Angiogenesis, the recruitment of existing blood vessels and the formation of new capillaries, is vital to tumor growth and establishment. This observation leads researchers to hypothesize that targeting angiogenesis was a viable strategy in cancer therapy [43]. The VEGF family of proteins is comprised of six members, VEGF-A through VEGF-E and placenta growth factor (PIGF)-1 and PIGF-2. VEGF-A, often simply called VEGF, is a potent ligand and regulator of proliferation and angiogenesis in both normal and tumor cells [44]. The VEGF binds to its receptor VEGFR2, a tyrosine kinase, resulting in activation of multiple intracellular signaling pathways and leading to increased endothelial cell proliferation, migration, survival, and ultimately angiogenesis. Not surprisingly, VEGF overexpression has been associated with worse outcomes related to tumor progression and poor overall survival in patients with CRC [45, 46].

Bevacizumab was the first antiangiogenic targeted drug to be approved for the treatment of CRC. Bevacizumab, a recombinant monoclonal antibody against VEGF-A, was heavily anticipated to demonstrate profound antitumoral effect; however, as a single agent the drug showed limited clinical benefit [47]. However, when used in combination with cytotoxic systemic chemotherapies, bevacizumab was shown to increase response rate in patients with mCRC leading to longer PFS and OS [48]. Bevacizumab works synergistically with 5-FU-based chemotherapy to induce apoptosis and inhibit tumor growth presumably by enhancing cytotoxic drug delivery to tumor cells [49]. Consequently, the use of bevacizumab in combination with FOLOX (5-FU, oxaliplatin, leucovorin)/FOLFIRI (5-FU, leucovorin, irinotecan) has been widely employed in the treatment of mCRC, and because of enhanced response rates in the neoadjuvant setting, a significant number of patients with potentially resectable mCRC are later able to undergo curative hepatic metastasectomy. The drug is well tolerated, but potential toxicities can be severe. GI perforation is the most feared and occurs in approximately 1.5% of patients on treatment. Other side effects include hypertension warranting medical therapy, increased risk for stoke and myocardial infarction, and impaired wound healing, though studies have shown that cessation of bevacizumab 4 weeks prior to major surgical intervention poses no increased risk of wound complications [50, 51]. The development of hypertension while on therapy has been associated with a positive treatment response [52, 53].

EGFR is a tyrosine kinase transmembrane receptor within the HER/ErbB family of proteins [54]. EGFR is overexpressed in tumors of epithelial origin and particularly in colon and rectal cancers. EGFR activation occurs when its primary ligands, epiregulin, amphiregulin, EGF, and TGF-α, bind to initiate activation of multiple signaling pathways. In particular, constitutive activation of the Ras/Raf/MAPK and the phosphatidylinositol 3-kinase (PI3K) signaling pathways lead to deregulated cellular proliferation, inhibited apoptosis, and cellular invasion (Fig. 19.5). EGFR activation therefore contributes not only to tumorigenesis , but is believed to be a key contributor to CRC metastatic progression [55].

Two classes of anti-EGFR therapies were developed to target this mechanism: tyrosine kinase inhibitors (TKIs) and monoclonal antibodies to EGFR. TKIs, such as gefitinib and erlotinib, competitively inhibit adenosine triphosphate (ATP) to block EGFR activity and preferentially suppress PI3K signaling in CRC [56]. However, efforts to utilize these agents in combination with standard chemotherapy regimens in CRC resulted in high rates of GI toxicity and were not further pursued.

Cetuximab was the first EGFR monoclonal antibody (mAb) to be FDA approved for treatment of metastatic CRC. These antibodies bind EGFR and block ligand binding, thereby inhibiting ligand-dependent autophosphorylation and inhibiting downstream MAPK and PI3K cell signaling. In their seminal paper, Cunningham et al. reported improved response, time to progression, and overall survival with single-agent cetuximab or when used in conjunction with irinotecan in mCRC patients with refractory disease [57]. Similarly, panitumumab was demonstrated to be well tolerated and improved response rates in patients with refractory mCRC and was FDA approved soon after [58]. Anti-EGFR antibody therapy is generally well tolerated. The most common side effect is an acneiform skin rash seen in 80–90% of patients on treatment [59]. Rashes tend to resolve once therapy is discontinued; however, studies have shown a positive correlation in overall survival and severity of skin rash suggesting that this side effect may serve as a forecaster of treatment response [58, 60].

Since their FDA approval in 2004, monoclonal antibodies to EGFR have been widely employed in the treatment of metastatic colon and rectal cancers. However, only few patients derive benefit to therapy with response rates only approaching 10% in most series and prompting efforts to investigate EGFR expression as a prognostic and predictive biomarker [57, 58, 60]. EGFR mutations while common in other tumors are rare in CRC [61, 62]. Early studies explored the association of increased EGFR gene copy or EGFR expression on improved overall survival (OS) and progression-free survival (PFS) in mCRC patients; however, these have not been demonstrated to be reproducibly associated with outcomes and are no longer used for patient selection [60, 63–65]. The investigation of downstream effectors of EGFR signaling, however, has led to the identification of the RAS proto-oncogene family as critical biomarkers in CRC.

The RAS family of proteins is comprised of three oncogenes (KRAS, NRAS, and HRAS) that code for proteins involved in EGFR-mediated cellular signaling and regulation of proliferation (Fig. 19.5). KRAS mutations occur in 30–40% of colorectal cancers, while NRAS mutations are much less rare with NRAS occurring in 5% of tumors [66–68]. HRAS is not seen in CRC, with mutation rates <0.1%. RAS gene mutations typically occur at codon 12 and less frequently in codons 13, 59, 61, 117, and 146 altering the GTP binding domain dynamics and resulting in constitutive activation of the oncogene. Like CRC, these mutations are observed at high frequency in large adenomas, but less so in smaller adenoma suggesting that they likely contribute more to adenoma growth and CRC carcinogenesis [33, 69, 70]. KRAS and NRAS mutations in colorectal cancers carry both predictive and prognostic significance. Retrospective analysis of several clinical trials of anti-EGFR mAbs revealed that wild-type KRAS was required for anti-EGFR antibody-mediated response in patients with colorectal cancer [71–75]. KRAS/NRAS mutations result in constitutive activation of the Ras-Raf-MAPK pathway, downstream of epidermal growth factor receptor (EGFR), rendering these tumors resistant to anti-EGFR therapies [71, 76–79]. Current guidelines require testing for KRAS and NRAS at the codons above prior to initiating anti-EGFR therapy and established the Ras-Raf-MAPK pathway as a principal target for the development of novel molecular therapeutic agents for the treatment of colorectal cancer [80–82].

While RAS mutation status clearly confers predictive relevance, its prognostic implications require further investigation. While KRAS mutations may be only weakly prognostic in stage II and III colorectal cancer, they may confer poor prognosis in metastatic CRC [83, 84]. Interestingly, exon 146 mutations, though rare, appear to confer better biology than their counterparts with codon 12, 13, or 61 mutations [85].

The BRAF oncogene , another potent modulator of the MAPK pathway, has recently emerged as a prognostic biomarker and novel therapeutic target in CRC. BRAF ^V600E is an activating mutation that accounts for approximately 90% of all BRAF mutations seen in colorectal cancer [79, 86]. As described above, oncogenic BRAF mutations result in tumor formation typically along the serrated pathway by the methylation of CpG islands that cause the silencing of critical tumor suppressor genes and accounting for 10–15% of CRC with discrete clinical characteristics and oncologic outcomes [87–92]. BRAF mutant tumors are more prevalent in women and in patients of advanced age, typically age >70 years [93–96]. Additionally these tumors tend to be proximal in location, MSI-high, of mucinous histology, serrated and poorly differentiated [93–100]. In sporadic colorectal cancers, BRAF mutation is seen in approximately 30–60% of MSI-high tumors but only 5–10% of microsatellite-stable (MSS) tumors [79, 90, 94, 101]. BRAF ^V600E mutations cause hypermethylation of the MLH1 gene promoter, resulting in loss of the tumor suppressor function and impeding DNA mismatch repair [102–105]. Tran et al. [106] defined BRAF mutant colorectal tumors as a distinct subtype when they delineated a unique pattern of metastatic spread. In their study of a cohort of 524 patients with metastatic colorectal cancer, 11% patients were found to harbor a BRAF mutation. In these patients, metastatic spread was more common via peritoneal disease (46% vs 24%) or distant lymph node metastasis (53% vs 38%) when compared to BRAF wild-type tumors and less likely to result in lung metastasis (35% vs 49%). Not surprisingly, BRAF mutation conferred poorer overall survival with a median of 10.4 months versus 34.7 months. These patients were also less likely to undergo metastasectomy with disease present at sites not amenable to resection. These data implicated BRAF mutation as a major driver of right-sided tumor biology and a potent contributor to poor prognosis observed between right-sided and left-sided colon cancers. Interestingly, the same prognostic implications may not hold for the rare patients with BRAF mutations occurring in rectal primaries, perhaps due to the limited epigenetic alterations seen in these tumors. Further studies are needed to explore this.

Patients with this mutation stand to benefit from effective targeted therapies. The BRAF oncogene codes for a serine/threonine kinase that acts downstream of KRAS in the MAPK pathway (Fig. 19.5). BRAF mediates its effect by activating mitogen-activated protein kinase kinase (MAPKK or MEK) initiating uninhibited EGFR-independent cellular proliferation [90, 107]. BRAF and KRAS/NRAS mutations are mutually exclusive in CRC, suggesting that BRAF is the principal effector of KRAS/NRAS in the MAPK pathway with equivalent effects on tumorigenesis [79, 94, 101]. Consequently, BRAF mutant tumors are able to escape the effects of anti-EGFR therapy similar to RAS mutation and are believed to be responsible for 12%–15% of RAS wild-type patients who fail anti-EGFR therapies [90, 94, 101].