Mini Review: The Application of Omics in Targeted Anticancer Biopharmaceuticals Development

Figure 1: The application of Omics technologies in the development of anticancer biopharmaceuticals.

Omics in Cancer Biomarker Identification

Intensive research showed that the malignance transformation of cancer is caused by the mutations of oncogene and tumor suppressor genes. Most cancers are signalled by molecular, pathway, and network markers. These biomarkers indicate the diseased cells, tissues, or individuals, so the investigation of biomarkers can benefit cancer diagnosis, prognosis, patient response prediction, and therapy development. The challenge in discovering biomarker is the lack of a complete understanding of cancer development. The highthroughput Omics technologies have generated a large amount of global molecular profiling, which opens the door to study the drivingmechanism of cancer.

Single molecular biomarker

Single molecular biomarkers include gene, mRNA, miRNA, protein, glycan, and metabolite, which are caused by gene mutations, transcription modifications and translation or post–translational modifications. Several Omics databases are available for single molecular biomarker study [12,21,22], including Entrez gene and Mouse Genome Databases for gene annotation [21,23], Unipart for protein annotation [24–26], and Gene Oncology annotation [27–29]. In addition, several Omics data analysis tools have been developed to map gene, mRNA or metabolite to biological protein, such as UniProt Knowledgebase [30,31], iProClass [32,33], Protein Information Resource [34], DAVID gene ID conversion tool [35], and Protein Identifier Cross–Reference [5,36,37].

The functional profiling of annotated genes and proteins has been successfully used to identify biomarkers of breast cancer, lung cancer, prostate cancer and other cancers. Previous breast cancer research showed that estrogen receptor α is an estrogen–inducible transcription factor, and its dysfunction accounts for the majority (70%) of breast tumor [12]. Human epidermal growth factor receptor 2 (HER2), a tyrosine kinase receptor, is over expressed in approximately 15–25% of breast cancer patients. A proteolytic fragment of α–antitrypsin has also been identified as a diagnostic and prognostic marker of breast cancer [21,38,39]. It is reported that the tumor suppressor proteins play a key role in DNA repair through the homologous recombination mechanism [14,40,41], so the germline mutations of breast cancer repressor genes (i.e. BRCA1 and BRCA2) are the fundamental defects in hereditary breast cancer and ovarian cancer. For lung cancer patients, the gain of chromosome 5p occurs in a high number (60%). Other alteration genes that are frequently involved in lung cancer are EGFR⁄HER1 and KRAS [42,43]. The proteomics analysis has indicated that pyrophosphatase 2 and Ezrin are the potential markers of metastatic prostate cancer [44]. In addition to functional genomics, epigenetics has been used to evaluate the aberration of DNA methylation in cancer cells. Jones and Baylin have observed the global and locus–specific (e.g., CpG islandrich promoters) hypomethylation in abnormal methylation of cancer cells [45,46]. Another study reported that PITX2 DNA methylation is an epigenetic marker to predict adjuvant tamoxifen therapy in the treatment of early stage cancer [47].

Network biomarker

Another classification of biomarker is network–based regulator (i.e. panel biomarker), such as cell signalling, biological pathway, and gene regulating network. The multiple network databases, e.g., Kyoto Encyclopaedia of Genes and Genomes (KEGG) [48], Pathway Interaction Database [49], and Pathway Commons [1,49,50], have been established for metabolic pathway and signalling pathway analysis. Mapping experimental data to network pathways is a key step in the functional interpretation of Omics data, which has been achieved by utilizing the systems of pathway analysis, such as Ingenuity IPA and Gene GO Meta Core [51]. Additional datasets, such as Intact and Molecular Interaction, have been built for proteinprotein interaction annotation to reveal cancer classification and indicate the mutation genes [52,53].

