2 Clarke Drive
Suite 100
Cranbury, NJ 08512
© 2024 MJH Life Sciences™ and OncLive - Clinical Oncology News, Cancer Expert Insights. All rights reserved.
More than a century has passed since the discovery of antibodies and thanks to a number of Nobel Prize-winning scientists, we have begun to realize their potential as therapeutic tools in cancer.
More than a century has passed since the discovery of antibodies and thanks to a number of Nobel Prize-winning scientists, we have begun to realize their potential as therapeutic tools in cancer. More than a dozen antibodies have been approved for a variety of different cancer indications, and their clinical successes have revolutionized treatment in many instances.
Although antibodies have yet to meet their full potential as single anticancer agents, new technologies for engineering improved antibody structures are paving the way for further success. Here, we provide a short guide to the past, present, and future of anticancer antibody therapy.Antibodies, or immunoglobulins (Igs), are proteins found in bodily fluids, produced by the B cells of the immune system, which coordinate immune responses against foreign substances. Each has a unique target (an antigen) that it recognizes on the surface of the foreign cell to which it binds, and then recruits other components of the immune system to destroy the cell.
Igs exist in five different forms—IgA, IgD, IgE, IgG, and IgM—that differ in their biological properties, functions, and location in the body. IgG is the predominant form in the blood and is composed of two identical light chains and two identical heavy chains, which are connected to form the distinctive Y-shaped molecule described by Rodney R. Porter and Gerald M. Edelman, who won a Nobel Prize for deciphering the structure of an antibody in 1972.
The protein chains found in the stem of the Y are very consistent in their composition, and this part of the antibody contains the fragment crystallizable (Fc) region, which engages and activates the effector cells of the immune system. Conversely, the protein chains in the two “arms” of the Y vary widely. This is known as the variable fragment (Fv) region and contains the two fragment antigen-binding (Fab) domains that determine which antigen the antibody binds.
Each B cell produces a specific antibody that recognizes a small part of a specific antigen. When they encounter their target antigen, they ramp up antibody production by creating an army of clones that are all secreting the same antibody. Since they are derived from a single B cell, these antibodies are referred to as monoclonal antibodies (mAbs).
Many researchers recognized the huge therapeutic potential of mAbs. The German scientist Paul Ehrlich, whose work in immunology was recognized with a Nobel Prize in 1908, famously dubbed them as “magic bullets” due to their ability to seek out and destroy specific toxins.
The third Nobel Prize-winning discovery came in the 1970s when César Milstein and Georges J.F. Köhler described the hybridoma technique, which allowed fusion of immortal myeloma cells to antibody-producing B cells taken from the spleen of a mouse immunized against a specific antigen. As a result of this advance, it became possible to produce large amounts of specific mAbs in the laboratory, thus spawning an explosion in antibody research as scientists were able to make mAbs directed against the antigen of their choice.
Early therapeutic antibodies met with limited success, mostly because they were produced in mouse cells, and therefore induced an immune response in a human host and were rapidly cleared from the body before they could do any good. To overcome these limitations, chimeric and humanized antibodies were developed, with a mixture of mouse and human components that are approximately 65% and 95% human, respectively. Finally, the ability to transfer human genes into the mouse genome to generate transgenic mice enabled the production of fully human antibodies.
Nowadays, therapeutic antibodies are typically produced using mammalian cell lines, which allow much larger-scale production. Chinese hamster ovary (CHO) cells are the most common cell line used, with 10 of the 13 approved anticancer antibodies made in CHO cells. Currently, antibodies are created in large bioreactors, with production levels in excess of 10g/L. However, recent research has shown that yeast and algae could be used in the future to generate therapeutic proteins like antibodies much more quickly and at a fraction of the cost.mAbs make ideal anticancer agents because they have the potential to specifically target tumor cells for destruction while sparing normal cells, thereby overcoming the significant toxicities associated with traditional cancer therapies.
