Introduction

Cancer is one of the most common non-communicable diseases among humans. It is characterized by an abnormal growth of cells, with ability to spread to other parts of the body [1]. The International Agency for Research on Cancer (IARC) reported about 18 million new cases of cancer comprising of 36 cancer types in 185 countries in 2018. Further, IARC reported about 9.6 million deaths due to cancer in 2018 globally [2]. Cancers can be categorized as carcinomas, lymphomas or leukemias, and sarcomas. Carcinomas are commonly associated with any type of epithelial cell of the skin or lining of organs, and about 90% of human malignancies are categorized as carcinomas [3]. Lymphomas and leukemias are malignancies of the immune cells and blood-forming cells, respectively. They constitute about 8% of human cancers. Sarcomas refer to any malignant tumor of connective tissues or non-epithelial cells, including bone, cartilage, fibrous, and muscle tissues. They account for less than 2% of reported cancer cases [4]. Cancers can also be further classified based on the affected cell type (e.g., erythroid leukemia, referring to red blood cell malignancy) or the affected tissues or organ (e.g., prostate and breast cancer). Although cancer continues to be a major global health issue that accounts for one in every seven deaths [5], there are only a few cancer types commonly diagnosed. Table 1 presents the ten most common cancers and their estimated cases of mortality in the USA in 2019. GLOBOCAN in 2018 reported that lung cancer is the most common cancer type, representing about 11.6% of global cancer cases. The mortality rate of lung cancer was reported to be the highest (18.4%), followed by colorectal cancer (9.2%) and stomach/liver cancer (8.2%) [6]. Generally, the treatment of cancer is carried out by surgery to remove specific cancerous tissues [7], radiation therapy [8], and chemotherapy [9]. Surgeries are usually recommended if the spread of the cancer is rapid and difficult to control using conventional treatment methods [10]. Cancer relapse is associated with the presence of circulating tumor cells in cancer patients and this can result in unsuccessful surgery and patient mortality [11]. Chemotherapy is the most common cancer treatment method where chemical-based drugs at high concentrations are administered to inhibit rapidly growing cancer cells [12]. Even though chemotherapy has proven to be successful in retarding cancer cell growth, the lack of selectivity and specificity induce severe side effects through the inhibition of normal healthy cell growth [13]. General side effects of chemotherapy include hair loss, loss of appetite, diarrhea or constipation, nausea, fatigue, mouth sores, and skin problems [14]. Radiation therapy is introduced to overcome the limitations of chemotherapy. The treatment utilizes high intensity radiation, targeted at the cancer-affected site or organ, to inhibit the growth of cancer cells. However, radiation therapy can lead to alterations in the DNA of neighboring healthy cells and this may lead to side effects [15]. Hyperthermia, cryosurgery and magnetic therapy are some of the novel methods developed to address the challenges of conventional cancer treatment approaches [16,17,18] though issues relating to toxicity, inhibition of healthy cells and cancer relapse persists [19, 20]. Targeted treatment of cancer hinges on the identification of specific molecular receptors on cancer cells to inhibit growth; the development of novel ligands to target biomarkers expressed by the cancer cells; and the creation of novel multifunctional carrier systems for targeted delivery of anticancer drugs to specific malignant sites. There are numerous oncotargets that have been discovered as receptors or biomarkers for targeted delivery of anticancer drugs to control the growth of cancer cells [21]. For example, the secretion of oncotargets such as vascular endothelial growth factor (VEGF) and glutathione peroxidases has been exploited for targeted delivery of anticancer drugs to inhibit cancer cell growth and modulate metastasis [22, 23]. Genomics, proteomics, and bioinformatics are effective tools to identify specific cancer inhibiting molecular targets [24, 25]. Moreover, MD simulations offer opportunities to structurally design novel ligands and optimize binding affinity for specific oncotargets binding. This article discusses the development of ligand-oncotarget bioaffinity systems via integrated structural modeling and MD simulations, making use of proteomics, genomic, and X-ray crystallographic resources, to catalyze targeted cancer therapy.

Table 1 The ten most common cancers and their estimated cases of mortality in the USA in 2019
Full size table

Oncotargets for Different Cancer Types

The efficacy and specificity of cancer diagnosis and treatment since the disease occurs at the cellular or tissue level [36, 37]. For example, chronic blood cancer is a common tumor disease among the elderly and it is attributed to the switching or reciprocal translocation of genetic materials between chromosomes 9 and 22. There is a biomarker known as ph1, expressed by 95% of chronic blood cancer patients and synthesized from the combination of the gene Bcr and oncogene Abl as an activated protein kinase [38], which is effective for diagnosis. By targeting and inhibiting the ph1 biomarker, the balance of cell proliferation and apoptosis is interrupted, and this facilitates the inhibition of tumor growth.

Several studies [39,40,41] have reported the potential of hypermethylated promoters as novel biomarkers for all human cancer diagnosis, prognosis, and therapy owing to their presence in specific CpG regions that affect cancer induction. CpG promoter is an ideal biomarker as it influences all cellular pathways with a cancer-type specific profile and it is associated with genes such as tumor suppressor, microRNAs, and DNA repair genes with growth inhibitory effects [41]. The CpG islands of the CpG promoter of these genes hypermethylate if the cell is undergoing transformation, suggesting its suitability as a cancer biomarker.

