Therapeutic Antibodies Against Cancer




Antibody-based therapeutics against cancer are highly successful and currently enjoy unprecedented recognition of their potential; 13 monoclonal antibodies (mAbs) have been approved for clinical use in the European Union and in the United States. Bevacizumab, rituximab, and trastuzumab had sales in 2010 of more than $5 billion each. Hundreds of mAbs, including bispecific mAbs and multispecific fusion proteins, mAbs conjugated with small-molecule drugs, and mAbs with optimized pharmacokinetics, are in clinical trials. However, deeper understanding of mechanisms is needed to overcome major problems including resistance to therapy, access to targets, complexity of biological systems, and individual variations.








  • Antibody-based therapeutics against cancer are highly successful in the clinic and currently enjoy unprecedented recognition of their potential.



  • Hundreds of mAbs, including bispecific mAbs and multispecific fusion proteins, mAbs conjugated with small-molecule drugs, and mAbs with optimized pharmacokinetics, are in clinical trials.



  • Challenges remain, and deeper understanding of mechanisms is needed to overcome major problems including resistance to therapy, access to targets, complexity of biological systems, and individual variations.



Key Points
Antibody therapy has its roots thousands of years ago; early forms of vaccination against infectious diseases were developed in China as early as 200 bc . However, the history of true antibody therapy began much more recently with the discovery that serum from animals immunized with toxins, for example, diphtheria toxin or viruses, is an effective therapeutic against the disease caused by the same agent in humans. This discovery resulted in the development of the serum therapy, which saved thousands of lives; von Behring, who in the 1880s developed an antitoxin that did not kill the bacteria but neutralized the toxin released into the body by the bacteria, was awarded the first Nobel Prize in Medicine in 1901 for his role in the discovery and development of a serum therapy for diphtheria. Although historically success with antibody (serum) therapy was initially mostly in the treatment of patients with infectious diseases, currently there is only one monoclonal antibody (mAb) approved for treatment of any infectious disease (palivizumab [Synagis]), which is for prevention of the infection and not for treatment of the already established infection. Initial attempts with serum therapy to treat patients with cancer were not successful. It was not until several decades ago that several revolutionary scientific discoveries were made, which allowed the development of recombinant therapeutic antibodies leading to the approval of the first anticancer therapeutic antibody, the mAb rituximab, in 1997 ( Table 1 ). Since then 13 mAbs have been approved for clinical use against cancer in the European Union and the United States, and 12 are on the market in August 2011; one of these, gemtuzumab ozogamicin (Mylotarg), has been withdrawn (see Table 1 ); by contrast, the first approved mAb-based therapeutic against an infectious disease (Synagis is for prevention) is still awaited. In 2010 sales of the top 4 recombinant therapeutic antibodies (bevacizumab, rituximab, trastuzumab, cetuximab) exceeded US$20 billion ( Table 2 ).

Table 1

Therapeutic monoclonal antibodies against cancer approved or in review in the European Union (EU) or United States (US) (information current as of August 2011)
























































































Name Trade Name Type Indication First Approved First EU (US) Approval Year
Rituximab MabThera, Rituxan Anti-CD20; chimeric IgG1 Non-Hodgkin lymphoma 1998 (1997)
Trastuzumab Herceptin Anti-HER2; humanized IgG1 Breast cancer 2000 (1998)
Gemtuzumab ozogamicin Mylotarg Anti-CD33; humanized IgG4 Acute myeloid leukemia NA (2000 a )
Alemtuzumab MabCampath, Campath-1H Anti-CD52; humanized IgG1 Chronic myeloid leukemia 2001 (2001)
Tositumomab + 131 I-tositumomab Bexxar Anti-CD20; murine IgG2a Non-Hodgkin lymphoma NA (2003)
Cetuximab Erbitux Anti-EGFR; chimeric IgG1 Colorectal cancer 2004 (2004)
Ibritumomab tiuxetan Zevalin Anti-CD20; murine IgG1 Non-Hodgkin lymphoma 2004 (2002)
Bevacizumab Avastin Anti-VEGF; humanized IgG1 Colorectal cancer 2005 (2004)
Panitumumab Vectibix Anti-EGFR; human IgG2 Colorectal cancer 2007 (2006)
Catumaxomab Removab Anti-EpCAM/CD3; rat/mouse bispecific mAb Malignant ascites 2009 (NA)
Ofatumumab Arzerra Anti-CD20; human IgG1 Chronic lymphocytic leukemia 2010 (2009)
Ipilimumab Yervoy Anti-CTLA-4; human IgG1 Metastatic melanoma 2011 (2011)
Brentuximab vedotin Adcetris Anti-CD30; chimeric IgG1; immunoconjugate Hodgkin lymphoma, systemic ALCL NA (2011)

Abbreviations: ALCL, anaplastic large-cell lymphoma; CD, cluster of differentiation; CTLA-4, cytotoxic T-lymphocyte antigen 4; EGFR, epidermal growth factor receptor; EpCAM, epithelial cell adhesion molecule; NA, not approved; VEGF, vascular endothelial growth factor.

Data from Dimitrov DS, Marks JD. Therapeutic antibodies: current state and future trends—is a paradigm change coming soon? Methods Mol Biol 2009;525:1–27.

a Voluntarily withdrawn from market in 2010.



Table 2

The 4 top-selling therapeutic antibodies in 2010 (in billion US$)







































Rank (2009) Name Target Type Company Sales
2 (3) Bevacizumab VEGF Humanized IgG Genentech
Roche
Chugai
6.973
3 (4) Rituximab CD20 Chimeric IgG Genentech Biogen-IDEC Roche 6.859
6 (7) Trastuzumab HER2 Humanized IgG Genentech Chugai Roche 5.859
18 (19) Cetuximab EGFR Chimeric IgG Eli Lilly BMS Merck Serono 1.791

The numbers denote ranking out of all therapeutic proteins in 2010; numbers in parentheses are for 2009. Currencies as of March 2012: 1 € = 1.33 US$; 1 CHF = 1.11 US$; 1 Yen = 0.01 US$; 1 DKK = 0.18 US$; 1 SEK = 0.15 US$.

Data from LaMerie Business Intelligence, Barcelona.


