What in the world is a mAb?: An Introduction to Therapeutic Use of Monoclonal Antibodies

Keeping on the theme of classes of drugs, I thought it would be interesting to discuss monoclonal antibodies (mAbs) next. Like peptides, mAbs are also biologics and composed of amino acids, but most of the similarities between these two drugs classes end there — the use and behavior of mAbs in patients is unique and fascinating (I think so, at least)!

The immune system is really complicated and I have personally found it challenging to distill this topic in a way that is both accessible to a layman and addressing all the important aspects of mAbs. I hope this is clear while still preserving the science well.

— SUMMARY —

Why we care: Therapies that are specific are safer and more predictable in humans. Monoclonal antibodies have been designed by the body to specifically bind things. We can take advantage of this to make therapies.

Take home points: Many monoclonal antibodies are already being used in the clinic for a variety of diseases. This is a proven therapeutic strategy. By specifically biding things, antibodies can be used in a variety of ways including getting the immune system to clear bad cells, inhibit molecular signaling, or deliver drugs specifically to the area it is needed.

— TABLE OF CONTENTS —

INTRODUCTION

STRUCTURE OF MABS
An introduction to mAb structure, isotypes, and vocabulary

HOW MABS WORK
How mAbs have a therapeutic effect in patients

MAJOR DEVELOPMENTS IN MAB TECHNOLOGY
Select important historical milestones

SELECTION AND PRODUCTION OF MABS
Overview of antigen selection, binding affinity, and two methods of producing and manufacturing mAbs.

MABS IN THE CLINIC
A list of market approved mAbs and a further dive into the use of mAbs for cancer, autoimmune disease, and Alzheimer’s disease.

REFERENCES

— INTRODUCTION —

Small molecules are what you probably think of first when you think of a drug: things like aspirin and ibuprofen are small molecules. The first drugs were small molecules, and they still dominate the market today. Biologics are therapies that are naturally derived: so things like peptides, proteins, and monoclonal antibodies as we will discuss. At their most simple, biologics are composed of carbohydrates (sugars), amino acids, and/or nucleic acids (DNA).

Monoclonal antibodies (also known as mAbs, immunoglobulins) are proteins which confer an organism’s immunity to antigens. Monoclonal antibodies have their effect at the cell surface: unlike the peptides we discussed previously, there is actually no desire for mAbs to be made cell permeable (which is great, because as we discussed, getting things into cells is hard)

Advantage of mAbs

• Highly specific for their target
• Off-target effects more predictable
• Can be dosed less frequently that some peptide, small molecules drugs
• Low risk of serious adverse events especially if fully human

Disadvantages

• Manufacturing costs higher than small molecules
• Product heterogeneity
• Possible immune reaction (risk proportional to the amount of non-human sections in the mAb)
• Not orally bioactive
• Structure can change in storage — e.g., aggregation, glycosylation

Tens of mAbs have received market approval for indications including cancers, transplant rejection, autoimmune diseases, and infectious diseases.

— STRUCTURE OF MABS —

Before we dig further, let’s get our vocab straight.

For each antigen (something in the body that can provoke an immune response), there is an antibody which is able to bind it specifically. This relationship between antigen and binding antibody confers the specificity that we take advantage of for therapies. The part of the antigen recognized by the antibody, and therefore the immune system, is known as the epitope. The shorter, outer arms are called light chains (referred to as Vl) and the longer arms the heavy chains (Vh). The blue part of the arms are consistent across antibodies (Fc), while the yellow tips are the variable regions and differ according to the specific antigen the antibody binds (Fv) Source: [1]

Light chain: The shorter arms seen on the outside of the antibody. There are two classes of light chains: kappa and lambda, although their function doesn’t differ significantly. These are about 200 amino acids long each.

Heavy chain: The longer arms in the antibody. The type of heavy chain defines the isotype of the antibody (more on this below in the constant region definition). A heavy chain is approx. 500 amino acids long.

Variable region: Approx. 120 AAs. Confers specificity for the antigen to be bound. Within this region, you have the hypervariable region and the framework region. The hypervariable region is also referred to as the complementarity determining region, as this region is of primary importance in specific antigen binding.

Constant region: Determines the mechanism with which the antigen is destroyed. There are five major categories/isotypes: IgA, IgD, IgE, IgG, and IgM which contain the alpha, delta, epsilon, gamma, or mu heavy chain respectively.

The five major subclasses of antibodies in mammals and their basic characteristics. kDa — kilodaltons, a measure of the size of a protein. 1 kDa is approximately 9 amino acids. Sources vary including [5]. IgG is believed to have such a long half-life due to a “protection receptor” which binds the antibody. Nearly all mAb therapies utilize IgG.

