancer 113: 500–509.

59.

De Stefano

Carlomagno

(2014) Beyond KRAS: Predictive factors of the efficacy of anti-EGFR monoclonal antibodies in the treatment of metastatic colorectal cancer. World J Gastroenterol 20:9732–9743.

60.

De Vos

Tetzner

Model

Weiss

Schuster

Distler

. (2009) Circulating methylated SEPT9 DNA in plasma is a biomarker for colorectal cancer. Clin Chem 55: 1337–1346.

61.

Ding

Merkulova-Rainon

Han

Tobelem

(2003) HGF receptor up-regulation contributes to the angiogenic phenotype of human endothelial cells and promotes angiogenesis in vitro. Blood 101: 4816–4822.

62.

Di Nicolantonio

Martini

Molinari

Sartore-Bianchi

Arena

Saletti

. (2008) Wild-type BRAF is required for response to panitumumab or cetuximab in metastatic colorectal cancer. J Clin Oncol 26: 5705–5712.

63.

Doherty

Smigiel

Junk

Jackson

(2016) Cancer stem cell plasticity drives therapeutic resistance. Cancers (Basel) 8: E8.

64.

Domingo

Ramamoorthy

Oukrif

Rosmarin

Presz

Wang

. (2013) Use of multivariate analysis to suggest a new molecular classification of colorectal cancer. J Pathol 229: 441–448.

65.

Downward

(2003) Targeting RAS signalling pathways in cancer therapy. Nat Rev Cancer 3: 11–22.

66.

Dubé

Rostom

Lewin

Tsertsvadze

Barrowman

Code

. (2007) The use of aspirin for primary prevention of colorectal cancer: a systematic review prepared for the US Preventive Services Task Force. Ann Intern Med 146: 365–375.

67.

Duh

Scherer

Tsui

Lerman

Zbar

Schmidt

(1997) Gene structure of the human MET proto-oncogene. Oncogene 15: 1583–1586.

68.

Duraturo

Liccardo

Cavallo

De Rosa

Grosso

Izzo

(2011) Association of low-risk MSH3 and MSH2 variant alleles with Lynch syndrome: Probability of synergistic effects. Int J Cancer 129: 1643–1650.

69.

Duval

Hamelin

(2002) Mutations at coding repeat sequences in mismatch repair-deficient human cancers: Toward a new concept of target genes for instability. Cancer Res 62: 2447–2454.

70.

Dvorak

(2002) Vascular permeability factor/vascular endothelial growth factor: a critical cytokine in tumor angiogenesis and a potential target for diagnosis and therapy. J Clin Oncol 20: 4368–4380.

71.

Eaden

Abrams

Mayberry

(2001) The risk of colorectal cancer in ulcerative colitis: a meta-analysis. Gut 48: 526–535.

72.

Ellis

Hicklin

(2008) VEGF-targeted therapy: mechanisms of anti-tumour activity. Nat Rev Cancer 8: 579–591.

73.

Eng

(2003) PTEN: one gene, many syndromes. Hum Mutat 22: 183–198.

74.

Engelman

Luo

Cantley

(2006) The evolution of phosphatidylinositol 3-kinases as regulators of growth and metabolism. Nat Rev Genet 7: 606–619.

75.

Erstad

Tumusiime

Cusack

(2015) Prognostic and predictive biomarkers in colorectal cancer: implications for the clinical surgeon. Ann Surg Oncol 22: 3433–3450.

76.

Esin

Yalcin

(2016) Maintenance strategy in metastatic colorectal cancer: A systematic review. Cancer Treat Rev 42: 82–90.

77.

Ewing

Hurley

Josephides

Millar

(2014) The molecular genetics of colorectal cancer. Frontline Gastroenterol 5: 26–30.

78.

Fan

Samuel

Evans

Xia

Zhou

. (2012) Overexpression of SNAIL induces epithelial-mesenchymal transition and a cancer stem cell-like phenotype in human colorectal cancer cells. Cancer Med 1: 5–16.

79.

Fan

Wey

McCarty

Belcheva

Liu

Bauer

. (2005) Expression and function of vascular endothelial growth factor receptor-1 on human colorectal cancer cells. Oncogene 24: 2647–2653.

80.

Fanali

Lucchetti

Farina

Corbi

Cufino

Cittadini

. (2014) Cancer stem cells in colorectal cancer from pathogenesis to therapy: controversies and perspectives. World J Gastroenterol 20: 923–942.

81.

Fang

Hutchinson

Deng

Green

(2016) Common BRAF(V600E)-directed pathway mediates widespread epigenetic silencing in colorectal cancer and melanoma. Proc Natl Acad Sci USA 113: 1250–1255.

82.

Ferlay

Shin

Bray

Forman

Mathers

Parkin . (2010) Estimates of worldwide burden of cancer in 2008: GLOBOCAN 2008. Int J Cancer 127: 2893–2917.

83.

Ferlay

Soerjomataram

Dikshit

Eser

Mathers

Rebelo

. (2015) Cancer incidence and mortality worldwide: sources, methods and major patterns in GLOBOCAN 2012. Int J Cancer 136: e359–e386.

84.

Fishel

Lescoe

Rao

Copeland

Jenkins

Garber

. (1993) The human mutator gene homolog MSH2 and its association with hereditary nonpolyposis colon cancer. Cell 75: 1027–1038.

85.

Fruman

Meyers

Cantley

(1998) Phosphoinositide kinases. Annu Rev Biochem 67: 481–507.