Dating back to mummies and up to the recent successes with ipilimumab, it has become axiomatic that the human immune system has an inherent capacity for antitumor activity. This notion was bolstered in the 1900s by the finding of spontaneous remissions recorded, often in sparse anecdotal findings, in nearly every stage and form of cancer, by the more common observation of spontaneous regressions of melanoma and renal carcinoma, the success of nonspecific immune-stimulants such as bacillus Calmette-Guérin or Coley toxin, and the increasingly targeted use of antibodies against antigens more specific to certain cell types. Indeed, the antibody specificity was perhaps the first and still the most powerful story supporting the ubiquitous catch-call of personalized medicine.


With all of the elegance of the specificity story and more than 35 years since the recipe for generating monoclonal antibodies by Kohler and Milstein, the clinical promise has been largely disappointing. With rare exceptions, these molecular missiles have not annihilated their target tumors and have fallen far short of the marvel of the antibiotic revolution. The rarity of cures should not dampen the substantial, if incremental, progress that has been made. Even in the age of single-nucleotide etiologies there is a strong case that cancer, by the time of its clinical visibility, consists of many broken parts; hence the growing argument that targeted therapies may parallel the breakthrough to cure with chemotherapy in the 1970s with the move toward not one, but a cocktail of simultaneous, combined agents. As in the case of combination chemotherapy, antibody therapy may come to use different effector pathways in this assault.


Therapeutic mAbs and other therapeutic proteins have been reviewed previously (see recent reviews and articles cited therein). Therefore, here the authors review the monoclonal antibodies used directly in treatment, shed some light on presumed primary mechanism of action, and survey use, from initial indication to the wider adoption based principally on clinical trials and trends. This line-up, with its wide spectrum of targets and mechanisms, may give some hope that the long trek may yet reach the originally envisioned summit. If not, these agents are undoubtedly part of the solution. This article focuses mainly on those native, unconjugated antibodies that directly affect solid tumors. Bevacizumab, though its antivascular action is indirect, has gained such wide application for solid tumors (and has been the subject of much controversy) that its inclusion seemed important. Although immune-conjugates have been well reviewed elsewhere and are not the present focus, brentuximab vedontin, as the first new indication for Hodgkin lymphoma in 30 years, warranted special inclusion. Its success represents a partial rescue of a paradigm after the first approved antibody-drug conjugate, gemtuzumab, was withdrawn in 2010 because of lack of efficacy and increased mortality. In the context of the present review it may also point to some limiting aspects of unconjugated tumor-directed antibodies which, as has been stated, have not delivered their quarter-century promise.


mAbs approved for clinical use


At present 13 mAbs are approved for clinical use in the European Union or the United States (see Table 1 ). One of the approved mAbs, gemtuzumab ozogamicin (Mylotarg), was withdrawn from the market because of lack of clinical benefit and safety reasons after a clinical trial in which a greater number of deaths occurred in the group of patients with acute myeloid leukemia (AML) who received Mylotarg compared with those receiving chemotherapy alone. Mylotarg, catumaxumab (Removab) (not yet approved in the United States), and the two radiotherapeutic mAbs, tositumomab (Bexxar) and ibritumomab tiuxetan (Zevalin), are not reviewed here.


Rituximab


The first candidate remains in many ways the poster child for both specificity and efficacy. Rituximab (MabThera, Rituxan), initially developed in San Diego in the late 1980s, and father to that region’s biotech explosion, was based on the finding of CD20 antigen on normal and malignant lymphocytes; it is not appreciably expressed at either pole of lymphocyte ontogeny—stem cells and plasma cells—or on other nonlymphoid cellular compartments. In contrast to many emerging cancer targets clearly connected with signal transduction circuitry, there is no clear consensus on the function of CD20. Nonetheless, the chosen antigen-antibody duo in CD20/rituximab rendered a striking clinical success and ushered in a continuing wave of similarly conceived agents, albeit with variant tactical goals and mechanisms of effect. It is interesting that only after many years afterward were clinical agents developed to target perhaps the ultimate tissue-specific bull’s eye: the individual epitope of each B-lymphocyte population, separating the malignant fiend from more than a million brethren lymphocytes by one signature antigen expressed on one malignant subspecies.


In 1997 rituximab was approved by the US Food and Drug Administration (FDA) for the treatment of relapsed indolent B-cell non-Hodgkin lymphoma (NHL). The antibody is a mouse-human chimera using murine variable regions to effect anti-CD20 specificity and human immunoglobulin (Ig) G1k constant region to facilitate effector function, including complement-mediated lysis and antibody-directed cellular cytotoxicity. Additional mechanisms include caspase activation, a “vaccinal effect” based on increased idiotype-specific T-cell response to follicular lymphoma, and upregulation of proapoptotic proteins such as Bax.


Rituximab’s well-known, early recognized, and sometimes fatal chief toxicity has been acute infusion reactions. Rare fatalities, occurring mainly during first infusion, have been considered secondary to a cytokine reaction; generally associated with flulike symptoms, they may progress to life-threatening hypotension, bronchospasm, and hypoxia, but can usually be controlled by stopping or adjusting of rates of infusion and proper premedication. Black-box events include tumor lysis syndrome, severe mucocutaneous reactions, and progressive multifocal leukoencephalopathy (PML) resulting in death.


Rituximab has demonstrated clinical activity across the spectrum of lymphoproliferative disorders, but the greatest impact has been in NHL, for which combinations and optimizations have sought to raise RRs and, ultimately, cure. Since its 1997 start with relapsed indolent NHL, rituximab has obtained the following additional indications for lymphoma per package insert: relapsed and refractory, follicular or low-grade, CD20-positive, B-cell NHL as single agent; previously untreated CD20-positive, follicular, B-cell NHL in combination with first-line chemotherapy; as single-agent maintenance therapy for patients achieving a partial or complete response to rituximab in combination with chemotherapy; for nonprogressing (including stable), CD20-positive, low-grade, B-cell NHL, as a single agent after first-line combination of cyclophosphamide, vincristine, and prednisone (CVP) chemotherapy; previously untreated CD20-positive, diffuse large B-cell NHL in combination with anthracycline-based chemotherapy, for example, in the workhorse, R-CHOP. It also has an oft-used indication for treatment of previously treated or untreated patients with CD20 chronic lymphocytic leukemia (CLL) in combination with fludarabine and cyclophosphamide (FC).