Some other terms you’ll hear in reference to antibodies: Fab (antibody-binding fragment of the antibody), Fv (the variable region of an anitbody, the bit that confers specificity to an antigen), Fc (tail region of the antibody).

— HOW MABS WORK —

A lovely overview of the broad therapeutic mechanisms of mAbs. A — mAb-mediated killing of the cell via ADCC, CMC, and opsonization (all three will be discussed below). B — Direct mAb activity on the ligand and receptor. Source: [8]

Unmodified mAbs can have two broad functions in vivo:

1. Target binding to suppress the biological activity of the target and its downstream effects or prevent a ligand binding; and
2. Target binding to mediate clearance of the cell

The function and type of cell death mechanism if applicable is determined by the “tail” of the antibody (shown in the below illustration as light blue). Antibody binding may trigger cell death or inhibit the target function. In some cases, mAbs can have multiple functions.

Note: I pestered a few mAb experts to explain to me exactly what determines whether the mAb acts merely as a blocker of a receptor/ligand or triggers a specific clearance mechanism, or does both, etc. They couldn’t tell me how besides “it’s experimentally derived” which is science code for “we throw it on some cells and see what happens”…

Target inhibition

A ligand is a molecule which binds another (usually larger) molecule in a biochemical reaction. Within the cellular membrane, there are proteins imbedded which are known as receptors. Specific ligands bind specific receptors, causing downstream biochemical effects in the cell (these can be referred to as a signaling cascade).

A common expression of a pathological cell is the up-regulation (increase) of expression of certain receptors which cause or facilitate the diseased state of the cell. For example: a cancer cell may have a receptor that, if bound by its ligand, increases the cell’s DNA synthesis capabilities and therefore allows the cell to replicate itself faster than normal. This receptor may be up-regulated in the cancer cell’s membrane, allowing for more receptor-ligand interactions, and therefore an increase in the proliferative activity of the cell.

A mAb which is engineered to bind the receptor’s binding site can block the ligand from binding the receptor, incapacitating the signaling pathway. The mAb can also be engineered to bind an allosteric binding site (not the main receptor-ligand binding site, but an alternative one which, when bound, changes the conformation of the receptor and makes it so the ligand is unable to bind). Alternatively, the mAb can bind the ligand itself, preventing its interaction with the receptor. In these roles, the mAb is functioning as an “antagonist” of the receptor. An agonist would be something that bound the receptor and increased its activity.

Transmembrane signaling

mAb binding to specific receptors may also induce the cell to apoptose or otherwise die in a manner that is not dependent on immune cell activity, as the following two mechanisms are. This is via intracellular signaling triggered by the binding of the antibody to the antigen.

Complement-dependent cytolysis (CDC)

The complement system is complicated: this video [6] explains antibody-mediated complement cytotoxicity well. I have summarized it below. Don’t be intimidated by all the acronyms!

The complement system is a subtype of the immune system. I really like molecular pathways so I went a bit further in-depth here than perhaps strictly necessary. To summarize: the antibody binds the target cell, the tail of the antibody is recognized sequentially by many proteins, named C1 to C8. This eventually causes many C9 proteins to come the to membrane and pierce through it to create a pore. These pores hurt the cell enough to cause it to die.

1. The C1 complex binds to IgG or IgM antibodies which are bound to the target cell.
2. This activates the C1-R and C1-S proteases, which are contained within the C1 complex. These activate another protein, C4, cleaving it into C4a and C4b.
3. C4b binds to the target membrane and recruits C2 to bind it. This protein complex is also cut by the proteases in the C1 complex, splitting C2 into C2a and C2b, activating the protein.
4. C4b/C2a remain bound at the cell membrane; this complex is now referred to as “C3 convertase”.
5. C3 is recruited to and cleaved by the C3 convertase into C3a and C3b; C3b remains and binds the cell surface. This trimer consisting of C3b/C4b/C2a can be referred to as “C3/C5 convertase”.
6. Multiple molecules of C3 are cleaved by the C3/C5 convertase into C3a and C3b. C3a fragments are released and C3b fragments continue to bind the target cell surface. These mediate phagocytosis (engulfment by another cell), but this is less relevant to antigen-mediated cell cytolosis (this is mostly for when the complement system is used to clear out invading bacteria)
7. C5 is bound to the C3/C5 convertase, cleaved into C5a and C5b. C5b recruits the membrane attack complex (MAC).
8. C5b associates with C6 and C7. C8 also binds and inserts itself into the membrane.
9. Multiple copies of C9 are recruited by this C5b/C6/C7/C8 complex. These form a pore in the membrane.
10. This membrane disrupts the normal functioning of the cell membrane and kills the target cell.