Some network biomarkers in various leading cancers have been discovered using cancer Omics. For instance, genomics studies have shown that: (i) the switcher between proliferation and differentiation occurs in the cellular networks of breast cancers [54]; (ii) twelve core signalling pathways alter in human pancreatic cancers [55]; (iii) the dysregulation of the transforming growth factor β plays a key role in epithelial ovarian cancers [56]; and (iv) Some cancer cells exhibit defective genetic–epigenetic networks in different parts of tumor tissues [57,58]. The proteomics study of breast and lung cancers showed that the co–activation of multiple receptor tyrosine kinases sustains cell proliferation in glioblastoma multiforme, indicating the importance of targeting multiple receptor tyrosine kinases [42,59,60]. In addition, the functional proteins can interfere with various signalling cascades, so the aberrant post–translational modification of regulating proteins in a network plays a critical role in the concogenesis of various malignancies.

All above progresses in network biomarker studies are attributed to the rapid advancement of systems biology that has accumulated a large amount of Omics data. Although various bioinformatics tools have been developed to facilitate functional analysis and data interpretation, it is still challengeable to extract key information from the big data. To solve this issue, various statistical frameworks, computational and mathematical formulas, and genome–scale models have been applied to infer cellular network [61–64]. For example, Alberghina et al. investigated the enhanced glycolysis and mitochondrial metabolic remodelling of cancer cells and identified multiple pathway biomarkers [65]. They characterized four basic cancer properties using the developed model, including enhanced cell proliferation, evasion from apoptosis, genomics instability, and instability of cancer cells to enter senescence, based on their modelling work.

MicroRNA

The microRNA (miRNAs), derived from non–coding primary messenger RNAs (mRNAs), are single–stranded RNAs of approximately 22 nucleotides in length [66]. The miRNAs bind to the complementary sequence of targeted gene and repress targeted mRNA through direct cleavage or translational repression. Therefore, the deregulation of miRNAs affects both gene expression and metabolic activity involved in cancer pathogenesis. Eventually, the identified predictive or prognostic miRNA signatures can be clinically translated to diagnose cancer disease. Over the last decade, the advanced next–generation sequencing technology and the accumulated transcriptomics knowledge have been utilized for effective and quantitative analysis of both mRNA and miRNA. For example, studies have found that miR–10b, miR–15, miR–16, and miR– 20 are anti–angiogenic miRNAs that target the mRNA of epidermal growth factor receptor (EGFR) [67,68]. In other studies, miR–500 has been identified as a potential diagnostic marker of hepatic cell carcinoma [21,69].

Taken together, Omics technologies have enabled the identification of a number of predictive biomarkers that can be used for early cancer detection. The discovered biomarkers also spurred the development of targeted cancer therapies, as discussed in the following sections.

Omics in Targeted Anticancer Biopharmaceuticals Development

Targeted therapy development

After discovering biomarkers and understanding the mechanism of cancer regulation, targeting the abnormal gene, pathway or metabolic activity plays a critical role in cancer therapy development. The typical targeting strategies are discussed below.

Target specific regulator: In cancer therapy development, it is very important but challengeable to target epigenetically and genetically abnormal molecules or pathways. The ideal targets should have the following features: unique to the specific cancer cell, cancer cell survival dependent, representative with high frequency, and feasibleto be targeted with medicine. In traditional therapy development,targeting specific regulator requires complicated experimentaldesign and systematic test. With the guidance of Omics, it is more efficient to predict the patient’s response to the targeted therapy, and,therefore, to define the rational targeting strategy in cancer treatment. One substantial target for cancer therapy is epidermal growth factorreceptor (EGFR), a tyrosine kinase receptor, which is over expressed n solid tumors in lung and colorectum [14]. The non–small–cell lung cancer (NSCLC) harbouring EML4–ALK mutation and EGFRmutation can be treated with tyrosine kinase inhibitors [70,71]. The HER2 proto–oncogene over expressed in breast cancer is also a tyrosine kinase receptor that can be treated with targeted therapy [1]. Another ideal target is the pathway deregulated by BCR–ABL1 that associates with estrogen receptor expression in chronic myelogenous leukemia [72]. For example, the kinase inhibitor targeting BCR–ABL1 fusion gene has been demonstrated as an efficient approach to treat the chronic myeloid leukemia tumor and gastrointestinal stromal tumor.