In 1997, the first mAb was approved for a cancer indication; rituximab (Rituxan) targeted the CD20 protein present on the surface of B cells and revolutionized the treatment of B-cell lymphomas, with response rates over 90% when used in combination with chemotherapy. There are now 13 FDA-approved mAbs, with at least four more in late-stage development (Table). They function as anticancer agents by:
FDA-Approved
Date of first approval
Antibody
Target
Tumor Type
Antibody Type
1997
Rituximab (Rituxan)
CD20
NHL, CLL
Chimeric (IgG1)
1998
Trastuzumab (Herceptin)
HER2
HER2-positive breast cancer; HER2-positive metastatic gastric or gastroesophageal junction carcinoma
Humanized (IgG1)
2001
Alemtuzumab (Campath)
CD52
B-cell CLL
Humanized (IgG1)
2002
Ibritumomab tiuxetan (Zevalin)
CD20
NHL
Antibody conjugate: mouse mAb (IgG1) ibritumomab conjugated to a radioactive isotope
2003
I-131 tositumomab (Bexxar)
CD20
NHL
Antibody conjugate: mouse mAb (IgG2) tositumomab conjugated to radioactive isotope
2004
Bevacizumab (Avastin)
VEGF
Metastatic CRC; advanced nonsquamous NSCLC; metastatic RCC; glioblastoma
Humanized (IgG1)
Cetuximab (Erbitux)
EGFR
Metastatic CRC; squamous cell carcinoma of the head and neck
Chimeric (IgG1)
2006
Panitumumab (Vectibix)
EGFR
Metastatic CRC
Human (IgG2)
Ofatumumab (Arzerra)
CD20
CLL
Human (IgG1)
2011
Ipilimumab (Yervoy)
CTLA-4
Metastatic melanoma
Human (IgG1)
Brentuximab vedotin (Adcetris)
CD30
HL; systemic ALCL
Antibody conjugate: chimeric mAb (IgG1) brentuximab conjugated to cytotoxic agent monomethyl auristatin E
2012
Pertuzumab (Perjeta)
HER2
HER2-positive metastatic breast cancer
Humanized (IgG1)
2013
Ado-trastuzumab emtansine (Kadcyla)
HER2
HER2-positive metastatic breast cancer
Antibody conjugate: humanized mAb (IgG1) trastuzumab conjugated to cytotoxic agent emtansine
In the Pipeline
In Phase III
clinical trials as of 2013
Elotuzumab
CS1
Multiple myeloma
Humanized (IgG1)
Farletuzumab
Folate receptor α
Ovarian
Humanized (IgG1)
Inotuzumab ozogamicin
CD22
ALL; NHL
Humanized (IgG4)
Naptumomab estafenatox (ABR-217620/Anyara)
5T4
RCC
Antibody conjugate: mouse Fab fragment conjugated to Staphylococcal enterotoxin E immunotoxin
Necitumumab
EGFR
NSCLC
Human (IgG1)
Nivolumab (BMS-936558)
PD-1
NSCLC; RCC; melanoma
Human (IgG4)
Obinutuzumab (GA101)
CD20
DLBCL; CLL; NHL
Humanized (IgG1)
Onartuzumab (MetMab)
c-Met
NSCLC; gastric cancer
Humanized (IgG1) Fab-Fc fusion
Racotumomab (Vaxira)
GM3
NSCLC
Mouse
Ramucirumab (IMC-1121B)
VEGFR2
Gastric, liver, and breast cancer; CRC; NSCLC
Human (IgG1)
ALCL indicates anaplastic large cell lymphoma; ALL, acute lymphoblastic leukemia; CLL, chronic lymphocytic leukemia; CRC, colorectal cancer; DLBCL, diffuse large B-cell lymphoma; HL, Hodgkin lymphoma; Ig, immunoglobulins; mAb, monoclonal antibody; NHL, non-Hodgkin lymphoma; NSCLC, non-small cell lung cancer; RCC, renal cell carcinoma.
