Targeted Therapies in Colorectal Cancer: Recent Advances in Biomarkers, Landmark Trials, and Future Perspectives

Simple Summary Colorectal cancer is one of the leading causes of cancer-associated deaths worldwide. From the 1960s to 2002, the standard systemic treatment consisted of cytotoxic chemotherapy. However, in the past 20 years, the development of targeted therapies has revolutionized care for CRC. Dozens of new drugs have been developed specifically targeting actionable and identifiable molecular biomarkers expressed by these tumors. This growing field, precision oncology, has been responsible for a dramatic drop of more than 50% in the CRC mortality rate in the United States. As research into this field exponentially expands, new targets are being uncovered, and new drugs are being developed. This review aims to summarize the recent advances in biomarkers, landmark trials, and future perspectives for colorectal cancer. Abstract In 2022, approximately 600,000 cancer deaths were expected; more than 50,000 of those deaths would be from colorectal cancer (CRC). The CRC mortality rate in the US has decreased in recent decades, with a 51% drop between 1976 and 2014. This drop is attributed, in part, to the tremendous therapeutic improvements, especially after the 2000s, in addition to increased social awareness regarding risk factors and diagnostic improvement. Five-fluorouracil, irinotecan, capecitabine, and later oxaliplatin were the mainstays of mCRC treatment from the 1960s to 2002. Since then, more than a dozen drugs have been approved for the disease, betting on a new chapter in medicine, precision oncology, which uses patient and tumor characteristics to guide the therapeutic choice. Thus, this review will summarize the current literature on targeted therapies, highlighting the molecular biomarkers involved and their pathways.


Introduction
Cancer is the second leading cause of death in the United States of America (USA) today, second only to heart disease [1]. In 2022, approximately 600,000 cancer deaths were expected; more than 50,000 of those deaths would be from colorectal cancer (CRC) [1]. Despite a recent increase in incidence in the younger population, which is not yet subjected to routine CRC screening [2], the CRC mortality rate in the US has decreased in recent decades, with a 51% drop between 1976 and 2014 [3]. This drop is attributed, in part, to the tremendous therapeutic improvements, especially after the 2000s, in addition to increased social awareness regarding risk factors and diagnostic improvement [3]. For patients who develop metastatic CRC (mCRC), approximately 50-60% of all patients with CRC [4], the standard treatment is antineoplastic agents aimed at improving and prolonging the quality of life [5]. 0.71; 95% CI, 0.65-0.77; p < 0.00001) and OS (HR, 0.85; 95% CI, 0.78-0.94; p = 0.0008) [29]. When the analysis was restricted to fluoropyrimidine monotherapy, the addition of bevacizumab led to a significant increase in PFS (HR, 0.57; 95% CI, 0.48-0.66; p < 0.00001) and OS (HR, 0.83; 95% CI, 0.70-0.98; p = 0.03) [29].
Ziv-aflibercept is a recombinant protein composed of extracellular parts of the vascular endothelial growth factor receptor 1 and 2 (VEGFR-1 and VEGFR-2), linked to the constant fragment (Fc) region of human immunoglobulin G 1 (IgG1) [30]. It acts as a VEGF receptor decoy, binding to VEGF factors, thus preventing their efficient action [31]. Its role is still less defined than that of bevacizumab, especially when it comes to first-line treatment, but apparently it presents itself as a good tool for patients who, despite treatment with bevacizumab, continue with disease progression [32].
Regorafenib is a multiple kinase inhibitor already approved for the treatment of patients with CRC refractory to other therapies worldwide [39]. Originally, Regorafenib was developed as a V-raf-1 murine leukemia viral oncogene homolog 1 (RAF1) inhibitor, but in preclinical experiments, its usefulness as a multi-targeting kinase inhibitor with a wide therapeutic range was observed [40,41]. In the multicenter, randomized, placebocontrolled, "Regorafenib monotherapy for previously treated metastatic colorectal cancer" (CORRECT) study, Grothey et al. evaluated the use of Regorafenib as monotherapy in patients with mCRC previously treated with standard therapies [42]. In this study, the median OS was 6.4 months in the Regorafenib group compared with 5 months in the placebo group (HR 0.77; 95% CI 0.64-0.94; p = 0·0052) [42]. The "Regorafenib for patients with metastatic colorectal cancer who progressed after standard therapy" study (CONSIGN study) confirmed these results and found an overall median PFS of 2.7 months (95% CI), 2.5 months for KRAS mutant tumors, and 2.8 months for KRAS wild-type tumors [42].
Over-expression of EGFR is found in 65-75% of patients with advanced colorectal cancer [48]. Epidermal Growth Factor (EGF) and EGFR levels are significantly higher in malignant zones of colorectal cancer specimens than in other nearby regions [49]. The presence of high levels of EGFR in CRC justifies the use of targeted therapies for this pathway [50]. Cetuximab and panitumumab are the most used treatments that fit this category for CRC [24].
Cetuximab is an IgG antibody that, after competitive binding to the external domain of the receptor, promotes the internalization and destruction of EGFR [51], resulting in an inhibition of cell growth, a decrease in the production of matrix metalloproteinase (MMP) and VEGF, and induction of apoptosis [49,52,53].
Cetuximab, in conjunction with other chemotherapies, improves progression control, as noted by the Phase III "Irinotecan and 5-FU/FA with or without cetuximab in the first-line treatment of patients with mCRC" (CRYSTAL) trial, which demonstrated superior median progression-free survival times with FOLFIRI plus cetuximab compared with FOLFIRI alone (8.9 vs. 8 months, HR 0.85; 95% CI), but with an OS without significant differences (HR 9.93; 95% CI) [54]. The use of only cetuximab after the combined use of FOLFOX plus cetuximab maintained a similarly elevated PFS, but with fewer adverse effects [54].
Both drugs, cetuximab and panitumumab, are considered first-line treatments for CRC [23] with no inferiority between them, as well as side effects, as shown in the phase III Panitumumab versus cetuximab in patients with chemotherapy-refractory wild-type KRAS exon 2 mCRC (ASPECCT) study, with an OS of 10.0 months for cetuximab and 10.4 months for panitumumab, 95% CI 0.839-1.113, HR 0.97, p < 0.0007 for noninferiority [61].

EGFR-Related Resistance Mechanism
Treatment using Anti-EGFR monoclonal antibodies (MoAb), such as cetuximab or panitumumab, is only effective in a portion of patients due to, among others, non-expression of EGFR and mutations, both in EGFR itself and at other points in the pathway, which can lead to the development of resistance [62][63][64][65].
Mutations directly in the EGFR or a low expression of EGFR can decrease the benefit of anti-EGFR therapies [66]. The most common mutation in patients with CRC is the substitution of serine to arginine at amino acid 492 (S492R) in the extracellular domain of EGFR, which causes a decrease in response to cetuximab and panitumumab [67]. The incidence of EGFR mutation varies across studies by region. In Greece, patients with non-Small-cell Lung Carcinoma (NSCLC) had an EGFR mutation rate of 28% [68], while in South Korea, the rate was 15% [69]. In colorectal cancer patients, the incidence in Western countries of EGFR mutation was 0.34% [70], while it was 12% in Japan and 22.4% in South Korea [71,72]. Patients from the Middle Eastern population with colorectal cancer had an EGFR mutation rate of 0% [73].
EGFR-targeted therapies, such as cetuximab and panitumumab, may also affect neighboring stromal cells near the cancer cells, promoting an increase in the secretion of EGF, inducing resistance against EGFR-targeted therapies through continuous mitogen-activated protein kinase (MAPK) signaling [65].

Resistance Mechanism-RAS Family
The RAS family, composed mainly of Kirsten rat sarcoma viral oncogene homolog (KRAS), neuroblastoma RAS viral oncogenes homolog (NRAS), and Harvey rat sarcoma viral oncogene homolog (HRAS), participates in the EGFR pathway, acting in the transducing and auto-inactivation of this pathway [66]. Mutations in this family are considered a major cause of resistance to anti-EGFR therapy [74][75][76][77]. The incidence of RAS mutations in CRC is about 53% [78], mainly in codons 12 or 13 of exon 2 (81-96% of all KRAS mutations), and 4-19% at exon 3 or exon 4 [66,79,80]. Therefore, the RAS genotype has become an essential factor in the therapeutic decision of patients with mCRC [48].
RAS inhibitors have recently been studied to overcome this obstacle [81]. Among the studies of RAS-targeted therapies, sotorasib, which acts selectively and irreversibly on KRAS-mutated glycine 12 to cysteine (G12C), is an example of a medication that promoted better responses to anti-EGFR in patients with RAS mutations, which is why it was approved by the Food and Drug Administration (FDA) in May 2021, although having no effect in patients with mutations at codons 12, 13, and 61 [82,83]. However, acquired resistance is anticipated, limiting its long-term clinical use due to secondary mutations causing resistance in sotorasib in vitro, and requiring a switch to two other drugs, BI-3406 plus Trametinib, to overcome this new obstacle [84]. Medications targeting other mutations, such as dasatinib, showed results in vitro but did not reproduce results in Phase IB/II studies of 77 patients with previously treated mCRC [85]. Some non-targeted drugs showed positive results in these patients [86]. Some studies indicate that simvastatin plus cetuximab reduces the proliferation of KRAS mutant cells and suppresses v-raf murine sarcoma viral oncogene homolog B1 (BRAF) activity in these patients [87,88], and that Metformin reverses acquired resistance in KRAS to anti-EGFR drugs [89].
PI3K inhibitors may contribute to the control of tumor growth, regardless of driver genotypes [91]. Mutations in PIK3CA lead to the permanent activation of the EGFR pathway, promoting resistance to EGFR-blocking effects and decreasing the response rate of anti-EGFR therapies, with a relative risk of 0.56 (0.38-0.82, 95% CI) [97]. Other drugs on the PI3K axis are being developed to act in cases of PI3KCA mutations, such as AKT inhibitors, mammalian target of rapamycin (mTOR) inhibitors, and dual inhibitors of PI3K and mTOR [93], but they are still in the initial phases [98].
In patients with mutated PI3KCA colorectal cancer, regular use of aspirin is associated with a blockade of the PI3K pathway, causing suppression of cancer cell growth and induction of apoptosis [99,100]. The post-diagnosis use is associated with a higher OS (HR = 0.83; 95% CI, 0.75-0.9) and superior colorectal cancer-specific survival (HR = 0.78, 95% CI, 0.66-0.92) [101]. In patients with PI3KCA WT, it is not associated with differences in these values [102] Phosphatase and Tensin homolog (PTEN) is a negative regulator of the PI3K/AKT pathway, and mutations are seen in 5.8% of CRC patients [103,104]. Loss of this regulator results in permanent activation of the PI3K/AKT pathway, generating resistance to anti-EGFR drugs [103,104]. PTEN-negative status is correlated with worse response rates and lower PFS than PTEN-positive status [105,106].
BRAF-inhibitor monotherapy in BRAFV600E-mutated CRC has low response rates, which is linked to incomplete inhibition of MAPK signaling in tumor cells [119,120]. BRAF inhibition generates a feedback activation of EGFR, causing an increase in the EGFR pathway and its consequences [121]. The addition of targeted therapies to BRAF and EGFR results in synergistic inhibition of BRAFV600E-mutant CRC [122,123]. Vemurafenib, the oral selective inhibitor of the BRAFV600 kinase, is suggested as a possibility for BRAFV600E mutation-positive patients by the NCCN [118], but studies vary in confirming its efficacy [124][125][126][127].
The phase III "Binimetinib, encorafenib, and cetuximab in BRAF600E-mutated CRC" (BEACON) trial showed the efficacy of another BRAF inhibitor, encorafenib, in dualtargeted EGFR and BRAF therapy as second-line systemic therapy for BRAFV600E-mutated CRC. In this study, the objective response rate (ORR) for the triple regimen (encorafenib and binimetinib, a MEK inhibitor, with cetuximab) was 26%, compared with 2% in the control group. The Median OS was 9 months for the triple regimen and 5.4 for the control group (p < 0.0001, 75) [123]. The use of triple therapy in clinical practice may be related to drug toxicity events, and drugs must be monitored constantly [128,129]. More recently, new local delivery technologies, nanocarriers (NCs), are being tested that propose to reduce these toxicity events by delivering medications precisely [129]. In the case of colorectal cancer, NCs that use CEA and CD44 receptors, both well-established biomarkers for CRC, have shown to be especially promising, with high specificity for tumor cells [130,131].
Patients with mCRC who underwent chemotherapy with FOLFOX or CAPEOX in the last 12 months and are BRAFV600E-mutation-positive have as therapeutic options recommended by the NCCN the use of encorafenib plus cetuximab or vemurafenib [118].

HER2
HER2 is the only member of the EGFR family that is not activated by ligands but by homo or heterodimerization with other ligand-bound receptors [133], resulting in the activation of signal transduction pathways (RAS-RAF-ERK and PI3K-PTEN-AKT), which control cell growth and differentiation [134]. High HER2 expression is found in 2-11% of CRC cases [135]. HER2 amplification and KRAS, NRAS, and BRAF mutations simultaneously are very unlikely in advanced CRC [136].
In addition to functioning as an oncogenic driver, HER2 is a mediator of resistance to anti-EGFR therapies [137]. In patients with de novo HER2 amplification, PFS (PFS 89 vs. 149 days; p = 0.0013) and OS (307 × 515 days) are lower than in patients without amplification, when both treated with anti-EGFR targeted therapies [137].
Targeted therapy for the HER2 pathway in cancers in general currently relies on three medications with different mechanisms: trastuzumab, an HER2-targeted monoclonal antibody that binds to the extracellular domain; pertuzumab, a recombinant humanized monoclonal antibody that inhibits the heterodimerization of HER2 with other HER2 receptors; and lapatinib, the tyrosine kinase inhibitor against EGFR1 and HER2 [137].
The HERACLES-A, a multicenter open-label phase II trial (HER2 Amplification for Colorectal Enhanced Stratification), used trastuzumab plus lapatinib in HER2-positive KRAS exon 2 WT advanced CRC patients who were resistant to first-line treatments, including cetuximab [138]. In this study, an ORR of 30% was found with a median duration of response of 9.5 months and a median PFS of 21 weeks (95% CI 16-32 weeks) [138]. HERACLES-B, a phase II study with only twelve patients treated and eight assessed for response, evaluated the combination of pertuzumab and ado-trastuzumab emtansine (T-DM1), an antibody-drug conjugate linking trastuzumab to a microtubule inhibitor chemotherapy, in patients with mCRC HER2-positive KRAS exon 2 WT. In that study, a decrease in tumor volume was found in 87% of patients [139].
The MyPathWay (pertuzumab plus trastuzumab for HER2-amplified metastatic colorectal cancer NCT02091141) trial is still ongoing [140]. However, interim data point to a PFS of 2.9 months, an OS of 11.5 months, and an overall response rate gain of 32% in patients with mCRC HER2-amplified treatment with double treatment of trastuzumab and pertuzumab [141].

Immune Checkpoints
Mismatch-repair deficiency (dMMR) is found in 15% of all colorectal cancer patients [142]. Eighty percent of patients with dMMR are sporadic cases, primarily due to methylation of the mutL homolog 1 (MLH1) promoter gene. The hereditary cases are due to germline mutations in the MLH1 and mutS homolog 2 (MSH2) genes [143]. These events result in the cells being unable to repair mutations, causing a significant accumulation of these, generating tumors with high microsatellite instability (MSI-H) [144,145]. MSI-H is associated with increased risk for CRC, particularly with high tumor mutational burden and higher numbers of tumor-infiltrating lymphocytes [145,146]. The presence of dMMR alters the chance of responsiveness to specific conventional chemotherapy regimens and encourages the possibility of adding other treatments [147].
The immune system identifies the high tumor mutational burden generated by the propensity to mutations promoting a series of secondary signals that regulate the immune system response [148,149]. The programmed death 1/programmed death 1-ligand (PD-1/PD-L1) receptor-ligand system and Cytotoxic T lymphocyte-associated antigen 4 (CTLA-4) play a fundamental role in this regulation, being responsible for the tolerance or not of the immune system to those cells [150,151]. Malignancies often have elevated levels of checkpoint inhibitors such as CTLA-4 and PD-L1, significantly higher in mCRC, which interferes with the host immune response and confers resistance of these tissues to the host immune system [152]. Co-high expression of tumor PD-L1 and CTLA-4 in CRC tissues is a negative predictor of OS (HR 3.86, 95% CI 1.71-8.51, p = 0.001 in the membrane-bound receptor form and HR 5.72, 95% CI 1.87-14.54, p = 0.004 in the soluble form), in addition to being an independent prognostic factor for poor disease-free survival [153].
Immune checkpoint inhibitors (ICI) act on co-inhibitory receptors, such as CTLA-4 and PD-1, on immune system cells, or their ligands, such as PD-L1, on tumor and immune cells, in order to prevent the resistance of these cells to the immune system, potentiating the cytotoxic killing of tumor cells [154].
PD-1 blockage using antibodies in patients with MSI-H or dMMR mCRC has become an efficient treatment option [155]. Pembrolizumab, a humanized IgG4 antibody, has been approved for the treatment of mCRC by the FDA [156]. The recently published "Health-related quality of life in patients with MSI-high or dMMR mCRC treated with first-line pembrolizumab versus chemotherapy" (KEYNOTE-177) randomized, open-label, phase III study compared Pembrolizumab versus chemotherapy for MSI-H or dMMR in mCRC patients, with a median follow-up of 44.5 months (IQR 39.7-49.8). Although the study did not obtain a median overall survival for patients treated with Pembrolizumab due to the planned alpha of 0.025 for statistical significance not being achieved, data for PFS were promising [157]. The median PFS for the pembrolizumab-treated group was 16.5 months (95% CI 5.4-38.1) versus 8.2 months (6.1-10.2) for chemotherapy-treated patients (HR 0.59, 95% CI 0.45-0.79) [157]. A lower rate of serious adverse events was also found in patients treated with pembrolizumab (16%) compared with those treated with traditional chemotherapy (29%), pointing to pembrolizumab as an efficient option for first-line therapy in patients with MSI-H or dMMR mCRC [157].
Nivolumab, another humanized monoclonal IgG4-based PD-1 antibody, has been FDA-approved for use alone or in combination with ipilimumab, the CTLA-4 inhibitor, in patients with mCRC who have progressed following first-line chemotherapy treatment [118]. The study that resulted in this approval was the "Nivolumab in patients with metastatic DNA dMMR or MSI-high CRC" (CheckMate-142) study; the phase II, open-label, multicenter study first investigated the use of Nivolumab in patients with MSI-H or dMMR mCRC after the use of first-line therapy, finding an objective overall response rate of 31.1% (95% CI 20.8-42.9%), median PFS of 14.3 months (95% CI 4.3-NE), and 12-month OS of 73.4% (95% CI 62-82) [158]. In this same study, Nivolumab plus low-dose ipilimumab for MSI-H/dMMR was evaluated as a possible first-line treatment in mCRC patients [159]. ORR was 69% (95% CI , and the disease control rate was 84% (95% CI 70.5-93.5), with a 13% complete response rate. No median PFS or median OS was found with a follow-up of 24.2 months (24-month rates of 74 and 79%, respectively) [159]. Figure 1 and Table 1 summarize the described targeted therapies and important clinical trials cited throughout this publication.

Conclusions
Understanding the mechanisms involved in the pathophysiology and evolution of colorectal carcinoma, as well as possible specificities that can influence clinical presentation and outcomes of patients, allows for the development of more precise care. In this context, the identification of molecular subtypes, and the elaboration of strategies aimed at each of these, provide better final results for patients, fewer side effects, and reduce loss of therapeutic adherence.

Conflicts of Interest:
The authors declare no conflict of interest.

ANCHOR-CRC trial
The Encorafenib, binimetinib, and cetuximab in subjects with previously untreated BRAF-mutant CRC trial; ASPECCT