A recent study by Whitwell et al. [42] has revealed the use of longitudinal multi-biomarker models to enhance the early diagnosis of ovarian cancer. The findings demonstrated the high sensitivity, 85.7%, and high specificity, 95.4%, of this model, which comprises of various overexpressed biomarkers, including CA125, HE4, CHI3L1, PEBP4, and/or AGR2 to target ovarian carcinoma. This provided a better detection of ovarian carcinoma at early-stage disease screening as compared to the conventional test that uses serum cancer antigen 125 (CA125) alone. The novel longitudinal model is capable of detecting cases of ovarian cancer, which were undiagnosable using CA125 test alone, showing its enhanced detection lead times and performance up to 1 year prior to clinical diagnosis. It was also demonstrated in another study [43] that combined biomarkers (both cancer CA125 and immune interleukin 7, IL-7 biomarkers) can perform better in the early detection of ovarian cancer with improved cancer detection times as shown in Fig. 1.

Fig. 1
figure 1

Schematics of interactions between immune effector and tumor cells. Reproduced with permission from [43], under the terms of the Creative Commons Attribution 4.0 International License

Full size image

Bladder cancer is one of the most common cancers globally with high recurrence rates and, therefore, requires an effective biomarker for targeted treatment of malignant bladder tumor. It has been demonstrated that the competitive glucose metabolism can serve as a specific target to kill tumor cells in addition to improving the immune system, especially the therapeutic efficacy of the immune checkpoint inhibitors (ICIs) in the case of metastatic bladder cancer [44]. This is attributed to the presence of overlapping metabolic phenotypes between the activated immune T cells and cancer cells, which leads to the metabolic competition that impacts immune performance. By targeting the glucose metabolism in tumor cells, especially bladder tumor cells, glycolysis-linked receptors can be effectively targeted and destroyed by various inhibitors of glucose metabolism [44]. This is probably due to the implication of both glycolytic signaling pathways and metabolites in the bladder tumor, offering abundant putative therapeutic targets for the anticancer strategy.

Different cancer cells overexpress their own unique biomarkers which are distinct from other cancer types due to varying genetic compositions. For example, the epidermal growth factor receptor (EGFR)/ERBB2 kinase is associated with breast carcinoma [45]; vascular epidermal growth factor receptor (VEGFR) kinase for renal carcinoma [46]; EGFR kinase and protein kinase ALK surface receptors are for non-small cell lung carcinoma (NSCLC) [47]; and BRAF kinase from melanoma [48]. Some other reported specific biomarkers or extracellular surface receptors of different tumor cells are provided in Table 2. The presence of these surface biomarkers helps increase the targeting specificity for diagnosis as well as boost drug therapeutic indices by distinguishing between healthy and malignant cells for enhanced targeted treatment strategies.

Table 2 Examples of some overexpressed biomarkers of various cancer cells
Full size table

Advanced Methods of Oncotarget Discovery

With advancements in both molecular and bioinformatics techniques, more gene data can be rapidly generated to probe genetic disorders, particularly the role of mutated genes in specific cancer cell formation and growth. This is crucial to identify novel tumor biomarkers for early cancer diagnosis and to deliver appropriate therapies. There are other computational approaches such as ligand–protein docking used to generate novel ligands for specific target binding as well as analysis of the binding characteristics to develop promising drug candidates for targeted delivery. Conventional methods often require pre-existing information about the protein sequences and structures and this creates significant challenges due to the similarities among protein structures [59]. Therefore, advanced technologies in detecting gene transcript diversity and the impact of individual protein isoforms of drug effects are important in the design and development of targeted therapies. Thomas et al. [60] described the efficacy of high-throughput genotyping in querying 238 known oncogenes via a powerful mutation distribution spanning 17 types of cancer across a thousand of human tumor samples. The findings indicated a new advancement in tumor genetics whereby multiple mutations associated with various tumor genes can be interrogated simultaneously on a large-scale mutation profiling in a ‘real-time’ scenario to detect previously unknown oncogene mutations as well as formulating cancer classification and cancer therapeutic interventions.

Intratumor heterogeneity plays a vital role in the discovery of cancer biomarkers and clinical treatments, especially for targeted therapies. Intra-tumor heterogeneity is described as solid tumors consisting of cells with distinct genomic alterations within the same tumor. With the advent of sequencing technologies such as next-generation sequencing (NGS) technique, certain features of intra-tumor heterogeneity have been used in routine pathologic evaluation to optimize therapeutic outcomes and patient care. For instance, Ding et al. [61] have sequenced the primary tumor and relapse genomes of several patients with acute myeloid leukemia and illustrated that the presence of a founding clone in the primary tumor has evolved in the relapsed tumor. By identifying the distinct genomic architectures and tumor evolution within multiple patients at different cancer stages (from primary to metastatic tumor and during therapy), it is possible to tackle the mechanism of tumor recurrence for a better targeted therapy [36]. Also, a more tractable method and robust biomarkers such as the ‘actionable mutations’ which are ubiquitous and dominant driver events in all cancer sites can be developed for therapeutic targeting applications, based on the trunk-branch model [62]. This is useful, particularly for tumors with lesser branched events.

