Antibody-drug conjugates in solid tumors; new strategy for cancer therapy

Abstract Antibody-drug conjugates (ADCs) have emerged as a novel class of anticancer treatment. ADCs are composed of three parts: a monoclonal antibody, a linker and a payload. A monoclonal antibody binds to the specific antigen present at the cancer cells, allowing selective delivery of the cytotoxic agents to the tumor site. Several ADCs are approved by the US Food and Drug Administration for the treatment of hematologic cancers and solid tumors with clinically meaningful survival benefit. However, the development of ADCs faces a lot of challenges and there is a need to get better understanding of ADCs in order to improve patient outcomes. Here, we briefly discuss the structure and mechanism of ADCs, as well as the clinical data of current approved ADCs in solid tumors.


Introduction
The current systemic therapeutic agents against solid tumors comprise chiefly chemotherapy, targeted therapy, immune checkpoint inhibitors (ICIs).Although these therapies have improved outcomes, better ways of treatment are expected.Antibody-drug conjugates (ADCs) are rapidly emerging class of therapeutic agents combining cytotoxic drugs and targeted antibodies via a chemical linker (Fig. 1).The concept of targeted chemotherapy was first proposed by a German scientist, Paul Ehrlich 100 years ago, known as 'magic bullet'.
The US Food and Drug Administration (FDA) firstly approved Gemtuzumab ozogamicin (Mylotarg ® ) in 2000 for the treatment of adults with CD33-positive acute myeloid leukemia (1).As for solid tumors, trastuzumab emtansine (T-DM1) was approved by FDA in 2013 for the treatment of metastatic breast cancer.As of 1 November 2023, the FDA has approved 13 ADCs for solid or hematologic cancers (6 for solid tumors) (Table 1, 2).In this review, we aim to provide a brief overview of the ADCs for solid tumors, their structure and mechanism of action, available clinical trial data.

Antibody
Monoclonal antibodies are the components that allow the ADCs to bind specifically to the target antigen or receptor on cancer cell.The antibodies for ADCs require an adequate binding affinity, efficient internalisation, low immunogenicity and a long half-life.Murine antibodies were often used in the early days of ADCs development, but their high immunogenicity made them likely to cause serious side effects.Currently, development is increasingly used chimeric, humanised and fully humanised antibodies with low immunogenicity.In the five types of Immunoglobulin (M, A, D, E, G), Immunoglobulin G (IgG) is the most commonly used in ADCs.IgGs are classified into IgG1-4, but IgG3 is usually not used because of its short half-life, and IgG1 is widely used due to its immunogenic functions such as antibody-dependent cellular cytotoxicity (ADCC), antibody-dependent cell-mediated phagocytosis (ADCP) and complement-dependent cytotoxicity (CDC) (2).Structure of antibody-drug conjugates (ADCs).ADCs have three components: a monoclonal antibody that binds to an antigen or receptor on cancer cells, a linker that connects the antibody and payload, and a payload that is a potent cytotoxic agent.ADCs bind to target antigens on the surface of cancer cell followed by internalisation.After lysosomal degradation, payloads are released into the cytoplasm and induces cell death via DNA damage or microtubule inhibition.Moreover, payloads pass through the cell membrane and show antitumor effect on surrounding cancer cells, called bystander effect.

Payload
Potent cytotoxic agents are the 'payloads' of ADCs.Many of the cytotoxic agents in ADCs are very low IC50 (the drug-concentration to inhibit the growth of 50% of cells), so that they cannot be used as drugs on their own because they exhibit strong toxicity.The payloads chiefly classified into tubulin-binding agents (auristatins or maytansinoids), DNA-targeting agents (calicheamicins, duocarmycin), topoisomerase I inhibitors (deruxctecan, .In these cytotoxic agents, auristatins are the largest family of payloads so far, due to their favorable biochemical characteristics.Auristatins include two types of derivatives: monomethyl auristatin E (MMAE) and monomethyl auristatin F (MMAF).While the MMAF shows low Furthermore, several types of alternative payloads are under development, such as radionucleotides, immune modulators, tyrosine kinase inhibitors and dual-distinct payloads.Moreover, the conjugation of payloads to bispecific antibodies is another strategy (4).The drug-antibody ratio (DAR) is the amount of payloads loaded to the antibody and is a critical factor to determine the efficacy of the ADCs.ADCs with higher DAR values increase potency in vitro but are usually less favorable in vivo due to the unstable structure and low binding capacity, resulting in faster plasma clearance.Therefore, in most of ADCs, their DAR values are in the range of 2-4.However, several ADCs have emerged with favorable pharmacokinetics profiles even in DAR values of around 8, such as trastuzumab deruxtecan (T-DXd) and sacituzumab govitecan (SG) (5,6).