86.

Fujita

Sugano

(1997) Expression of c-Met proto-oncogene in primary colorectal cancer and liver metastases. Jpn J Clin Oncol 27: 378–383.

87.

Gabrilovich

Chen

Girgis

Cunningham

Meny

Nadaf

. (1996) Production of vascular endothelial growth factor by human tumors inhibits the functional maturation of dendritic cells. Nature Med 2: 1096–1103.

88.

Galon

Mlecnik

Bindea

Angell

Berger

Lagorce

. (2014) Towards the introduction of the ‘Immunoscore’ in the classification of malignant tumours. J Pathol 232: 199–209.

89.

Gangemi

Paleari

Orengo

Cesario

Chessa

Ferrini

. (2009) Cancer stem cells: a new paradigm for understanding tumor growth and progression and drug resistance. Curr Med Chem 16: 1688–1703.

90.

Gann

Manson

Glynn

Buring

Hennekens

(1993) Low-dose aspirin and incidence of colorectal tumors in a randomized trial. J Natl Cancer Inst 85: 1220–1224.

91.

Garassino

Farina

Rossi

Martelli

Torri

(2008) Should KRAS mutations be considered an independent prognostic factor in patients with advanced colorectal cancer treated with cetuximab? J Clin Oncol 26: 2600.

92.

Garrido

Cabrera

Concha

Glew

Ruiz-Cabello

Stern

(1993) Natural history of HLA expression during tumour development. Immunol Today 14: 491–499.

93.

Garrido

Ruiz-Cabello

Cabrera

Pérez-Villar

López-Botet

Duggan-Keen

. (1997) Implications for immunosurveillance of altered HLA class I phenotypes in human tumours. Immunol Today 18: 89–95.

94.

Goel

Nagasaka

Arnold

Inoue

Hamilton

Niedzwiecki

. (2007) The CpG island methylator phenotype and chromosomal instability are inversely correlated in sporadic colorectal cancer. Gastroenterology 132: 127–138.

95.

Goldstein

Zeichner

Bartnik

Neustadter

Flowers

(2015) Metastatic colorectal cancer: a systematic review of the value of current therapies. Clin Colorectal Cancer 15: 1–6.

96.

Goossens-Beumer

Zeestraten

Benard

Christen

Reimers

Keijzer

. (2014) Clinical prognostic value of combined analysis of Aldh1, Survivin, and EpCAM expression in colorectal cancer. Br J Cancer 110: 2935–2944.

97.

Greaves

Maley

(2012) Clonal evolution in cancer. Nature 481: 306–313.

98.

Groden

Thliveris

Samowitz

Carlson

Gelbert

Albertsen

. (1991) Identification and characterization of the familial adenomatous polyposis coli gene. Cell 66: 589–600.

99.

Guadamillas

Cereo

Del Pozo

(2011) Overcoming anoikis – pathways to anchorage-independent growth in cancer. J Cell Sci 124: 3189–3197.

100.

Guerrieri

Gesuita

Ghiselli

Lezoche

Budassi

Baldarelli

(2014) Treatment of rectal cancer by transanal endoscopic microsurgery: experience with 425 patients. World J Gastroenterol 20: 9556–9563.

101.

Gunderson

Jessup

Sargent

Greene

Stewart

(2010) Revised TN categorization for colon cancer based on national survival outcomes data. J Clin Oncol 28: 264–271.

102.

Hagland

Berg

Jolma

Carlsen

Søreide

(2013) Molecular pathways and cellular metabolism in colorectal cancer. Dig Surg 30: 12–25.

103.

Hallberg

Rayter

Downward

(1994) Interaction of RAS and RAF in intact mammalian cells upon extracellular stimulation. J Biol Chem 269: 3913–3916.

104.

Haraldsdottir

Bekaii-Saab

(2013) Integrating anti-EGFR therapies in metastatic colorectal cancer. J Gastrointest Oncol 4: 285–298.

105.

Hawkes

Cunningham

(2010) Relationship between colorectal cancer biomarkers and response to epidermal growth factor receptor monoclonal antibodies. J Clin Oncol 28: e529–e531.

106.

Herrera

Islam

Herrera

Martín

García

Silva

. (2013) Functional heterogeneity of cancer-associated fibroblasts from human colon tumors shows specific prognostic gene expression signature. Clin Cancer Res 19: 5914–5926.

107.

Hicklin

Ellis

(2005) Role of the vascular endothelial growth factor pathway in tumor growth and angiogenesis. J Clin Oncol 23: 1011–1027.

108.

Hofheinz

Wenz

Post

Matzdorff

Laechelt

Hartmann

. (2012) Chemoradiotherapy with capecitabine versus fluorouracil for locally advanced rectal cancer: a randomised, multicentre, non-inferiority, phase 3 trial. Lancet Oncol 13: 579–588.

109.

Hohenberger

Weber

Matzel

Papadopoulos

Merkel

(2009) Standardized surgery for colonic cancer: complete mesocolic excision and central ligation - technical notes and outcome. Colorectal Dis 11: 354–364, discussion 364–365.

110.

Horcic

Koelzer

Karamitopoulou

Terracciano

Puppa

Zlobec

. (2013) Tumor budding score based on 10 high-power fields is a promising basis for a standardized prognostic scoring system in stage II colorectal cancer. Hum Pathol 44: 697–705.

111.

