Antibody-Drug Conjugates: Possibilities and Challenges
-
Nejadmoghaddam, Mohammad Reza
-
Nanotechnology Research Center, Faculty of Pharmacy, Tehran University of Medical Sciences, Tehran, Iran
-
Nanobiotechnology Research Center, Avicenna Research Institute, ACECR, Tehran, Iran
-
Minai-Tehrani, Arash
-
Nanobiotechnology Research Center, Avicenna Research Institute, ACECR, Tehran, Iran
-
Ghahremanzadeh, Ramin
-
Nanobiotechnology Research Center, Avicenna Research Institute, ACECR, Tehran, Iran
-
Mahmoudi, Morteza
-
Nanotechnology Research Center, Faculty of Pharmacy, Tehran University of Medical Sciences, Tehran, Iran
-
Dinarvand, Rassoul
Nanotechnology Research Center, Faculty of Pharmacy, Tehran University of Medical Sciences, Tehran, Iran, Tel: +98 21 64121014, 22432020, Fax: +98 21 66959052, 22432021, E-mail: dinarvand@tums.ac.ir
-
Nanotechnology Research Center, Faculty of Pharmacy, Tehran University of Medical Sciences, Tehran, Iran
-
Department of Pharmaceutics, Faculty of Pharmacy, Tehran University of Medical Sciences, Tehran, Iran
-
Zarnani, Amir-Hassan
Reproductive Immunology Research Center, Avicenna Research Institute, ACECR, Tehran, Iran, Tel: +98 21 64121014, 22432020, Fax: +98 21 66959052, 22432021, E-mail: zarnania@gmail.com
-
Department of Immunology, Faculty of Public Health, Tehran University of Medical Sciences, Tehran, Iran
-
Reproductive Immunology Research Center, Avicenna Research Institute, ACECR, Tehran, Iran
-
Immunology Research Center, Iran University of Medical Sciences, IUMS, Tehran, Iran
Abstract: The design of Antibody Drug Conjugates (ADCs) as efficient targeting agents for tumor cell is still in its infancy for clinical applications. This approach incorporates the antibody specificity and cell killing activity of chemically conjugated cytotoxic agents. Antibody in ADC structure acts as a targeting agent and a nanoscale carrier to deliver a therapeutic dose of cytotoxic cargo into desired tumor cells. Early ADCs encountered major obstacles including, low blood residency time, low penetration capacity to tumor microenvironment, low payload potency, immunogenicity, unusual off-target toxicity, drug resistance, and the lack of stable linkage in blood circulation. Although extensive studies have been conducted to overcome these issues, the ADCs based therapies are still far from having high-efficient clinical outcomes. This review outlines the key characteristics of ADCs including tumor marker, antibody, cytotoxic payload, and linkage strategy with a focus on technical improvement and some future trends in the pipeline.
 
Introduction :
Similar to conventional cancer treatments such as chemotherapy and radiotherapy, antibody immunotherapy and targeted therapies based on nanoparticulate structures are not safe and efficacious as often claimed; therefore, alternative therapies are urgently needed. In this regard, Antibody Drug Conjugates (ADC) technology that could bring forth a new generation of cancer therapeutics was the main focus of this study. ADCs are monoclonal antibodies (mAbs) connected by a specified linkage to antitumor cytotoxic molecules. The main components of an ADC and mechanism of its action are further demonstrated in figure 1.
In ADC technology, the specificity of an antibody for its immunogenicity is exploited to home a chemically supertoxic agent into tumor cells, while administration of unconjugated drug alone is not suitable due to its high toxicity. Therefore, ADCs can be further defined as prodrugs requiring the release of their toxic agent for their activation that commonly happens after ADC internalization into the target cell 1. From the standpoint of nanomedicine, the antibody in ADC structure acts as a self-targeting nanoscale carrier 1-3, thus, it could overcome the issues associated with nanomedicines based on synthetic nanomaterials such as cellular internalization, clearance, sterical hindering of binding to the epitopes and failing to release into targeted cells 4.
The first experimental design on ADC subject dates back to more than 50 years ago 5. However, the use of ADCs for cancer therapy has achieved considerable success in recent years after the introduction of four clinically approved ADCs such as Brentuximab vedotin 6,7, Trastuzumab emtansine 8-11, Inotuzumab ozogamicin 12 and Gemtuzumab ozogamicin 12,13 used for the treatment of patients with lymphoma (HL and ALL), HER2-positive, CD22-positive AML and CD33-positive ALL cancers, respectively. Likewise, a great deal of effort has also been made by the pharmaceutical companies to overcome the technological barriers associated with ADCs 14,15, whereby there are 160 ADCs undergoing preclinical trials 16 and 70 more under various stages of clinical evaluation (Table 1).