Whew! That was a journey…

Antibody-dependent cell-mediated cytotoxicity (ADCC)

A simplification of the ADCC pathway. NK — natural killer cell; Fc — constant region of the antibody; CD16 — the specific receptor on the surface of natural killer cells which recognizes the

Here, antibodies (generally, IgGs), bind the target cell. The constant region of this antibody is recognized by natural killer cells, which then release cytotoxic factors to induce the target cell to apoptose. Specifically, the antibody constant region binds FcRs, a receptor found on white blood cells including natural killer cells (NK, pictured above, kills targeted cells by puncturing them using perforin and help lysis using enzymes such as granzymes), and also granulocytes (contain granules within the cell body which contain materials to kill target cells), monocytes (can become macrophages), and macrophages (“big eaters”, eats bad things to clear them away, such as cancer cells but also cellular debris and pathogens).

Opsonization is where the antibody binding signals triggers the above cells to increase phagocytosis and lysis.

Conjugates

mAbs have also been explored to deliver cytotoxic compounds or drugs to target cells. These are referred to as antibody drug conjugates (ADCs). Chemotherapy is quite broad in that it effects any cell which replicates quickly, a defining characteristic of cancer cells, but also many healthy cells which causes off-target effects and side effects for the patients. Here, the cytotoxic agent is attached to the mAb with a linker that inhibits the toxin’s activity until the ADC is internalized by the cell. Then, the mAb and drug separate (for example, some ADCs have a linker that dissolves in high pH environments, as are found in the endosomes of cells).

Radioisotopes have also been linked to mAbs. Radioimmunotherapy, as these treatments are referred to, deliver a radioactive isotope to target cells in order to specifically eradicate them. Ideally, the radioisotope is only activated once in the target cell, as to minimize off-target damage while circulating.

— MAJOR DEVELOPMENTS IN MAB TECHNOLOGY —

1961 — First description of the structure of an antibody by Edelman et al

1975 — Method of creating monoclonal antibodies using hybridoma technique in vitro described by Köhler et al

1986—The first monoclonal antibody received market approval by the US Food and Drug Administration: muromonab (OKT3), indicated for in kidney transplant rejection. This was a murine mAb

1991 — Method of creating monoclonal antibodies using phage display described by Barbas et al

2002 — First humanized mAb, adalimumab, received market approval in the US for rheumatoid arthritis

2016 — The first biosimilar mAb was approved in the US, Inflectra (infliximab-dyyb) for multiple indications. (A biosimilar is a generic version of a branded drug , for example Bayer Aspirin is the branded drug while the grocery store’s local Aspirin is the generic. This is possible once the patent covering the drug has expired) [2]

— SELECTION AND PRODUCTION OF MABS —

The process of manufacturing mAbs is actually a bit more interesting than, say, a sequence of chemical reactions to make a small molecule (apologies to my past chemistry professors!), so I’ll overview the main methods briefly here.

Antigen Target

The characteristics of a good candidate antigen to target with a mAb for therapeutic uses:

• Membrane or extracellular expression if a receptor
• Large density and consistency of expression for receptors and ligands
• Highly expressed on the target cell type
• Low or no expression on non-target cells (if the antigen is highly expressed on other, critical cell types, this can result in side effects or severe adverse events when used in patients)
• Specificity of action if a ligand

Binding Affinity

The biological activity of a mAb is dependent largely on its binding ability and strength to its target. A strongly binding mAb may even be able to displace other less strongly bound molecules. The binding affinity is defined and quantified by the dissociation constant (Kd).

The ligand-protein complex LP is composed of x molecules of L and y molecules of P (here shown 1:1). The energy required to dissociate the complex into the constituent molecules is the dissociation constant.

SI units used in Kd values

The dissociation constant is the ligand concentration in molar units (M) where half of the total quantity of the two proteins are bound at equilibrium. Therefore, lower numbers are better here as it means that even in very low ligand concentrations, the proteins bind at a quick rate. One paper found the average Kd across 1,450 rabbit mAbs as 66 pM, with values from 10 pM to 200 pM. [9]

Design and manufacturing: hybridoma method

The hybridoma method was the first method of generating mAbs, as described in 1975 by Köhler et al.

