|Home | About | Journals | Submit | Contact Us | Français|
Antibodies have emerged as important therapeutics for cancer. Recently, it has become clear that antibodies possess multiple clinically relevant mechanisms of action. Many clinically useful antibodies can manipulate tumour-related signalling. In addition, antibodies exhibit various immunomodulatory properties and, by directly activating or inhibiting molecules of the immune system, antibodies can promote the induction of anti-tumour immune responses. These immunomodulatory properties can form the basis for new cancer treatment strategies.
The concept of using antibodies to selectively target tumours was proposed by Paul Ehrlich over a century ago1. The advent of hybridoma technology in 1976 enabled the production of monoclonal antibodies2. Due to their murine origins, these monoclonal antibodies typically were immunogenic in humans and possessed poor abilities to induce human immune effector responses, thereby limiting their clinical applicability. Later advances in antibody engineering provided flexible platforms for the development of chimeric, humanized and fully human monoclonal antibodies which satisfactorily addressed many of these problems (FIGURE 1).
Over the past decade, the effectiveness of antibodies in treating patients with cancer has been realized with increasing frequency (TABLE 1). Many of these antibodies are specific for antigens expressed by the tumour itself. Antibodies conjugated to radioisotopes or chemotherapeutic drugs have shown therapeutic efficacy primarily in hematological malignancies, whereas unconjugated antibodies targeting growth factor receptors, such as epidermal growth factor receptor (EGFR) and human epidermal growth factor 2 (HER2, also known as ERBB2/NEU) are commonly used for the treatment of non-leukaemic cancers. In addition to antibodies that target tumour antigens, antibodies that target the tumour microenvironment slow tumour growth by enhancing host immune responses to self-tumour antigens or curtailing pro-tumourigenic factors produced in the tumour stroma.
Here, we highlight important features of anti-tumour antibodies, with a focus on how such antibodies promote immune effector mechanisms to control tumour growth.
Antibodies are grouped into five classes based on the sequence of their heavy chain constant regions: IgM, IgD, IgG, IgE and IgA. Of the five classes, IgG is the most frequently used for cancer immunotherapy and is the focus of this Review. Antibodies can be subdivided into two distinct functional units: the fragment of antigen binding (Fab) and the constant fragment (Fc). The Fab contains the variable region, which consists of three hypervariable complementarity determining regions (CDRs) that form the antigen binding site of the antibody and confer antigen specificity. Antibodies are linked to immune effector functions via the Fc fragment, which is capable of initiating complement-dependent cytotoxicity (CDC), binding to Fc γ-receptors (FcγR) and binding to the neonatal Fc receptor (FcRn) (FIGURE 2).
Subclasses of IgG, most notably IgG1 and IgG3, are potent activators of the classical complement pathway. The binding of two or more IgG molecules to the cell surface leads to high-affinity binding of C1q to the Fc domain, followed by activation of C1r enzymatic activity and subsequent activation of downstream complement proteins. The result of this cascade is the formation of pores by the membrane attack complex (MAC) on the tumour cell surface and subsequent tumour cell lysis. In addition, production of the highly chemotactic complement molecules C3a and C5a leads to the recruitment and activation of immune effector cells, such as macrophages, neutrophils, basophils, mast cells and eosinophils3. These properties have been extensively reviewed elsewhere4.
FcγRs can transduce activating signals through immunoreceptor tyrosine-based activation motifs (ITAMs), or deliver inhibitory signals through immunoreceptor tyrosine-based inhibitory motifs (ITIMs). The major inhibitory FcγR is the single chain FcγRIIB (also known as CD32) whereas most Fc-dependent stimulatory signals are transduced by FcγRI (also known as CD64) and FcγRIIIA (also known as CD16A), both of which require an accessory ITAM-containing γ-chain to initiate signal transduction5. FcγRI is a high-affinity receptor expressed by macrophages, DCs, neutrophils and eosinophils5. FcγRIIIA is the primary activating FcγR expressed by natural killer (NK) cells, dendritic cells (DCs), macrophages and mast cells, and is required for NK cell-mediated antibody-dependent cell-mediated cytotoxicity (ADCC)5. FcγRIIIB (CD16B) is a glycophosphatidyl inositol (GPI)-anchored protein that, unlike FcγRIIIA, does not contain the common gamma chain and is exclusively expressed on human neutrophils.