‘Cell competition’ has been reported to have strength in triggering cancer cell killing and it can possibly be modified to develop novel anticancer treatments with tumor-suppressive behaviors [63]. Cell competition is a type of cell-to-cell communication whereby competent cells are capable of eliminating neighboring cells within the microenvironment in order to survive, expand, colonize, and grow into targeted tissues. Many reported studies have demonstrated that gene mutation in cell signaling, cell growth, and endocytosis can induce cell competition whereby wild-type cells behave as winners against mutant cells to prevent tumor growth [64,65,66]. Cell competition occurs via several mechanisms, including mechanical interactions and molecular exchange. Mechanical interaction is utilized to remove cells with a higher crowding sensitivity due to enhanced p53 activation and subsequent apoptosis [67, 68]. On the other hand, molecular exchange is based on the capability of cells in capturing and sequestering growth factors such as Wnt ligands [69] or TGFβ [70], in order to activate the death-triggering pathways in cells receiving lower signals. Studies by Rhiner, Lopez-Gay, Soldini, Casas-Tinto, Martin, Lombardia, and Moreno [71] and Madan, Pelham, Nagane, Parker, Canas-Marques, Fazio, Shaik, Yuan, Henriques, Galzerano, Yamashita, Pinto, Palma, Camacho, Vieira, Soldini, Nakshatri, Post, Rhiner, Yamashita, Accardi, Hansen, Carvalho, Beltran, Kuppusamy, Gogna, and Moreno [72] showed a correlation between the Flower isoforms expression in Drosophila and the human cancer progression by investigating the human Flower protein expression and mouse cancer models. This further supports the promising potential of cell competition as a target to develop novel anticancer approaches.

In addition, a network-based method is emerging due to its effectiveness in discovering cancer drug target at the isoform level as shown in Fig. 2. Ma et al. [73] reported credible detection of target genes at the isoform level by integrating the isoform co-expression (IIC) networks with perturbed genes. This cancer-type-specific IIC network was built according to the isoform expression data integrated from both gCSI and CCLE datasets. Perturbed isoforms of target genes were selected based on the response of the corresponding genes to the applied drug. Further examinations via docking tests and comparison of proteomics data and expression status were carried out to determine the major protein isoform drug targets. This method is able to discover targets for cancer drugs as well as improve the drug targeting capability. The study also illustrated that major isoforms of targets can be identified to study the mechanism of action of the drug. Besides that, tumor classifications based on the evolution of different tumors using evolutionary index (Evo-index) and ecological index (Eco-index) provide promising implications in cancer target discovery [74].

Fig. 2
figure 2

Network-based method for the identification of cancer biomarkers. Reproduced with permission from ©Yang et al. [75], under the Creative Commons Attribution License.

Full size image

Furthermore, mathematical modeling is often employed together with computational methods to investigate the specific targets of different cancer cells. These methods can be used to study the biological mechanisms and interactions between tumor cells and complex immune systems as well as the effects on tumor cell biomarker expression [76,77,78]. As a result, numerous stochastic biomarker-network models have been developed and reported for the diagnosis of diverse cancer cells [79,80,81]. These networks are built up of a library of cancer-associated proteins and other protein types and are exploited to determine the most specific targets or biomarkers for cancer detection.

Multi-region exome sequencing (Mseq) is another promising approach used to examine therapeutic targets by studying the cancer evolution based on multiple mutational profiles of spatially different regions of a tumor. Loga et al. [82] reported the application of MSeq in identifying the extreme intra-tumor heterogeneity and evolution of mismatch repair deficiency (dMMR) gastro-oesophageal cancer. The evolution analysis and MSeq data revealed the promising potential of MSeq in the discovery of biomarkers for highly heterogenous cancers. MSeq is capable of estimating the truncal mutation loads more accurately and robustly to prevent sampling biases and illusion of clonality of driver mutations in order to develop effective immunotherapy biomarkers such as ‘mutation burden’ [82]. In short, advanced technologies in understanding the cellular and genetic characteristics of biological systems are a promising development in detecting powerful tumor biomarkers or therapeutic targets to boost the cancer target discovery, diagnosis, and therapy.