The majority of approved antibodies are based on full-length IgG. While there have been some astounding clinical successes, they are not able to offer a cure as single agents and, indeed, their clinical efficacy usually results from combining them with traditional anticancer agents such as chemotherapy. “Naked” full-length antibodies have failed to reach their potential for a number of reasons, including their large size, limited functionality, counteractive effects of the tumor microenvironment, suboptimal interaction with immune effector cells, and redundancy of the cell-signaling pathways they target.In order to improve the clinical efficacy of anticancer antibodies, a number of variations on the canonical IgG structure have been designed. In fact, noncanonical versions now comprise nearly half of all anticancer antibodies in development. Most advanced are conjugated antibodies, in which the antibody is joined to either radioactive chemicals (radioimmunoconjugates), toxins (immunotoxins), or cytotoxic drugs (antibody-drug conjugates [ADCs]). Rather than directly mediating the destruction of tumor cells, the antibody acts as a targeting device. Four conjugated antibodies are currently approved by the FDA (Table).There are a number of different innovative approaches to engineering the structure of IgG itself, with the aim of improving its pharmaceutical properties, such as stability or antigen affinity, the length of time it is active in the body, or functions (ADCC or CDC). The posttranslational addition of sugar molecules, such as glucose and fucose, to IgG plays a significant role in its function and antigen binding. Antibodies can be produced with alterations in these sugar molecules through a process called glycoengineering. For example, the anti-CD20 antibody obinutuzumab (GA101) has been designed to lack fucose molecules and is currently undergoing phase III testing in patients with various B-cell malignancies (NCT01332968, Challenges Still Abound in Antibody Therapeutics NCT01414855, NCT01059630).Solid tumors are often resistant to antibody therapy due to poor tumor penetration by the antibody. This has been the driving force behind research into the use of smaller antibody fragments, including single-chain variable fragments (scFv), which is a fusion protein of the variable regions of the heavy and light chains, lacking the Fc domain, and Fab fragments. They are less immunogenic than whole Igs, and their smaller size allows faster and deeper penetration into solid tumors. Antibody fragments are often joined together to form multimeric structures, which increases the strength of antigen binding. Two formats have been particularly intensively studied: tandem single-chain variable fragments (scFvs) and diabodies.mAbs are bivalent, meaning they have two “arms,” and at the end of each arm is an antigen-binding domain (Fab). Another significant focus for antibody design has been the creation of bispecific antibodies, which combine fragments of two different mAbs so that they are able to bind two different antigens. Typically, bispecifics are designed to target tumor cells and immune effector cells, but those targeting two different tumor-associated antigens are also being examined. More than 40 different formats of bispecific antibodies have been designed in research laboratories. These include bispecific T-cell engagers (BiTES), a type of tandem scFv that binds the CD3 antigen on immune cells and tumor-associated antigen on cancer cells. Blinatumomab (MT103) is furthest along in development, in phase II trials in B-cell lymphomas (NCT01466179, NCT01741792, NCT00274742).Bispecific antibodies typically lack the Fc domain, but those that retain the Fc domain have also been designed and are called trifunctional abs (or triomAbs). Like bispecifics, these are able to bind tumor and immune cells, but additionally mediate Fc-dependent functions such as ADCC. Catumaxomab (Removab) was approved in the European Union in 2009 for the treatment of malignant ascites and is currently undergoing phase II trials for gastric and ovarian cancer in the United States (NCT01815528, NCT01504256, NCT01246440).Conversely, sometimes only a single antigen-binding site is beneficial. For example, where cell-surface receptors are activated by dimerization, mAbs with two antigen-binding sites can potentially further activate oncogenic signaling pathways by bringing two receptor molecules into proximity. Therefore, monovalent antibodies have also been designed to overcome this issue. Onartuzumab (MetMAb) is a monovalent antibody targeting the c-Met receptor, currently undergoing phase III testing in non-small cell lung cancer and gastric cancer, and phase II testing in glioblastoma (NCT01887886, NCT01662869, NCT01632228).Finally, antibody mixtures incorporating numerous antibody types are also being evaluated. These include oligoclonal antibodies, which are mixtures of different mAbs directed against the same target. For example, MM-151 is a mixture of three fully human EGFR mAbs that is currently undergoing phase I testing in advanced solid tumors (NCT01520389).
Jane de Lartigue, PhD, is a freelance medical writer and editor based in Davis, California.
Related Content: