Thursday, 24 April 2014

Antibody Drug Conjugates (ADCs) a new type of compound that potentially cures end-stage terminal patients

How does a new class of cancer drug, called Antibody Drug Conjugate, eliminate 70 lethal tumors within a terminal patient in just two weeks.  

Before (left) and after (right) treatment scans. The image on the left 
shows the extent of metastatic disease (spread of cancer) in the 
patient (70 tumours from Non-Hodgkin's Lymphoma). The scan on 
the right demonstrates complete elimination of tumours two weeks 
after treatment (the black blobs in the scan on the right are normal 
(these are the brain, kidneys, and the bladder).
Various media outlets recently reported that a 47 year old man in the UK with only weeks to live, made a full recovery from his terminal Non-Hodgkin’s Lymphoma cancer following treatment with a recently approved drug called Brentuximab vedotin (Adcetris). His body, as seen in the scan on the left, was riddled with approximately 70 fatal tumours. These tumours had spread to distant sites throughout his body, yet within a space of two weeks all these tumours had disappeared. The drug responsible for this remarkable recovery is part of a new class of immunotherapeutics called Antibody Drug Conjugates (ADCs). Although the therapy itself wasn’t a particularly pleasant experience (apparently the patient didn’t feel well for the first few days), it gave the patient a choice; death within a couple of weeks or a chance to be in complete remission. ADCs are also being developed for cancer of the lung, colon, prostate, brain and other solid tumours as well as leukaemia. Approximately 40 ADCs are currently undergoing clinical trials (see the enclosed list below with details and links to clinical trials).

In this article I will discuss the following:

  1. What is Brentuximab vedotin (Adcetris)?
  2. What are Antibody Drug Conjugates (ADC)?
  3. Different technologies and compounds that make up an ADC
  4. The biotech and pharmaceutical companies that are developing ADCs for other cancers
  5. List and links to current clinical trials investigating ADCs for other cancers

What is Brentuximab vedotin (Adcetris)?

Brentuximab vedotin or its marketing / brand name, Adcetris is a cancer treatment for lymphoma (both, Hodgkin’s and Non-Hodgekin’s Lymphoma types).
This cancer therapy is a completely new class of drug that has been in development since the early part of the 21st century and is called Antibody Drug Conjugate (ADC), although some of the components that make up this targeted cancer drug have been around for a bit longer.
Essentially ADCs are made up of 3 parts:

1.     A monoclonal antibody that specifically targets cancer cells
2.     A highly toxic compound (e.g. monomethyl auristatin E (MMAE))
3.     Technology that links the above two entities

It is the combination that makes up this new class of novel compounds, such as Brentuximab vedotin; compounds that have added a second dimension to the monoclonal antibody treatment paradigm. Arming exquisitely specific monoclonal antibodies with a toxic payload is a truly innovative- and quite possibly effective way of treating cancer as the tumour cells get hit with a “double whammy”. In the case of Brentuximab vedotin this means that the monoclonal antibody component will target a receptor (called CD30) that is important to the tumour cell, while subsequently (once the antibody has been delivered to the interior of the cell), the antibody will release its toxic payload (a cytotoxic, microtubule-disrupting agent, called monomethyl auristatin E (MMAE)) that will then literally cause the cancer cell to self-destruct. Thankfully, ADCs are gaining acceptance in the oncology community and are expected to become a major contributor to improved cancer therapy.

What are Antibody Drug Conjugates (ADCs); a more scientific explanation

Antibody Drug Conjugates (ADCs) consist of a monoclonal Antibody (mAb) or antibody
Antibody Drug Conjugate ADC schematic diagram
Various schematic representations of an Antibody Drug Conjugate (ADC). 
Note the difference in size; IgG antibody is approximately 150 kDa in size, 
while the attached drug is around 100 times smaller. Attaching drugs to 
an antibody can cause a conformational change that may inhibit function. 
fragment which is chemically coupled (i.e. conjugated) to a cytotoxic drug via a synthetic linker (e.g. disulfide or  non-cleavable thioether linker chemistry). In general, ADCs deliver deactivated cytotoxins specifically to cancer cells. Once in the tumour cell, the cytotoxins are released from the ADC, regaining their full cytotoxic activity, which in turn leads to rapid cell death. The concept and mechanistic aspects of ADCs are easy to understand and relatively straightforward, however, the design and synthesis of a fully functional and effective ADC entity is remarkably challenging. As such, several aspects of ADC design need careful consideration in order to optimise stability in circulation and the targeted local or intracellular effect. 

