STTT Review: ADC drug development history, challenges and next generation development directions

Antibody-drug conjugate (ADC) consists of
a linker, payload, monoclonal antibody (mAb).
It combines the advantages of high specific targeting ability and strong killing effect to achieve accurate and efficient killing of cancer cells, and has become one of
the hot spots in the research and development of anti-cancer drugs 。 Since the first ADC drug, Mylotarg:registered: (gemtuzumab ozogamicin), was approved by the FDA in 2000, as of December 2021, a total of 14 ADC drugs have been approved worldwide for hematologic malignancies and solid tumors, in addition to more than 100 ADC candidates currently in various stages
of clinical trials.

Recently, the Nature journal signaltransduction and targeted therapy published an in-depth review published by Tongji Medical College of Huazhong University of Science and Technology, which reviewed the history and mechanism of ADCs, and discussed the key components of ADCs and their influence mechanism on ADCs activity
.
The review also details approved ADC drugs and other promising (phase III clinical) drug candidates, and discusses current challenges and future prospects
for the development of next-generation ADCs.

1.
Timeline of important events for ADC drug development and approval

As early as the early 20th century, Paul Ehrlich first proposed the concept of the "magic bullet" and hypothesized that certain compounds could directly enter certain desired targets in the cell to cure diseases
.
In theory, these compounds should be effective at killing cancer cells, but harmless
to normal cells.

In 2000, the U.
S.
Food and Drug Administration (FDA) first approved the ADC drug Mylotarg:registered: (gemtuzumab ozogamicin) for adult acute myeloid leukemia (AML), marking the beginning of
the era of ADC-targeted therapy for cancer.
Figure 1 depicts the landmark events
of ADC drugs from infancy to mature development over the past hundred years.
With the continuous expansion of targets and indications, ADCs are leading a new era of targeted cancer therapy and are expected to replace traditional chemotherapy drugs
in the future.

Figure 1 Important timeline for ADC drug development and approval over a century (Image source: Reference[1])

1.
Composition of ADC drugs

The ADC consists of
an antibody, payload, and chemical linker.
The ideal ADC drug remains stable in the blood circulation, reaches the therapeutic target with precision, and ultimately releases the cytotoxic payload
in the vicinity of the target, such as cancer cells.
Each element affects the ultimate efficacy and safety of an ADC, and in general, ADC development needs to consider all of these key components, including target, antibody, cytotoxic payload, linker, and conjugation method selection
.

Figure 2 Structure and characteristics of ADC drugs (Image source: Reference [1])

Target antigen

The target antigen expressed on tumor cells is the direction in which ADC drugs recognize tumor cells, and it also determines the mechanism by which cytotoxic payloads are delivered into cancer cells (e.
g.
, endocytosis).

Therefore, choosing the appropriate target antigen is the primary consideration for
the ADC.

First, to reduce off-target toxicity, the targeted antigen should be expressed only or mainly in tumor cells, and not or very little
in normal tissue.
For example, HER2 expression is approximately 100 times higher in some types of tumors than in normal cells, which laid a solid foundation
for the development of ado-trastuzumab emtansine, fam-trastuzumabderuxtecan, and disitamab vedotin.

Second, the target antigen should be nonsecretive, as circulating antigens can cause the ADC to bind beyond the tumor site, leading to reduced tumor targeting and safety issues
.

Third, the ideal antigen of interest should be internalized after binding to the corresponding antibody so that the ADC-antigen complex can enter the cancer cell and then pass through the appropriate intracellular transport pathway and release the cytotoxic payload
.

The target antigens currently approved for ADC drugs are usually specific proteins overexpressed by cancer cells, including HER2, Trop2, Nectin4 and EGFR targets in solid tumors, and CD19, CD22, CD33, CD30, BCMA and CD79b
targets for hematological malignancies.
Driven by basic research in oncology and immunology, the selection of ADC target antigens has gradually expanded from traditional tumor cell antigens to targets in the tumor microenvironment (e.
g.
, in the stromal and vasculature).

Emerging evidence from preclinical and clinical studies suggests that components of the neovascular system, subthelial extracellular matrix, and tumor matrix may be valuable target antigens
for ADC drug development.

Figure 3 Tumor cells and tumor microenvironment (vasculature and stromal) targets available for ADCs (Image source: Reference[1])

antibody

Tumor-targeted antibodies are critical
for specific binding between the target antigen and the ADC.
In addition to a high binding affinity to the target antigen, the ideal antibody fraction should promote effective internalization, exhibit low immunogenicity and maintain a long plasma half-life
.

