Antibody-Drug Conjugates: A New Paradigm for Cancer Treatment: Part I

Sunday, January 31, 2016

Introduction

Antibody-drug conjugates (ADCs) are a novel class of therapeutic agents that have widely been investigated for their potential applications in the treatment of cancer. In simple term, ADCs consist of a cytotoxic drug (0.3 to 1 kDa) connected by a linker to a monoclonal antibody (150 kDa), which directs it to the target cancer cells with abundant cell surface associated antigens1. This phenomenon, thus, enables the restriction of the cell killing activity of cytotoxic drugs to the cancer cells with high specificity, while protecting the healthy cells from their harmful side effects2. Various classes of cytotoxic drugs have been bonded to monoclonal antibodies using different linker technologies to form ADCs. Table 1 represents some of the ongoing clinical trials evaluating this drug targeting technology.

Table 1. Selected ADCs in Clinical Trials

In most cases ADCs rely on the targeting of cancer cell surface associated antigens expressed in higher magnitudes and homogeneously on cancer cell surfaces relative to healthy tissue. This type of ADCantigen complex is typically internalized inside cells by receptormediated endocytosis for intracellular drug release3.

In an alternative approach, ADCs have been developed to target markers expressed on tumor neo-vasculature. Since angiogenesis is a common feature of all malignancies for tumor growth, tumors express common markers of angiogenesis. The potential of targeting antigens expressed on tumor neo-vasculature has been established in preclinical and clinical applications4.

ADCS - Early Development

An example of early ADC is BR96-doxorubicin, targeted against the LeY tetra-saccharide antigen expressed on human cancer cells. This ADC consisted of eight doxorubicin molecules conjugated to a chimeric antibody via an acid labile hydrazone linker. The linker was intended to remain stable under the near neutral pH condition found in circulation but then cleave and release the drug once the ADC was internalized into the more acidic intracellular environment. The clinical development of this ADC was unsuccessful for a number of reasons that are now better understood. Firstly the hydrazone linker was found to be less stable in circulation than expected resulting in the premature release of the drug. Also the target antigen is also found on some sensitive nontumor cells resulting toxicity, and finally the doxorubicin was likely insufficiently potent given the small dose delivered to the tumors.

ADCS - Approved Products

Subsequent research activities led to the development and subsequent commercialization of gemtuzumab ozogamicin (Mylotarg®), an anti-CD33 ADC for the treatment of acute myeloid leukemia. Gemtuzumab ozogamicin consisted of a highly potent calicheamicin derivative linked covalently to the lysine residues of the anti-CD33 antibody at an average DAR of 2 to 3 by the acid labile hydrazone linker and a hindered disulfide bond7,8. However, eventually the mAb-drug hydrazone linker was found unstable in circulation, and released 50% of conjugated drug in 48 hours9. Although gemtuzumab ozogamicin appeared promising in clinical trials and was approved for commercial use by US Food and Drug Administration (FDA) in 2000, the product was withdrawn from the market after 10 years due to fatal hepatotoxicity and efficacy concerns10.

In 2011, brentuximab vedotin (Adcetris®; SGN-35), developed by Seattle Genetics, was approved by FDA for the treatment of Hodgkin Lymphoma. Adcetris® comprises of the chimeric anti-CD30 monoclonal antibody cAC10 and the cytotoxic drug monomethyl auristatin E (MMAE)11. The antibody is covalently attached to an average of four MMAE molecules per antibody through the protease cleavable valine-citruline dipeptide linker12.

In 2013, ado-trastuzumab emtansine (Kadcyla®, T-DM1) was approved for the treatment of HER2-positive metastatic breast cancer. Kadcyla®, developed by Genentech, is composed of the anti-HER2 humanized monoclonal antibody trastuzumab, linked covalently by the non-cleavable disulfide linker to the cytotoxic maytansine derivative, DM1, at an average DAR of 3.513.

Design of ADCS

There are three distinct elements in the design of an ADC - (1) the monoclonal antibody; (2) the drug or the payload; (3) the linker that connects the drug with the antibody.

 Figure 1. Structure of an ADC

Monoclonal Antibody (mAb)

The selection of the antibody obviously influences the success of a given ADC. The antibody must have high specificity for the targeted cancer antigen to circumvent crossreactivity with other antigens or non-specific binding, and, thus, avoid resulting toxicity and unwarranted elimination of ADC from the circulation. Secondly, the antibody must have high binding affinity for the target antigen (Kd <10 nM) to ensure good tumor localization9. Thirdly, antibodies with longer half lives are preferred in the design of ADCs, in order to maintain effective plasma concentration for longer period of time in the body. In addition, binding of the antibody to the target antigen on the surface of cancer cell may elicit antibody-dependent cellular toxicity (ADCC) and complement-dependent cytotoxity (CDC), leading to cell death14.