Considerations include isotype selection during the antibody engineering phase. Depending on the choice of IgG isotype, different mechanisms of effector action will be elicited in vivo. While, IgG1 and IgG3 are highly active in initiating Antibody Dependent Cell-Mediated Cytotoxicity (ADCC), and Complement Dependent Cytotoxicity (CDC), IgG2 and IgG4 are much less capable in evoking such an immune response. Interestingly, most ADCs currently in development are of the IgG1 isotype. Only a small number of IgG2 and IgG4 isotypes are utilised in ADC development, while IgG3 is not used used at all (probably for reasons of instability). Although, some of these isotypes belong to the IgG2 or IgG4 class, they are likely to have been modified in the hinge region in order to exert greater control in vivo, e.g. over mAb half-body formation.

In addition, linker technology and type of conjugated toxin (e.g. duocarmycin derivatives, doxorubicin, maytansine, etc…) play an important role in how (or even if) cancer cells will be killed. For example, duocarmycins disrupt the DNA of tumour cells at any phase of the cell cycle, unlike many other toxins that are conjugated to ADCs which only attack tumour cells in a dividing state. Another characteristic to take into consideration is the level of efficacy in treating tumour cells that are multi-drug resistant (again, pre-clinical tests have shown that duocarmycins are able to overcome resistance).

How do monoclonal antibody properties govern ADC function?

Compared to conventional cancer treatments or drugs, monoclonal antibodies (mAbs) used as pharmacological entities have physical characteristics and modes of action that are particular to this class of therapeutics. mAbs are highly specific (targeting very precisely). Dependent on the backbone design, they can induce powerful immune responses all by themselves. However, appropriate selection of antigen targets for antibody cancer therapy requires a comprehensive understanding and analysis of tumour-associated antigen expression.  Unfortunately, truly tumour specific antigens have not been identified. As such, mAbs used as a cancer therapy
A schematic overview of the various types of monoclonal antibodies used
in cancer therapy. 

are often targeted against tumour-associated antigens rather than antigens unique to just tumour tissue. This means that tumour associated antigens are often highly expressed on cancer cells while only a limited expression occurs within normal tissues, or that they are expressed at early stages of development (i.e. in the foetus) but not in adults (i.e. temporal difference in expression).  Monoclonal antibodies used in cancer therapy are derived from various starting materials. Fully human antibodies are predominantly generated either with the use of transgenic mice and subsequent conventional  hybridoma technology, or from single-chain variable fragment phage display display techniques combined with a prefabricated human constant region. Humanized antibodies are made by replacing the Complementarity-Determining Regions (CDRs) of a human IgG antibody with the CDRs of a mouse antigen-specific monoclonal antibody. In order to minimise loss of target affinity, one or more amino acid residues from the Framework Regions (FRs) are also often incorporated. Chimeric antibodies are created by joining the antigen binding variable heavy- and light-chain domains (VL and VH) of a mouse monoclonal antibody specific for a particular antigen with the constant region domains (CH1, CH2, and CH3) of a human monoclonal antibody.Please note that I will discuss these types of antibodies in more detail in a video in the near future, please follow me @PvanUden on Twitter or on Facebook at if you want to be notified on the day when the video is being released.

Other aspects that require a sound understanding when designing mAbs are level of homogeneity of target antigen expression within tumour tissues as well as its physiological role in tumour development. Also, in contrast to mAb therapies (mAbs without being conjugated to a drug) in which a very slow internalisation process of the antigen-mAb complex  is preferred in order to elicit ADCC or CDC immune responses, rapid internalisation is desirable for ADCs delivering toxins into the cancer cell and for antibodies whose action is primarily based on downregulation of cell surface receptors. Hence, target antigen and antibody isotype selection are both important factors to consider in the design of an ADC.

In summary, the safest and most efficacious mAbs used in ADC development are those that target antigens which are expressed selectively and homogeneously  at a high density on the surface of malignant cells, given that intracellular concentrations of cytotoxic compounds inside cancer cells (i.e. killing of cells) is directly related to the level of antigen expression and the efficiency by which the ADCs are internalised.