The binding of IgG antibodies to tumour cells enables the recognition of these targets by immune effector populations that express Fcγ receptors, such as NK cells, neutrophils, mononuclear phagocytes and dendritic cells. Cross-linking of FcγRs on these cells promotes ADCC and tumour cell destruction (FIGURE 2). Following tumour cell lysis, antigen-presenting cells can present tumour-derived peptides on MHC class II molecules and promote CD4+ T cell activation. Additionally, in a process known as cross-presentation, tumour-derived peptides can be presented on MHC class I molecules, resulting in activation of CD8+ cytotoxic T cells (see below).
FcRn are structurally distinct from FcγR, and are related to MHC class I molecules. They bind both IgG antibodies and albumin in a pH-dependent manner, with optimal binding occurring at acidic pH. The role of FcRn in the passive transfer of maternal humoral immunity from mother to foetus has been well characterized6. FcRn also plays an important role in the maintenance of serum IgG and can contribute to the long half-life seen with this isotype (FIGURE 2). FcRn expressed on the vascular endothelium can bind IgG by its Fc domain, returning it to the circulation, or protecting it from transcytotic lysosomal catabolism en route to the lymphatics6. FcRn–IgG interactions could be useful therapeutically, and efforts to manipulate the binding of monoclonal IgG to FcRn to enhance the serum half-life of IgG are underway7. In addition to regulating the serum half-life of human IgG, FcRn may also contribute to antibody-mediated antigen presentation8.
Many of the tumour-expressed targets for therapeutic antibodies are growth factor receptors that show increased expression during tumourigenesis. By blocking ligand binding and/or signalling through these receptors, monoclonal antibodies may serve to normalize growth rates, induce apoptosis and/or help sensitize tumours to chemotherapeutic agents9. In addition, antibodies that target the tumour microenvironment and inhibit processes such as angiogenesis have shown therapeutic promise.
Members of the epidermal growth factor receptor (EGFR) family, including EGFR (also known as ERBB1), HER2 (also known as ERBB2), HER3 (also known as ERBB3), and HER4 (also known as ERBB4), are frequently overexpressed in solid tumours and are the target of many currently used therapeutic antibodies. Cetuximab, a chimeric EGFR-specific IgG1 monoclonal antibody, functions by preventing binding of activating ligand10 and by preventing receptor dimerization, a critical step for initiating EGFR-mediated signal transduction11. Panitumumab, a fully humanized IgG2 isotype antibody that is specific for EGFR, works by a similar mechanism as cetuximab12, but unlike cetuximab, it does not promote ADCC. Both of these agents have been used as second- or third-line therapy for the treatment of metastatic colorectal cancer12. Cetuximab, in contrast to the EGFR-specific antibody panitumumab, is often used in combination with other chemotherapeutic regimens. Combining cetuximab therapy with folinic acid, 5-fluorouracil and irinotecan (FOLFIRI) chemotherapy has been shown to prolong progression-free survival in patients with metastatic colon cancer whose tumours harbour wild type v-Ki-ras2 Kirsten rat sacrcoma viral oncogene homologue (KRAS) alleles13. In contrast, these therapeutic agents are ineffective when KRAS is mutated13. A fully human anti-EGFR antibody, Necitumumab (IMC-11F8) was recently described and shown to be well tolerated in patients with advanced solid malignancies14.
In addition to targeting the complete form of EGFR, efforts are underway to target a truncated form of the receptor, EGFRvIII (an in frame deletion of exons II–IV), which is found in patients with glioblastoma, head and neck cancer and non-small cell lung carcinoma15. A phase I study using the EGFRvIII-targeting monoclonal antibody 806, which targets EGFR as it dimerizes following ligand binding16, showed good antibody penetration of tumour tissue and no significant toxicities in patients with metastatic disease17.