Clinical efficacy of the ADCs arises following accurate selection of four parameters including tumor targeting, antibody, cytotoxic payload, and method of antibody linkage to the payload. The precise selection of each parameter can be achieved through the knowledge gained from the previous studies and established ADCs, and is discussed here.
Tumor markers in ADCs
The important aspects of tumor markers in ADCs are demonstrated in figure 2. An antigen with expression pattern slightly greater in tumor cells compared to healthy cells is sufficient to induce ADC activity. However, like other targeted drug delivery systems, the number of cell surface tumor markers can be a key determinant of ADC activity 17. The targets for ADC do not necessarily intervene in cell growth. ADCs tumor-suppressive function is mainly mediated through tumor marker potency for ADC internalization compared to the inhibition by blocking the cell growth 1,18-20. However, target biological roles such as those involved in cell division pathway (e.g. CD30 and CD70 tumor necrosis factor signaling) can be considered as an advantage for ADC efficacy. Accordingly, the currently employed targets and their biological roles are listed in table 1.
For instance, glembatumumab vedotin is an ADC against an extracellular domain of non-metastatic B melanoma-associated glycoprotein (GPNMB) that is aberrantly expressed in various carcinoma including hepatocellular 21, melanoma 22, gliomas 23, and two specific breast cancer types, Basal-Like Breast Cancer (BLBC) and Triple Negative Breast Cancer (TNBC) 24. The GPNMB do not represent a high relative level of expression in all aforesaid carcinoma. One important property that may make GPNMB a potential therapeutic target for ADCs, originates from its biological role in MAPK/ERK pathway, as GPNMB expression can be upregulated by MAPK/ERK inhibitors 25.
From the structure standpoint, a relevant antigenic determinant on cell surface membranes, termed Extracellular Domain (ECD), is required as an immunizing agent for antibody generation 19. However, the potential of ECD to be shed into the circulation must be considered. The shed ECDs can potentially bind to ADC and consequently reduce the targeted delivery into the tumor cells 19.
A further concern in the selection of the target for ADC is related to the homogeneity or heterogeneity expression of the tumor marker on the tumor cell surface. Homogenous expression of the tumor targets has been demonstrated to be more in favor of ADC targeting than those expressed heterogeneously 26. However, heterogeneous antigen expression can particularly be beneficial for those ADCs that possess bystander killing activity 26-28. Bystander killing activity is referred to the potency of therapeutics delivery system in killing neighboring cells independently of targeted therapy assignment. This effect can be raised through reactive oxygen species or some cytotoxic metabolites that may be excreted from the tumor-targeted cells 26-29. As a result, recycling capability of a tumor marker would enhance bystander killing activity as it may promote leakage of ADC and metabolites to the neighboring cells. However, according to the reports, an extra recycling property is not desirable as in further Bystander activity (Ba), the greater side effects are predicted 30,31.
The promising future of the ADCs supports extensive studies to look for a potent ADC target with a wide range of expression, from earliest cell recognizable lineage to maturation. This represents an exquisitely selective target that covers all types of malignancies. CD19 is a good example of such target that is highly expressed in B-cell and the vast majority of Non-Hodgkin lymphomas (NHLs), and B-cell Acute Lymphoid Leukemia (B-ALL) (99%) 7,32-35. As shown in table 1, CD19 has been marked as a target to produce ADCs, including SAR3419 7,34,35, SGN-CD19A 32, MDX-1206 36, and ADCT-402 33.
Antibodies in ADCs
Antibody component in ADCs undertakes both roles including being a carrier and targeting agent. The main aspects of the antibody in ADCs are demonstrated in figure 3. High specificity of targeting and minimal immunogenicity are the main characteristics for Ab component in ADCs. These prevent antibody cross reactions to other antigens, avoiding both toxicity and removal/elimination of the ADC before reaching to the tumor. The high affinity of the Ab for efficient uptake into target cells is another important factor in ADC design 30,54-56. To the best of our knowledge, there is no substantial report about optimal or even minimum required binding affinity (Kd) of antibody component. In figure 4, a binding affinity less than 10 nM (Kd<10 nM) is commonly needed for the Ab component and accordingly for an effective ADC, based on frequency distribution histogram. The affinity of the antibody to its immunogen can affect the property of antibody which is termed as receptor-mediated antibody internalization. Receptor-mediated antibody internalization is a key mechanism underlying antibody endocytosis that is induced through antibody binding to its specific antigen 77. It has been reported that, alternative antibodies against the same immunogen can exhibit different rates of internalization 19. Rapid internalization can raise both ADC efficacy and safety simultaneously, since it reduces the opportunity of the ADC for off-target release 1,98.