In the early stages of ADC drug development, antibodies of mouse origin are mainly used, with a high
rate of development failure due to severe immunogenicity-related side effects.
With the advent of recombinant technology, murine antibodies are mostly replaced
by chimeric antibodies and humanized antibodies.
Currently, ADCs are increasingly employing fully humanized antibodies
with significantly reduced immunogenicity.
Of the 14 approved ADC drugs, only Brentuximab vedotin used chimeric antibodies
.

As the main component of immunoglobulins in serum, the antibodies currently used in ADC drugs are mostly immunoglobulin G (IgG) antibodies, including four subtypes
of IgG1, IgG2, IgG3 and IgG4.
IgG1 is a commonly used subtype of ADC because IgG1 is most
abundant in serum.

The efficiency of the internalization of antibody-antigen complexes depends mainly on the binding affinity between the antibody and the tumor cell surface antigen, and higher affinity usually leads to faster internalization
.
However, high-affinity antibodies may reduce the penetration
of antibodies into solid tumors.
Treatment of solid tumors is more complex than hematological tumors because there is a binding site barrier (BSB) in solid tumors, and the extremely high affinity between antibodies and antigens can cause ADCs to become trapped outside the blood vessels after they bind but less penetrate into tumor cells
far away from the blood vessels.
Therefore, a reasonable affinity between antigen and antibody should be optimized to balance rapid uptake and anti-cancer efficacy
of target cells.

In addition to binding affinity, another factor that affects tumor penetration is the size of
the antibody.
The large molecular weight of IgG antibodies (about 150 kDa) is usually a large obstacle
to penetrating the matrix in capillaries and tumor tissue.
Therefore, early ADC drugs mainly target hematologic malignancies
.
To make ADCs better at treating solid tumors, researchers have tried to miniaturize
antibodies by removing FC fragments.
Miniaturized antibodies not only retain high affinity and specificity, but also penetrate blood vessels more easily into solid tumors, thereby greatly improving the killing effect
on solid tumors.
However, it was also found that this change leads to a decrease
in half-life.
Therefore, various factors
should be considered when designing ADCs with miniaturized antibodies.

Figure 4 Stromal changes in cancer (Image source: Reference [2])

 Linker

The linker in the ADC links the antibody to the cytotoxic drug
.
It has a great impact on the stability of the ADC, the payload release curve, and is critical
to the final efficacy of the ADC drug.
Ideal Linker should not induce ADC aggregation and prevents premature release of the payload in plasma and can be released efficiently within cancer cells
.
Depending on the metabolic fate in the cell, two types of linkers are used in most ADC drugs, including cleavable and non-cutable linkers
.

Cleaveable linkers take advantage of environmental differences between the systemic circulation and tumor cells to accurately release free cytotoxic drugs, which can be further divided into chemical cleavage linkers (hydrazone and disulfide bonds) and enzymatic cleavage linkers (glucuronide bonds and peptide bonds).

Another cleaveable connector
that is sensitive to reducing glutathione (GSH).
GSH plays a vital role
in maintaining intracellular redox balance during cell survival, cell proliferation, and differentiation.
The concentration of GSH in the blood is much lower than the intracellular concentration
in cancer cells.
Thus, this type of linker can remain stable in the blood system while specifically releasing the active payload
in cancer cells.

Peptide-based linkers are sensitive to lysosomal proteases and have been used in many ADCs
.
Lysosomal proteases, such as cathepsin B, are often overexpressed in cancer cells, allowing for accurate drug
release near the tumor.
In addition, due to the presence of protease inhibitors in the blood, enzyme-lyzed linkers are generally stable in the systemic circulation, so enzyme-cleavable linkers reduce the risks
of premature breakage.
Of the approved ADC drugs, 9 out of 14 use peptide-based linkers
.

Non-cutable joints (e.
g.
, thioether or maleimidohexanoyl) are inert to the chemical and enzymatic environments commonly found in the body, and the biggest advantage of non-cutable joints due to improved plasma stability is their lower off-target toxicity, but the bystander effect of the payload is affected
.

payload

The payload is a cytotoxic warhead
that exerts itself after internalization of the ADC into cancer cells.
Because only about 2% of ADCs can reach the target tumor site after intravenous administration, the compounds payloaded in the ADC need to be highly toxic (IC50 in the nM and pM ranges).

In addition, these compounds should remain stable under physiological conditions and have functional groups
that can covalently bind to antibodies.
Currently, cytotoxic payloads for ADCs mainly include potent tubulin inhibitors, DNA damaging agents, and immunomodulators
.