Target core pathway: As previously reviewed, the global genomics analysis showed that a dozen core signalling pathways were deregulated in the progression to metastatic pancreatic cancer [1]. The Cancer Genome Atlas (TCGA) project has identified frequent genomic and epigenetic aberrations that deregulate the p53, retinoblastoma RB, PI3–kinase, receptor tyrosine kinase (RTK) core signally pathways of glioblastoma multiforme [73]. It is found that angiogenesis, a process of neovascular formation during carcinogenesis, is found to correlate with vascular endothelial growth factor (VEGF), fibroblast factor, Notch, transforming growth factor–β, Hedgehog, and Wnt signalling [4]. These findings indicate that it might be promising totreat cancer by targeting the deregulated core pathways regardless of the progression mechanisms of metastatic cancer. The development of common therapeutic approach by targeting core pathway relies onthe integration of multi–level Omics analysis. For example, various powerful Omics data analysis tools, including KEGG, Pathway Commons, Ingenuity, and Pathway Recognition Algorithm using Data Integration on Genomic Models, have been developed [74]. It is worthy to point out that some regulatory pathways (e.g., MYC, KRAS and TP53) are remarkably resistant to therapeutic interventions due to genetic instability [61]. So, it is critical to develop a second line herapy to counteract the recurrent disease and improve the clinical efficiency of cancer therapy.

Regulate glycolysis: Warburg effect is referred as the phenomenon that the high rate of aerobic glycolysis followed by anaerobic respiration producing lactic acid dominates the energy production in malignant tumor. Glycolysis is an important metabolic process that provides carbon and energy to maintain cell growth and cellular function. Multiple Omics studies show that the glycolysis in cancer cells is abnormal [10,61], so it will be effective to target glycolysis. As the terminal enzyme of glycolysis, pyruvate kinase is the key regulator that determines cellular energy level, redox homeostasis, and cell proliferation. A strong correlation between transcriptional activation of oncofetal M2 isoform of pyruvate kinase (PKM2) and cancer metabolism has been reported [75]. PKM2 switches metabolism to lactate fermentation and promotes tumorigenesis. Most PKM2 positive cancer cells are signalled by epidermal growth factor eceptor (EGFR) at the early–stage of cancer development. Therefore, regulating PKM2 is essential to control cancer cell proliferation. The activity of PKM2 is tightly regulated and controlled at multiple Omics levels (i.e., tanscript omics, proteomics and metabolomics), which responds to environmental stresses or toxins. PKM2 is a specific metabolism hallmark of cancer cells directly controlling cancer cell progression, so it is recognized as an attractive target for novel cancer therapy development.

Target mitochondria: Mitochondria are semi–autonomous organelles that generate more than 95% of cellular energy and control various cellular metabolic activities. Previous mitochondrial metabolomics study uncovered the correlation between tumorigenesis and multiple abnormal metabolic activities. For example, the production of reactive oxygen species, decreased oxidative phosphorylation, increased glycolysis, and deregulated cellular energetics are considered as typical metabolic biomarkers of cancer cells [10]. The mitochondrial genomics analysis showed that the D–loop (a common mtgenome mutation) is an excellent signature for early detection of solid tumours. All these exciting findings have increased the enthusiasm for designing or identifying novel cancer therapeutic agents directly targeting mitochondria. As Verschoor et al. reviewed, the numerous mitochondrial databases, including MitoInteractiome, MitoP2, HMPDb, and MitoMiner, provide the platforms for new therapy development [76].

Screen small molecule drug: The leading small molecule drug compounds for pre–clinical studies can be screened by integrating transcriptomics data, toxic genomics data, and computational approach [77]. The available drug databases include Drug Bank [78,79], KEGG, NCBI Pub hem, Connectivity Map [80,81], and Comparative Toxic genomics Database [82,83]. The Omics–based drug screening can overcome the disadvantages (i.e., expensive, time consuming and laborious) of traditional in vivo animal experiments and in vitro drug screening.