In addition to rapid internalization as a prerequisite for an antibody, the route by which antibody is internalized should be also considered, because it can potentially influence ADC processing 99. For instance, Clathrin-coated Pit-mediated receptor internalization (caveolae pathway), at least in some cases, has been reported to traffic ADC to the cells. In caveolae pathway, ADC is directed to the Golgi or endoplasmic reticulum (Non-proteolytic compartments) instead of endosomes or lysosomes (Proteolytic compartment of the cells) 118. ADC’s traffic to the non-proteolytic compartments may impede its proteolytic process to release effective metabolites 6. Antibody capability to induce receptor mediated internalization is somewhat a mandatory requirement in design of new generation of ADCs. Antibody with low internalization rate has no desired therapeutics index even for the tumors expressing high levels of surface antigen 99. To compensate inefficient internalizing of ADC, a much more potent drug and high stable linkage chemistry (linkage between the antibody and drug moiety) are required that would be discussed in next sections.
Optimal pharmacokinetic (PK) properties including longer half-life is another aspect of the antibody component in ADC design 30,54,55. It has been reported that Ab with longer half-life show high elimination and rapid clearance of the ADC in plasma 136. As shown in table 1, it is not compulsory for a mAb itself to represent therapeutic activity in the ADC. However, therapeutic activity of the mAb is a desirable property besides killing activity mediated by the cytotoxic payload 137,138.
Antibody therapeutic activity is usually mediated via immune-mediated effector functions such as Antibody-Dependent Cellular Cytotoxicity (ADCC), Antibody-Dependent Cellular Phagocytosis (ADCP), Complement Dependent Cytotoxicity (CDC), and cytokine signaling modulation in terms of inhibition or induction (Table 1). Such therapeutic activities can be further employed to design ADCs with enhanced cell killing activity 8-11,43,120-123. According to the obtained data in table 1, isotype 1 immunoglobulin (IgG1) seems to be prone to induce immunotherapeutic activity.
In this regard, many attempts have been made to engineer mAbs with therapeutic activity. For instance, the Fc domain affinity of anti-CD19 targeting antibodies for the FcγRIII has been enhanced, either by Fc glycoengineering approaches, e.g. MEDI-55 150 and MDX-1342 151 or amino acid substitution, e.g. XmAb5574 152 and XmAb 5871 or MOR-208 35,153. Such modification resulted in an increase of ADCC activity in antibody. To the best of our knowledge, the above engineered antibodies have not been used for designing ADCs yet. However, there are some reports of ADCs which have employed a combination/fusion of two engineered antibody fragments. Such fusion antibodies are termed as bispecific Antibody (bsAb), while ADCs designed from the bsAbs were named bispecific ADC (bsADC) 154. Blinatumomab and AFM11 are typical bispecific antibodies, two fusions of anti-CD19 scFv and anti-CD3 scFv, which were engineered to enhance CD19-positive cells killing activity through induction of T or NK cytotoxic immune effector cells 35,155. A derivative of blinatumomab has been also constructed to induce the controlled T cell activation, named ZW38 156. The ZW38 was conjugated to a microtubule cytotoxic agent for the preparation of a novel class of bsADC capable of mediating T cell cytotoxicity 156. Another bsADC, B10v5x225-H-vc-MMAE (Monomethyl auristatin E-MMAE), has been recently developed to simultaneously target EGFR and c-MET which are two tyrosine kinases receptors correlated with tumor growth and metastasis 157,158. B10v5x225-H-vc-MMAE contains a bsAb from fusion of anti c-MET Fab fragment and anti-EGFR scFv that was engineered to represent low affinity to EGFR which is a ubiquitous tissue antigen 157. The side effect of B10v5x225-H-vc-MMAE can be avoided to some extent due to attenuated affinity toward EGFR receptors in healthy cells 157. Bridging a rapidly internalizing protein with a tumor specific marker is also another recent method to construct bsAb, e.g., anti HER2 crosslink to prolactin cytoplasmic domain receptor 159 with the ability to improve internalization and cell killing activity of the bsADC.