Linker
Linkers connect the antibody to the cytotoxic payload and define several characteristics of ADCs.Ideal linkers should ensure ADCs stability in the bloodstream but release payloads at the tumor site.
The conjugation site has a significant impact on the stability and pharmacokinetics of ADCs.Site-specific binding produces homogeneous ADCs whose properties can be tailored to maximise therapeutic range (7).
Linkers can be categorised into cleavable and non-cleavable.Cleavable linkers release the payload based on some factors of cancer cell such as pH, reducing agents and lysosomal protease.While cleavable linkers have the advantage of providing a bystander effect, they have the disadvantage of increased toxicity in normal tissues.Conversely, non-cleavable linkers are more stable in plasm, so that they release payload only after complete lysosomal degradation of the antibody and limit premature release of payload.Whereas these features of non-cleavable linkers offer less toxicity, bystander effect is not expected because of the extremely low cell membrane permeability.Among the 13 approved ADCs, cleavable linkers are the most common (11 of 13) and noncleavable linkers are only two (trastuzumab emtansine, belantamab mafodotin).

Mechanism of action
The action of ADCs begins with the recognition and binding of target antigens on the surface of cancer cell.After that, the ADCs are internalised through endocytosis with formation of early endosomes.The early endosomes mature into late endosomes, which then fuse with lysosomes.Whereas ADCs with cleavable linkers release their payload depending on the localisation of enzymes and other factors, the ADCs with non-cleavable linkers release their payloads when antibodies are degraded in lysosomes.The payloads are released into the cytosol and induces cell death via DNA damage or microtubule inhibition.In addition, after the payloads are released intracellularly, they may pass through the cell membrane and exert an antitumor effect on surrounding cancer cells, referred to as the bystander effect.The bystander effect is important because tumors are heterogeneous populations of cancer cells and target antigens are not expressed by all cancer cells.This bystander effect may explain the efficacy of T-DXd in patients with 'HER2-low' breast cancer.Furthermore, the ADCs display anticancer effect through immunogenic reactions including ADCC, ADCP and CDC.
In addition, it is used for an adjuvant therapy for residual disease in early-stage HER2-positive BC after neoadjuvant therapy, on the basis of KATHERINE trial (9).In this phase III trial, T-DM1 (for 14 cycles) was compared with trastuzumab for patients with HER2positive early breast cancer who were found to have residual invasive disease at surgery after neoadjuvant therapy including a trastuzumab and a taxane (with or without anthracycline).3 years invasive disease free survival was 88.3% (HR 0.50, 95%CI 0.39-0.64).Grade ≥ 3 TEAEs were reported that a thrombocytopenia (5.7%) and hypertension (2.0%).Assigned therapy were completed in 71.4 (T-DM1 group) and 81.0% of patients (trastuzumab group).T-DM1 was discontinued due to thrombocytopenia (4.2%), elevated blood bilirubin level (2.6%), elevated aspartate aminotransferase level (1.6%), elevated alanine aminotransferase level (1.5%), peripheral sensory neuropathy (1.5%) and decreased ejection fraction (1.2%).

Enfortumab vedotin
Nectin-4 is a cell-adhesion molecule that may associated with cancer cell growth and proliferation and overexpressed in urothelial cancer and breast cancer.EV is an ADC consisting of a fully human monoclonal nectin-4 antibody and monomethyl auristatin E (MMAE) (10).A phase III trial EV-301 was conducted to evaluated EV as compared with investigator's choice chemotherapy (docetaxel, paclitaxel or vinflunine) in patients with locally advanced or metastatic urothelial cancer who had previously treated with a platinum-based chemotherapy and PD-1/PD-L1 inhibitor.The mPFS was 5.6 months (HR 0.62, 95% CI 0.51-0.75)and the mOS was 12.9 months (HR 0.70, 95% CI 0.56-0.89).The most common grade ≥ 3 treatmentrelated adverse events (TRAEs) were maculopapular rash (7.4%), fatigue (6.4%) and decreased neutrophil count (6.1%).