Role of Genomics, Proteomics, and Bioinformatics

The central theme of cancer medicine is prevention, detection, and treatment; and for these goals to be achieved, a clear understanding of the genomic and proteomic influences at the molecular level is critical. Such a knowledge is required to develop solutions in the treatment of cancer, especially in the areas of early detection and personal medicine [83, 84]. Oncotargets are an essential part of this understanding and genomics play a crucial role in their discovery, to revolutionize biomedical research with the promise of early diagnosis of cancer, identification of cancer risk, selection of required treatment therapy, and the detection of relapse as shown in Fig. 3 [85]. Two major categories of genomic oncotargets are prevalent: DNA and RNA biomarkers. DNA biomarkers include the measurement of serum DNA [sDNA] concentrations as an indication of the presence of cancer, and this can be used in the process of staging during detection and treatment of cancer. Researchers have demonstrated the feasibility of using serum DNA concentrations as biomarkers for cancer detection. A study conducted by Gal et al. [86] showed a four-fold difference in the median levels of circulating sDNA between breast cancer patients and healthy controls (medians of 221 and 63 ngml−1, respectively) using a real-time polymerase chain reaction (PCR) quantitative method [86]. The researchers found that 72% of patients with breast cancer (69 out of 96) had DNA concentrations of more than 100 ngml−1, while only 12.5% of the control group had the same amount of sDNA concentration (3 out of 24). Again, using a receptor operating characteristic (ROC) curve analysis, the researchers demonstrated that there was a strong relationship between the DNA concentration and malignancy by observing an area of 0.92 under the ROC curve. Another group of researchers, Sozzi and coworkers, performed a similar experiment in lung cancer patients and discovered a similar correlation of circulating plasma DNA and cancer by using a ROC curve [87]. They obtained their results by comparing plasma DNA extracted from 84 patients with radically resected primary non-small cell lung cancer (NSCLC) (stages I–III) and that of 43 healthy blood donors as control.

Fig. 3
figure 3

Pipeline for the validation and discovery of biomarker candidates for disease diagnosis. Reproduced with permission from [85], ©Elsevier, 2008

Full size image

Apart from circulating DNA in plasma/serum being used as biomarkers, mutations in oncogenes, tumor suppressor genes, and mismatch repair genes can also serve as DNA biomarkers [85]. This is made possible by the discovery of the HRAS gene, a gene that causes a glycine-to-valine mutation at codon 12, and subsequent identification of similar mutations in the family members of the KRAS and NRAS genes. Other genomic biomarkers have also been discovered. The RAS family encodes small proteins with enzymatic GTPase activities that couple external growth signals governing cell proliferation, and thus, a mutation in the RAS gene encodes a protein with normal guanosine triphosphate (GTP] binding, but no GTPase activity, causing the resulting RAS protein to be activated all the time as seen, leading to unregulated cell division [84, 88]. It is worth noting that the gain-of-function mutations of the RAS oncogene occur in approximately 30% of all human cancers and its isoform. The KRAS gene is the most studied and most frequently mutated oncogene and constitutes 86% of RAS mutations [89]. Recent studies have shown strong indications that the KRAS gene (as a cancer biomarker) and its expressed KRAS protein are potential targets for drug discovery and development for KRAS-targeted therapies [90]. Mutations can also occur in the tumor suppressor genes of cell to result in cancer. The detection of this defect can serve as an oncotarget. The pRb, p53, and p21 genes that make up the family of tumor suppressor genes encode proteins that restrain cell division. A defect in the RB1 gene (the gene that encodes pRb), for example, causes retinoblastoma in children and results in blindness if not removed.

RNA biomarkers have also gained traction in recent times, especially in the discovery of the functional and expression roles of piwiRNA (piRNA), messenger RNA (mRNA), small nucleolar RNA (snoRNA), long non-coding RNAs (lncRNAs), circular RNA (circRNA), and microRNAs (miRNAs), that can regulate gene expression by various mechanisms thus presenting a great potential for the development of diagnosis and therapeutic measures for several types of cancers [91,92,93]. For instance, miRNA in particular has been demonstrated as a biomarker for cancer, either causing malignancy in certain types of cancer or acting as a tumor suppressor in others. A study conducted by Akao et al. [94] found a significantly reduced expression of miRNA-143 and miRNA-145 in colon cancer cells and that by transfecting each precursor miRNA into the cells, a notable growth inhibition of human colon cancer DLD-1 and SW480 cells was observed, indicating that the miRNA-145 acts as a tumor suppressor [94]. Furthermore, another supporting study published by Link et al. [95] found that miR-21 expression was higher in stool samples from patients with colorectal neoplasia than those with normal colonoscopies, demonstrating the use of miR-21 as an oncotarget [95]. These studies demonstrate the great role that genomics play or will play in the future of cancer classification, early detection, prognosis prediction, and therapeutic decision-making.

Cancer detection, diagnosis, treatment, monitoring, and prognosis have been bolstered by protein biomarkers under the broad area of proteomics. The underlying theme of oncological proteomics relates to understanding distinct changes that occur during neoplasm and the available technologies for probing these changes. The prospects of this can be seen with the many approvals obtained from the US Food and Drugs Administration for the use of single proteins as cancer biomarkers. For instance, the human epidermal growth factor receptor-2 HER2/NEU, EGFR, and KIT are used clinically to determine whether or not breast cancers, colon cancers, or gastrointestinal stromal tumors would respond to trastuzumab, cetuximab, or imatinib, respectively, as membrane protein-targeting drugs. Similarly, breast cancer cells that are ER positive or PR positive inform clinicians which hormonal therapy will be more effective for treatment [83, 96, 97]. Identification of protein biomarkers is done by techniques and assays including differential in-gel electrophoresis (DIGE), two-dimensional polyacrylamide gel electrophoresis (2D-PAGE), multi-dimensional protein identification technology (MudPIT), reverse phase micro arrays, mass spectrometry, and field effect transistors (FET) that are coupled with molecular probes. FET is a promising technique as it increases protein marker sensitivity to several orders of magnitude [83]. FET-based protein detection, for example, has been demonstrated by Ohno et al. [98] where they immobilized immunoglobulin E (IgE] aptamers onto a graphene surface FET and showed selective electrical detection of IgE protein [98]. Such a technique will play a huge role toward rapid detection of protein oncotargets.