Linker technology and toxins used in ADC synthesis 

ADC cleavable linkers hydrazone disulfide bond peptide coupling drugs antibody
Various types of cleavable linkers are depicted in the diagram.
Early ADC development utilised hydrazone linkers, whereas more
recent developments include the disulfide bond and amide bond
linker technology. 
Attaching a drug to an antibody requires a linker that is stable in circulation in order to match the long half-life of an antibody in serum, while it should simultaneously be able to release the active form of the drug following antigen mediated internalisation by a tumour cell. Linker chemistry can be categorised on the basis of their inherent drug release mechanism. Cleavable linkers, the most common category in clinical development, release the active form of a drug as a result of either acidic and reducing conditions or enzymatic cleavage of the labile bond, while the non-cleavable linkers release the drug once the antibody is degraded in the lysosome following internalisation. For example, hydrazone bonds release the conjugated drug in the lysosome as a result of acidic conditions, whereas disulfide bonds release the attached toxic payload following intracellular reduction. The use of amide / peptide bonds has improved serum stability of ADCs considerably, while it permits rapid enzymatic cleavage once an ADC has been internalised by a cancer cell. An interesting example is the valine-citruline (peptide) based linker, which shows a substantially improved stability profile in serum (> 9 days) when compared with hydrazone-doxorubicin (43 hours).  Please see different linkers in the figure on the left.

Drugs used to create ADCs

Schematic diagram representing an Antibody 
Drug Conjugate (ADC) attached to Doxorubicin 
via a hydrazone linker (e.g. Milatuzumab-Dox 
developed by Immunomedics)
A range of tumouricidal drugs are used to create ADCs. Given that all major classes of chemotherapeutic drugs are associated with dose-limiting toxicities, some of these have since been tried and tested (and in some cases rejected) as an antibody drug conjugate. The first generation of experimental ADCs often incorporated cytotoxic drugs such as Methotrexate (an inhibitor of dihydrofolate reductase), Vinblastine (a plant-alkaloid and microtubule inhibitor), or Doxorubicin (a DNA intercalating drug that inhibits topoisomerase II), drugs which are still commonly used on their own to treat various types of cancers. Although, encouraging results were obtained initially in pre-clinical studies with these early ADCs, they required high doses to be administered in order to achieve significant anti-tumour activity. To increase the potency of these first generation ADCs, a range of methods were employed including increasing the drug:antibody ratio by utilising branched linkers or direct conjugation. While these methods were able to increase the potency of early ADCs to some extent, results from phase I and phase II clinical trials have demonstrated that these early types of conjugated drugs are unlikely to yield objective responses within cancer patients.  

Given that conjugation of a drug alters the pharmacokinetics as well as the pharmacodynamics of a drug, some drugs, that were initially deemed either too toxic for use in humans or that were cleared too rapidly from circulation, have since been re-examined for use as a cytotoxic agent to arm chimeric-, humanized-, or fully human- antibodies.  

Broadly speaking these cytotoxic compounds are either anti-microtubule agents or DNA minor groove binders which are biologically active at an extremely low dosage (ng/Kg). This level of potency places these compounds in the most potent class of advanced cancer drugs.

The first ADC licensed for clinical use, Gemtuzumab Ozogamicin (Mylotarg; Pfizer (previously Wyeth)) was approved by the FDA in 2000 for the treatment of relapsed acute myelocytic leukemia in adults. The cytotoxic compound that is arming this ADC is called N-acetyl-g-calicheamicin which binds in the minor groove of DNA and consequently causes double strand DNA cleavage which results in target cell death.
A highly potent and more recent version of such a minor groove binder are a class of drugs called Duocarmycins (developed by Synthon, e.g. ADC SYD985).

The cytotoxicity of maytansine analogues, such as DM1, DM4, or monomethyl auristatin E (MMAE) and monomethyl auristatin F (MMAF) compounds is realised because of their ability to obstruct cell division by inhibiting tubulin. This inhibition of tubulin arrests target cells in the G2/M stage of the cell cycle which results in apoptosis (cell death). These very potent drugs on their own (which kill tumour cells at sub-nanomolar concentrations), will indiscriminately destroy both healthy and diseased tissue by stopping mitosis. Hence, accurate targeting is required to realise the full potential of these new cancer drugs.