Cytotoxic payloads in ADCs
Briefly, cytotoxic payloads for new generation of ADCs should meet many of the criteria as outlined in figure 5. Antibody component in ADCs is incapable of carrying a large number of cytotoxic payload due to its structure. Therefore, the cytotoxic payload in the new generation of ADCs must be highly super-toxic to eradicate majority of the tumor cells even with minimal payload delivery 160. The rate of mAb uptake by tumor cells is approximately less than 0.003-0.08% of injected dose per gram in a tumor 54,55. Furthermore, low expression and poor internalizing activity of the most tumor-associated antigens can cause negligible ADC delivery to the tumor target cells. Hence, ADCs equipped with highly super-cytotoxic payload are imperative, because they must show therapeutic effect while having limited release. According to the reports, a highly cytotoxic agent should exhibit an IC50 of about 10 nM or less obtained from an examination with KB cells upon a 24-hr exposure time 30,54,55,161. A highly super cytotoxic payload can be originated from plant, animal or microorganisms; in this regard, the most important issue can be the finding of cytotoxic payloads with negligible immunogenic potential in the body. In new generation of ADCs, such cytotoxic payloads are likely to be chemical anti-cancer drugs since experimental evidence confirmed that they are less immunogenic than glycol/peptide cytotoxic agents when circulating in the blood. Some anticancer drugs such as doxorubicin (DOX), mitoxantrone, and etoposide are impaired under hypoxic condition; a condition appeared in solid cancer cell population 162,163. Hence, needless to say, those drugs may not be considered as cytotoxic payloads.
Taking a look at current cytotoxic drugs (Table 1) shows that they generally affect DNA synthesis or cell division to block cell proliferation (mitosis) 38,98. Monomethyl auristatin derivatives which bind to tubulin and are able to inhibit microtubule assembly/polymerization (IC50=10-500 pM) 32 are the most commonly used cytotoxic drugs in ADC design with approximately 50% share of the field (Table 1). Maytansinoids derivatives (~30%), pyrrolobenzodiaze-pine (~7%), camptothecin analogs (~6%), n-acetyl-γ-calicheamicin (4%), duocarmycin (DUO) (~3%) and doxorubicin (~1%) are the other abundant cytotoxic payloads (Table 1). The above cytotoxic compounds are 100 to10000 folds more potent in vitro than typical chemotherapeutic agents and are chosen based on their different actions on cancer and noncancerous cells. DNA modulators have significant effects on malignant cells as they are divided more rapidly than normal cells 163.
Furthermore, a cytotoxic agent of the ADC is better to be studied in an in vitro condition to determine whether it is a substrate, inhibitor or inducer of metabolizing enzymes (e.g., cytochrome P-450 isozymes (CYPs), and some transporter enzymes like P-glycoprotein) 98. Such studies help to elucidate the in vivo factors that may be contributed to the elimination/enhancement of the cytotoxic agent 27,98,164. New studies to introduce new payloads focused on agents against Tumor-Initiating Cells (TICs) 27,164. Such payloads assist to widen the target area and to circumvent potential resistance of cancer cells. Pyrrolobenzodiazepines (PBDs), derivatives of naturally occurring tricyclic antibiotics, duocarmycins, anthracyclines, α-amanitin (a bicyclic octapeptide from the fungus Amanita), and topoisomerase inhibitors including SN-38 are categorized as TIC payloads 1,164.
Rovalpituzumab tesirine is one example of ADC with PBD as a payload (Table 1), that has been reported to have a potency to eliminate pulmonary neuroendocrine TICs at subpicomolar level in vivo 83.
The cytotoxic payload should be also stable during preparation or storage and circulation in the blood. Cytotoxic payloads that are not fully stable can potentially be converted to undesirable drug forms during conjugation or storage. Solubility of the cytotoxic agent in aqueous solution is another important criterion in ADC design. Antibody is considered a protein and its conjugation to the cytotoxic agent must be performed in aqueous solutions with minimal organic cosolvents 163,165. Extreme hydrophobicity of payload can potentially change antibodies biological properties, resulting in hydrophobic aggregation of the antibody either during conjugation process or storage 163. The hydrophilicity of cytotoxic payloads will affect cell membrane permeability of parent ADC or its metabolites which may also be beneficial in term of bystander activity 17,26,163,166. However, the ability of cytotoxic payloads to form hydrophobic metabolite after intercellular cleavage of ADC is preferable since the metabolites with more hydrophobic group show better blood clearance and safety 165. According to the reports, about 95-99% of ADC molecules are metabolized before binding to tumor cells 160. This may raise safety concern as it can enhance the potential cytotoxic side effects of ADC. Thereby, the use of cytotoxic payloads with well-characterized metabolite profiles can be an advantage to enhance ADC safety in particular 1,2,167-169.