In contrast to EGFR, HER2 has no known ligand and antibodies targeting this receptor function primarily to inhibit receptor homo- and hetero-dimerization and internalization, rather than by blocking ligand-binding18. HER2 is gene-amplified and overexpressed in approximately 30% of invasive breast cancers and is overexpressed, although rarely gene-amplified, by some adenocarcinomas of the lung, ovary, prostate and gastrointestinal tract18. Trastuzumab, a humanized IgG1 antibody, is used for the treatment of invasive breast cancer that exhibits gene amplification and overexpression of HER2. Trastuzumab monotherapy showed a 35% objective response rate in patients with metastatic breast cancer not previously receiving chemotherapy19. The mechanisms of action by which trastuzumab exerts its anti-tumour effects include inhibition of receptor dimerization, endocytic destruction of the receptor and immune activation20. Another HER2-directed antibody, pertuzumab, binds at a distinct site from trastuzumab and sterically inhibits receptor dimerization21. Synergistic anti-tumour effects of combination therapy with pertuzumab and trastuzumab have been reported in pre-clinical models22.
A new HER3-targeted antibody, MM-121, is currently being developed and has been shown to specifically bind HER3, inhibit growth of mouse xenograft tumours and block heregulin-dependent signalling through the protein kinase AKT, leading to tumour cell death23. Efforts to target HER4 are underway; however, the biological significance of HER4 expression in cancer is poorly understood. HER4 has been reported to be both upregulated and downregulated in cancer, presumably due to the presence of many isoforms and its prognostic value is yet to be determined24. Treatment with a monoclonal antibody targeting selected HER4 isoforms resulted in decreased proliferation of two tumour cell lines; mechanistically, this was due to inhibition of HER4 phosphorylation and cleavage, and the downregulation of HER4 expression24.
Strategies to target critical events within the tumour microenvironment have demonstrated therapeutic benefit in preclinical and clinical settings. For example, many solid tumours express vascular endothelial growth factor (VEGF), which binds to its receptor on the vascular endothelium to stimulate angiogenesis. Bevacizumab, a VEGF-specific humanized monoclonal antibody, blocks binding of VEGF to its receptor and is approved for the treatment of breast, colorectal and non-small cell lung cancer in combination with cytotoxic chemotherapy25. Efforts to target VEGF receptors (VEGFRs) by other molecules are also underway. Ramucirumab, a fully human monoclonal antibody against VEGFR2, has been shown to inhibit growth of human xenografts in mice26. A multi-center phase III clinical trial investigating the effect of combination therapy with ramucirumab and the chemotherapy agent docetaxel in women with HER2-negative metastatic breast cancer is currently underway27. Similarly, efforts to target VEGFR1 with the fully human antibody IMC-18F1 are currently underway and have shown preclinical promise28.
The increasing therapeutic use of bevacizumab has led to an increase in bevacizumab-resistant tumours due to upregulation of other proangiogenic mediators such as platelet-derived growth factor (PDGF). PDGF-receptor (PDGFR)-signalling is important in maintaining the endothelial support system, which stabilizes and promotes the growth of new blood vessels29. Blockade of PDGFR-signalling via a PDGFRβ-specific human antibody has been shown to synergize with anti-VEGFR2 therapy in preclinical models and suggests the utility of anti-PDGFRβ therapy in the setting of bevacizumab resistance30.
In addition to directly targeting tumour cells, numerous antibody-based therapeutic strategies have been developed to target cells of the immune system with the goal of enhancing anti-tumour immune responses. Here, we consider the targeting of immunoregulatory co-receptors, antibody-based strategies aimed at reversing tumour-mediated immunosuppression and Fc domain modulation to alter the specificity of Fc receptor-targeting and activation.
CD40 is a member of the tumour necrosis factor (TNF) receptor (TNFR) family and is expressed by B cells, DCs, monocytes and macrophages. Engagement of CD40 on antigen-presenting cells leads to the upregulation of costimulatory molecules, production of pro-inflammatory cytokines and facilitates cross-presentation of antigens31. Many tumours have been shown to express CD40, including carcinomas of the ovary, nasopharynx, bladder, cervix, breast and prostate, and the engagement of CD40 can lead to a direct anti-tumour effect in some tumours in vivo32. In one study, the effect of a fully human anti-CD40 agonist monoclonal IgG2 (CP-870,893) was assessed in 29 patients with advanced solid tumours. The results demonstrated four partial responses (all patients with melanoma), with one complete resolution33. A second anti-CD40 antibody, dacetuzumab (SGN-40), is a humanized IgG1 that can induce tumour cell apoptosis as well as ADCC, and was recently shown to exert its anti-tumour effects by inducing Fc-mediated phagocytosis of tumour cells by macrophages34. Early clinical trials suggest that dacetuzumab exhibits promise in the treatment of patients with non-Hodgkin’s lymphoma35.