Figure 5 Representative small molecule payloads used in ADC drugs (Image source: Reference [1])

Coupling method

In addition to selecting the antibody, linker, and payload, the method by which the small molecule fraction (i.
e.
, the linker plus payload) is attached to the antibody is important
for the successful construction of the ADC.
Typically, the presence of lysine and cysteine residues on antibodies provides a reactive site for conjugation, and early ADC drugs are often by random coupling
of lysine or cysteine residues.
However, antibodies typically contain approximately 80–90 lysine residues, of which 40 are usually reactive
.
By random coupling to lysine residues, different amounts of small molecule toxins of DAR(0-8) may attach to the antibody, resulting in a wide
drug-antibody ratio (DAR) distribution.
Furthermore, since lysine residues are distributed throughout the light and heavy chains of the antibody, conjugation reactions near the antibody-antigen recognition site may reduce the binding
of the ADC to the target.

Cysteine-based reactions provide another way to
couple.
Typically, IgG1 antibodies have both interchain and intrachain disulfide bonds
.
Interstranded disulfide bonds are exposed to the outside of the antibody and are easily reduced to expose free cysteine residues, providing usable sites for linker payload conjugation to the antibody
.
Due to the limited number of binding sites and the unique reactivity of thiol groups, the use of cysteine as the ligation site helps reduce the heterogeneity
of the ADC.
Depending on the reduction rate, products with DARs of 2, 4, 6, and 8 may produce better homogeneity compared to lysine residue-coupled products
.
This is by far the most commonly used coupling method
in commercial products.
However, it is worth noting that opening interchain disulfide bonds may disrupt the integrity
of the antibody.

The random coupling of lysine and cysteine residues can cause many problems
.
The stability of this coupling is sometimes insufficient, which leads to premature payload release, which can lead to off-target toxicity
.
In addition, it is difficult to guarantee that the payload is attached to a consistent site on the antibody, increasing the difficulty
of mass and uniform DAR value control.
To reduce the heterogeneity of ADCs, several site-specific coupling strategies
have been developed in new ADCs.

Figure 6 Specific coupling strategy of ADC (Image source: Reference [1])

2.
The mechanism of action of ADC

ADCs combine "specific" targeting and "efficient" cancer cell
killing.
Such drugs are like precision-guided "biological missiles", which can accurately destroy cancer cells, improve the treatment window, and reduce off-target side effects
.

Figure 7 Mechanism of action of ADCs to kill cancer cells (Image source: Reference[1])

The anticancer activity of ADCs is also involved in the role
of ADCC, ADCP, and CDC.
The antibody Fab fragment of some ADCs can bind to the epitope of the antigen of virus-infected cells or tumor cells, while the FC fragment binds to the FCR on the surface of killer cells (NK cells, macrophages, etc.
), thereby mediating the direct killing effect (Figure 7, lower left).

In addition, the antibody component of the ADC can specifically bind to the epitope antigen of cancer cells, inhibiting downstream signal transduction of antigen receptors (bottom right of Figure 7).

3.
Research progress of ADC

From the perspective of drug composition and technical characteristics, ADC drugs can be subdivided into three generations
.

Figure 8 Evolution of ADC drug development (Image source: Reference [1])

First generation ADC

Early ADCs conjugated primarily from conventional chemotherapy drugs to mouse-derived antibodies through non-cleavable linkers
.
These ADCs are not more potent than free cytotoxic drugs and are highly immunogenic
.
Later, the combination of more potent cytotoxic agents with humanized mAbs greatly improved efficacy and safety, so the first generation of ADCs gained market approval (including gemtuzumabozogamicin and inotuzumab ozogamicin).

In both products, humanized mAbs of the IgG4 isoform are used and conjugated
to potent cytotoxic gallithromycin via acid-unstable linkers.
However, the system also has significant flaws;

1) Linker instability: For example, acidic conditions may occur in other parts of the body, and the couplings in the first generation ADC can also be found to be slowly hydrolyzed in the body cycle (pH 7.
4, 37°C), resulting in uncontrolled release of toxic payloads and unexpected off-target toxicity
.

2) Easy to aggregate: The payload used in the first generation is hydrophobic and easy to cause antibody aggregation, resulting in some defects such as short half-life, fast clearance speed and immunogenicity
.

3) Drug heterogeneity: The conjugation of first-generation ADCs is based on random coupling of lysine and cysteine residues, resulting in a highly heterogeneous mixture
of DARs.
As a result, first-generation ADCs exhibit suboptimal treatment windows that require further improvement
.