Hurwitz

Fehrenbacher

Novotny

Cartwright

Hainsworth

Heim

. (2004) Bevacizumab plus irinotecan, fluorouracil, and leucovorin for metastatic colorectal cancer. N Engl J Med 350: 2335–2342.

112.

Hutchins

Southward

Handley

Magill

Beaumont

Stahlschmidt

. (2011) Value of mismatch repair, KRAS, and BRAF mutations in predicting recurrence and benefits from chemotherapy in colorectal cancer. J Clin Oncol 29: 1261–1270.

113.

Ionov

Peinado

Malkhosyan

Shibata

Perucho

(1993) Ubiquitous somatic mutations in simple repeated sequences reveal a new mechanism for colonic carcinogenesis. Nature 363: 558–561.

114.

Ishikawa

Akishima-Fukasawa

Ito

Akasaka

Yokoo

Ishii

. (2008) Group for cancer biological, histopathologic determinants of regional lymph node metastasis in early colorectal cancer. Cancer 112: 924–933.

115.

Jasperson

Tuohy

Neklason

Burt

(2010) Hereditary and familial colon cancer. Gastroenterology 138: 2044–2058.

116.

Jass

(2007) Classification of colorectal cancer based on correlation of clinical, morphological and molecular features. Histopathology 50: 113–130.

117.

Jass

Barker

Fraser

Walsh

Whitehall

Gabrielli

. (2003) APC mutation and tumour budding in colorectal cancer. J Clin Pathol 56: 69–73.

118.

Jayson

de Haas

Delmar

Miles

Shah

van Cutsem

. (2011) Evaluation of plasma VEGFA as a potential predictive pan-tumor biomarker for bevacizumab. Proceedings of 2011 European Multidisciplinary Cancer Congress; 23–27 Sep 2011; Stockholm, Sweden. Eur J Cancer 47(Suppl.1): S96.

119.

Jiang

Zhang

Jiang

Chen

(2016) Lgr5 expression is a valuable prognostic factor for colorectal cancer: evidence from a meta-analysis. BMC Cancer 15: 948.

120.

Jubb

Harris

(2010) Biomarkers to predict the clinical efficacy of bevacizumab in cancer. Lancet Oncol 11: 1172–1183.

121.

Kalikaki

Politaki

Souglakos

Apostolaki

Papadimitraki

Georgoulia

. (2014) KRAS genotypic changes of circulating tumor cells during treatment of patients with metastatic colorectal cancer. PLoS ONE 9: e104902.

122.

Karapetis

Khambata-Ford

Jonker

O’Callaghan

Tebbutt

. (2008) K-RAS mutations and benefit from cetuximab in advanced colorectal cancer. N Engl J Med 359: 1757–1765.

123.

Karakas

Bachman

Park

(2006) Mutation of the PIK3CA oncogene in human cancers. Br J Cancer 94: 455–459.

124.

Karsa

Lignini

Patnick

Lambert

Sauvaget

(2010) The dimensions of the CRC problem. Best Pract Res Clin Gastroenterol 24: 381–396.

125.

Kawakami

Zaanan

Sinicrope

(2015) Implications of mismatch repair-deficient status on management of early stage colorectal cancer. J Gastrointest Oncol 6: 676–684.

126.

Kemper

Grandela

Medema

(2010) Molecular identification and targeting of colorectal cancer stem cells. Oncotarget 1: 387–395.

127.

Kim

Bae

Song

Cho

Lee

Kang

(2015) Clinicopathologic, molecular, and prognostic implications of the loss of EPCAM expression in colorectal carcinoma. Oncotarget 7: 13372–13387.

128.

Kinzler

Nilbert

Vogelstein

Bryan

Levy

. (1991) Identification of FAP locus genes from chromosome 5q21. Science 253: 661–665.

129.

Koda

Reszec

Sulkowska

Kanczuga-Koda

Sulkowski

(2004) Expression of the insulin-like growth factor-I receptor and proapoptotic BAX and BAK proteins in human colorectal cancer. Ann NY Acad Sci 1030: 377–383.

130.

Komori

Hirai

Kanemitsu

Shimizu

Sano

Ito

. (2010) Is ‘depth of submucosal invasion 4 or 1/4 1,000 microm’ an important predictive factor for lymph node metastases in early invasive colorectal cancer (pT1)? Hepatogastroenterology 57: 1123–1127.

131.

Kosinski

Chan

Zhang

Tsui

. (2007) Gene expression patterns of human colon tops and basal crypts and BMP antagonists as intestinal stem cell niche factors. Proc Natl Acad Sci USA 104: 15418–15423.

132.

Kowanetz

Ferrara

(2006) Vascular endothelial growth factor signaling pathways: therapeutic perspective. Clin Cancer Res 12: 5018–5022.

133.

Lai

Killingsworth

Lee

(2015) Gene of the month: PIK3CA. J Clin Pathol 68: 253–257.

134.

Langan

Mullinax

Ray

Raiji

Schaub

Xin

. (2012) A pilot study assessing the potential role of non-CD133 colorectal cancer stem cells as biomarkers. J Cancer 3: 231–240.

135.

Lee

Chan

(2011) Molecular prognostic and predictive markers in colorectal cancer: current status. Curr Colorectal Cancer Rep 7: 136–144.

136.

Chen

Papadopoulos

Goodman

Bjerregaard

Laurberg

. (2009) Sensitive digital quantification of DNA methylation in clinical samples. Nat Biotechnol 27: 858–863.

137.

Pei

Zheng

(2014) Transitions between epithelial and mesenchymal states during cell fate conversions. Protein Cell 5: 580–591.