To put this in context, highly potent cytotoxics such as calicheamicins, maytansinoids, auristatins, and duocarmycins are 100 to 1000 times more potent than the first generation drugs used to create ADCs (e.g. Doxorubicin, or Vinblastine, etc...). Most of these potent cytotoxic compounds had failed clinically as free drugs because they were simply too toxic for use in humans. Hence, chemically coupling them to monoclonal antibodies to precisely target tumour tissue provides a means to clinically exploit the potency of these drugs. Below I have included diagrams of the structures of drugs and linkers that are currently being investigated in clinical trials, including some examples.


The duocarmycin analogues are extremely cytotoxic members of a small group of natural products that are able to exert their mode of action at any phase in the cellular cycle (Duocarmycins were  first isolated from Streptomyces bacteria in 1988). These synthetic small-molecules are DNA minor groove binding alkylating agents that cause irreversible alkylation of DNA. This alkylation of DNA disrupts the nucleic acid architecture, which eventually leads to tumour cell death. As mentioned earlier, unlike tubulin binders, which will only attack tumour cells when they are in a mitotic state, Duocarmycins work at any phase of cell cycle. Recent research suggests that DNA damaging agents, are more efficacious in killing cancer cells than tubulin binders, especially solid tumours.

Schematic diagram representing an Antibody Drug Conjugate (ADC) attached to 
Duocarmycin (e.g. an ADC such as SYD985 developed by Synthon)
Duocarmycins have a potency in the low picomolar range (i.e. extremely little is needed to kill cells), which maximizes the cell-killing potency of antibody-drug conjugates that utilise this compound. In pre-clinical tests a new ADC linked to Duocarmycin called SYD985 (Synthon) outperformed another ADC conjugated to DM1 (Kadcyla developed by Roche/Genentech) in a breast cancer study. Both ADCs utilise an anti-HER2 monoclonal antibody called Trastuzumab to which their respective drugs are conjugated.

In addition, Duocarmycins have shown activity in a variety of Multi-Drug Resistant (MDR) tumour cells (e.g. potent cytotoxicity has been observed in cells that express the P-glycoprotein (P-gp) efflux pump). Multi-drug resistance can be a significant problem in the clinical setting (particularly in end-stage terminal cancer patients). Compounds which are less susceptible to these mechanisms of drug resistance are likely to be more successful in the prolonged successful treatment of terminal cancer patients. 


Calicheamicins also bind to the minor groove of DNA which results in double strand breaks in DNA and apoptosis (cell death). Again, this DNA binder is also extremely cytotoxic and is active at sub picomolar concentrations.

Schematic diagram representing an Antibody Drug Conjugate (ADC) attached to 
Calicheamicin (e.g. ADCs such as Inotuzumab ozogamycin and Gemtuzumab 
ozogamycin (Mylotarg) originally developed by Wyeth (now part of Pfizer))
An example of an ADC that is chemically coupled to Calicheamicin is called Gemtuzumab ozogamacin (Mylotarg). In this instance calicheamicin is conjugated to a humanized anti-CD33 mAb via a hydrazone linker (see figure on the left).

This compound is extremely potent and demonstrated antigen-specific activity in preclinical models at doses of approximately 100 μg/kg.

Gemtuzumab ozogamacin received accelerated approval from the FDA in 2000 for the treatment of CD33-positive Acute Myeloid Leukemia (AML). However, this ADC was voluntarily withdrawn from the market in 2010. 

Maytansinoids DM1 and DM4

Schematic diagram representing an Antibody Drug Conjugate 
(ADC) attached to DM1 via a non-cleavable linker
Encouraging clinical trial data have been reported for Trastuzumab Emtansine (also known as Kadcyla or T-DM1), an antibody-drug conjugate that utilises a monoclonal antibody called  Trastuzumab which is chemically linked to a maytansinoid drug. Maytansine, including its analogs (maytansinoids), are potent microtubule-targeting compounds that inhibit proliferation of cells that are in the mitotic phase of the cell cycle. DM1 and DM4 are benzoansamacrolides which are derived from ansamitocin. These derivatives differ in steric hindrance around the disulfide bridge. Antibody-maytansinoid conjugates which consist of maytansinoids (DM1 or DM4) that are attached to tumor-specific antibodies can be seen to the left and right of the text in this section. The maytansine linkers are chemically coupled through the amino groups of mAb lysine residues.