Rituximab has found off-label use in the clinic in all or nearly all malignant (and many nonmalignant) settings where B cells are presumed to participate in pathogenesis, and has been the subject of many scholarly reviews. Common use spans from aggressive to low-grade lymphoproliferative disorders including: combination with chemotherapy for induction in second-line therapy for relapsed lymphoma anticipating autologous transplant ; combination with chlorambucil for indolent and with bendamustine in treatment of relapsed or refractory CLL ; induction for Burkitt lymphoma; use for gastric and nongastric mucosa-associated lymphoid tissue (MALT tumors), mantle cell tumor, primary cutaneous B cell, splenic marginal zone NHL, Waldenström’s macroglobulinemia/lymphoplasmacytic lymphoma. Its uses have been tailored to mutational status of del(17p) and del(11q) with refractory CLL (National Comprehensive Cancer Network [NCCN] guidelines: http://www.nccn.org/index.asp ) and combined in cocktail fashion with other antibodies such as alemtuzumab for refractory lymphoid malignancies.


The evolution of treatment for CLL mirrors, in many ways, that of NHL as it leads from purines to chemoimmunotherapy, and most recently to novel anti-CD20 antibodies. Conventional treatment of CLL evolved from alkylators to purine analogues when it was demonstrated that fludarabine (F) yielded greater efficacy with better complete response (CR), progression-free survival (PFS), and overall survival (OS) rates than with chlorambucil as primary therapy. Subsequently, the combination of fludarabine with cyclophosphamide (FC) showed better CR and PFS than F. Based on the activity of rituximab (R) alone as a front-line agent, it was added to FC (ie, FCR) and compared with FC alone; in a phase III randomized trial the combination FCR demonstrated better OR, CR, and PFS, establishing both the regimen and the concept of chemoimmunotherapy in this setting as the upfront standard of care.


Ofatumumab


Unfortunately, the activity of rituximab as a single agent is only modest, and duration of response in relapsed disease is generally measured in months. This shortcoming was part of the impetus to develop newer anti-CD20 targeted antibodies with a goal of improving such characteristics as binding affinity, specificity and effector function, and efficacy. Ofatumumab (ofa), a fully human monoclonal IgG1 antibody, binds to a unique epitope, induces considerably higher complement-dependent cytotoxicity (CDC) than rituximab, and shows activity in rituxan-refractory B-cell lymphoma.


Based on these potential biological advantages and modest early-phase clinical activity, ofa was tested against CLL, which was either refractory to fludarabine and alemtuzumab or refractory to fludarabine with disease considered too bulky for efficacy with alemtuzumab. The drug was well tolerated, though complicated by infections in 25% of the patients, but the impressive clinical results including median OS of 13.7 or 15.4 months, within 2 high-risk groups, respectively, contributed to the approval of ofa for disease refractory to fludarabine and for those who have failed alemtuzumab.


Given the potential advantages of ofa versus rituximab and with FCR established as standard of care in the front line, substituting ofa for rituxan in the so-called O-FC regimen was tested in a multinational, randomized phase II trial in treatment-naïve patients. Of the 2 tested doses, the higher dose arm yielded a CR rate of 50%. It remains unclear as to how this should be positioned with respect to such other findings as the initial randomized phase III trial that established FCR as standard of care. The precedent of combining permutations of purine analogues, alkylators, and antibodies including newer regimens such as ofa/bendamustine continues to inform ongoing studies.


Ipilimumab


The novel treatment agents for melanoma, vemurafenib (a B-raf inhibitor) and ipilimumab (an antibody against cytotoxic T-lymphocyte antigen 4 [CTLA-4]), represent perhaps the most significant advance in oncology in several years. How they will fit into tactical treatment strategies and with respect to conventional dacarbazine, interleukin (IL)-2, and a new gp100-based vaccine is a welcome and exciting challenge after decades without appreciable progress. Blockade of the CTLA-4 has been the subject of long and intensive investigation.


Among the most active immune inhibitory pathways is the CD28/CTLA-4:B7-1/B7-2 receptor/ligand grouping, which modulates peripheral tolerance to tumors and outgrowth of immune-evasive clones. Inhibition is both toward the overexpressed self targets and via upregulation of inhibitory ligands on lymphocytes. Thus, blockade of CTLA-4 has potential for both monotherapy and in synergy with other therapies that enhance presentation of tumor epitopes to the immune system. Genetic ablation of CTLA-4 leads to a massive and lethal lymphoproliferative disorder. Antibody blockade of CTLA-4 induces potent antitumor activity through enhancing effector cells and concomitantly inhibiting T-cell regulatory activity.


Given that this inhibition is not tumor specific, it is not surprising that other tumors including ovarian cancer, prostate cancer, and renal cell cancer have demonstrated durable remissions.


In a recent phase III trial, patients with melanoma refractory to chemotherapy or IL-2 who received ipilimumab had improved OS compared with those receiving the gp100 peptide vaccine, and on this basis received FDA approval in 2011.


Ipilimumab holds an FDA indication for the treatment of unresectable or metastatic melanoma, with NCCN guidelines that largely elucidate specific contexts consistent with this approval, including use as single agent for unresectable stage III in-transit metastases, local/satellite and/or in-transit unresectable recurrence, incompletely resected nodal recurrence, limited recurrence or metastatic disease, and disseminated recurrence of metastatic disease in patients with good performance status.


Based on its mechanism of unleashing the immune recognition and effector system, there was rationale to test the interactive effects with tumor-specific antigen. Specifically, the melanoma antigen, gp100, overexpressed on this tumor and among the antigens presented in the appropriate genetic major histocompatibility complex (MHC) context (HLA∗A201), represented a prime vaccine candidate. In a phase III randomized trial, increased RRs were seen when vaccine was added to IL-2 compared with IL-2 alone (16% vs 6%, P = .03); PFS was also significantly improved with a trend toward improved OS. Questions arose, nonetheless, as to whether gp100 vaccine was an appropriate control in the aforementioned phase III trial for ipilimumab. Another phase III randomized clinical trial treating previously untreated patients with metastatic melanoma compared ipilimumab (every 3 weeks for 4 doses followed by maintenance every 3 months) with and without dacarbazine as the standard control; improved OS was seen, including a difference at 3 years of nearly 21% versus 12%.