Bioinformatic tools are essential to process and extract meaningful information from the large proteomic and genomic data. It is particularly useful for the discovery of sensitive and specific biomarkers in cancer research [96]. The importance of bioinformatics in understanding biological macromolecules and their relation to the activity of diseases, especially cancer, cannot be underrated. Public genomic datasets like GEO (https://www.ncbi.nlm.nih.gov/geo/) provide large micro-array data of complementary DNA that has been used by researcher to discover new biomarker for diseases. For instance, micro-array datasets GSE19665, GSE33006, and GSE41804 from GEO databases were used by Li et al. [99] together with the open source bioinformatics tools STRING and Cytoscape to identify a total of 273 deferentially expressed genes (DEGs) and 16 hub genes involved in the carcinogenesis of hepatocellular carcinoma (HCC) and have been used as potential biomarkers for diagnosis and treatment [99]. Finally, apart from proteomics, which is well researched in the development of cancer drugs targeting expressed proteins, genomics research and applications in cancer therapy has a number of key challenges, both scientific and non-scientific, that have to be addressed in order to fully exploit the knowledge of cancer genomics to develop targeted therapeutics and informative biomarkers. The challenges mostly relate to logistical and regulatory issues associated with data acquisition and privacy as well as scientific challenges, such as the incorporation of early preneoplastic or intra-tumoral heterogeneity. The genomic data sharing policy of the National Institute of Health (NIH), for example, requires that publishers only make genomic data available after a period of time post publication or after an embargo period has been expended [84]. These policies, particularly, become bottlenecks in fields, including pediatric cancer research, that deal with rare tumor types and require a quick flow of available data to increase understanding of the disease characteristics and create new and improved theranostic strategies. Furthermore, it has been shown in recent times that genomic data can raise serious privacy concerns among individuals [100,101,102].

Role of Molecular Dynamics Simulation and Experimental Validation

Molecular dynamics (MD) simulations are well-established techniques for investigating the time-resolved motions of biological macromolecules at the atomic level. It provides information about the characteristics, stability, and binding features through fluctuations in the relative positions of atoms in protein and DNA macromolecules as a function of time [103, 104]. An understanding of these molecular motions is important in the area of drug discovery and provides the foundation for experimental validations, helping to refine experimentally determined structures of macromolecules (for instance, biomarkers) and performing conformational analysis and protein homology modeling as well as docking of biomolecular complexes.

MD simulations initially start with an understanding of the biological system followed by the acquisition of structural information from X-ray crystallographic or nuclear magnetic resonance (NMR) spectroscopy. This structural information contains all the possible atoms and their initial spatial coordinates. Publicly available databases such as the protein databank (RCSB PDB—with about 159,230 submitted structures as of January 2020), which collect structures from X-ray crystallography and NMR spectroscopy, are the usual sources for such initial structures [105, 106]. The availability of these structures plays a major role in the modeling of numerous oncotargets, existing ligands, or druggable molecules. Force fields are used to calculate the accelerations of atoms, new coordinates, and velocities at each time-step of the simulation using special software, such as NAMD [107], CHARMM [108], AMBER [109], and GROMACS [110]. Rampogu et al. [111] applied molecular docking and molecular dynamics techniques to obtain a potential ligand that can repress cancer angiogenesis and growth [111]. The study demonstrated the role of MD simulations in predicting the behavior of macromolecules under various conditions in silico. We have recently proposed that aptamers can be a potential oncotargeting molecular probe for targeted inhibition of specific cancer cells [112]. The article also recommended that MD simulations can play a significant role in designing aptamers as probes to specifically target cancer cells for improved treatment and reduced side-effect to the normal healthy cells. Danquah et al. [113] proposed an integrated structural MD simulation approach to design RNA and DNA aptamers for targeting the outer membrane domain of CD19 protein as a biomarker for acute lymphoblastic leukemia [113]. Several other applications of MD simulations have been demonstrated by researchers. For instance, Kumar et al. [114] used MD simulations to detect the phenotypic characteristics of Single Nucleotide Polymorphism. The authors used a long time scale simulation of 200 ns to study the stability of a mutated Aurora-kinase protein and demonstrated the atomic alterations relating to the mutant protein in comparison with the native form using several assessment techniques including RMSD and RMSF that show unique structural conformations and adaptations associated with the expression of cancer. MD simulations have been used to refine top hit molecularly docked protein–protein or protein–ligand complexes in cancer research. Weako et al. [115] successfully examined the stability and efficacy of the unbound form of MetAP2 (an isoform of Methionine Aminopeptidases that is upregulated in various cancers), its complexes with fumagillin, spiroepoxytriazole, and the two best promising compounds (N-(4-[2-(2,5-Dimethylbenzyl)-1-pyrrolidinyl]sulfonyl phenyl and 1-(4-Benzyl-1-piperazinyl)-2-[3-(2-methylphenyl)-1,2,4-oxadiazol-5-yl] ethanone) obtained from molecular docking with compounds from the non-academic version of the OTAVA Chemical Library (https://www.otavachemicals.com). Further, Reddy et al. [116] also used MD simulation to explore the stability of gefitnib (a highly effective kinase inhibitor) and its derivative with Epidermal Growth Factor Receptor (EGFR) after docking in AutoVina [117] and found the best protein–ligand hits based on the highest binding affinity. MD simulations are a highly beneficial tool in drug discovery processes for identifying novel drugs for the treatment of non-small cell lung cancer.