Conjugated maytansinoids (once released) potently inhibits breast cancer cell proliferation at sub-nanomolar concentrations, by arresting the cells in the mitotic pro-metaphase / metaphase. Given that T-DM1 utilises a non-cleavable linker it is thought that drug release from the ADC happens as a result of degradation of the antibody (Trastuzumab) inside lysosomes. Blocking of cell cycle progression occurs in concert with the internalization and intracellular processing of ADCs, which induces abnormal mitotic spindle organisation and suppresses microtubule dynamic instability. It is thought that microtubule depolymerisation only occurs at much higher drug concentrations.

Schematic diagram representing an Antibody Drug Conjugate (ADC) 
attached to DM4 via a cleavable linker (e.g. SAR3419 developed by 
Sanofi Pasteur or IMGN853 from Immunogen)

Genentech has licensed the drug and linker technology (DM1 and N- sucinimidyl 4-(maleimidomethyl) Cyclohexane, a thioether linkage via lysine residues) for their antibody from ImmunoGen and Seattle Genetics. Trastuzumab emtansine is targeted for use in patients with advanced HER2-positive breast cancer. Clinical trial data indicates that Trastuzumab Emtansine is stable in circulation for at least 7 days after administration and that it is superior to standard treatment.  
In addition to non-cleavable Maytansinoid drug conjugates, cleavable linkers in combination with DM4 are also being studied in clinical trials (E.g. IMGN853 and SAR3419). SAR3419 consists of a humanized antibody that is coupled to DM4 using a cleavable hindered disulfide linker. SAR3419 is targeted at patients with lymphoma (several lymphoma types, please see list and links to clinical trials at the bottom of this article). SAR3419 is currently undergoing phase II clinical trials.

IMGN853, developed by ImmunoGen, is composed of an anti-FOLR1 antibody conjugated to the cytotoxic maytansinoid, DM4, via a disulfide-containing linker, SPDB, derived from the experimental antibody drug conjugate M9346A-sulfo-SPDB-DM4. IMGN853 is targeted at patients with ovarian cancer or other solid tumours that over-express FOLR1 (this is also known as Folate Receptor alpha), including Non-Small Cell Lung Cancer (NSCLC). The linker in IMGN853 serves a dual purpose. It is meant to keep the DM4 stably attached to the antibody while the compound is in the bloodstream, but also to optimise release of the drug once it has been internalised. Pre-clinical studies have demonstrated that this combination of linker and maytansinoid drug is superior to other constructs that utilise either DM1 or DM4 as the conjugated drug.

The Auristatins (MMAE and MMAF)

The auristatin analogs, MonoMethyl Auristatin E (MMAE) and MonoMethyl Auristatin F (MMAF) are derived from pentapeptides, called dolastatin 10, found in D. auricularia, a small sea mollusc). The bind to tubulin and inhibit mitosis. Dolastatin 10 is much more potent than Vinblastine and is cytotoxic (kills cells) at subnanomolar concentrations.

Schematic diagram representing an Antibody Drug Conjugate (ADC) 
attached to Mono-MethylAuristatin E (MMAE) via a stable valine-
citrulline dipeptide linker (e.g. Glembatumomab vedotin (CDX-011)

which is being developed by Celldex Therapeutics)
The majority of ADCs that are currently undergoing clinical trials belong to this class of drugs. Most developers that use MMAE as the conjugated drug, utilise the valine-citrulline (vc) linker to chemically attach MMAE to the monoclonal antibody. Following internalisation by a tumour cell, the linker is cleaved by lysosomal enzymes (e.g. Cathepsin B), which will subsequently release free MMAE. Given that MMAE is able to cross cellular membranes, local bystander killing (cells in close proximity) may occur, even if those cells do not express the antigen to which the conjugated antibody is targeted.

Brentuximab vedotin (see the start of this article) is such a vc-MMAE ADC. As mentioned, impressive results have been obtained with this particular ADC in various clinical trials (including complete remissions in terminal cancer patients and objective tumour response rates of 50% were seen in patients treated at the Maximum Tolerated Dose (MTD)(1.8 mg/kg). Interestingly, the antibody on its own was shown to have no effect.