The cluster of well-identified side effects induced by CTLA-4 inhibition has been referred to as immune-related adverse events (IRAEs). These unique adverse effects are likely a direct effect of impairing immune tolerance, and include colitis/diarrhea, dermatitis, hepatitis, uveitis, nephritis, inflammatory myopathy, and endocrinopathies. Although these reactions have gained a black-box designation for occasional severe and even fatal instances, they are generally manageable and reversible with treatment guidelines that include systemic corticosteroids. These toxicities may be prolonged, suggestive of sustained release from immune tolerance, and perhaps a different response profile including long periods of stable disease, and correlation of toxicity with efficacy. In one report with high-risk melanoma, ipilimumab-treated patients who experienced high-grade IRAEs had a significantly higher rate of tumor regression than those without IRAEs (36% vs 5% of patients).


Based on a mechanism of action clearly different from IL-2, which increases responsiveness to immune targets and is nonoverlapping with chemotherapy, earlier phase trials and future efforts will focus on combinations of vaccines, chemotherapy, and other immune modulators. Furthermore, given the prolonged time course of side effects and the resulting requirement for prolonged steroids, timing of its use with respect to IL-2 and vaccines will be the subject of much attention.


Trastuzumab


The human epidermal growth factor receptor 2 (HER2) is overexpressed in 20% to 30% of invasive breast cancers and is associated with a worse prognosis. Trastuzumab is a humanized mAb targeting HER2, which was approved by the FDA in 1998 as the first monoclonal for a solid tumor indicated for patients with invasive breast cancer that overexpresses HER2. It is now a standard part of the treatment of HER2-positive tumors in both metastatic and adjuvant settings. Because tumors that overexpress HER2 receptor respond better across the range of studies, considerable effort has been expended to accurately assess receptor status.


HER2 is part of a family of transmembrane tyrosine kinase receptors that normally regulate cell growth and survival, differentiation, and migration. It consists of an extracellular binding domain, a transmembrane segment, and an intracellular tyrosine kinase domain. The receptor is activated generally by homodimerization or heterodimerization, but not always activated through ligand binding; it can dimerize and thus activate, independent of ligand through either overexpression or mutation. Thus activated by overexpression, signal-transduction cascades act to promote a host of progrowth activities including proliferation, survival, and invasion. Such signal transduction is mediated through the RAS-MAKP pathway, inhibiting cell death through the m-TOR pathway. In addition, it inhibits the PI3K pathway, reducing PTEN phosphorylation and AKT dephosphorylation and thus increasing cell death.


The human IgG1 is capable of inducing antibody-dependent cell-mediated cytotoxicity (ADCC) in vitro and of recruitment of effector cells in animal studies. An immune mechanism is suggested by the increased lymphoid infiltration into tumor after preoperative administration of trastuzumab. There is also evidence that it causes regression of vasculature by modulating angiogenic factors.


As a single agent in metastatic breast cancer, and receptor status using earlier immunohistochemistry (IHC) expression criteria, trastuzumab produced RRs of 11% to 26%. From the earliest studies, though time has sharpened the assessment, it has been clear that the best results occur in tumors that overexpress HER2. The breakthrough trial for trastuzumab in metastatic disease came in a randomized phase III trial when it was used in combination with chemotherapy for HER2-positive patients. As first-line therapy for metastatic disease, patients were given either chemotherapy alone or in combination. Patients were given an anthracycline and cyclophosphamide or, paclitaxel (if they had previous anthracycline in an adjuvant setting). Results showed not only improvement in response rate (RR) and progression-free interval but also in OS. Trastuzumab was subsequently showed to have efficacy and safety with a variety of other chemotherapeutics including docetaxel, vinorelbine, and doxil in nonrandomized trials.


As for the adjuvant setting, large randomized trials established significant benefits from the addition of trastuzumab to both anthracycline and nonanthracycline regimens for early breast cancer. Four major adjuvant trials including more than 13,000 women with HER2-positive early breast cancer used different adjuvant regimens with trastuzumab; in these studies overall, trastuzumab reduced the 3-year risk of recurrence by about half in this population. On this basis, trastuzumab has become part of standard adjuvant therapy. Both the international Consensus Group and NCCN recommend its use for women with HER2-positive, node-positive tumors as well as for node-negative disease when the primary tumor is larger than 1 cm.


Trastuzumab combined with chemotherapy has also shown improvement in pathologic responses and event-free survival when used in the neoadjuvant setting before surgery. In a randomized phase III trial, patients with advanced gastroesophageal and gastric adenocarcinoma tumors that overexpressed HER2 showed a significant increase in OS when trastuzumab was added to their chemotherapy. Trastuzumab now has an FDA indication for use in combination with cisplatin and fluorouracil (5FU) or capecitabine for first-line treatment of gastric and gastroesophageal tumors that overexpress HER 2.


The most significant toxicity associated with trastuzumab is cardiomyopathy, ranging from subclinical decreases in left ventricular ejection fraction to cardiac failure manifesting as congestive heart failure. The risk is greatest when administered concurrently with anthracyclines. Use following anthracyclines was associated commonly with asymptomatic cardiac dysfunction, but most severe decreases recovered with time. Close monitoring of clinical status and cardiac function, sequential rather than concomitant use, and development of nonanthracycline regimens have all been objectives.


Bevacizumab


The discussion of bevacizumab here is asymmetric in bulk and breadth compared with the other antibodies, owing to its conceptual and actual application in many tumor types, its unique mechanism, and toxicity profile. Bevacizumab is a humanized IgG1 mAb that binds to and neutralizes the ligand vascular endothelial growth factor (VEGF) rather than binding the cell-surface receptor. In fact many tissues and most malignancies produce VEGF whose native function, whether acting from a distance or in an autocrine loop, operates through binding and activation of the VEGF receptor. The latter includes an extracellular binding domain and a cytoplasmic kinase domain. Following VEGF binding, the otherwise inactive monomer receptor undergoes dimerization, autophosphorylation of the tyrosine kinase domain, and downstream activation of many of the usual signal transduction suspects including MAPK and protein kinase C pathways, which mediate proliferative events—in this setting, endothelial proliferation and angiogenesis ; such neoangiogenesis is required by tumors once they grow to greater than 2 mm.