Cytotoxic payload should present a dominant functional group suitable for linkage to the antibody component of ADC 34. If a dominant functional group does not exist on the cytotoxic agent, at least, it should be amenable to modification, in which a desired substituent is introduced on appropriate sites 170.
The copy number and heterogeneity of antigen expression are the other important issues that must be considered in the selection of cytotoxic agent 30,31. More expression of target antigen may be a reason to apply a cytotoxic agent with low potency. Typically, payloads that promote the bystander effect in cancer cells are more desirable to design ADCs directed for the antigens expressed heterogeneously 26.
The ability to choose specified cytotoxic payloads with mechanism of action compatible with standard of care has been reported to facilitate clinical success of the ADCs in biopharmaceutical market. For instance, microtubule disrupting payloads are commonly chemotherapeutic drugs that are used for the treatment of cancers, including breast, ovarian and prostate cancer 54,55 (Table 1). Both availability in the market and reasonable cost can be alternative rationale for choosing a cytotoxic payload in ADC design 1.
Linking cytotoxic payloads to antibodies in ADCs
One of the dynamic research fields in ADC design is the study of the methods that are correlated with antibody conjugation to cytotoxic payloads, as it has a great role on balancing between ADC therapeutic efficacy and toxicity 30,31,54. The key concerns in linkage chemistry are demonstrated in figure 6. Conjugation site on antibody component, a well-defined Drug to Antibody Ratio (DAR), homogeneity and linkage stability are the important parameters that need to be considered in conjugation.
In general, interchain disulfide bridges and surface-exposed lysines are the most currently used residues on the antibody for conjugation to cytotoxic payloads, respectively (>50 vs. >30%) (Table 1). Hydroxyl groups on carbohydrate structures are the other residues in antibodies that have been rarely used as conjugation sites for ADC (The schematic linkage in figure 6 is an example of this strategy) 1,171.
Theoretically, the linkage of cytotoxic payloads to the surface-exposed lysine of mAb occurs after reduction of ~40 lysine residues on both heavy and light chain of mAb 172 and it results in 0-8 cytotoxic payload linkages per antibody and heterogeneity with about one million different ADC species 30,173. Cysteine conjugation occurs after reduction of four interchain disulfide bonds and results in eight exposed sulfhydryl groups. Linking drugs per antibody can differ from zero to 8 molecules, generating a heterogeneous population of ADC (Greater than one hundred different ADC species) 30.
Due to low stability and safety properties of the pharmaceutical products with heterogeneous contents, they are complex to be accurately predicted in terms of efficacy or therapeutic window 27,30. Therefore, improvement of conjugation methods to achieve homogeneous ADC is very crucial.
In this case, it is possible to reduce just two of four interchain mAb’s disulfide bonds of cysteine residues through carefully mild reduction conditions, as interchain disulfide bridges are more prone to reduction than intrachain disulfide bridges 171,174,175. However, such mild reduction is not easily possible in practice and a diverse number of cysteines may be reduced (0-4), resulting in a heterogeneous mixture of ADC 30,173. Hence, the production of homogeneous ADCs through payload conjugation with native residues can be laborious. To overcome this limitation, many site-specific conjugation approaches have been developed, in which a known number of cytotoxic payloads are constantly conjugated to defined sites on mAbs. Some of the approaches are explained below:
1. A conjugation through engineered cysteine residues that neither damages antibody fab region nor interferes with Fc-mediated effector functions, called THIOMAB technology 173,176. In THIOMAB technology, the heavy chain alanine 114 is substituted with two or more reactive cysteine residues at a predefined site for conjugation with cytotoxic payload 173. Anti-TENB2 ADC is an example that is prepared by THIOMAB technology and is currently in phase I trial (Table 1).
2. Re-engineering of mAb is able to incorporate with unnatural amino acids, e.g. selenocysteine 177, acetyl-phenylalanine 178, and para-azidomethyl-l- phenylalanine 42.