However, it should be noted that computational simulations do not displace experimental methods as both are relevant to realize the full potential of MD simulations in the discovery of cancer oncotargets. Techniques such as surface plasmon resonance (SPR] analysis and circular dichroism spectroscopy are well suited to probe target binding relationships and thus serve as indispensable tools to verify computational results [118]. For instance, Suenaga et al. [119] used SPR analysis to verify the binding free energies of the interaction between the SH2 domain of the growth factor receptor-binding protein 2 (Grb2) and ErbB receptor-derived phosphotyrosyl peptides after performing MD simulations in Amber [120]. The comparative analysis indicated that the computationally estimated binding free energies were larger than that of the experimental data; however, both were highly consistent. Further analysis indicated that the dipartites in the data were as a result of the short simulation timescale of 1 ns and hence can be resolved by performing MD simulations at a higher timescale. Furthermore, Amiri et al. [121] also used various spectroscopic and electrochemical methods in addition to MD simulations to investigate the interaction between three oxovanadium (IV) Schiff base complexes (anticancer drugs) with bovine serum albumin (BSA). Experimentally, the authors used UV–visible absorption spectroscopy, fluorescence quenching, and circular dichroism to study the molecular interaction of the complexes and compared it to the computational data. They observed a good agreement between the experimental and computational results.

Latest Trends in Targeted Cancer Treatments

Currently, there are a significant number of cancer treatment technologies that are under extensive preclinical research or in transition states to clinical studies. Nanomedicines and genetic modification methods are widely employed in lab-based research that targets cancer cells and inhibits their growth. Rao et al. [122] reported that nanosized hyaluronic acid can be a potential therapeutic agent carrier for targeted tumor treatment. The efficient receptor-binding features and the biodegradable and biocompatible properties of hyaluronic acid conjugates make them effective carriers for controlling delivery of drug to target cancer cells. For example, biodegradable hyaluronic acid conjugates have been demonstrated to possess the ability to deliver small interfering ribonucleic acid to targeted cancer cells for treatment [122]. Plant indole-3-acetic acid (IAA) hormone activated horseradish peroxidase (HRP) enzymes are used for in vitro anticancer agents. However, utilizing these enzymes in direct prodrug therapy has led to adverse effects in animal models. The plant enzymes are noticed to contain isoenzyme mixtures with a heterogeneous pattern of glycosylation that are not compatible with human systems. Bonifert et al. [123]. demonstrated that the recombinant variants of HRP can be a potential anticancer drug for targeted treatment of cancer complications. Two HRP isoenzymes C1A and A2A were recombinantly produced in Pichia pastoris strain after α‐1, 6‐mannosyltransferase (OCH1) was knocked out to limit hypermannosylation, producing a glycol-engineered strain. The results revealed that the recombinant isoenzyme HRP C1A is a promising targeted cancer treatment agent [123]. Dong et al. [124] showed that hydrophobic metal–organic framework can be a significant nanoplatform for targeted delivery of anticancer drugs. The authors introduced a novel strategy to fabricate camptothecin@zeolitic imidazolate framework-8 that is modified with Arginine-glycine-aspartic acid (RGD) as a hydrophobic nanoplatform-based drug delivery system. The results showed that the metal–organic nanoplatform framework possesses an excellent pH-mediated controlled and targeted delivery properties and improves the intracellular generation of reactive oxygen species (ROS) to inhibit cancer growth [124]. Gavande et al. (2016) discussed the potential of DNA repair targeted therapy as a next-generation cancer treatment method. It is worthy to note that cancer cells have the ability to repair DNA damages that are induced by therapies, affecting the efficacy of treatment modalities. Thus, the development of anticancer agents to target pathways of DNA and protein repair is useful for targeted and efficient cancer treatment compared to conventional chemotherapy [125]. Wang and Mooney [126] recently described biomaterial-assisted immune cell modulation as a potential targeted cancer treatment method. They reported that dendritic cells, tumor-associated macrophages, B cells, T cells, myeloid-derived suppressor cells and natural killer cells can be targeted and modulated using immuno-materials as a cell-targeted immunomodulated treatment strategy. These biomaterial-based targeted cancer treatment method can address several existing treatment limitations, including side effects and reduced patient response and are a promising replacement for chimeric antigen receptor-T cell therapy and immune checkpoint blockade therapy [126]. Even though, nanomedicines and genetic modification methods are proposed to be promising, there are some limitations. The lack of risk assessment procedures, and in vivo studies showing alterations in the anticancer ability and potential toxicity-based side effects are some of the challenges hindering the transition of these treatment approaches [127, 128]. The combinations of the latest, a novel and advanced aptamers and CRISPR technologies along with pre-existing methods may improve the cancer targeting ability and address some of the existing limitations.