138.

Liao

Lochhead

Nishihara

Morikawa

Kuchiba

Yamauchi

. (2012) Aspirin use, tumor PIK3CA mutation, and colorectal-cancer survival. N Engl J Med 367: 1596–1606.

139.

Lim

Counter

(2005) Reduction in the requirement of oncogenic RAS signaling to activation of PI3K/AKT pathway during tumor maintenance. Cancer Cell 8: 381–392.

140.

Lim

(2005) The role of CD24 in various human epithelial neoplasias. Pathol Res Pract 201: 479–486.

141.

Lin

Tseng

Gocke

Reil

Azad

. (2015) Mutational profiling of colorectal cancers with microsatellite instability. Oncotarget 6: 42334–42344.

142.

Liu

(1998) The human hepatocyte growth factor receptor gene: Complete structural organization and promoter characterization. Gene 215: 159–169.

143.

Locker

Hamilton

Harris

Jessup

Kemeny

Macdonald

. (2006) ASCO 2006 update of recommendations for the use of tumor markers in gastrointestinal cancer. J Clin Oncol 24: 5313–5327.

144.

Loupakis

Ruzzo

Cremolini

Vincenti

Salvatore

Santini

. (2009) KRAS codon 61, 146 and BRAF mutations predict resistance to cetuximab plus irinotecan in KRAS codon 12 and 13 wild-type metastatic colorectal cancer. Br J Cancer 101: 715–721.

145.

Loupakis

Pollina

Stasi

Ruzzo

Scarrozzi

Santini

. (2009) PTEN expression and KRAS mutations on primary tumors and metastases in the prediction of benefit from cetuximab plus irinotecan for patients with metastatic colorectal cancer. J Clin Oncol 27: 2622–2629.

146.

Lucci-Cordisco

Risio

Venesio

Genuardi

(2013) The growing complexity of the intestinal polyposis syndromes. Am J Med Genet A 161: 2777–2787.

147.

Lugli

Iezzi

Hostettler

Muraro

Mele

Tornillo

. (2010) Prognostic impact of the expression of putative cancer stem cell markers CD133, CD166, CD44s, EpCAM, and ALDH1 in colorectal cancer. Br J Cancer 103: 382–390.

148.

Lugli

Karamitopoulou

Zlobec

(2012) Tumour budding: a promising parameter in colorectal cancer. Br J Cancer 106: 1713–1717.

149.

Luo

(2014) Predictive and prognostic biomarkers with therapeutic targets in advanced colorectal cancer. World J Gastroenterol 20: 3858–3874.

150.

Lupini

Bassi

Mlcochova

Musa

Russo

Vychytilova-Faltejskova

. (2015) Prediction of response to anti-EGFR antibody-based therapies by multigene sequencing in colorectal cancer patients. BMC Cancer 15: 808.

151.

Tang

Thompson

Mobley

Clark

Sarkaria

. (2015) InsR/IGF1R pathway mediates resistance to EGFR inhibitors in glioblastoma. Clin Cancer Res 22: 1767–1776.

152.

Malumbres

Barbacid

(2003) RAS oncogenes: the first 30 years. Nat Rev Cancer 3(6): 459–465.

153.

Mandal

Becker

Strebhardt

(2015). Stamping out RAF and MEK1/2 to inhibit the ERK1/2 pathway: an emerging threat to anticancer therapy. Oncogene 35: 2547–2461.

154.

Manfredi

(2010) Hereditary hamartomatous polyposis syndromes: understanding the disease risks as children reach adulthood. Gastroenterol Hepatol (NY) 6: 185–196.

155.

Mani

Guo

Liao

Eaton

Ayyanan

Zhou

. (2008) The epithelial-mesenchymal transition generates cells with properties of stem cells. Cell 133: 704–715.

156.

Marisa

de Reyniès

Duval

Selves

Gaub

Vescovo

. (2013) Gene expression classification of colon cancer into molecular sub-types: characterization, validation, and prognostic value. PLoS Med 10: e1001453.

157.

Markowitz

Bertagnolli

(2009) Molecular origins of cancer: Molecular basis of colorectal cancer. N Engl J Med 361: 2449–2460.

158.

Markowitz

Wang

Myeroff

Parsons

Sun

Lutterbaugh

. (1995) Inactivation of the type II TGF-beta receptor in colon cancer cells with microsatellite instability. Science 268: 1336–1338.

159.

Martini

de Santis

Braccini

Gulluni

Hirsch

(2014) PI3K/AKT signaling pathway and cancer: an updated review. Ann Med 46: 372–383.

160.

Matsumura

Kubota

Taiyoh

Fujiwara

Okamoto

Ichikawa

. (2013) HGF regulates VEGF expression via the c-Met receptor downstream pathways, PI3K/Akt, MAPK and STAT3, in CT26 murine cells. Int J Oncol 42: 535–542.

161.

Matz

(2012) Use of a SILS port in transanal endoscopic microsurgery in the setting of a community hospital. J Laparoendosc Adv Surg Tech A 22: 93–96.

162.

Medema

Vermeulen

(2011) Microenvironmental regulation of stem cells in intestinal homeostasis and cancer. Nature 474: 318–326.

163.

Miyaki

Konishi

Tanaka

Kikuchi-Yanoshita

Muraoka

Yasuno

. (1997) Germline mutation of MSH6 as the cause of hereditary nonpolyposis colorectal cancer. Nat Genet 17: 271–272.

164.

Mlecnik

Tosolini

Kirilovsky

Berger

Bindea

Meatchi

. (2011) Histopathologic-based prognostic factors of colorectal cancers are associated with the state of the local immune reaction. J Clin Oncol 29: 610–618.

165.

Moodie

Willumsen

Weber

Wolfman

(1993) Complexes of Ras.GTP with Raf-1 and mitogen-activated protein kinase. Science 260: 1658–1661.

166.

Moore

Lemischka

(2006) Stem cells and their niches. Science 311: 1880–1885.

167.

Moura

Cavaco

Leite

(2015) RAS proto-oncogene in medullary thyroid carcinoma. Endocr Relat Cancer 22: R235–252.

168.

Najdi

Holcombe

Waterman

(2011) WNT signaling and colon carcinogenesis: beyond APC. J Carcinog 10: 5.

169.

Nakamura

Mitomi

Kikuchi

Ohtani

Sato

(2005) Evaluation of the usefulness of tumor budding on the prediction of metastasis to the lung and liver after curative excision of colorectal cancer. Hepatogastroenterology 52: 1432–1435.

170.

Nicolaides

Papadopoulos

Liu

Wei

Carter

Ruben

. (1994) Mutations of two PMS homologues in hereditary nonpolyposis colon cancer. Nature 371: 75–80.

171.

Niessen

Hofstra

Westers

Ligtenberg

Kooi

Jager

. (2009) Germline hypermethylation of MLH1 and EPCAM deletions are a frequent cause of Lynch syndrome. Genes Chromosom Cancer 48: 737–744.

172.

O’Brien

Pollett

Gallinger

Dick

(2007) A human colon cancer cell capable of initiating tumour growth in immunodeficient mice. Nature 445: 106–110.

173.

Ogino

Shima

Meyerhardt

McCleary

Hollis

. (2012) Predictive and prognostic roles of BRAF mutation in stage III colon cancer: results from intergroup trial CALGB 89803. Clin Cancer Res 18: 890–900.

174.

Ohm

Gabrilovich

Sempowski

Kisseleva

Parman

Nadaf

. (2003) VEGF inhibits T-cell development and may contribute to tumor-induced immune suppression. Blood 101: 4878–4886.

175.

Okuyama

Nakamura

Yamaguchi

(2003) Budding is useful to select high-risk patients in stage II well-differentiated or moderately differentiated colon adenocarcinoma. Dis Colon Rectum 46: 1400–1406.

176.

Ong

Chong

McArt

Chan

Tan

Kumar

. (2015) The prognostic value of the stem-like group in colorectal cancer using a panel of immunohistochemistry markers. Oncotarget 6: 12763–12773.

177.

Osumi

Shinozaki

Osako

Kawazoe

Oba

Misaka

. (2015) Cetuximab treatment for metastatic colorectal cancer with KRAS p.G13D mutations improves progression-free survival. Mol Clin Oncol 3: 1053–1057.

178.

Ozawa

Ichikawa

Zheng

Oshima

Miyata

Nakazawa

. (2014) Prognostic significance of CD44 variant 2 upregulation in colorectal cancer. Br J Cancer 111: 365–374.

179.

Pai

Jayachandran

Koong

Chang

Kwok

. (2012) BRAF-mutated, microsatellite-stable adenocarcinoma of the proximal colon: an aggressive adenocarcinoma with poor survival, mucinous differentiation, and adverse morphologic features. Am J Surg Pathol 36: 744–752.

180.

Pancione

Remo

Colantuoni

(2012) Genetic and epigenetic events generate multiple pathways in colorectal cancer progression. Patholog Res Int 2012: 509348.

181.

Park

Choe

Lee

Jung

Park

Kim

(2005) Tumor budding as an indicator of isolated tumor cells in lymph nodes from patients with node-negative colorectal cancer. Dis Colon Rectum 48: 292–302.

182.

Parsons

Wang

Samuels

Bardelli

Cummins

de Long

. (2005) Colorectal cancer: mutations in a signalling pathway. Nature 436: 792.

183.

Patriarca

Macchi

Marschner

Mellstedt

(2012) Epithelial cell adhesion molecule expression (CD326) in cancer: a short review. Cancer Treatment Reviews 38: 68–75.

184.

Peeters

Douillard

van Cutsem

Siena

Zhang

Williams

. (2013) Mutant KRAS codon 12 and 13 alleles in patients with metastatic colorectal cancer: assessment as prognostic and predictive biomarkers of response to panitumumab. J Clin Oncol 31: 759–765.

185.

Peeters

Price

Cervantes

Sobrero

Ducreux

Hotko

. (2010) Randomized phase III study of panitumumab with fluorouracil, leucovorin, and irinotecan (FOLFIRI) compared with FOLFIRI alone as second-line treatment in patients with metastatic colorectal cancer. J Clin Oncol 28: 4706–4713.

186.

Peifer

(2002) Developmental biology: colon construction. Nature 420: 274–275, 277.

187.

Pietilä

Ivaska

Mani

(2016) Whom to blame for metastasis, the epithelial-mesenchymal transition or the tumor microenvironment? Cancer Lett pii: S0304–3835(16)00005–7.

188.

Pino

Chung

(2010) The chromosomal instability pathway in colon cancer. Gastroenterology 138: 2059–2072.

189.

Prall

(2007) Tumour budding in colorectal carcinoma. Histopathology 50: 151–621.

190.