3. Site-specific enzyme-mediated conjugation to genetically engineered antibody is as follows:
Incorporating a thiolated sugar analogue, 6-thiofucose, to the antibody carbohydrate that introduces new chemically active thiol groups using fucosyltransferase VIII 179,
Providing a ketone reactive group on antibody glycosylation site by glycotransferases 180,
Introducing an aldehyde reactive group on the antibody using sialyltransferase 181 or formylglycine-generating enzyme 182,
Genetically introducing specific glutamine tags to antibody whereby payloads with a primary amine group can be linked to the γ-carbonyl amide group of glutamine tags. Such reaction is catalyzed by a microbial transglutaminase which is capable of recognizing glutamines tags from naturally glutamines residues 73,183-185,
Providing LPXTG tagged antibodies (A penta-peptide as a substrate for transpeptidation reaction) as specific linkage sites for the oligo-glycine-containing payloads, which are mediated by Staphylococcus aureus Sortase A enzyme 186,
Conjugation of phosphopantetheine-linked payloads to the serine residues of the peptide-tagged antibodies via phosphopantetheinyl transferases catalysis 187,
4. Chemoenzymatic site direct conjugation, e.g., providing two azide groups at asparagine 297 (Asn-297) residue in antibody constant region (Fc) is linked with cytotoxic payloads using copper-mediated click reaction 188. The azide functional groups are formed in a selective hydrolysis reaction that is mediated by an Endo-beta-N-acetylglucosaminidase (EndoS) chemo-enzyme.
ADC as a potential targeted delivery system must be passed through all hurdles, including blood circulation, antigen binding, internalization, payload release, and eventual payload action. An unstable linkage can lead to premature release of the payload, before reaching the site of action 98. Therefore, reasonable chemical stability must be considered in the design of chemical linkage between cytotoxic payload and antibody.
Although a direct linkage between cytotoxic and antibody components has generally shown more stability in circulation 1,98, conjugation reactions are mostly created with linkers in comparison with direct linkage between cytotoxic and antibody component (Table 1). The choice of proper linkers has been discussed in the related publications devoted to the progress of ADCs 30,31,54,189,190. As shown in table 1, about 50% of the ADCs are using Valine-Citrulline peptidyl (VC) linker. N-succinimidyl 4-(2-pyridyldithio) butyrate (SPDB) (18%), acid-labile hydrazine (10%), maleimidomethyl cyclohexane-1-carboxylate (MCC), maleimidocaproyl (MC) (10%), N-succinimidyl 4-(2-pyridyldithio pen-tanoate (SPP) and carbonate (3%) linkers are other employed linkers.
Limited drug-linker designs for more than 70 current ADC clinical trials (Table 1) are a dilemma regarding linkage chemistry that may restrict simultaneous development of ADCs against both hematological and solid tumors. Generally, the properties of linkers can be altered by the cytotoxic payload release mechanism 191. Cytotoxic payload in ADC technology must be released into the cell to exert its therapeutic activity, thus ADC linkers should be chosen based on their stability to keep ADC intact during circulation and capable of cleaving inside the targeted cell 191,192. Linker stability is defined based on lack/low level of cleaving agents (e.g., protease or reductive agents) in the bloodstream compared to the cytoplasm 163.
The current linkers used in ADCs are also broadly classified as cleavable and noncleavable linkers based on where they are cleaved into the cytoplasm. Cleavable linkers are those containing a conditional cleavage sites sensitive to be cleaved immediately after ADC internalization, such as VC, SPDB, SPP, and hydrazine which can be triggered through protease reactions, glutathione reduction, and acidic pH, respectively 163,164. Noncleavable linkers are stable from early to late endosome transition and their cytotoxic partner is just released by degradation of antibody in lysosomes, e.g. MCC and MC linkers that link Ab to the payload via thioether linkage 190.
Characteristics of ADC target such as copy number, internalization rate and level of homogeneity should be considered in conjugation method and linker selection. For instance, ADC with disulfide-linkage has been shown to have more cytotoxic activity than the same ADC with thioether linkage when they were directed to the tumor cell lines expressing a low copy number of targeted antigen 17.
Cleavable linkers may increase the possibility of bystander effect 27. Hence, it is logical to use cleavable linkers in designing ADCs directed for the antigen that is heterogeneously expressed in tumors 26.
In vivo adverse effects of ADCs are influenced by the use of cleavable or noncleavable linkers. As in the case of tubulin inhibitor payloads, which is linked through cleavable linkers to the antibody component, e.g. SPDB-DM4 (Ravtansine-DM4), or VC-MMAE, peripheral neuropathy can be frequently observed, whereas noncleavable linkers often trigger hematological toxicity, possibly due to an increased dose and interactions with Fcγ receptors on hematopoietic cells 164.
The type of linker plays an important role in ADC catabolite products with regard to processing into targeted cells or metabolizing by clearance mechanisms. The type of ADC catabolites may influence some ADC features such as IC50, Maximum Tolerated Dose (MTD) 192,193, and kill Multidrug Resistance (MDR) expressing cells 192,194.
 
Conclusion :
ADC is considered exciting and promising antibody-based therapeutics to improve cancer therapy. Growth in the number of registered ADCs in clinical trials (Table 1) represents the pharmaceutical industry interest in investment for research and development in the field, as it has been stated by others 14,15.