Cytotoxic Drugs

The choice of the cytotoxic drug plays a critical role in the performance of an ADC. Since only 0.003-0.08% of an injected antibody dose reaches each gram of tumor, cytotoxic drugs with sub-nanomolar potency are preferred for ADCs9.

 Figure 2.

The two major categories of cytotoxic drugs currently employed in the manufacture of ADCs are the microtubule inhibitors and the DNA damaging agents.

Microtubule inhibitors which include auristatins and maytansines bind to tubulin and inhibit tubulin polymerization, causing cell cycle arrest and eventually cell death5,9,15. Auristatins and maytansines have found successful applications in commercially approved ADCs. MMAE, a highly potent auristatin, has been used in Adcetris® by Seattle Genetics for the treatment of Hodgkin Lymphoma16. DM1, a highly potent maytansinoid, was used by Genentech in Kadcyla® for the treatment of metastatic breast cancer17.

DNA alkylating agents, which include calicheamicin, anthracyclines and duocarmycins, bind to the minor groove of DNA, and cause DNA strand scission, alkylation or cross-linking9. Gemtuzumab ozogamicin (Mylotarg®) and inotuzumab ozogamicin consist of a derivative of calicheamicin linked to a humanized monoclonal antibody targeted against CD33 or CD22, respectively18.

Linker Technology

Development of suitable linkers is critical to maintain stability of ADCs in circulation, and to ensure release of cytotoxic drug at the intended cancer cells. Stable linkers help to avoid associated toxicities resulting from unconjugated cytotoxic drug in the event of premature release before reaching the target. However, it is also equally important that the linker releases the drug to enable cell killing once the ADC reaches the target cells.

There are two major broad classes of linkers: cleavable and noncleavable linkers.

Cleavable Linkers

Acid-Labile Hydrazone Linkers

These linkers were employed in the first generation of ADCs, and had found application in gemtuzumab ozogamicin (Mylotarg®), the first ADC to receive FDA approval. Hydrazone linkers are expected to be stable at the near neutral physiological pH, and then undergo hydrolysis at the acidic environment of the cellular compartments, releasing the calicheamicin prodrug14. The enediyne drug is eventually activated by reductive cleavage, presumably by the glutathione, of the disulfide bond6. However, these hydrazone linkers have been associated with potential toxicity issues resulting from non-specific release of the drug14,19.

Peptide Based Linkers

These linkers employ peptide bonds to connect an antibody with a cytotoxic drug, and have been reported to be superior to the hydrolytically labile hydrazone linker in terms of stability in circulation and specificity to the target. Once the ADC is internalized, the drug is released by hydrolysis by the intracellular proteases such as cathepsin B. Adcetris®, currently approved by FDA for the treatment of Hodgkin Lymphoma, utilizes dipeptide based valine citruline linker to achieve targeted delivery to the CD30 positive cancer cells. The amide bond between the citrulline residue and the p-aminobenzyl carbamate portion of the linker is cleaved, upon internalization of the ADC, by lysosomal proteases, releasing the unmodified MMAE20. The uncharged MMAE then diffuses from the CD30-positive lymphoma cells into their surrounding microenvironment, in concentration high enough to kill neighboring tumor cells, thereby exerting a bystander effect14,21.

Disulfide Linkers

These linkers release the cytotoxic drug upon internalization of the ADC, where the disulfide bonds in these linkers get cleaved in the cytosol by the intracellular concentration of the reducing agent glutathione14. The circulating stability of the linker’s disulfide bond can be further enhanced by the addition of methyl groups to the adjacent carbon atoms. These additional methyl groups improve the disulfide bond stability by introducing a degree of steric hindrance to the approach of a reducing agent22.

Non-Cleavable Linkers

These thioether containing linkers provide greater stability in the circulation relative to the cleavable linkers, resulting in greater safety margin of highly potent cytotoxic drugs. Such linkers are postulated to release drug by the intracellular proteolytic degradation14. An example of the successful use of a non-cleaveable linker is Kadcyla®, approved by FDA in 2013 for the treatment of metastatic breast cancer. This ADC consists of the cytotoxic drug T-DM1 attached to an anti-HER2 antibody at the lysine residues via a non-cleavable linker. Despite the use of a non-cleavable linker, the intracellular release of T-DM1 is achieved. In this case, the intracellular release of T-DM1 is achieved not by the linker cleavage but by the complete proteolytic degradation of the antibody. The charged metabolite lysine-SMCC-T-DM1 does not permeate into the neighboring cells to induce a bystander effect.