Sacituzumab govitecan
TROP2 (trophoblast antigen 2) is a transmembrane glycoprotein coded by the gene TACSTD2, which primarily acts as intracellular calcium signal transducer (17).While TROP2 is overexpressed in many human epithelial tissues, the expression increases in tumor tissue compared to normal tissues and has been associated with poor prognosis.This overexpression occurs in various tumor types including breast cancer, NSCLC, colon cancer, esophaegeal squamous cancer, thyroid cancer and hepatobiliary cancer.
SG, formerly IMMU-132, is composed of a humanised TROP2 antibody and the active metabolite of the topoisomerase I inhibitor irinotecan (SN-38).

Tisotumab vedotin
Tisotumab vedotin is an ADC consisting of a monoclonal antibody against tissue factor (TF) and MMAE (20).TF is a transmembrane glycoprotein physiologically expressed on fibroblasts and subendothelial cells and serves as an initiator of the extrinsic coagulation cascade.Conversely, TF is frequently overexpressed on several solid tumors, including cervical cancer and related to tumor growth, metastasis and angiogenesis.

Datopotamab deruxtecan
Datopotamab deruxtecan (Dato-DXd) is a novel ADC composed of humanised anti TROP2 IgG1 monoclonal antibody, a cleavable linker, a topoisomerase I inhibitor payload and has a DAR of 4 (22).Although both of Dato-DXd and SG are the TROP2 ADCs, the two ADCs are different in some points.Whereas 90% of the payload of SG is released in 3 days, just 5% of the payload of Dato-DXd is released after 21 days.Because of this pharmacokinetic feature, Dato-DXd is administered every 3 weeks in contrast to SG dosed on Days 1 and 8.
TROPION-Lung08 is a phase III trial to evaluate Dato-DXd plus pembrolizumab comparing with pembrolizumab as first-line therapy for patients with a PD-L1 TPS of ≥50%.AVANZAR, a phase III study, examined the survival benefit of Dato-DXd plus durvalumab plus carboplatin against pembrolizumab plus platinumbased chemotherapy as first-line treatment.

Patritumab deruxtecan (HER3-DXd)
Human epidermal growth factor receptor 3 (HER3) is a tyrosine kinase receptor that expressed in various solid tumors, including breast cancer and NSCLC (25,26).It is reported that high expression of HER3 is associated with poor prognoses and resistance mechanism of EGFR TKI.

Future perspective
The development of ADCs has the following directions; stabilisation in vivo (increase hydrophilicity, site specific conjugation), enhanced efficacy (increase DAR, dual payload, bystander effect, combination therapy) and reduction of toxicity (tumor specific antigen, tumor site specific payload release, dose-optimisation).Improvements in antibodies, linkers, payloads and conjugation are underway to achieve these goals.

Efficacy enhancement
In general, increasing hydrophilicity of ADCs result in improving stability in circulating plasma and anticancer effect.One of the such strategy is that the incorporation of polyethylene glycol (PEG) as a side chain in a linker contributes to increase hydrophilicity and decrease plasma clearance (29,30).Although the PEGylate linker has been a valid approach to improve the pharmacokinetics properties of ADCs, there are some limitations such as hypersensitivity, nonbiodegradablity and accelerated blood clearance.
Currently, polysarcosine (PSR), an intermediate and byproduct of glycine synthesis and degradation, is considered a promising alternative to PEG in terms of biodegradablility and ease of synthesis and versatility (31,32).
While payload masking linkers are impressive methods for ADCs hydrophilicity, hydrophilic payloads are another approach to ADCs stabilisation and enables high DAR.For example, β-D-glucuronylmonomethylauristatin E (MMAU) is the novel payload which enable high DAR = 8 ADCs with stability and efficacy due to its hydrophilicity (33).
The therapeutic effect will be enhanced by increasing the number of payloads in the ADCs.Most ADCs linkers used so far have linear structures, allowing only a single payload attachment.However, branched linkers enable the loading of two or more payload molecules while maintaining the ADCs' stability (34,35).
The use of dual payload is another approach.Since in conventional cancer chemotherapy, combining two or more different anticancer agents have enhanced the therapeutic effect, it is expected that loading multiple payloads in ADCs would also enhance the antitumor effect.Several studies have been reported about dual payloads ADCs with antitumor activity, such as MMAE and MMAF (36), MMAE and pyrrolobenzodiazepine (37), MMAF and PNU-159682 (38).
Moreover, the combination therapy of ADCs with other antitumor agents, including cytotoxic drugs, antiangiogenic agents, targeted antibodies and ICIs, is currently investigated (43).