CTLA4, a homologue of CD28, is a negative regulator of T cell activation that binds CD80 (also known as B7.1) and CD86 (also known as B7.2) with higher affinity than CD28. CTLA4 blockade enhanced rejection of CD86-negative tumours, implying that this was not through direct effects on the tumour, and reduced the growth of established tumours36. Furthermore, CTLA4 blockade resulted in protection from secondary tumour challenge, suggesting the development of immunological memory36. Subsequent studies showed that treatment with a CTLA4-specific antibody can prevent and reverse antigen-specific CD8+ T cell tolerance in a CD4+ T cell-dependent manner37. Recent work has revealed that enhancement of T cell effector functions, combined with the inhibition of regulatory T cells (TREG), might be responsible for the anti-tumour effects of CTLA4 blockade38. These preclinical data show that anti-CTLA4 treatment can enhance adaptive immunity and promote tumour regression.
Two CTLA4-specific monoclonal antibodies have been studied in clinical trials. Tremelimumab (IgG2 isotype) and ipilimumab (IgG1 isotype) can induce delayed disease regression in patients with metastatic melanoma39. In addition, ipilimumab is being investigated as a treatment option for men with hormone-refractory prostate cancer40. On a cautionary note, dose-dependent, immune-related toxicity has been reported with CTLA4-specific antibodies; these side-effects include rash, vitiligo, diarrhoea, colitis, nephritis with azotemia, hypophysitis and hepatitis41. Antibodies targeting OX40, B7-H1 and CD137 have also shown pre-clinical promise and are described in Table 2, but have not yet undergone extensive clinical testing.
Tumour cells and the surrounding stroma can produce profoundly immunosuppressive cytokines and growth factors. Among the best characterized of these factors is transforming growth factorβ (TGFβ), which can inhibit T cell activation, differentiation, and proliferation42. TGFβ has been shown to promote tumour escape from the immune system, and high plasma levels of TGFβ correlate with a poor outcome in various malignancies42. GC-1008 is a fully human anti-TGFβ antibody that binds to all three isoforms of TGFβ43. Recruitment is currently underway to a clinical trial GC-1008 in patients with metastatic kidney cancer or malignant melanoma (clinicaltrials.gov, NCT00899444).
The production of immunosuppressive factors such as TGFβ can result in the accumulation of suppressive CD4+CD25+FOXP3+ TREG cells44, which is associated with poor clinical outcomes45. Treatment with CD25-specific monoclonal antibody to deplete TREG cells has shown remarkable potential in pre-clinical models of various malignancies46 and has been shown to suppress tumour formation and metastasis in a mouse model of breast cancer47. Clinically, the humanized IgG1 CD25-specific monoclonal antibody daclizumab is well tolerated, resulted in depletion of TREG cells in patients with metastatic breast cancer and may synergize with peptide vaccines targeting human telomerase reverse transcriptase and the anti-apoptotic protein survivin48. However, anti-CD25 therapy could potentially deplete effector T cells that have upregulated CD25 upon activation. In addition, further studies are required to understand the role of CD25-negative TREG in the setting of cancer48.
As discussed below, numerous pre-clinical studies suggest that ADCC is a major mechanism of action of several monoclonal antibodies used in cancer immunotherapy. Efforts to modify the Fc domain primary structure using computational and high-throughput screening have resulted in Fc domains with higher affinity for FcγRIIIA and an enhancement of ADCC49. The new CD20-specific antibodies, ocrelizumab and AME-133, both contain mutated Fc domains resulting in enhanced ADCC compared to rituxumab50. Modification of Fc domain oligosaccharide content provides another mechanism for enhancing ADCC. Most of the currently used therapeutic antibodies are highly fucosylated due to the nature of the cell lines used for manufacturing. However, antibodies with defucosylated oligosaccharides display a significant enhancement in ADCC in vitro and enhanced in vivo anti-tumour activity50. Phase I trials of defucosylated antibodies specific for CC chemokine receptor 4 (CCR4), which is expressed by some lymphoid neoplasms and used by TREG cells to facilitate their migration to the tumour microenvironment45, have shown promise and early data suggest efficacy at significantly lower doses than conventional therapeutic antibodies50.