Puglisi

Sgambato

Saulnier

Ravanelli

Barba

Boninsegna

. (2009) Isolation and characterization of CD133+ cell population within human primary and metastatic colon cancer. Eur Rev Med Pharmacol Sci 13(Suppl 1): 55–62.

191.

Ratner

Miller

(2015) A RASopathy gene commonly mutated in cancer: the neurofibromatosis type 1 tumour suppressor. Nat Rev Cancer 15: 290–301.

192.

Rojas

Santos

(2002) Ras genes and human cancer: different implications and different roles. Curr Genom 3: 295–311.

193.

Roth

Tejpar

Delorenzi

Yan

Fiocca

Klingbiel

. (2010) Prognostic role of KRAS and BRAF in stage II and III resected colon cancer: results of the translational study on the PETACC-3, EORTC 40993, SAKK 60–00 trial. J Clin Oncol 28: 466–474.

194.

Rowland

Dias

Wiese

Kichenadasse

McKinnon

Karapetis

. (2016) Meta-analysis comparing the efficacy of anti-EGFR monoclonal antibody therapy between KRAS G13D and other KRAS mutant metastatic colorectal cancer tumours. Eur J Cancer 55: 122–130.

195.

Rustgi

(2007) The genetics of hereditary colon cancer (review). Genes Dev 21: 2525–2538.

196.

Sadanandam

Lyssiotis

Homicsko

Collisson

Gibb

Wullschleger

. (2013) A colorectal cancer classification system that associates cellular phenotype and responses to therapy. Nat Med 19: 619–625.

197.

Sagar

(2011) Colorectal stents for the management of malignant colonic obstructions. Cochrane Database Syst Rev 11: CD007378.

198.

Samamé Pérez-Vargas

Biondini

Maggi

Gariboldi

Gloghini

Inno

. (2013) Role of cMET in the development and progression of colorectal cancer. Int J Mol Sci 14: 18056–18077.

199.

Samowitz

Curtin

Lin

Robertson

Schaffer

Nichols

. (2001) The colon cancer burden of genetically defined hereditary nonpolyposis colon cancer. Gastroenterology 121: 830–838.

200.

Samuels

Wang

Bardelli

Silliman

Ptak

Szabo

. (2004) High frequency of mutations of the PIK3CA gene in human cancers. Science 304: 554.

201.

Sartore-Bianchi

Di Nicolantonio

Nichelatti

Molinari

De Dosso

Saletti

. (2009) Multi-determinants analysis of molecular alterations for predicting clinical benefit to EGFR-targeted monoclonal antibodies in colorectal cancer. PLoS ONE 4: e7287.

202.

Sauer

Liersch

Merkel

Fietkau

Hohenberger

Hess

. (2012) Preoperative versus postoperative chemoradiotherapy for locally advanced rectal cancer: results of the German CAO/ARO/AIO-94 randomized phase III trial after a median follow-up of 11 years. J Clin Oncol 30: 1926–1933.

203.

Schmoll

van Cutsem

Stein

Valentini

Glimelius

Haustermans

. (2012) ESMO Consensus Guidelines for management of patients with colon and rectal cancer. A personalized approach to clinical decision making. Ann Oncol 23: 2479–2516.

204.

Schölch

García

Iwata

Niemietz

Betzler

Nanduri

. (2016) Circulating tumor cells exhibit stem cell characteristics in an orthotopic mouse model of colorectal cancer. Oncotarget doi: 10.18632/oncotarget.8373. [Epub ahead of print].

205.

Segelov

Thavaneswaran

Waring

Desai

Robledo

Gebski

. (2016) Response to cetuximab with or without irinotecan in patients with refractory metastatic colorectal cancer harboring the KRAS G13D mutation: Australasian Gastro-Intestinal Trials Group ICECREAM Study. J Clin Oncol 34: 2258–2264.

206.

Sehgal

Coffey

(2014) Historical development of mesenteric anatomy provides a universally applicable anatomic paradigm for complete/total mesocolic excision. Gastroenterol Rep (Oxf) 2: 245–250.

207.

Seo

Kim

(2015) Prognostic implications of adhesion molecule expression in colorectal cancer. Int J Clin Exp Pathos 8: 4148–4157.

208.

Seow

Yip

Fifis

(2016) Advances in targeted and immunobased therapies for colorectal cancer in the genomic era. Onco Targets Ther 91: 1899–1920.

209.

Sgourakis

Lanitis

Gockel

Kontovounisios

Karaliotas

Tsiftsi

. (2011) Transanal endoscopic microsurgery for T1 and T2 rectal cancers: a meta-analysis and meta-regression analysis of outcomes. Am Surg 77: 761–772.

210.

Siani

Pulica

(2014) Stage I-IIIC right colonic cancer treated with complete mesocolic excision and central vascular ligation: quality of surgical specimen and long term oncologic outcome according to the plane of surgery. Minerva Chir 69: 199–208.

211.

Sideris

Papagrigoriadis

(2014) Molecular biomarkers and classification models in the evaluation of the prognosis of colorectal cancer (review). Anticancer Res 34: 2061–2068.

212.

Simons

Hughes

Smits

Khalid-de Bakker

de Bruïne

Carvalho

. (2013) A novel classification of colorectal tumors based on microsatellite instability, the CpG island methylator phenotype and chromosomal instability: implications for prognosis. Ann Oncol 24: 2048–2056.

213.

Simpson

Anyiwe

Schimmer

(2008) Anoikis resistance and tumor metastasis. Cancer Lett 272: 177–185.