The design of an ADC might seem to be not very complex, while several issues must be taken into consideration to complete ADC’s potential as a therapeutic agent for cancer. This might be the main reason for the condition that small number of ADCs have reached the market (Table 1). The major issues associated with the development of ADCs seem to be originated from the factors that interfere with ADCs efficacy and off-target cytotoxicity. The precise selection of all four parameters, i.e. tumor marker, antibody, cytotoxic payload, and linkage strategy would be required to prepare a successful ADC.
With regard to ADC tumor markers, they do not have to be involved in tumor growth 1,18,20,31. Therefore, ADC can present therapeutic application in a broad range of tumors. However, an ADC tumor marker should meet at least three criteria of considerable expression level in tumor cells vs. normal cells, presenting cell surface immunogen, and being capable of performing ADC internalization.
High specificity, adequate affinity, and receptor-mediated internalization are the major aspects of antibody choice. Efforts to optimize antibody component would be a great idea to translate into improved ADCs. In fact, some major ADCs’ weaknesses including, low efficiency 156, low internalization 159, off-target effect due to the target expression in normal tissues 157, and heterogeneity expression of the target in the tumors can be overcome via antibody improvement. Antibody engineering technology for production of alternative bsAbs to design more efficient ADCs (bsADCs) has been proven in several preclinical models 156,157,159. The rationale behind this technology is the fact that the aforesaid ADC’s weaknesses can be solved through ADC designs (bsADCs) operating from improved antibody (bsAb) in terms of affinity, specificity, internalization activity, by enhancing the therapeutic activity or decreasing ADC’s side effects.
Another main concern in the development of ADCs is related to the study of finding cytotoxic payloads that are potent enough with confined DAR (Up to 7 drugs per antibody) 195 to exert therapeutic activity. Having reasonable aqueous solubility, non-immuno-genic, as well as stability in storage and bloodstream is a common criterion for choosing cytotoxic payloads.
In contrast, the introduction of innovative methods to modify ADCs cytotoxic payloads with versatile functional groups (e.g. thiol, amine groups) is the other interesting subject, as it eases the conjugation process.
One further challenge of ADCs is associated with the limitation of linkage and conjugation chemistry to link an optimized number of the payloads to the antibody in predefined location homogeneously.
Interdisciplinary and multidisciplinary works and related studies such as recombinant DNA technology, bioconjugation, and chemistry are the hopeful strategies to get the purpose of achievement in site-specific conjugation and homogeneous ADCs 73,173,176-187,196,197.
Based on promising reports from research to synthesize homogeneous ADCs, it is likely that the first ADC products constructed using site-specific conjugation will be made for cancer therapy that may hold the promise about the future use of ADCs.
Taken together, despite challenges in ADC design, the future of ADCs seems to be much promising as more clinical trials and basic researches conducted on existing ADCs would pave the way to tackle issues regarding tumor marker, antibody, cytotoxic payload, and linkage strategy.
 
Acknowledgement :
This review study was supported as a Ph.D., program by a grant from Nanotechnology Research Center, Faculty of Pharmacy, Tehran University of Medical Sciences (TUMS) (grant no. 92-03-159-25467). We further acknowledge the numerous labs, authors, and publications that we were unable to cite in this review due to space restrictions.
 
Conflict of Interest :
The authors declare that they have no competing interests.
 
Figure 1. Schematic representation of ADC, showing the main components of an ADC and its cell cytotoxicity mechanism. Clinical efficacy of ADCs is determined by fine-tuning combination of tumor antigen, targeting antibody, cytotoxic payload and conjugation strategy (a). ADC binds to tumor target cell surface antigens (b) leading to trigger a specific receptor mediated internalization (c). The internalized ADCs are decomposed to release cytotoxic payloads inside the tumor cell either through its linkage/linker sensitivity to protease, acidic, reductive agents or by lysosomal process, leading to cell death (d).
|
Figure 2. Main considerations in selecting tumor markers for ADC design and development.
|
Figure 3. Main considerations in producing antibodies for ADC design and development.
|
Figure 4. Kd frequency distribution (a) and histogram data (b) of current ADC in clinical development (Table S1, n=13). Antibody affinities (Kd) that have been used in current ADC in clinical development were classified into either ≤10 nM or ≥10 nM groups. The average Kd and standard deviation of ≤10 nM group was 1.12 and 1.3 and for ≥10 nM group was 39.9 and 28.2, respectively. Median Kd of ≤10 nM group and ≥10 nM groups was 0.7 and 40.5, respectively. Average Kd was significantly different between two groups (p<0.05). The frequency distributions of Kd in ≥10 nM groups are more than ≤10 nM groups (a).