Conjugation Chemistry

Random

Random conjugation takes place at the epsilon amino group of lysine or the cysteine residues derived by the reduction of interchain disulfide bonds19. For lysine based ADCs, such as Kadcyla®, theoretically conjugation can take place on the 40 unique solvent exposed lysine residues per antibody by direct acylation or with hetero-bifunctional, cross-linking agents. This phenomenon may result in one million different ADCs9. These conjugates are necessarily regioisomers, differing not only in drug distribution but also in lysine positions conjugated to drug molecules.

Cysteine based conjugation takes place by alkylation at the thiol groups produced by the reduction of four interchain disulfide bonds9,19. Cysteine conjugation results in ADC species carrying 0, 2, 6 or 8 drug molecules per antibody. Still, cysteine based conjugation can result in a heterogeneous mixture of conjugated species that may contain positional isomers, and a somewhat loosely bound antibody structure held together by non-covalent forces9,23.

The ADC species, produced by lysine or cysteine conjugation, differs in their stability profiles as well in their pharmacokinetic properties, the detailed discussion of which is beyond the scope of this article.

Site Specific

Site-specific conjugation has been investigated as a means to control the heterogeneity of ADC species produced by random conjugation. In site-specific conjugation cytotoxic drug molecules are introduced at specific sites of the antibody, producing either 2 or 4 drug molecules conjugated to the antibody9. This ratio is chosen for optimal cytototoxity and pharmacokinetic behavior.

Site-specific conjugation is primarily investigated under three broad categories:

Engineered Cysteine Residues

Cysteine residues were engineered at specific sites of the antibody to overcome the heterogeneity challenges associated with random conjugation. However, such engineered free cysteine residues could react intermolecularly with other cysteine residues to form dimers or intramolecularly with other cysteine residues to result in incorrect disulfide bonds or disulfide scrambling. Such reactions may affect the efficacy and safety of ADCs intended for patient administration.

The phage display-based method (PHESELECTOR) was subsequently employed by Genentech to introduce reactive cysteine residues at specific sites on the antibody without any effect on the folding and conformation of native protein Such antibodies with engineered cysteine residues have been referred to THIOMABs, which have shown highly homogeneous DAR values and excellent profiles in preclinical trials3,9,20.

Unnatural Amino Acids and Selenocysteine

This technology involves insertion of reactive functional groups on the antibody such as the twenty-first amino acid, selenocysteine, and the unnatural amino acid, acetylphenylalanine (pAcPhe). Selenocysteine contains a selenium atom in place of a sulfur atom unlike the traditional cysteine counterpart, making it a more reactive nucleophile, while reacting with electrophiles like maleimide during conjugation. The keto functional group of an engineered pAcPhe residue on trastuzumab was employed for site specific attachment to an auristatin F derivative using oxime chemistry9,11.

Enzymatic Conjugation

Enzymatic conjugation employs two enzymes, glycotransferase and transglutaminase, for the site-specific conjugation. The enzyme glycotransferase is mutated in the B4Gal-T1 pocket, to facilitate the transfer of the reactive functional group, 2-keto-Gal from the C-2 position of this modified galactose moiety. Human IgG antibodies are degalactosylated to G0 at the Asn-297 of the Fc fragment to achieve site specific conjugation with the 2-keto-Gal functional group, which then conjugates with cytotoxic drugs having reactive functional groups9,11.

Glutamine residues, alternatively called glutamine tags, were engineered at specific residues in the Fc domain of an anti-epidermal growth factor antibody. These engineered glutamine residues such as LLQG were conjugated with the free amino groups of cytotoxic drugs such as AcLys-vc-monomethylauristation D in the presence of the enzyme transglutaminase to form a covalent amide bond and a resulting average DAR of 22,9.

Conclusions

The science of ADCs continues to evolve as a rapidly progressing drug targeting technology for the treatment of cancer. The last few decades have witnessed enormous progress in the development of targeting antibodies, potent cytotoxic drugs, and drug-linker technologies to facilitate improved ADC stability, potency and targeting efficiency.