214.

Singh

Settleman

(2010) EMT, cancer stem cells and drug resistance: an emerging axis of evil in the war on cancer. Oncogene 29: 4741–4751.

215.

Sinicrope

Sargent

(2012) Molecular pathways: microsatellite instability in colorectal cancer: prognostic, predictive, and therapeutic implications. Clin Cancer Res 18: 1506–1512.

216.

Smith

Bhowmick

(2016) Role of EMT in Metastasis and Therapy Resistance. J Clin Med 5: E17.

217.

Sood

McClain

Maitra

Basu-Mallick

Seetharam

Kaubisch

. (2012) PTEN gene expression and mutations in the PIK3CA gene as predictors of clinical benefit to anti-epidermal growth factor receptor antibody therapy in patients with KRAS wild-type metastatic colorectal cancer. Clin Colorectal Cancer 11: 143–150.

218.

Steinman

Hawiger

Liu

Bonifaz

Bonnyay

Mahnke

. (2003) Dendritic cell function in vivo during the steady state: a role in peripheral tolerance. Ann NY Aca Sci. 987: 15–25.

219.

Subramaniam

Ramalingam

Houchen

Anant

(2010) Cancer stem cells: a novel paradigm for cancer prevention and treatment. Mini Rev Med Chem 10: 359–371.

220.

Tabernero

van Cutsem

Lakomý

Prausová

Ruff

van Hazel

. (2014) Aflibercept versus placebo in combination with fluorouracil, leucovorin and irinotecan in the treatment of previously treated metastatic colorectal cancer: prespecified subgroup analyses from the VELOUR trial. Eur J Cancer 50: 320–331.

221.

Tachezy

Zander

Gebauer

Marx

Kaifi

Izbicki

. (2012) Activated leukocyte cell adhesion molecule (CD166)-its prognostic power for colorectal cancer patients. J Surg Res 177: e15–e20.

222.

Tahover

Uziely

Salah

Temper

Peretz

Hubert

(2013) Hypertension as a predictive biomarker in bevacizumab treatment for colorectal cancer patients. Med Oncol 30: 327.

223.

Taieb

Zaanan

Le Malicot

Julié

Blons

Mineur

. (2016) Prognostic effect of BRAF and KRAS mutations in patients with stage III colon cancer treated with leucovorin, fluorouracil, and oxaliplatin with or without cetuximab: a post-hoc analysis of the PETACC-8 trial. JAMA Oncol 14: 1–11.

224.

Takahari

Yamada

Okita

Honda

Hirashima

Matsubara

. (2009) Relationships of insulin-like growth factor-1 receptor and epidermal growth factor receptor expression to clinical outcomes in patients with colorectal cancer. Oncology 76: 42–48.

225.

Tateishi

Nakanishi

Taniguchi

Shimoda

Umemura

(2010) Pathological prognostic factors predicting lymph node metastasis in submucosal invasive (T1) colorectal carcinoma. Mod Pathol 23: 1068–1072.

226.

Tejpar

Bokemeyer

Celik

Schlichting

Sartorius

van Cutsem

(2011) Association of KRAS G13D mutations on outcome in patients with metastatic colorectal cancer treated with first-line chemotherapy with or without cetuximab. J Clin Oncol 30: 3570–3577.

227.

Tejpar

Celik

Schlichting

Sartorius

Bokemeyer

van Cutsem

(2012) Association of KRAS G13D tumor mutations with outcome in patients with metastatic colorectal cancer treated with first-line chemotherapy with or without cetuximab. J Clin Oncol 30: 3570–3577.

228.

Testa

Bellacosa

(2001) AKT plays a central role in tumorigenesis. Proc Natl Acad Sci USA 98: 10983–10985.

229.

Therkildsen

Bergmann

Henrichsen-Schnack

Ladelund

Nilbert

(2014) The predictive value of KRAS, NRAS, BRAF, PIK3CA and PTEN for anti-EGFR treatment in metastatic colorectal cancer: a systematic review and meta-analysis. Acta Oncol 53: 852–864.

230.

Thibodeau

Bren

Shaid

(1993) Microsatellite instability in cancer of the proximal colon. Science 260: 816–819.

231.

Thiery

Acloque

Huang

Nieto

(2009) Epithelial mesenchymal transitions in development and disease. Cell 139: 871–890.

232.

Todaro

Gaggianesi

Catalano

Benfante

Iovino

Biffoni

. (2014) CD44v6 is a marker of constitutive a reprogrammed cancer stem cells driving colon cancer metastasis. Cell Stem Cell 14: 342–356.

233.

Tsai

Yang

(2013) Epithelial-mesenchymal plasticity in carcinoma metastasis. Genes Dev 27: 2192–2206.

234.

Tveit

Guren

Glimelius

Pfeiffer

Sorbye

Pyrhonen

. (2012) Phase III trial of cetuximab with continuous or intermittent fluorouracil, leucovorin, and oxaliplatin (Nordic FLOX) versus FLOX alone in first-line treatment of metastatic colorectal cancer: the NORDIC-VII study. J Clin Oncol 30: 1755–1762.

235.

Tyler

Fox

Desai

Perry

Glasgow

(2013) Outcomes and costs associated with robotic colectomy in the minimally invasive era. Dis Colon Rectum 56: 458–466.

236.

Ueno

Murphy

Jass

Mochizuki

Talbot

(2002) Tumour ‘budding’ as an index to estimate the potential of aggressiveness in rectal cancer. Histopathology 40: 127–132.