Many of this antibody’s common toxicities are related to its impact on microvasculature, including hypertension, proteinuria, rare bowel perforation, impaired wound healing, and bleeding. Other than the rare bowel perforation, these can generally be managed without necessitating cessation of therapy. Although there will naturally be some specificity of side effects and adverse reactions dependent on the drugs with which bevacizumab is paired, with some notable exceptions toxicities are generally neither drug-combination specific nor tumor specific.


More severe and fatal consequences of bevacizumab have been the subject of several meta-analyses and reports of large-institution experience. In perhaps the largest of these, fatal adverse events (FAEs) were considered in a meta-analysis of more than 10,000 patients with various solid tumor types, comparing regimens with and without the addition of bevacizumab. The overall incidences of FAEs were 2.5%, and among these nearly a quarter were attributable to hemorrhage, about half related to neutropenia, and a smaller amount to perforation. There was increased RR attributable to combining bevacizumab with taxanes or platinum but not with other agents, nor were there significant tumor-specific increases. In another large meta-analysis bevacizumab was associated with high-grade congestive heart failure in breast cancer, with an overall incidence of 1.6%. Yet a third large meta-analysis identified a 12% risk of thromboembolic events. Of note, a pooled analysis of phase II and phase III trials did not show an increase in venous thromboembolic events (VTEs), which is important to recognize given a baseline of tumor-associated VTEs of around 10% with or without this agent. Massive hemoptysis, which has been linked to large central lesions at risk for cavitation, was avoided in these circumstances, and more generally in squamous cancer where this risk is increased. Bowel perforation occurred with an incidence of less than 2% in a large institution with a treated population of more than 1400 patients; it was generally managed without the need for surgical intervention.


Bevacizumab demonstrated no to small RRs as monotherapy and, with such exceptions as maintenance regimens and single-agent use with recurrent glioblastoma, its predominant clinical role lies in combination with chemotherapy. In 2004, based on improvement of RRs, PFS, and OS, bevacizumab, when combined with chemotherapy in metastatic colorectal cancer, became the first antiangiogenic agent approved for clinical use. Since then it has gained indications for metastatic breast cancer, metastatic renal cancer, metastatic (as well as advanced or recurrent) non–small cell lung cancer (NSCLC), and glioblastoma. Increasing use of bevacizumab is also being seen with hepatocellular and ovarian cancer.


Colorectal cancer


At this time bevacizumab has an indication in metastatic colorectal cancer in both first-line and second-line settings. The initial approval followed its use with bolus irinotecan, 5FU, and leucovorin (IFL) whereby addition of bevacizumab significantly improved RR and median survival (20 vs 16 months) compared with chemotherapy only. While bolus IFL has fallen out of general use because of its toxicity profile, studies have supported the value of bevacizumab in combination with more widely used treatments including FOLFIRI (FOL, leucovorin plus F, 5FU and IRI, irinotecan [Camptosar]), and 3 oxaliplatin-containing regimens. In addition, when bevacizumab was added to 5FU/leucovorin in the absence of irinotecan or oxaliplatin, RRs were approximately doubled and median survival improved in comparison with chemotherapy alone.


Efforts to apply bevacizumab in the adjuvant setting for colorectal cancer moved from initial enthusiasm to disappointment. As already noted, bevacizumab had shown favorable impact in metastatic disease in several settings including in combination with IFL (irinotecan, 5FU, and leucovorin) for metastatic colorectal cancer. Borrowing the prevailing paradigm for chemotherapy, which attempts to apply results in metastatic disease to adjuvant use on the presumption of potential elimination of micrometastases, bevacizumab was studied in the adjuvant setting for colorectal cancer. Two recently published phase III trials, unfortunately, did not show the sought-for benefit. When bevacizumab was added (for 12 months) to FOLFOX (folinic acid [FOL], fluorouracil [F], and oxaliplatin [OX]) (for 6 months), it failed to meet its primary end point of improving 3-year disease-free survival. In a second phase III trial the combination of bevacizumab with FOLFOX actually led to a slight but significant decrease in OS.


Non–small cell lung cancer


The role of bevacizumab in NSCLC was initially established in a phase III trial as first-line therapy for advanced, nonsquamous NSCLC including those with malignant effusions and metastatic disease. Patients received paclitaxel and carboplatin with or without bevacizumab; those patients receiving bevacizumab then continued it as monotherapy for an additional 6 cycles unless disease progressed. The objective RR more than doubled, and there was an increase in PFS and OS. At 2 years, the survival rate was 23% in the group treated with bevacizumab versus 15% without. In another phase III trial with a similarly defined patient population, on the addition of bevacizumab to gemcitabine and cisplatin (also with maintenance bevacizumab in the concurrent bevacizumab group) an increase in PFS did not translate into improved OS ; the investigators suggested this may have been due to the wide availability of secondary therapies. Testing with a current standard, pemetrexate, is under way but has not yet ripened to a point to give clinical guidance.


The story for use of bevacizumab in advanced metastatic breast cancer (MBC) has been tumultuous, tracking a course from early excitement and widespread use to an FDA withdrawal; understandably this raised public furor from a highly engaged population. In the first phase III trial to assess impact in newly diagnosed patients with MBC, bevacizumab was either added to chemotherapy (weekly paclitaxel) or not; the bevacizumab arm doubled the PFS and significantly improved RR. These striking results led to accelerated FDA approval and its wide adoption. Unfortunately, neither of 2 phase III postapproval studies, one trial with docetaxel and the other with capecitabine, a taxane, or an anthracycline, confirmed this magnitude of benefit, and no trial has shown an improvement in OS.


Renal cancer


For metastatic renal cancer, 2 phase III trials demonstrated improved OS when bevacizumab was added to interferon-α in first-line treatment. In one of these trials the initially reported PFS with bevacizumab of 10.2 months was nearly doubled, but only a nonsignificant and clinically small difference of OS was reported in the final analysis. In the second phase III trial, with a similar dose schedule as the first, bevacizumab plus interferon-α improved RR and PFS compared with monotherapy with interferon, but did not reach significance for OS.