Second generation ADC

The second generation of ADCs, represented by Brentuximabvedotin and Ado-trastuzumab emtansine, was approved
after optimizing the mAb sibling, payload, and linker.
These two ADCs feature:

1) Using IgG1 isotype mAbs, it is more suitable for bioconjugation
with small molecule payloads and high cancer cell targeting capabilities.

2) Higher toxic payload, improved water solubility and coupling efficiency
.
More payload molecules can be loaded onto each mAb without inducing antibody aggregation
.

3) Linker's improvements enable better plasma stability and uniform DAR distribution
.
Overall
.

Improvements in all three elements will improve the clinical efficacy and safety
of second-generation ADCs.
However, there are still many unmet needs, such as insufficient therapeutic windows due to off-target toxicity, and aggregation or rapid clearance
in those ADCs with high DAR.
When DAR exceeds 6, the ADC exhibits high hydrophobicity and tends to reduce ADC potency
due to faster in vivo clearance.
In this case, it is also necessary to optimize DAR through site-specific coupling, as well as continuous optimization of mAbs, adapters, and payloads
.

Third generation ADC

The third generation of ADCs is represented
by polatuzumabvedotin, enfortumab vedotin, fam-trastuzumab deruxtecan, and later approved ADCs.

1) Homogeneous DAR (2 or 4), ADCs with consistent DAR show less off-target toxicity and better pharmacokinetic efficiency
.

2) Completely humanized antibodies instead of chimeric antibodies to reduce immunogenicity
.

In addition, antigen-binding fragments (Fabs) are being developed to replace intact mAbs in many candidate ADCs, as Fabs are more stable in the systemic circulation and may be more easily internalized
by cancer cells.
In addition, more effective payloads have been developed: such as PBD, microtubulolysin, and immunomodulators
with novel mechanisms.
Although there are no updates to the Linker types in the third generation, some new entities have been developed to combine various payloads
.
To avoid interference with the immune system and improve retention time in the blood circulation, more hydrophilic Linker combinations, such as PEGIZATION
, are used in the third-generation ADCs.
Hydrophilic Linker can also be used to balance the high hydrophobicity of certain cytotoxic payloads such as PBD, where ADCs are often prone to aggregation
.
Overall, third-generation ADCs have lower toxicity and higher anti-cancer activity as well as greater stability, enabling patients to receive better anti-cancer treatments
.

4.
Current challenges and next-generation ADCs

As can be seen from approved drugs and drug candidates in development, the specificity and cytotoxicity of next-generation ADCs are getting better
than previous generations.
However, many challenges remain in the development of ADCs, including the complexity of pharmacokinetics, insufficient tumor targeting and payload release, and drug resistance
.

Major challenges

Complex pharmacokinetic profile

After ADC administration (primarily by intravenous infusion), three main forms may be present in the systemic circulation, namely intact ADC, naked antibody, and free payload
.
In the typical pharmacokinetic profile of ADCs, the concentration of conjugated ADCs and bare antibodies continues to decrease
with internalization and antibody clearance of the ADC.
Factors affecting antibody clearance include mononuclear phagocyte systems and Fc receptor (FcRn)-mediated recycling
.
By binding to the endocytic ADC with high affinity, FcRn outputs the ADC to an extracellular compartment for recycling
.
As a result, antibodies including conjugated ADCs and naked antibodies typically have longer half-lives
than conventional small molecule drugs.

The free cytotoxic payload is mainly metabolized in the liver and excreted through the kidneys (urine) or feces, which may lead to liver and kidney impairment
.
All of these factors, combined with high patient-to-patient variations, make it difficult to model PK and PD to characterize the clinical features of ADCs and assist in the design of new ADCs
.

Inevitable side effects

Among the 14 approved ADCs, the most common serious side effect (grade 3 or higher) is hematotoxicity, including neutropenia, thrombocytopenia, leukopenia, and anemia
.
Hematotoxicity, as well as hepatotoxicity and gastrointestinal reactions, may be associated with
premature release of cytotoxic payloads into the blood circulation.
This is consistent with
conventional chemotherapy drugs, which primarily affect healthy cells that proliferate rapidly.
In addition, the immune response induced by the antibody moiety of the ADC may cause secondary injury, leading to nephrotoxicity
.
Based on recent clinical observations, potential pulmonary toxic effects (e.
g.
, ILD) during ADC therapy should be noted, particularly in
anti-HER2 ADCs.
In clinical trials of T-DM1 and DS-8201, several deaths have been reported to be associated
with ILD.
However, the detailed mechanism of action of ILD remains unclear
.
There has been speculation that one of the possible causes may be related to
poor uptake of ADCs in healthy lung cells and free payload released by ADCs.
Due to the richest blood flow and longest residence time in the lungs, poor uptake of ADCs and free payload in the blood occur most often in the lungs to induce ILD
.
This requires next-generation ADCs to be optimized to minimize side effects
.
Adverse reactions should be closely monitored during medication to prevent or give supportive care
.