1. Immunization of an animal (usually a mouse) against the antigen/epitope of choice. Whatever you inject them with needs to be capable of producing an immune reaction
2. B cells (the type of cell which produces antigens in vivo against foreign invaders and other things that the body fights with the immune response) in the mouse create antibodies which bind the antigen you introduced. We can refer to these B cells as the “antibody factory”
3. Harvest B cells from the mouse
4. Fuse the harvested B cells with myeloma cells (cancerous B cells, which are immortalized, AKA, they replicate a lot unlike a non-cancerous cell). You fuse with the myeloma cells to make the mouse B cell “the antibody factory” keep its antibody production up longer than it would normally. This fusion product is called a hybridoma
5. The hybridoma produces the identical multiples of the antibody of interest in vitro for the researcher to use!

This method will produce murine mAbs, which are unideal for therapeutic use. (see: Advantages and disadvantages). To get fully human mAbs (hu-mAbs), you can use transgenic mice (genetically modified mice) where the DNA encoding the mouse’s immune system has been replaced with the DNA encoding for the human immune system. Therefore, the B-cells isolated from the mouse to create the hybridoma secrete human mAbs.

Design and manufacturing: Phage display method

Mouse-derived portions of a mAb increase the chance of the patient developing human anti-mouse antibodies. Source: [4]

First off — bacteriophages (commonly called phages) look like little robots, which I find fantastic. These little robotic creatures are specially-evolved viruses which infect bacteria, hence the name.

Phage display is a relatively high-throughput method of identifying and expanding antibodies for an antigen of interest.

I personally find this method confusing to wrap your head around in the first go; hopefully this simplification below introduces it in an accessible manner. Much of this is adapted from [3]

1. Immobilize the target peptide, antigen, epitope etc onto a surface (e.g., a membrane) so that the binding portion is accessible to molecules
2. Prepare a library of antibodies, usually from B cell lines or synthetically designed libraries. “Library” here means basically having a large number of different antibodies
3. Litigation of the cDNA for the variable regions (the bits that confer the specificity to the epitope) of the antibodies from the antibody library into the bacteriophage plasmid encoding the structural protein of the phage. A few concepts to explain here: a) cDNA: DNA synthesized from a strand of RNA. b) Variable region: the part of the antibody that holds the specificity to the specific peptide/epitope. Shown in blue in the above antibody illustration. c) Plasmid: a short circular DNA loop which encodes for a small amount of proteins. Here, the plasmid would code for the proteins which create the outer structure of the phage. d) Litigation of peptides into the plasmid: this is kind of like copying and pasting. You can “copy” the DNA sequence encoding the peptide of interest, and “paste” it into the plasmid encoding the structure of the phage. Then, when the plasmid is transcribed from DNA to protein, your peptide of interest will be encapsulated into the structure of the phage!
4. Expression of the phage. Your antibody variable region should be anchored to the coat of the phage, but free to bind and interact with other materials (the epitope-binding region should be viable).
5. The phages displaying the various types of antibodies is added to the dish where your target peptide is. This is allowed to incubate for a certain amount of time, and then the membrane is washed multiple times to remove the phages which did not bind to the peptide of interest.
6. The phages which have bound to the peptide are removed, expanded using bacteria, and then the incubation process is repeated. This repetition allows the isolation of the antibodies which have the strongest binding affinity to the peptide.
7. After multiple repeats, you can sequence the DNA of the phage (which contains the sequence for the antibody fragment) to identify the antibody fragment which had the strongest binding affinity for the target.

— MABS IN THE CLINIC —

Here is a list of all of the market approved mAbs by the EMA and FDA as of 2017: link. I haven’t checked the list for completion, but at minimum this list provides a good overview of mAb market.

Suffixes for the technical name of the mAbs vary depending on their type. Mouse: -omab; chimeric: -ximab; humanized: -zumab; and fully human: -umab. The potential immunogenicity of a mAb is correlated with the amount of mouse DNA it contains. Source: [4]

The first mAbs were murine (mouse) mAbs. While easier to produce using the hybridoma method, murine mAbs are not clinically viable as they are seen as foreign to the patient’s immune system. In the best case, this results in a very short half-life of the mAb as it is cleaved and degraded by the patient’s immune system. In worse cases, this can trigger severe side effects for the patient. As a rule, the less mouse the mAb is, the less potential there is for a negative immune reaction.

mAbs are being used in many different classes of disease. I have outlined a few below that I find interesting!

Cancer

mAbs have been of particular interest in cancer. The idea of a “magic bullet” [10] for cancer is obviously very alluring — killing cells is not necessarily difficult, the difficult part of designing cancer therapies is developing drugs which are good at killing just cancer cells.

[8] is a great overview of the varying therapeutic mAb strategies used in cancer. mAbs have been used to target cancer cells for immunomediated clearance, to block receptors which promote cellular proliferation and survival, and deliver chemotherapies and other drugs specifically to cancer cells.