Glioblastoma


For recurrent glioblastoma, adding bevacizumab to irinotecan increased RR. Nevertheless, in the notoriously difficult setting of recurrent glioblastoma both alone and in combination with irinotecan, bevacizumab showed respectable RRs of 28% and 38%, respectively ; it holds an indication for use as monotherapy in this setting despite the absence of a demonstrated improvement in OS.


Ovarian cancer


The benefit of bevacizumab in ovarian cancer was assessed in the setting of first-line use with paclitaxel and carboplatin in a large trial for stage III or IV epithelial ovarian, primary peritoneal, or fallopian tube cancer following maximal cytoreduction. Of the 3 arms in this phase III trial, representing chemotherapy only, concurrent bevacizumab and chemotherapy, and concurrent bevacizumab and chemotherapy followed by maintenance bevacizumab, the latter improved PFS but not OS. First-line use for advanced and high-risk early-stage disease treated with paclitaxel and carboplatin, with and without bevacizumab, showed significant improvement in median survival without improving OS; a subset analysis suggested that adding bevacizumab may be more beneficial among women with a poorer prognosis.


Small studies in hepatocellular with bevacizumab alone or with gemcitabine and oxaliplatin showed RRs sufficient to generate further interest and more definitive study.


Bevacizumab has a unique profile of toxicities and adverse reactions. Some preclinical studies had suggested that VEGF-targeted therapies could unfavorably alter the biology of the neoplasms, for example, by upregulating proinflammatory pathways and factors that are associated with metastasis, but a pooled meta-analysis of 5 randomized phase III trials did not show altered disease progression following bevacizumab. Although the clinical data are too scant to explain the unpredicted disappointments such as failure in the adjuvant setting for colorectal cancer, numerous hypotheses such as the foregoing, some readily testable, have been suggested. As in most contexts in oncology, risk/benefit analysis is important to decision making, and the risk in some clinical settings where bevacizumab is considered often pits treatment against the prospect and probability of imminent death. It is notable, therefore, that while a recent meta-analysis of 16 randomized trials in advanced cancer showed nearly a 1.5-fold increase in fatal adverse events, the absolute values were 2.5% versus 1.7% in the respective presence or absence of bevacizumab. Nevertheless, these same numbers gather increased clinical sway in adjuvant settings where the risks and benefits are markedly different.


Cetuximab


Cetuximab is a recombinant chimeric antibody that derives specificity from its murine Fv portion and effector functions from human IgG1 constant regions. The primary mechanism of impact is through disruption of the signal transduction pathway of the endothelial growth factor receptor (EGFR). Nevertheless, selection based on IHC expression of EGFR expression or somatic mutations of the EGFR tyrosine kinase domain, as in the response of NSCLC to small-molecule tyrosine kinase inhibitors, do not predict response of colorectal cancer to EGFR antibodies. Wild-type K-ras, on the other hand, is necessary for effect.


Cetuximab has been studied alone and in combination, predominantly with colorectal cancer and head and neck cancer. In colorectal cancer, cetuximab as monotherapy showed improvement in OS compared with best supportive care (BSC) in patients previously treated with a fluoropyrimidine, irinotecan, and oxaliplatin. This study also demonstrated improved quality of life and the association of rash with a favorable outcome. Cetuximab as monotherapy or combined with irinotecan both showed clinically significant activity in patients with metastatic disease who were refractory to irinotecan, but the combination showed superior RR, time to progression, and median survival. In another study, the combination of cetuximab and irinotecan also showed improvement in RR and PFS in patients previously treated with oxaliplatin and fluoropyrimidines for metastatic disease. In combination with FOLFIRI as first-line therapy for metastatic disease it showed increased OS in patients with wild-type K-ras. The data showing advantage in the first line when combined with oxaliplatin are not as clear. In one study the addition of cetuximab to FOLFOX showed significant improvement of RR only in the wild type K-ras subpopulation but in another, more recent trial, no advantages were shown when added to oxaliplatin, even in the wild-type K-ras group.


In squamous cell head and neck cancer, cetuximab showed improvement in OS when added to radiation compared with radiation alone for locally and regionally advanced disease. The advantage did not extend to those with marked functional compromise or who were older than 65 years. Here, too, response was improved in those with acneiform rash.


As first-line treatment in patients with recurrent or metastatic squamous cell carcinoma of the head and neck, cetuximab plus platinum-5FU chemotherapy improved OS compared with platinum-based chemotherapy plus fluorouracil alone.


Despite 2 recent phase III trials in NSCLC, the role of cetuximab in lung cancer remains unclear. These randomized trials compared doublets of standard chemotherapy with and without cetuximab in the first-line setting for metastatic disease, and may suggest different clinical guidance. In the FLEX trial, a randomized phase III multinational study, patients with IIIB (malignant pleural effusion) and IV, who expressed EGFR, received cisplatin and vinorelbine with or without cetuximab. Patients who received cetuximab had significant but clinically modest increased OS at 11.3 months versus 10.3 months with chemotherapy alone. First-cycle rash in this study was substantially associated with OS, with the median with rash at 15 months compared with 8.8 months without the rash. In another phase III randomized trial studying same-stage patients in first-line treatment, without restrictions on EGFR expression, cetuximab combined with taxol/carboplatin did not improve PFS compared with chemotherapy alone; a small increase in OS for cetuximab of less than 2 months did not reach statistical significance.


Panitumumab


Panitumumab, an IgG2 class antibody to the EGFR receptor, was the first fully human antibody to be approved by the FDA in 2006 for the treatment of patients with EGFR-expressing, metastatic colorectal carcinoma with disease progression on or following fluoropyrimidine-, oxaliplatin-, and irinotecan-containing chemotherapy regimens. It received regulatory approval for use as monotherapy in refractory disease based on prolonging disease-free survival. Given the similarities to cetuximab, efforts have focused on where to place each and in which clinical contexts and sequences, although they have notably not been compared in a face-to-face randomized phase III trial. The close relation to cetuximab, both biological and clinical, provides a helpful context for review. Like cetuximab, panitumumab binds to the receptor, the dyad is internalized, and the downstream signal transduction is blunted. Its activity cannot be reliably shown to depend on the overexpression of EGFR. However, downstream signal transduction by constitutively activated K-ras abrogates its effect, and its use according to American Society of Clinical Oncology guidelines, consistent with clinical trials, is limited to tumors with wild-type K-ras. There is more recent evidence that mutations in B-raf may also predict no response to either cetuximab or panitumumab. Biologically, its difference from cetuximab in being fully human may underlie its significant reduction in infusion reactions. Despite its design to be more activating of cell-mediated cytotoxicity and ADCC, neither of these activities nor its efficacy compared with cetuximab has been demonstrated. Its toxicities, which include a predictable rash in almost all cases, as well as frequent diarrhea and malaise, parallel those of cetuximab, as does the positive association of rash with clinical impact.