There are various novel oncotargets and methods that are under preclinical and clinical trial to demonstrate efficacy for targeted cancer treatment applications. Vassilev et al. [129] validated the toxicity and biodistribution of chimeric oncolytic adenovirus named ONCOS-102 to obtain a regulatory approval for Phase I clinical studies. The toxicology and biodistribution profile were evaluated using 300 hamsters and revealed no adverse effects on body weight, hematology, food consumption, histopathology, clinical chemistry parameters and bio-accumulation even after repeated administration of ONCOS. The study confirmed that ONCOS-102 is highly safe with enhanced double targeting and granulocyte–macrophage colony-stimulating factor (GM-CSF) expressing property for advanced NCT01598129 cancer-type therapy [129]. Massacesi et al. [130] showed that phosphoinositide 3-kinase (PI3K) inhibitor, which inhibits the PI3K enzyme in PI3K/protein kinase B (Akt)/mTOR pathway, responsible for cell cycle regulation, can be a novel targeted anticancer agent. These inhibitors have been formulated into pan-PI3K buparlisib inhibitor (BKM120) and PI3Kα-selective alpelisib inhibitor (BYL719) by Novartis Oncology as a superior anticancer agent [130]. The article reported some glitches in the clinical development of PI3K inhibitors and the challenges in overcoming target enrichment and stratification based on the activation state of PI3K pathway [130]. Mirzaei et al. [131] discussed the potential of boron neutron capture therapy as an important targeted cancer treatment modality. This therapy is carried out by irradiating a stable boron-10 isotope with low thermal energy neutrons to obtain stripped nuclei of lithium-7 and helium-4, which eventually causes toxic reactions in cancer cells. The clinical and preclinical studies of this therapy showed efficacy in curing head, neck, thyroid cancer, melanoma and brain tumors [131]. Furthermore, polyethylene glycol (PEG) formulated anticancer drugs named as PEGylated advanced drug delivery systems are also under clinical development for effective targeted cancer treatment. The retention time enhancement property of PEGylation protects therapeutic agents such as enzymes, liposomes, small molecular drugs, proteins and nanoparticles from degradation in biological fluids. Additionally, PEGylation is proven to alter the pharmacokinetics of anticancer drug to enable adjustment to target site pathophysiological environment without any toxic reactions [132]. Li et al. [133] revealed via various cell and animal studies that combinations of the cancer stem cell targeting agents and chemotherapy are highly beneficial targeted cancer treatments. These targeting agents, along with chemotherapy, possess the ability to block signaling pathways of self-renewal, reduce drug and ATP-binding cassette transporter efflux expression, promote cancerous stem cell differentiation and modulate epigenetic aberrations. Improvements in the pharmacokinetics, efficacy and safety of combinatorial therapy are the objectives of ongoing research in this area [133]. Belfiore et al. [134] discussed that ligand-functionalized liposomes can be used as a targeted cancer agent. It is noteworthy that ligand-directed liposomes can be engineered to possess active targeting ability toward cell receptors of the tumor cell surface, facilitating uptake of anticancer drugs into the tumor or tumor-associated stromal cells with enhanced selectivity [134]. However, translation from preclinical to clinical studies faces several hurdles, including cytotoxicity, ineffective characterization of active anticancer drugs, and the nonexistence of in vivo preclinical tumor models to evaluate the performance of anticancer drugs.

In spite of various hurdles from lab to preclinical and clinical studies, there are few targeted cancer treatments and prevention methods that are currently available on the market. Cancer vaccines such as T cell vaccines (TPIV200) from Marker Therapeutics, Inc. are currently available on the market to prevent ovarian and breast cancers. This vaccine consists of five peptide antigens of folate receptor alpha (FRα), which is overexpressed in the surface of ovarian and triple-negative breast cancer cells [135]. Similarly, Panacea Pharmaceuticals and 3 M Drug Delivery Systems recently developed a cancer vaccine, which can prevent cancers via the hollow microstructure transdermal system. This microstructure system is proven to possess enhanced ability to deliver cancer vaccines into highly vascularized dermis with enhanced reproducibility of direct delivery [136]. Hill et al. [137] reported that imatinib, lapatinib, erlotinib and sorafenib are the four types of tyrosine kinase inhibitors available on the market for the targeted treatment of cancer. However, economic barriers affect the mass production of these drugs [137]. Moreover, antibody–drug conjugates such as Polatuzumab Vedotin-piiq, Brentuximab Vedotin, Ado-trastuzumab emtansine, Inotuzumab ozogamicin, and Gemtuzumab ozogamicin are approved by FDA and are available on the market for targeted inhibition of cancer cells [138]. Additionally, Herceptin® for targeting human epidermal growth factor receptor-2 protein for breast and stomach cancer treatment [139], Zelboraf® for targeting BRAF protein for metastatic melanoma treatment [140], and Gleevec® for targeting BCR-ABL protein for promoting leukemia cell growth [141] are anticancer drugs that are currently available on the market for effective targeted cancer treatment.

Future Outlook

It is speculated that the combination of genomic, proteomics, bioinformatics, and MD simulations, along with advancements in the field of nanomedicine and genetic engineering, will be beneficial in developing the next-generation targeted treatment modalities for cancer [142]. Bioinformatic tools are currently used to probe the genomics as well as proteomics of cancer patients and will be essential in designing and evaluating the molecular efficiency of patient-specific targeted treatment strategies. The combinatorial efforts of bioinformatics and MD simulations form a novel computational oncological field, which reduces the number of experiments and ease the optimization process via in silico approaches [143, 144]. The development in these computational oncological approaches has led to the emergence of machine learning methods to facilitate precise and controlled delivery of cancer drugs at the target site. Ding et al. [145] suggested that the combination of omics data and machine learning approaches can be useful in precision oncology to identify specific tumor-treating drugs and prescribe optimal regimens for clinical treatment. They also reported that machine or deep learning approaches are beneficial in extracting genomic information and train classifiers to predict optimal cancer drugs for improved therapy [145]. Bibault et al. [146] recommended that big data and machine learning will be beneficial in combining electronic health record and genomic phenotypic profiles to generate high-quality precision medicine in radiation oncology [146]. Similarly, Majumder et al. [147] showed that machine learning methods can help in the chemical screening of cancer-cell-specific lethality via a network of degrading specific protein target. In addition, they discussed that machine learning-based computational chemical screening approaches are useful in recognizing pathways of protein-targeted networks, the anticancer potency of specific compounds with enhanced cell-specific activity, and drug combinations for targeted inter- or intra-type heterogenous cancer treatment [147].

The introduction of nanotechnology-based approaches has shifted the conventional therapeutic approaches of cancer into the molecular and genetic level, targeting cancer cells without altering the metabolism of normal healthy cells. Nano-aptameric sensors have been developed for early and rapid diagnosis of cancer [148, 149]. Dehghani et al. [150] discussed the ability of aptamer-based biosensors and nanosensors in monitoring vascular endothelial growth factor (VEGF) for the effective cancer diagnosis [150]. Further, Hao et al. [151] showed that aptameric graphene-based nanosensors are beneficial for high-sensitivity detection of biomarkers of lung cancer with improved stability [151]. Liu et al. [152] revealed that multifunctional aptamer-based nanoparticles are highly potent for targeted delivery of cancer drugs against circumvent resistant cancer cells [152]. Ouyang et al. [153] recently demonstrated via in vitro and in vivo studies that DNA nanostructures with guidance control and warhead can be highly beneficial in the aptamer-mediated delivery of anticancer drugs into targeted cancer cells via precision-guided missile approach [153]. Thus, nanoparticles with aptamers that are designed via computational approaches are proposed to a promising next-generation strategy for cancer theranostic applications.

In recent times, nanorobots are recommended for targeted cancer theranostic applications. Nanorobots designed with nanoparticles, anticancer drugs, and biosensors can detect cancer biomarkers released by specific cancer cells, navigate to the target cell, internalize, and release the drug payload to inhibit cancer cell growth [154]. Li et al. [155] reported a novel DNA nanorobot that can function as a self-assembled cancer treatment agent via in vivo molecular trigger. The nanorobots are fabricated using a DNA origami method with nucleolin-targeting DNA aptamer and thrombin protein for programmed delivery of the drug payload at the tumor-associated endothelial cells. The nanorobots can deliver the thrombin in tumor-affected blood vessels after intravenous administration to cause necrosis and inhibit tumor growth by inducing intravascular thrombosis. Further, the nanorobots proved to be immunologically safe in mice and Bama miniature pigs, showing the efficacy of these nanosized molecular robots in targeted cancer therapy [58]. Iron-palladium-based mobile magnetic nanocatalysts [156], intelligent DNA nanorobot with enhanced protein lysosomal HER2 degradation [157], and magnetically actuated multifunctional nanoplatforms [158] are the other nanorobots that are under extensive studies for targeted cancer cell inhibition. Correspondingly, multicompartmental nanoformulations are introduced as a mechanism to encapsulate multiple anticancer drugs in a single platform for controlled release programs. Such a system may contain anticancer phytochemicals, nanoparticles, nanosized conventional enzyme inhibitors and genes in a single polymer formulation such as dendrimers and coated with aptamers for programmed targeting and release of drug payloads [159,160,161]. This novel nanoformulation concept along with nanorobotics and nanomedicine approaches can be a highly efficient, targeted cancer treatment opportunity to improve existing cancer treatment methods.

Conclusion

Cancer continues to be a major life-threatening disease among humans. The complex process of identifying the type of cancer, delivering anticancer drugs to target cancer cells, avoiding toxicity toward healthy cells, and the prevalence of lateral cancer emergence are some of the challenges hindering complete eradication of the disease. The discovery and development of oncotarget-based drugs could offer promising opportunities that would catalyze the treatment of cancer. Computational in silico techniques such as genomics, proteomics, and bioinformatics as well as MD simulation tools for probing ligand-target interactions play a key role in identifying and designing targeting strategies and smart cancer drug formulations and their interactions with cancer and tumor cells. Additionally, the combination of in silico methods with treatment methods such as chemotherapy, nanomedicine, or gene therapy would offer improved treatment modalities. However, challenges including chronic toxicity to healthy cells and chances for cancer re-emergence via a minimal residual disease (MRD) must be addressed to create more promising oncotarget-mediated treatment approaches.