An important case study cancer mAb is Herceptin (trastuzumab), developed by Genetech and approved by the FDA in 1998. A humanized IgG antibody, it binds human epidermal growth factor receptor-2 (HER-2) positive breast cancer cells. HER2 is over-expressed in some breast cancer cell types, promoting excessive cellular growth. HER-2 is bound by epidermal growth factors (EGFs), triggering this excessive growth. The mechanism of Herceptin is not completely understood, but it is thought to block the binding of the ligand, and to trigger ADCC clearance.

Autoimmune diseases

Autoimmune disease is a class of disease where the patient’s immune system attacks normal and healthy cells. The first biosimilar mAb (see: Major Developments in mAb Technology above), infliximab, is indicated for autoimmune diseases. Infliximab is a chimeric IgG which binds tumor necrosis factor alpha (TNF-alpha), a chemical messenger/ligand involved in multiple pro-inflammatory and pro-immune pathways. By binding TNF-alpha, infliximab inhibits the activation of these pro-inflammatory pathways and the pathological increase of immune reactions.

Alzheimer’s disease [11–12]

Alzheimer’s disease is perhaps not an initially obvious candidate for mAb therapies, but mAbs have actually been explored heavily for use in Alzheimer’s.

It is generally thought that the aggregation of misfolded amyloid-beta (a-beta) peptides is a significant contributor to the characteristic neurodegeneration and general neuronal dysfunction of Alzheimer’s. mAbs can cross the blood-brain barrier (at a very low efficacy: [12] cites 0.1%) and bind the plaques (aggregated misfolded a-beta peptides) or free a-beta, promoting their immune clearance.

Soluble a-beta is thought to circulate though the entire body, not just reside in the brain. Therefore another theory, the “peripheral sink hypothesis”, theorizes that by clearing a-beta circulating in the peripheral system (AKA not the brain) using techniques such as mAbs, a-beta in the brain will “leak out” of the brain, into the periphery to reach equilibrium between the amount of a-beta on each side of the blood brain barrier.

mAbs targeting a-beta have thus far failed to show patient benefit in clinical trials. None have received market approval. As I mentioned in my previous post on peptides, Alzheimer’s is a challenging disease to hit.

— — —

I hope this was helpful! Feel free to email me at celine.halioua@gmail.com if you have any questions, notice a mistake, or just want to chat science!

— REFERENCES —

[1] Folz I, Karow M, Wasserman S. Evolution and Emergence of Therapeutic Monoclonal Antibodies What Cardiologists Need to Know. Circulation. 2013;127(22): 2222–2230.

[2] Brennan, Paul. FDA Approves Second Biosimilar, First MAb Biosimilar for US Market. Regulatory Affairs Professional Society, April 5, 2016. Available at: https://www.raps.org/regulatory-focusTM/news-articles/2016/4/fda-approves-second-biosimilar,-first-mab-biosimilar-for-us-market.

[3] Clementi et al. Phage Display-based Strategies for Cloning and Optimization of Monoclonal Antibodies Directed Against Human Pathogens. International Journal of Molecular Sciences. 2012;13: 8273–8292.

[4] Catapano A and Papadopoulos N. The safety of therapeutic monoclonal antibodies: Implications for cardiovascular disease and targeting the PCSK9 pathway. Atherosclerosis. 2013;228(1): 18–28.

[5] GenScript. Antibody Services Technical Resources. Available at: https://www.genscript.com/antibody_technical_resources.html

[6] The Complement System. April 16, 2009. Available at: https://www.youtube.com/watch?v=vbWYz9XDtLw

[7] Robinson A, Watson W, Leslie K. Targeted treatment using monoclonal antibodies and tyrosine-kinase inhibitors in pregnancy. Lancet Oncology. 2007;8(8): 738–743.

[8] Weiner G. Building better monoclonal antibody-based therapeutics. Nature Reviews Cancer. 2015;15(6): 361–370.

[9] Landry JP, Ke Y, Yu GL, Zhu XD. Measuring Affinity Constants of 1,450 Monoclonal Antibodies to Peptide Targets with a Microarray-based Label-Free Assay Platform. Journal of Immunological Methods. 2015;417: 86–96.

[10] Eccles S. Monoclonal antibodies targeting cancer: ‘magic bullets’ or just the trigger? Breast Cancer Research. 2000;3(2): 86–90.

[11] Prins N and Scheltens P. Treating Alzheimer’s disease with monoclonal antibodies: current status and outlook for the future. Alzheimer’s Research and Therapy. 2013;5(56): 1–6.