Panitumumab has shown activity when used both in monotherapy and in combination. A large phase III trial showed improved RR and decreased tumor progression when used as monotherapy compared with BSC in patients refractory to oxaliplatin and irinotecan-based therapies. In combination chemotherapy with FOLFOX, panitumumab in the first line improved both RR and PFS in contrast to cetuximab, which had mixed results as previously noted. When combined with FOLFIRI versus FOLFIRI alone after failure with 5FU-based chemotherapy (the majority with oxaliplatin), addition of panitumumab significantly improved PFS.


Brentuximab Vedotin


The first-in-class antibody drug conjugate (ADC), brentuximab vedontin, received accelerated FDA approval in August 2011 for treatment of relapsed or refractory Hodgkin lymphoma and systemic anaplastic large-cell lymphoma (ALCL). Approval was based on impressive RR rather than demonstrable survival improvement in rather dire clinical circumstances: in Hodgkin lymphoma after failure of at least 2 prior systemic regimens in autologous stem cell candidates, and for ALCL after failure of at least 1 multiagent regimen.


The antibody is a chimerized IgG, which targets CD30 and thus delivers its antimitotic payload, monomethyl auristatin (vedontin). CD30 is only minimally expressed in normal tissue but densely expressed in both Hodgkin lymphoma and ALCL.


As optimistically as progress with treatment of Hodgkin disease (HD) is viewed, approximately 15% to 30% of patients do not achieve long-term remissions on conventional therapy, and despite autologous stem cell transplantation (ASCT) many of these subsequently perish while still in young adulthood.


In the pivotal phase I and subsequent phase II trials RR data were rightly greeted with excitement. In the phase II trial, for patients with HD, all with prior transplants, 75% achieved an objective response including 32% complete responders who have not yet reached median duration of response. For 58 patients with relapsed or refractory systemic ALCL, a CR was reached in 56% of patients the median duration of which, likewise, has not yet been reached. Moreover, retreatment has been successfully used to maintain complete remissions and, though the number of patients was small, a retrospective look across 3 studies demonstrated provocatively high responses with retreatment.


It has been suggested that these impressive results may include several mechanisms, from apoptosis by ligating CD30, to cytotoxic, to the bystander effect on surrounding tissue. The remarkable results with HD were particularly impressive considering the minimal responses achieved using the unconjugated anti-CD30 antibody. Beyond the direct impact on the malignant cell, it has been suggested that the local effect on the tumor-supporting cellular milieu was also a factor. By bulk, malignant Reed-Sternberg cells represent a minority of the masses in HD, which otherwise consist largely of inflammatory cells recruited by chemokines that in turn support the tumor cells that recruited them. The shrinkage of these masses might thus be understood, in part, to be due to the bystander effect caused by local diffusion of the cytotoxic agent into this local environment. Although other differences no doubt exist, this may be one factor to explain greater responses than achieved for similar antibody-toxin conjugates, for example, with trastuzumab in HER2-positive cancer where tumor masses are predominantly composed of malignant cells.


Alemtuzumab


Alemtuzumab is a humanized anti-CD52 IgG1 monoclonal antibody. Early studies demonstrated its efficacy in refractory disease, leading initially to approval by the FDA in 2001 for treatment of fludaribine-refractory CLL; subsequent trials demonstrated its use as front-line monotherapy for B-cell CLL. Antibodies to CD52 induce complement-mediated lysis and antibody-directed cellular toxicity through this target that is not only expressed on CLL and lymphomas but also on both normal B cells and normal T cells, neutrophils, and monocytes. This large spectrum of targets accounts not only for positive aspects such as off-label uses with T-cell lymphoproliferative disorders such as peripheral T-cell lymphoma, T-cell prolymphocytic leukemia, cutaneous T-cell lymphoma, mycosis fungoides, and Sezary syndrome, but also for negative consequences such as heightened infusion reactions and significant vulnerability to opportunistic infections.


Based on increased acute toxicity and prolonged myelosuppression, alemtuzumab has not supplanted the more B-cell–specific rituximab either as monotherapy or in combination with cytotoxic chemotherapy. First-line treatment for CLL generally uses fludaribine as the cornerstone, often in combination with cyclophosphamide and rituximab. Second-line therapy with alemtuzumab added to fludarabine and cyclophosphamide demonstrated substantial efficacy in a recently reported phase II trial. Response rates have ranged in the area of 30% to 50% in the relapsed setting and 80% to 90% in previously untreated patients with CLL. In a large multicenter study, patients with refractory or relapsed CLL, previously exposed to alkylating agents and having failed fludarabine, had an overall RR of about one third, nearly all partial responses; median survival for responders was 32 months. Alemtuzumab has received particular attention in high-risk settings, including 17p deletions and p53 defects known to be resistant to standard agents including chlorambucil, purine analogues and rituximab. One study demonstrated nearly 50% overall RRs and favorable OS in the 17p-deletion cytogenetic group. Alemtuzumab has been shown to achieve CR in the setting of p53 mutation and resistance to chemotherapy, and in one study of fludarabine-refractory disease, even within a small subset with the presence of p53 mutations or deletions (predictors of poor response to conventional therapy), responses occurred in 40% with a median response duration of 8 months. In a phase II trial with subcutaneous alemtuzumab, efficacy with fludarabine-refractory CLL did not vary with 17p deletion, mutated p53, 11q deletion, or mutated p53.


Combination of alemtuzumab with rituximab has not gained traction, based on results of FCR that are hard to compete with and significant infectious complications. In a study of 32 patients with relapsed or refractory disease, for example, while slightly more than 50% showed response a similar percentage also developed infections, including 27% cytomegalovirus antigenemia. In a recent phase II trial, alemtuzumab was added to conventional FCR yielding 70% CR and 18% partial response in high-risk patients, results considered comparable with historic FCR-treated high-risk patients. Based on nearly 60% CRs in the subset with 17p deletion it was suggested, however, that this may nonetheless have a useful front-line role before allogeneic stem cell transplantation.