References

  1. Wang Z, Guravaiah N, Ning C, He Y, Yao L, et al. Antibody-Drug Conjugates: The Forefront of Targeted Chemotherapy for Cancer Treatment. J Drug Des Res. 2015; 2(3): 1016-1024
  2. Chari RVJ, Miller ML, Widdison, WC. Antibody-Drug Conjugates: An Emerging Concept in Cancer Therapy. Angew Chem Int Ed. 2014; 53(15): 3796-3827
  3. Peters C, Brown S. Antibody-drug conjugates as novel anti-cancer chemotherapeutics. Biosci Rep. 2015; 35(4): pii: e00225. doi: 10.1042/BSR20150089
  4. Casi G, Neri D. Antibody-drug conjugates: Basic concepts, examples and future perspectives. J Cont Rel. 2012; 161(2): 422-428
  5. Senter, PD. Potent antibody drug conjugates for cancer therapy. Curr Opin Chem Biol. 2009; 13(3): 235-244
  6. Bouchard H, Viskov C, Garcia-Echeverria C. Antibody-drug conjugates - A new wave of cancer drugs. Bioorg & Med Chem Lett. 2014; 24: 5357-5363
  7. Beck A, Haeuw JF, Wurch T, Goetsch L, Bailly C, Corvaïa N. The next generation of antibodydrug conjugates comes of age. Discov Med. 2010; 10(53): 329-339
  8. Lambert JM. Drug-conjugated antibodies for the treatment of cancer. Br J Clin Pharmacol. 2013; 76(2): 248-262
  9. Panowksi S, Bhakta S, Raab H, Polakis P, Junutula JR. Site-specific antibody drug conjugates for cancer therapy. MAbs. 2014; 6(1): 34–45
  10. Trail PA. Antibody Drug Conjugates as Cancer Therapeutics. Antibodies. 2013; 2(1): 113-129
  11. Sochaj AM, Swiderska KW, Otlewski, J. Current methods for the synthesis of homogeneous antibody-drug conjugates. Biotechnol Adv. 2015; 33(6 Pt 1): 775-784
  12. Van de Donk NW, Dhimolea E. Brentuximab vedotin. MAbs. 2012; 4(4): 458–465
  13. Kim MT, Chen Y, Marhoul J, Jacobson, F. Statistical Modeling of the Drug Load Distribution on Trastuzumab Emtansine (Kadcyla), a Lysine-Linked Antibody Drug Conjugate. Bioconjugate Chem. 2014; 25(7): 1223-1232
  14. Iyer U, Kadambi VJ. Antibody drug conjugates - Trojan horses in the war on cancer. J Pharmacol Toxicol Methods. 2011; 64(3): 207-212
  15. Alley SC, Okeley NM, Senter PD. Antibody-drug conjugates: targeted drug delivery for cancer. Curr Opin Chem Biol. 2010; 14(4): 529-537
  16. Mei M, Thomas S, Chen R. Management of Relapsed or Refractory Hodgkin Lymphoma with Second-Generation Antibody–Drug Conjugates: Focus on Brentuximab Vedotin. BioDrugs. 2014, 28(3): 245-251
  17. Ballantyne A, Dhillon S. Trastuzumab emtansine: first global approval. Drugs. 2013; 73(7): 755-765
  18. Ricart AD. Antibody-drug conjugates of calicheamicin derivative: gemtuzumab ozogamicin and inotuzumab ozogamicin. Clin Cancer Res. 2011; 17(20): 6417-6427
  19. Drachman JG, Senter PD. Antibody-drug conjugates: the chemistry behind empowering antibodies to fight cancer. Hematology Am Soc Hematol Educ Program. 2013; 2013:306- 310
  20. Sapra P, Shor B. Monoclonal antibody-based therapies in cancer: advances and challenges. Pharmacol Ther. 2013; 138(3): 452-469
  21. Deng C, Pan B, O’Connor OA. Brentuximab Vedotin. Clin Cancer Res. 2013; 19(1): 22-27
  22. Shefet-Carasso L, Benhar I. Antibody-targeted drugs and drug resistance--challenges and solutions. Drug Resist Updat. 2015; 18: 36-46
  23. Goswami S, Wang Wei, Arakawa T, Ohtake S. Developments and Challenges for mAb-Based Therapeutics. Antibodies 2013; 2(3): 452-500

Acknowledgements

The author would like to acknowledge Mark Bolgar for his valuable inputs about antibody-drug conjugates.

Potential Conflicts of Interest

Author is a full time employee of Bristol-Myers Squibb, and declares no conflict of interest.

  • <<
  • >>

Join the Discussion