Tumor targeting and payload release

ADCs have a much larger molecular weight than traditional cytotoxic drugs, and the drug has limited
efficiency in penetrating tumors.
Current research shows that only a small fraction of ADCs input to the patient can reach tumor cells, so the efficacy
of the payload needs to be considered when designing the ADC.

Drug resistance

Another challenge in ADC development is drug resistance
.
Tyrosine kinase inhibitor (TKI) resistance typically involves escape mutations
at drug targets.
However, the resistance mechanism of ADCs has not been fully characterized
.
ADC resistance is more complex and diverse
.
Current evidence suggests that tumors can develop ADC resistance in a variety of ways, such as reducing antigen expression levels, altering intracellular transport pathways, and resistance to payloads
.

Next-generation ADCs

1) Use ADCs to target mutant proteins

Current studies have shown that ADC internalization and intracellular transport pathways have a key impact on
the cytotoxic activity of ADCs.
Mutant proteins typically have higher levels of ubiquitination and are more easily internalized and degraded
than wild-type proteins.
This means that if an ADC is used to target mutant proteins, it may lead to a significant clinical response
.
It is conceivable that ADCs targeting cancer-causing mutant proteins, such as certain EGFR mutants, can maximize the tumor specificity of the treatment to the level
of selective TKIs.

2) Dual epitope or dual target ADC

Advances in bispecific antibody technology have opened up more possibilities
for ADC innovation.
These ADC designs can improve antibody internalization and increase tumor specificity
.
Therapies currently under development have been exploring these possibilities
.
Bispecific ADCs targeting different sites on the same antigen can improve receptor aggregation and lead to rapid internalization of
the target.
In addition, bispecific ADCs dual-targeting HER2 and LAMP-3 have shown better lysosomal aggregation and load delivery
in preclinical experiments.

3) Use two different payload combinations

A dual-payload ADC using two different cytotoxic agents as payloads to reduce drug resistance
.
By accurately controlling the ratio of the two drugs and delivering two synergistic payloads to cancer cells, more effective efficacy
can be achieved.
And with the application of payloads of two different mechanisms, the incidence of drug resistance will be significantly reduced
.
For example, a homogeneous anti-HER2 ADC containing both MMAE and MMAF was designed and exerted more significant antitumor activity
in xenograft mouse models than co-administration of corresponding single-payload ADCs.

4) Peptide conjugates (PDCs)

Another ADC development strategy is to abandon the traditional structure of mAbs and choose to couple payloads to smaller molecular weight polypeptide fragments
.
The main goal of these strategies is to reduce the molecular weight of the ADC, thereby improving penetration efficiency and payload delivery
to tumor tissue.
For example, PEN-221 is an ADC formed by DM-1 binding to a polypeptide chain targeting somatostatin receptor 2
.
Its molecular weight is only 2 kDa, which is much lower than the 150 kDa IgG molecules
in traditional ADCs.
The current technical challenge with such ADCs is that they may be cleared
quickly in the plasma.
However, if we can overcome this obstacle, it has potential to treat inaccessible tumors, including tumors with poor vascular innervation and central nervous system tumors
.

5) Develop non-internalized ADCs

Traditionally, to deliver payloads into cancer cells, ADCs require mAbs
with high internalization capacity.
However, mAbs are often difficult to spread into
solid tumors due to the antigen barrier.
Therefore, non-internalized antibodies
can be developed for ADCs.
It is based on the principle
that the payload is released directly outside the cell under reducing conditions in the tumor microenvironment and then spreads inside the cancer cell leading to cell death.

Finally, there are still many opportunities
for innovation in payload selection.

5.
Conclusion

A variety of ADC therapies have been successfully developed to benefit
tens of thousands of cancer patients.
The approval of 14 ADC drugs and the excellent clinical performance of multiple ADCs have also drawn more attention to the field, which is important
for this relatively young but highly complex field.
With the continued efforts of researchers in these fields, it is not difficult to imagine that future ADCs will show more surprises
in cancer-targeted therapies.

References

[1] Antibody drug conjugate:the “biological missile” for targeted cancer therapy.
Signal Transduction and Targeted Therapy (2022) 7:93

[2]  The matrix in cancer.
Nat Rev Cancer.
2021 Feb 15.
doi:10.
1038/s41568-020-00329-7.
Epub ahead of print.
PMID: 33589810.