Optimising the safety of ADCs
The toxicity of ADCs varies widely among individuals due to organ functions, comorbidities, pharmacogenomics and so on.Several methods have been considered to optimise the dose of ADCs.

Antibody modifications
When target antigens are expressed in both tumor cells and normal tissues, on-target adverse events are increased.Probody-drug conjugates (PDC) are antibody prodrugs designed to remain intact in normal tissue by masking antigen-binding regions, while they are activated by protease in the tumor microenvironment.As a result, this approach enables safer binding to target antigens expressed in both cancer and normal cell (44).
First-in-Human studies of CX-2029 and Praluzatamab ravtansine (CX-2009) revealed relevant improvements in the safety profiles.CX-2029 is a PDC comprising an anti-CD71 antibody conjugated to MMAE.CD71 is previously considered as undruggable ADC target because of broad expression in both of tumor and normal tissue.However, the phase I study of CX-2029 revealed the tolerability of targeting therapy for CD71 (45).
Praluzatamab ravtansine (CX-2009) is a PDC consisting of an anti-CD166 antibody and DM4.Although CD166 is also widely expressed in tumor and normal tissues, the results of phase I study of praluzatamab ravtansine showed tolerable safety profiles (46).
In addition, bispecific antibodies, binding to two different antigens, increase, selectivity resulting in the reduction of the off-target toxicity (39).

Modifying conjugation technology or drug/linker chemistry
Tandem linker.One drawback associated with cleavable linkers is the potential to release payloads in plasma circulation before tumor targeting.If the payloads are released systemically prematurely, the efficacy of the remaining circulating ADCs may be reduced, leading to off-target toxicity.Tandem cleavable linkers are novel linkers that incorporate a β-glucuronide moiety and require tandem enzymic cleavage events (47).These linkers reduce payloads release during circulation and off-target toxicity.
Silencing the Fc portion of the ADCs.Whereas the Fc domain of the antibody induces immunogenicity, the internalisation of ADCs into non-targeted cells via Fcγ receptors on immune cells may provoke off-target toxicities.By silencing the Fc domain, Fc-mediated offtarget cytotoxicity was reduced (48).
Novel site specific conjugation technologies.Conventional antibody conjugates have been constructed through cysteine or lysine residue side chains.These approaches were stochastic, non-specific conjugation and generated heterogenous ADCs.To overcome the drawback of non-specific conjugation, intense research to realise site specific conjugation has been conducted.Chemo-enzymatic methods include transglutaminase and glycan-mediated conjugation, and chemical methods include selective reduction of disulfides and N-terminal amine modifications (49).

Dose-optimisation strategy
Five dose-optimisation strategies are known to minimise toxicity and maximise efficacy.These include body weight-based dose capping, treatment duration capping, dose schedule, response-guided dosing and randomised dose-finding (50).
Body weight-based dose-capping.Body weight has been generally considered as significant factor to adjust drug dosage.Body weightbased dose-capping was adapted for EV, ADCs used in urothelial cancer patients (51).The recommended dose of EV was 1.25 mg/kg, but the maximum dose was decided as 125 mg.

Treatment duration capping.
Limiting the treatment duration is widely adopted when using taxanes or platinum salts to minimise chronic toxicities such as peripheral neuropathy (52,53).
Polatuzumab vedotin, ADCs for diffuse large B cell lymphoma patients, is used up to six cycles.Predicted risk of developing grade ≥ 2 peripheral neuropathy with polatuzumab vedotin increases by ≥50% when eight versus six treatment cycles are administered.These data supported the approval of polatuzumab vedotin for patients with relapsed and/or refractory diffuse large B cell lymphoma at a dose of 1.8 mg/kg every 3 weeks for a maximum of six cycles (54).Dose schedule.Dose frequency is fundamental for the safety profile of a drug.Dose fractionation offers the almost same cumulative dose with a lower peak plasma concentration (Cmax) compared with a single dose and reduce toxicity related to the Cmax.Within the FDA approved ADCs, 4 ADCs (gemtuzumab ozogamicin, inotuzumab ozogamicin, SG and EV) are administered weekly dosing (51,(55)(56)(57).
Response-guided dosing.Response-guided dosing is a flexible strategy to modify the dose of drug based on initial response of individual patient.This method is used in inotuzumab ozogamicin, consisting of anti-CD22 antibody and calicheamicin.The subsequent dose is decreased in patients with complete response after an initial dose of inotuzumab ozogamicin (58).
Randomised dose-finding study.Although conventional Phase II/III studies have been arranged to evaluate primary clinical efficacy and safety, these studies often lack data on multiple dose levels.Randomised dose-finding study is conducted to reveal an optimal dose level based on data of multiple dose levels.DESTINY-Lung02 is an example in which the optimal dose of T-DXd (5.4 vs 6.4 mg/kg) for NSCLC was assessed (16).The confirmed ORR was 49.0% (95%CI, 39.0-59.1)and 56.0%(95% CI, 41.3-70.0),median DOR was 16.8 months (95% CI, 6.4-not estimable [NE]) and NE (95% CI, 8.3-NE) with 5.4 and 6.4 mg/kg, respectively.Grade ≥ 3 TEAEs were observed in 38.6 and 58.0% of patients with 5.4 and 6.4 mg/kg, respectively.ILD occurred in 12.9 and 28.0% of patients with 5.4 and 6.4 mg/kg, respectively.Based on this result, the approval dose of T-Dxd was 5.4 mg/kg for NSCLC.
Randomised dose-finding studies might be reasonable for other ADCs and indications, particularly for those in which fatal adverse events like as ILD and/or considerable grade ≥ 3 toxicities are observed either in trials or clinical practice.

Inverse targeting strategy
The inverse targeting concept is another strategy to reduce off-target toxicity by payload-binding selectivity enhancers (PBSE).PBSE is payload-binding fragments which is co-administered with ADCs to bind unconjugated payload and decrease distribution of payload into non-targeted cells.Preclinical study showed that anti-MMAE antibody fragment and anti-maytansinoid single domain antibody increased the therapeutic window of MMAE-based ADCs and DM4based ADCs, respectively (59,60).

Pharmacogenomics
Pharmacogenomics is the term that has been used to mean the study of how a person's genetic characteristics alter their response to drug.Such genomic technologies are expected to clarify in which patients the toxicity is enhanced (61).
As an example of pharmacogenomics, UGT1A1 polymorphisms is known to cause delayed metabolism of SN-38.SN-38, the active metabolite of irinotecan, is glucuronosylated and excreted by UGT1A1 (62).UGT1A1 has a genetic polymorphism, and patients homozygous or heterozygous for either UGT1A1 * 6 or UGT11 * 28 are known to have a delayed metabolism of SN-38 resulting in higher incidence of toxicity.This correlation was observed as well in the study of ADCs.SG, loading SN-38 as payload, reported that the toxicity was increased in patients with deleterious UGT1A1 polymorphisms (63).

Conclusions
In summary, we have reviewed the structure and mechanism of ADCs, clinical data of approved ADCs for solid tumors.Based on the ability to deliver novel payloads selectively to targeting cancer cells, they have enabled more effective treatment with fewer side effects.Some ADCs show efficacy in multiple types of solid tumors, which suggests that ADCs may enable cancer treatment based on not only tumor types but also cell surface biomarkers.Further development on ADCs will bring new strategies for cancer treatment.

Figure 1 .
Figure1.Structure of antibody-drug conjugates (ADCs).ADCs have three components: a monoclonal antibody that binds to an antigen or receptor on cancer cells, a linker that connects the antibody and payload, and a payload that is a potent cytotoxic agent.ADCs bind to target antigens on the surface of cancer cell followed by internalisation.After lysosomal degradation, payloads are released into the cytoplasm and induces cell death via DNA damage or microtubule inhibition.Moreover, payloads pass through the cell membrane and show antitumor effect on surrounding cancer cells, called bystander effect.

Table 2 .
Selected clinical trial data