237.

Van Cutsem

Köhne

Hitre

Zaluski

Chang Chien

Makhson

. (2009) Cetuximab and chemotherapy as initial treatment for metastatic colorectal cancer. N Engl J Med 360: 1408–1417.

238.

Van Cutsem

Tabernero

Lakomy

Prenen

Prausová

Macarulla

. (2012) Addition of aflibercept to fluorouracil, leucovorin, and irinotecan improves survival in a phase III randomized trial in patients with metastatic colorectal cancer previously treated with an oxaliplatin-based regimen. J Clin Oncol 30: 3499–3506.

239.

Van Lier

Wagner

van Leerdam

Biermann

Kuipers

Steyerberg

. (2010) A review on the molecular diagnostics of Lynch syndrome: a central role for the pathology laboratory. J Cel. Mol Med 14: 181–197.

240.

Vasen

Watson

Mecklin

Lynch

(1999) New clinical criteria for hereditary nonpolyposis colorectal cancer (HNPCC, Lynch syndrome) proposed by the International Collaborative group on HNPCC. Gastroenterology 116: 1453–1456.

241.

Vasen

Möslein

Alonso

Aretz

Bernstein

Bertario

. (2008) Guidelines for the clinical management of familial adenomatous polyposis (FAP). Gut 57: 704–713.

242.

Vermeulen

De Sousa Mello

van der Heijden

Cameron

de Jong

Borovski

. (2010) WNT activity defines colon cancer stem cells and is regulated by the microenvironment. Nat Cell Biol 12: 468–476.

243.

Vermeulen

Todaro

de Sousa Mello

Sprick

Kemper

Perez Alea

. (2008) Single-cell cloning of colon cancer stem cells reveals a multi-lineage differentiation capacity. Proc Natl Acad Sci USA. 105: 13427–13432.

244.

Vivanco

Sawyers

(2002) The phosphatidylinositol 3-Kinase AKT pathway in human cancer. Nat Rev Cancer 2: 489–501.

245.

Vogelstein

Fearon

Hamilton

Kern

Preisinger

Leppert

. (1988) Genetic alterations during colorectal-tumor development. N Engl J Med 319: 525–532.

246.

Vojtek

Hollenberg

Cooper

(1993) Mammalian RAS interacts directly with the serine/threonine kinase Raf. Cell 74: 205–214.

247.

Vousden

Prives

(2009) Blinded by the light: the growing complexity of p53. Cell 137: 413–431.

248.

Wang

Kevans

Mulcahy

O’Sullivan

Fennelly

Hyland

. (2009) Tumor budding is a strong and reproducible prognostic marker in T3N0 colorectal cancer. Am J Surg Pathol 33: 134–141.

249.

Warne

Viciana

Downward

(1993) Direct interaction of RAS and the amino-terminal region of Raf-1 in vitro. Nature 364: 352–325.

250.

Webber

Kauffman

O’Connor

Goddard

(2015) Systematic review of the predictive effect of MSI status in colorectal cancer patients undergoing 5FU-based chemotherapy. BMC Cancer 15: 156.

251.

Yasuda

Inomata

Shiromizu

Shiraishi

Higashi

Kitano

(2007) Risk factors for occult lymph node metastasis of colorectal cancer invading the submucosa and indications for endoscopic mucosal resection. Dis Colon Rectum 50: 1370–1376.

252.

Yin

Fan

Sheng

Xia

Xie

(2013) Effect of vascular endothelial growth factor C and collagen triple helix repeat containing 1 expression on prognosis of rectal carcinoma patients. Zhonghua Wei Chang Wai Ke Za Zhi 16: 673–675.

253.

Bardia

Wittner

Stott

Smas

Ting

. (2013) Circulating breast tumor cells exhibit dynamic changes in epithelial and mesenchymal composition. Science 339: 580–584.

254.

Yuan

Cantley

(2008) PI3K pathway alterations in cancer: variations on a theme. Oncogene 27: 5497–5510.

255.

Zeuner

Todaro

Stassi

De Maria

(2014) Colorectal cancer stem cells: from the crypt to the clinic. Cell Stem Cell 15: 692–705.

256.

Zhang

Settleman

Kyriakis

Takeuchi-Suzuki

Elledge

Marshall

. (1993) Normal and oncogenic p21Ras proteins bind to the amino-terminal regulatory domain of c-Raf-1. Nature 364: 308–313.

257.

Zhang

Sum

Volpert

van de Woude

(2003) Hepatocyte growth factor/scatter factor mediates angiogenesis through positive VEGF and negative thrombospondin 1 regulation. Proc Natl Acad Sci USA 100: 12718–12723.

258.

Zlobec

Bihl

Foerster

Rufle

Lugli

(2012) The impact of CpG island methylator phenotype and microsatellite instability on tumour budding in colorectal cancer. Histopathology 61: 777–787.

259.

Zlobec

Hädrich

Dawson

Koelzer

Borner

Mallaev

. (2014) Intratumoural budding (ITB) in preoperative biopsies predicts the presence of lymph node and distant metastases in colon and rectal cancer patients. Br J Cancer 110: 1008–1013.

260.

Zlobec

Minoo

Terracciano

Baker

Lugli

(2011) Characterization of the immunological microenvironment of tumour buds and its impact on prognosis in mismatch repair-proficient and -deficient colorectal cancers. Histopathology 59: 482–495.