Many mechanisms have been proposed to explain the clinical anti-tumour activity of unconjugated tumour antigen-specific monoclonal antibodies. As discussed above, the ability of some antibodies to disrupt signalling pathways involved in the maintenance of the malignant phenotype has received wide attention. However, the ability of antibodies to initiate tumour-specific immune responses has been less well recognized. Here, we describe these mechanisms and discuss the potential for using antibodies to manipulate the host immune response to tumours. We focus on three mechanisms: ADCC, complement-dependent cytotoxicity (CDC), and the induction of adaptive immune responses.
Several studies have established the importance of Fc-FcγR interactions to the in vivo anti-tumour effects of certain monoclonal antibodies in murine models and clinical trials. A seminal paper demonstrated that the anti-tumour activities of trastuzumab and rituximab were reduced in FcγR-deficient mice as compared with wild-type mice51. The role of FcγR in the anti-tumour response has been further supported by the finding that polymorphisms in the gene encoding FcγRIII, which lead to higher binding of antibody to FcγRIII, are associated with high response rates to rituximab in patients with follicular non-Hodgkin lymphoma52. A separate study that compared clinical responses to rituximab in patients with follicular lymphoma implicated both FcγRIII and FcγRIIB as playing a role in responses to rituximab53. More recent findings show that FcγR polymorphisms are associated with clinical responses to other antibodies, including trastuzumab54 and cetuximab55. Patients with breast cancer who responded to trastuzumab with complete or partial remission have been found to have a higher capability to mediate in vitro ADCC in response to trastuzumab than did patients whose tumours failed to respond to therapy54.
ADCC enhancement through Fc domain modification has shown promise in the development of next generation antibodies. For example, a CD19-specific antibody with increased FcγRIIIA binding affinity mediated significantly increased ADCC compared to its parental antibody and rituximab56. The in vivo infusion of this high affinity antibody efficiently cleared malignant B cells in cynomolgus monkeys (Macaca fascicularis)57.
Most clinically approved monoclonal antibodies that mediate ADCC also activate the complement system. However, alemtuzumab, which recognizes CD52 expressed on mature B and T cells, activates human complement and demonstrates anti-tumour activity in patients with chronic lymphocytic leukemia, but does not mediate ADCC58. The relationship between complement activation and therapeutic activity is also suggested from studies with several clinically approved monoclonal antibodies. The anti-CD20 monoclonal antibody rituximab has been found to be dependent in part on CDC for its in vivo efficacy. In a preclinical therapy model, anti-tumour protection by rituximab was completely abolished in C1q knockout mice59. It was also demonstrated that depletion of complement reduced the therapeutic activity of rituximab in a xenograft model of human B cell lymphomas60. The importance of rituximab-mediated CDC is supported by the demonstration that genetic polymorphisms in the C1qA gene correlate with clinical response to rituximab therapy in patients with follicular lymphoma61.
Optimization of antibody-based complement activities can enhance anti-tumour activity. For example, a CD20-specific antibody, ofatumumab, which mediates improved CDC was approved for the treatment of patients with chronic lymphocytic leukemia (CLL) in 2009. This fully human antibody binds a different epitope than rituximab with more optimal binding kinetics and induces potent tumour cell lysis through improved activation of the classical complement pathway62. An initial study in patients with refractory CLL showed a 50% response rate to ofatumumab, suggesting a higher efficacy than rituximab in the setting of CLL, although this higher response rate may not solely be due to enhanced CDC62. Several studies indicate that both CDC and ADCC can contribute to monoclonal antibody-induced tumour cell lysis. However, the relative clinical importance of each mechanism, and whether the mechanisms are synergistic, additive or antagonistic, remains uncertain. For example, in a mouse model of lymphoma, depletion of complement enhances NK cell activation and ADCC, thus improving the efficacy of the antibody63.
