Antibody-drug conjugate (ADC) has achieved great progress in the past decade. So far, ten ADC drugs have been approved by the FDA, of which five were approved in 2019 and 2020, and more than 80 are under active clinical studies.
ADC for solid tumors has made a satisfactory breakthrough in clinical practice. Until November 2019, only Kadcyla had indications for solid tumors. With the approval of Padcev and Enhertu by FDA at the end of 2019 and the approval of Trodelvy in April 2020, there are currently four FDA-approved ADC drugs for solid tumors. At present, ADC in the late clinical stage is mainly aimed at solid tumors.
In 2009, calcitamicin, golden glucagon and medanesin were the main cytotoxic drugs used in the development of ADC. Ten years later, these molecules are still being used and optimized for better stability and hydrophilicity. In addition, new types of cytotoxins such as polybrominated biphenyls, dimycin, and camptothecin derivatives have been developed.
In the past 10 years, antibody engineering optimization introduced more site-specific binding to improve structural uniformity and stability, and the second and third generation ADCs were introduced into clinic to broaden the treatment index. In preclinical studies, dozens of biological conjugation techniques based on cysteine residues, unnatural amino acids or molecular engineering models were proposed.
There are two key factors to the success of ADC. The first depends on a linker between mAb and cytotoxic drugs. The mAb-linker-payload system was originally designed with a cleavable linker that is stable under physiological conditions and cleavage rapidly after endocytosis of tumor cells in order to selectively deliver cytotoxins to the tumor and limit the effects of off-target toxicity. The linker is sensitive to lysosome conditions (protease, acidity, and reducing medium).
The second key factor in ADC is the grafting of a powerful cytotoxic agent onto antibodies. In fact, the first generation of ADC was characterized by a low treatment index because of the low efficacy of payloads, such as anthracyclines, resulting in very limited efficacy at a maximum tolerable dose (MTD) of 100 mg/kg. This phenomenon can be explained by the low antigen density on the cancer cell surface, and the proportion of ADC that can reach the tumor cell surface is very poor compared with the injection volume.
1. Mylotarg, Besponsa and the first generation of cleavable linkers
Mylotarg (Gemtuzumab Ozogamicin) was approved by the FDA for treating acute myeloid leukemia (AML) in 2000, which is formed by conjugating calicheamicin, a powerful DNA lytic with low nanomolecular activity, to a mutant anti-CD33 IgG4 gene mutation getuzumab through a cleavable linker. This ADC with an average DAR of 1.5 is a complex mixture containing about 50% unbound McAbs. After ADC internalization, the hydrazone bond can be hydrolyzed in acidic environment in vivo to release the precursor of calicheamicin, and then reduced to free active calicheamicin by glutathione. The latter binds to DNA grooves and undergoes bergman cyclization to produce highly active double radicals, resulting in sequence selective double strand cleavage.
Theoretically, hydrazone compounds should be stable in blood circulation under physiological pH value and selectively hydrolyzed under more acidic conditions (endosome pH 5.0-6.5, lysosome pH 4.5-5.0). However, Mylotarg linkers showed some instability, causing calicheamicin to be released prematurely in the plasma circulation, which explains its highly toxic characteristics and the reasons subsequent voluntary withdrawal by Pfizer in 2010. Mylotarg was approved by the FDA in 2017, benefiting from clinical knowledge accumulated in recent years to use at lower dosages and tailored delivery regimens to different patient populations.
A similar ADC drug has been developed and used to conjugate calicheamicin with inotuzumab, which is anti-CD22 IgG4, approved in 2017 for the treatment of acute lymphoblastic leukemia (ALL).
2. Kadcyla and second generation uncleavable linkers
In previous findings, an alternative strategy for designing linker is necessary for ADC development. Therefore, the immunogen research team focused on the use of delivery systems that connect mesanthin derivatives and combine glutathione-sensitive disulfide bonds. These innovative chemically unstable linkers are designed to allow controlled release in the presence of glutathione (GSH). The cytoplasmic concentration of glutathione in cancer cells is about 1000 times higher than in plasma. In addition, the careful positioning of the two methyl groups close to the disulfide bond enables the release kinetics to be controlled. Therefore, the high concentration of reducing molecules in the tumor should have ensured the selective release of the payload in the tumor environment. This type of linker has not been approved on the market yet.
3. Adcetris, Polivy and the second generation of cleavable linker
At the same time, Seattle Genetics conjugated dostatin derivatives (such as monomethyl yellow E or MMAE) to CD30 IgG1-resistant cysteine residues to produce Adcetris (Brentuximab vedotin). After mild and partial reduction of disulfide bonds between chains, anti-CD30 mAb (cAC10) was bioconjugated to cleavable heterobifunctional maleimide linker. The maleimidocaproyl-valine-citrulline-p-aminobenzyloxycarbonyl conjugation includes valine-citrullinated peptide trigger (ValCit) that is sensitive to lysosomal cathepsin B, and p-aminobenzyl alcohol (PAB) which allows MMAE to be released after internalization in CD30-positive tumor cells.
Adcetris was approved by the FDA in 2011 for the treatment of anaplastic large cell lymphoma and Hodgkin’s lymphoma. After internalization of CD30-dependent ADC, the cleavage linker can be degraded, and the released MMAE can destroy the target cells and spread to the plasma membrane, reaching and killing neighboring tumor cells. This phenomenon is called bystander killing effect, and the released MMAE can kill CD30-positive and CD30-negative tumor cells. From this, it is inferred that the bystander killing effect explains the special in vivo efficacy of Adcetris in the treatment of patients with heterogeneous lymphoma. Recently, Neri and his team demonstrated that the linker ValCit-PAB used in ADC can also be cleaved before internalization, helping to dispel the notion that ADC must target internalized antigens in order to work.
Similarly, the second-generation linekr (MC-VC-PAB) is also used to conjugate MMAE to anti-CD79b polatuzumab (Polivy) and anti-Nectin 4 enfortomab (Padcev), which was approved by FDA in June and December 2019, respectively.
However, there are two limitations to the molecular structure of Adcetris, Padcev, and Polivy. The first is related to the instability of Maleimide used for biological conjugate, which can carry out reverse Michael reaction in plasma, resulting in partial unconjugation, and is then slowly transferred to albumin, which may be related to off-target toxicity. The second involves Adcetris ‘ValCit’ trigger, the substrate of carboxylesterase 1C in the plasma circulation and other proteases secreted by neutrophils, such as elastase, which partly explains the adverse blood effects associated with Adcetris (such as neutropenia).