|
Figure 5. Main considerations in choosing cytotoxic payloads for ADC design and development.
|
Figure 6. Main considerations for linking cytotoxic payload to antibodies in ADC design and development.
|
Table 1. Current ADCs in clinical development based on targeting antigens with an overview of their properties
Not available (n/a), Relapsed B-cell non-Hodgkin's lymphoma (B-NHL), Acute myeloid leukemia (AML), Mertansine (DM1), Calicheamicin (calich.), N-succinimidyl 4-(N-maleimidomethyl) cyclohexane-1carboxylate (SMCC), Hydrazone acetyl butyrate (AcBut), Uterine Serous Carcinoma (USC), Tumor-Associated Antigen (TAA), Valine-citrulline-seco (vc-seco), Renal Cell Carcinoma (RCC), clear cell Renal Cell Carcinoma (ccRCC), Mantle-Cell Lymphoma Diffuse (MCLD), Non Small-Cell Lung Cancer (NSCLC), Receptor tyrosine kinases (RTKs), Recurrent Glioblastoma Multiforme (GBM), Transmembrane Protein (TP), CD27 ligand (CD27L), Epidermal growth factor receptor variant III (EGFRvIII), Glioblastoma multiforme (GBM), Epithelial Ovarian Cancer (EOC), Head and Neck Squamous Cell Carcinomas (HNSCC), Auristatin F-hydroxypropylamide (AF-HPA), Polyacetal-based polymer (Fleximer®), Non-natural amino acid linker para-acetyl-phenylalanine (pAcF), Amberstatin, a short polyethylene glycol (PEG) spacer terminated by an alkoxyamine (AS269).
|
Contd table 1.
B Cell Receptor (BCR), Chronic Lymphocytic Leukemia (CLL), Prostate-specific membrane antigen (PSMA), Maleimido-[short PEG]-Lys- PABOCO-20-O (CL2A), Metastatic colorectal cancer (mCRC), Carcinoembryonic Antigen Related Cell Adhesion Molecule 5 (CEACAM5), Trophoblast cell-surface antigen 2 (Trop-2), Tumor-Associated Calcium Signal Transducer (TACSTD2), Gastric Antigen 733-1 (GA733-1), Malignant Pleural Mesothelioma (MPM), Platinum-resistant ovarian cancer (P-OC).
|
Contd table 1.
Target sodium phosphate transporter 2b (NaPi2b), Transmembrane cell surface receptor guanylyl cyclase C (GCC), Delta-like protein 3 (DLL3), polyethylene glycol spacer (PEG8), Selective Catalytic Reduction (scr), Metastatic Urothelial Cancer (MUC), B-Cell Maturation Antigen (BCMA), DX-8951 a derivative of the camptothecin analog exatecan (DXd).
|
Contd table 1.
Folate receptor 1(FOLR1), Maleimidocaproyl-valine-citrulline- (MC-VC-PABC), Carbohydrate antigen 125 (CA-125), Mucin 16 (MUC16), A high molecular weight mucin-type glycoprotein (CanAg), Erythropoietin producing hepatoma A2 receptor (EphA2 or EPHA2), Ectonucleotide pyrophosphatase/phosphodiesterase family member 3 (ENPP3), Poliovirus receptor related protein 4 (PVRL4), 2 N-terminal Leucine-Rich Repeat (LRR), Human Tissue Factor (TF), Stem Cell Factor Receptor c-Kit (SCFR).
|
Contd table 1.
Hepatocyte Growth Factor Receptor (HGFR), Structural homolog of the urokinase-type Plasminogen Activator Receptor (uPAR), Tumor-associated antigen (C4.4a), Lung Squamous Cell Carcinoma (LSCC), Fibroblast growth factor receptor type 2 (FGFR2), Ovarian Cancers (OC), Trophoblast Glycoprotein (TPBG), metastatic Castration-Resistant Prostate Cancer (mCRPC), -transmembrane epithelial antigen of the prostate-1 (STEAP1), Anti-solute carrier family 39 zinc transporter member 6 (SLC39A6; LIV-1; ZIP6), Anti-Endothelin B Receptor (ETBR), Auristatin-0101 (Aur0101).
|
Contd table 1.
Lymphocyte antigen 6 complex locus E (Ly6E), Antigen-Presenting Cell (APC), α subunit of the interleukin-2 receptor (IL-2R alpha), Lysosome-Associated Membrane Protein 1 (LAMP1).
|
|