The general use of alemtuzumab for consolidation in the community setting cannot yet be recommended, although the question remains to be settled and is the subject of significant investigation. A phase III trial in which alemtuzumab was used as consolidation to fludarabine ± cyclophosphamide was stopped prematurely because of severe infections; nevertheless, minimal residual disease was durably reduced by consolidation and PFS was significantly improved after median follow-up of 48 months. Although there was a trend toward shorter response duration in comparison with historic groups receiving intravenous alemtuzumab, patients receiving subcutaneous treatment showed reliable decreases in graded measures of residual disease. Although alemtuzumab consolidation improved both CR and minimal residual disease (MRD)-negative rates, in a study of 102 patients initially treated with induction fludarabine and rituxan there were 5 deaths from infection, and 2-year PFS and OS were not improved. Efforts have been under way over the past decade to unravel genomic complexity in CLL. Such understanding will inform trial design and, undoubtedly, the value of consolidation will depend on identification of molecular diagnostic settings where improvements of MRD-negative status translate into improved OS.




mAbs in clinical and preclinical development


Hundreds of mAbs are in thousands of clinical trials ; 2239 entries for planned, ongoing, or completed clinical trials were retrieved from http://www.clinicaltrials.gov by searching with cancer AND therapy AND monoclonal antibodies as of August 2011, of which 270 are in phase III. A significant number of all new medicines are mAbs against cancer (see also http://www.phrma.org/research/new-medicines ). At least 1 to 3 different antibodies are being developed at different companies for each relevant therapeutic target. However, some molecules are targeted by many more mAbs; for example, the insulin-like growth factor receptor type I (IGF-IR) is targeted by more than 10 different mAbs. During the last decade and especially in the last several years, the number of clinical trials with therapeutic antibodies has increased dramatically. However, this increase has been largely due to an increase in the number of indications for the same antibodies, especially in combination with other therapeutics. The number of targets and corresponding antibodies in preclinical development and in the discovery phase has also increased significantly during the past decade.


Second- and third-generation mAbs are being developed against already validated targets. The improvement of already existing antibodies also includes an increase (to a certain extent) of their binding to Fc receptors for enhancement of ADCC and half-life, selection of appropriate frameworks to increase stability and yield, decrease of immunogenicity by using in silico and in vitro methods, and conjugation to small molecules and various fusion proteins to enhance cytotoxicity. A major lesson from the current state of antibody-based therapeutics is that gradual improvement in the properties of existing antibodies and identification of novel antibodies and novel targets is likely to continue in the foreseeable future.


One area where one could expect conceptually novel antibody-based candidate therapeutics, even though within the current paradigm, is going beyond traditional antibody structures. At present, all FDA-approved anticancer therapeutic antibodies (see Table 1 ) and the vast majority of those in clinical trials are full-size antibodies, mostly in IgG1 format of about 150 kDa. A fundamental problem for such large molecules is their poor penetration into tissues (eg, solid tumors) and poor or absent binding to regions on the surface of some molecules (eg, on the human immunodeficiency virus envelope glycoprotein) that are accessible by molecules of smaller size. Therefore much work, especially during the last decade, has been aimed at developing novel scaffolds of much smaller size and higher stability (see, eg, recent reviews in Refs. ). Such scaffolds are based on various human and nonhuman molecules of high stability and could be divided into 2 major groups for the purposes of this review: antibody-derived and others. Here the advantages of antibody-derived scaffolds, specifically those derived from antibody domains, and binders selected from libraries based on engineered antibody domains (eAds) are briefly discussed; an excellent recent review describes the second group.


First, their size (12–15 kDa) is about an order of magnitude smaller than the size of an IgG1 (about 150 kDa). The small size leads to relatively good penetration into tissues and the ability to bind into cavities or active sites of protein targets that may not be accessible to full-size antibodies. Second, eAds may be more stable than full-size antibodies in the circulation and can be relatively easily engineered to further increase their stability. For example, some eAds with increased stability could be taken orally or delivered via the pulmonary route, or may even penetrate the blood-brain barrier, and retain activity even after being subjected to harsh conditions such as freeze-drying or heat denaturation. In addition, eAds are typically monomeric, of high solubility, and do not significantly aggregate or can be engineered to reduce aggregation. Their half-life in the circulation can be relatively easily adjusted from minutes or hours to weeks by making fusion proteins of varying size and changing binding to the neonatal Fc receptor (FcRn). In contrast to conventional antibodies, eAds are well expressed in bacterial, yeast, and mammalian cell systems. Finally, the small size of eAds allows for higher molar quantities per gram of product, which should provide a significant increase in potency per dose and reduction in overall manufacturing cost. However, despite all these advantages there is still no candidate therapeutic based on such scaffolds in phase III clinical trial as of August 2011.


Research on novel antibody-derived scaffolds continues. The authors identified a scaffold based on the variable region of heavy chain that is stable and highly soluble. It was used for construction of a large-size (20 billion clones) eAd phage library by grafting complementarity-determining regions (CDR3s and CDR2s) from 5 of the authors’ other Fab libraries and randomly mutagenizing CDR1. It was also proposed to use engineered antibody constant domains (C H 2 of IgG, IgA and IgD, and C H 3 of IgE and IgM) as scaffolds for construction of libraries. Because of their small size and the domain’s role in antibody effector functions, these have been termed nanoantibodies, the smallest fragments that could be engineered to exhibit simultaneous antigen binding and effector functions. Several large libraries (up to 50 billion clones) were constructed and antigen-specific binders successfully identified. The authors have recently engineered C H 2-based scaffolds with high stability by introducing an additional disulfide bond and by shortening C H 2. It is possible that these and other novel scaffolds under development could provide new opportunities for identification of potentially useful therapeutics.

Only gold members can continue reading. Log In or Register to continue

Stay updated, free articles. Join our Telegram channel

Mar 1, 2017 | Posted by in HEMATOLOGY | Comments Off on Therapeutic Antibodies Against Cancer

Full access? Get Clinical Tree

Get Clinical Tree app for offline access