Targeted anticancer biopharmaceuticals development

The public NIH Roadmap Epigenomics Mapping Consortium and the Personal Genome Project aim to create human genome maps integrated with highly comprehensive phenotype data [42]. The systems biology knowledge can provide insights into cancer cause and, therefore, guide the development of novel therapies. Scientists and clinicians are dedicated to providing the highest life quality for patients diagnosed with cancer by improving prevention, diagnosis and survivorship. Developing novel biopharmaceuticals and increasing the clinical efficiency of various therapeutic proteins represent an efficient way in cancer therapy.

The ever–increasing efficiency of cancer genomics has accelerated the development of targeted anticancer therapeutic biomedicines. Biopharmaceuticals represent a significant amount of targeted therapeutics, including monoclonal antibody (mAb), recombinant therapeutic protein, and nucleic acid–based biomedicine [84]. More than one hundred biopharmaceuticals have been approved by the FDA and are available on the market. Several dozen of pathwaytargeted drugs have been applied for clinical treatment of breast, ovarian, colon, and other cancers. In addition, thousands of pipeline biopharmaceuticals are in the clinical trial stage.

mAb targets specific receptors of tumor cells to mediate cytotoxicity via antibody dependent cell cytotoxicity or complement dependent cytotoxicity. Table 1 highlights some mAbs for cancer treatment on market. Breast cancer is the second leading cause of death among women, which kills nearly 40,000 Americans in 2011, and an estimated 300,000 new cases of breast cancer are expected to be diagnosed each year [85]. The monoclonal antibody Trastuzumab (Brand name: Herceptin) developed by Genentech is the first commercially available biopharmaceutical to target HER2. It has been proved that humanized Trastuzumab improves the survival rate of breast cancer patients by interfering the concogenes to which cancer cells addict [5]. The fusion protein MM–302 is a therapeutic protein candidate for HER2 positive breast tumor under clinical evaluation [14]. The patients with colorectal cancer, the fourth leading cancer death, can be treated by inhibiting VEGF receptor using chimeric antibody Cetuximab (Erbitux, products of Bristol–Myers–Squibb, Merck and ImClone systems) [86]. Other cancers, such as head and neck cancers, can be treated by Cetuximab. The human mAb, Bevacizumab (Avastin) developed by Genentech, has been approved for clinical use in treating metastatic colorectal cancer by inhibiting VEGF. Also, Bevacizumab is used in the clinical treatment of NSCLC, breast cancer, ovarian cancer, and kidney cancer [4]. The anti–EGFR receptor therapeutic protein, Panitumumab (Vectibix), has been developed by Amgen and Pfizer to treat colorectal cancer.

  

    
Name / 
      
Brand name 
    
Description 
    
Company 
    
Cancer treated 
  
  

    
Bevacizumab /
      
Avastin
    
Inhibit VEGF, humanized mAb
    
Genentech (Roche)
    
Colorectal cancer, non-small-cell lung cancer, breast cancer,    glioblastoma, kidney cancer
  
  

    
Cetuximab / Erbitux
    
Inhibit EGFR, chimeric mAb
    
Bristol-Myers-Squibb, Merck and ImClone systems
    
EGFR expressing metastatic colorectal cancer, head cancer and neck    cancer
  
  

    
Etanercept / Enbrel
    
Inhibit tumor necrosis factor, fusion protein
    
Amgen
    
Target tumor necrosis factor
  
  

    
Panitumumab / Vectibix
    
Inhibit EGFR, humanized mAb
    
Amgen & Pfizer

    
Metastatic colorectal cancer
  
  

    
Rituximab / Rituxan
    
Anti-CD20 IgG1, chimeric mAb
    
Biogen Idec
    
Leukemia, transplant    rejection, and autoimmune    disorders 
  
  

    
Trastuzumab / Herceptin
    
Anti-Her 2, humanized mAb
    
Genentech (Roche)
    
Metastatic breast cancer
  

Table 1: The major components used in our systems. The list cost is for guidance only.