Schematic diagram representing an Antibody Drug Conjugate (ADC) 
attached to MMAF via a non-cleavable Maleimido Caproyl linker (e.g. 
SGN-CD19A developed by Seattle Genetics)
In addition, biotech companies have developed non-cleavable maleimidocaproyl (mc) linked conjugates of the auristatin analog called MMAF. The mc linker attaches MMAF to solvent accessible thiols present in mAb cysteines. As such, it is thought that MMAF is released following degradation of the antibody in the lysosomal compartment (i.e. after the ADC has been internalised by a cancer cell). Interestingly, the MMAF drug is unable to cross cellular membranes and as a consequence bystander killing is unlikely to occur.
Examples of MMAF conjugated ADCs would be SGN-CD19A or SGN-75 (although SGN-75 has been discontinued by Seattle Genetics and is superseeded by SGN-CD70A (a PBD dimer based ADC (pyrrolobenzodiazepines (PBD) dimers, are derived from a toxin originally isolated from various Streptomyces and are developed by Spirogen)). SGN-CD19A is targeted at patients with various types of lymphoma and is currently undergoing phase I clinical trials (please see list with links at the bottom of this article).

More ADC variants are expected to enter phase I clinical trials in the near future. Some of the new compounds to look out for are Duocarmycins developed by Synthon (as mentioned earlier in the text), but also compounds such as PBD dimers developed by Spirogen which has signed licensing agreements with Genentech and Seattle Genetics. Apparently, they also have a number of additional collaborations with unnamed pharma and biotech companies. Spirogen says that its PBDs can incorporate a wide variety of linker and conjugation chemistries. Two PBD based ADCs are currently in Phase 1 trials, including SGN-CD33A from Seattle Genetics.
Another company, called Heidelberg Pharma, has developed a potent RNA polymerase inhibitor from a mushroom with the name Amanita phalloides, while Viventia Biotechnologies has developed a de-immunised form of a bouganin protein toxin that is derived from the leaves of Bougainvillea spectabilis.

Given the rapid developments in this area, it is likely that I will write a more detailed article on ADCs utilising the drugs briefly mentioned in the previous paragraph. If you would like to be informed on the day it is published, please follow me at Twitter @PvanUden. 

List of Antibody Drug Conjugates (ADC) for the treatment of cancer currently (2014) undergoing clinical trials. Please see the list
with links to clinical trials below.


ADC & clinical trial link
1 IGN523
2 DEDN6525A (RG7636)
3 Anti-MUC16 (DMUC5754A, RG7458)
4 Anti-NaPi2b (DNIB0600A, RG7599)
5 Anti-STEAP1 (DSTP3086S, RG7450)
6 Pinatuzumab vedotin (Anti-CD22, DCDT2980S, RG7593)
7 Polatuzumab vedotin (Anti-CD79b, DCDS4501A, RG7596)
8 Trastuzumab emtansine [Kadcyla], trastuzumab-MCC-DM1, T-DM1
9 IMGN853
10 SGN-CD19A
11 SGN-CD70A (superseding SGN-75)
12 Brentuximab vedotin (SGN35, ADCETRIS)
13 SGN-CD33A
15 ASG-22ME / ASG-22M6E
16 ASG15E-13-1 (ASG-15ME)
17 Inotuzumab ozogamicin 
18 SAR3419
19 SAR566658
20 SYD985
21 HuMax-TF-ADC (TF-011-MMAE)
22 BT-062
23 Glembatumomab vedotin (CDX-011)
25 IMGN289 (J2898A)
26 Lorvotuzumab mertansine
27 Milatuzumab-dox
28 SGN-75
29 AGS-16M8F
30 BAY-94-9343
31 BIIB015
32 IMGN529 (K7153A)
33 IMGN853 (M9346A)
34 IMMU-130 (hMN-14-SN38 / Labestuzumab-SN-38)
35 IMMU-132 (hRS7-SN38 ADC)
36 MLN0264
37 AMG 595
38 AMG 172
39 PF-0626350
40 SC16LD6.5
41 Gemtuzumab ozogamicin
42 ASG-5ME
43 Bay 79-4620

Please note: Any medical or scientific information published on this website is not intended as a substitute for informed medical advice from a physician and you should not take any action before consulting with a health care professional. For more information, please read my terms & conditions.