Early research on the anti-tumour effects of therapeutic antibodies focused on the potential value of passive immunotherapy provided through ADCC and CDC. Interestingly, studies have suggested that the maximal clinical benefit of rituximab is not apparent until months after initiation of therapy, suggesting a role for the adaptive immune system in mediating the long-term benefit of monoclonal antibodies64. There is growing evidence to suggest a role for cross-presentation in the induction of adaptive immune responses following antibody therapy. DCs are capable of presenting peptides from engulfed apoptotic cells on MHC class I molecules to elicit antigen-specific CD8+ T cell responses65. In one study, DC loaded with killed allogeneic melanoma cells induced cytotoxic T-lymphocytes (CTL) that were specific for various melanoma antigens and killed melanoma cell lines in vitro66. Another group reported a similar finding using a head and neck squamous cell carcinoma cell line67. Induction of adaptive immunity was linked to antibody therapy in a study showing that tumours coated with antibodies were able to enhance cross-presentation of tumour antigens in a FcγR-dependent fashion68, and blockade of the inhibitory receptor FcγRIIB was shown to enhance cross-presentation by DCs69. ADCC mediated by monoclonal antibodies may trigger cross-presentation by DCs and lead to adaptive immune responses, as DCs can engulf the resultant apoptotic tumour cells and subsequently present tumour antigens on MHC class I and II molecules (FIGURE 3). This dual presentation leads to direct tumour cytotoxicity by CTLs and the generation of CD4+ T cells, which can prime B cells for the production of tumour-specific host antibodies. In addition, cross-presentation may be mediated by phagocytosis of dying antibody-coated tumour cells through FcγR68. It is important to note that DC presentation of engulfed tumour antigens can lead to either immunity or tolerance based on the tumour microenvironment. Accordingly, strategies aimed at targeting tolerizing factors in the microenvironment may synergize with tumour-directed antibody therapy by enhancing cross-presentation and breaking local tolerance.
The anti-tumour efficacy of many therapeutic antibodies can be enhanced by their use in combination with other immunomodulatory approaches such as chemotherapy, radiotherapy, targeted-therapy agents, vaccines or other immunomodulators.
Although chemotherapy has been traditionally thought to be immunosuppressive, recent evidence has challenged this view70. Bevacizumab in combination with FOLFIRI resulted in a 10.6 month median duration of progression free survival compared to 6.2 months with FOLFIRI alone71. Similarly, trastuzumab in combination with chemotherapy showed a 50 percent objective response rate compared to 35 percent with chemotherapy alone in patients with metastatic breast cancer72. One small study found that combined treatment with trastuzumab and paclitaxel induced humoral and cellular anti-HER2 immune responses that were associated with favorable clinical outcomes in patients with advanced breast cancer73. The induction of HER2-specific CD4+ T cells and humoral immune responses indicated that an adaptive immune against HER2 was induced by this treatment. Similar benefits are also well documented with rituximab, where inclusion of rituximab in a standard chemotherapy regimen resulted in an overall response rate of 76 percent compared to 63 percent with chemotherapy alone in patients with diffuse large B-cell lymphoma74. The benefit of adding rituximab to chemotherapeutic regimens was also reported in follicular and mantle cell lymphoma75.
Although radiation therapy has long been viewed as being immunosuppressive, it has been hypothesized that radiation can result in enhanced anti-tumour immunity76. Based on this understanding, investigators have begun to combine radiation and antibody therapies. A phase III clinical trial showed that combining the EGFR-targeting antibody, cetuximab, to a primary radiotherapy regimen significantly improved the overall 5 year survival of patients with advanced squamous-cell carcinoma of the head and neck, compared with radiotherapy alone77. The safety and efficacy of neoadjuvant bevacizumab, which targets VEGF, with chemoradiotherapy in advanced rectal cancer was evaluated in a phase I/II clinical study and appeared to be safe and potentially effective78. The concentrations of VEGF and proinflammatory cytokines such as IL-6 in the blood of these patients significantly correlated with clinical outcomes. However, these studies have not tested the possibility that combinations of radiation with antibody therapy can induce host-protective adaptive immune responses. In a phase II clinical study, which reported the use of a recombinant cancer vaccine combined with standard definitive radiotherapy in patients with localized prostate cancer, it was shown that combination treatment can generate the development of T cells directed against other tumour-associated antigens not present in the vaccine79. Recent evidence suggests that radiation therapy might activate innate immune effector cells through toll-like receptor (TLR)-dependent mechanisms, thereby augmenting the adaptive immune response to cancer80.
Improved understanding of the cytokines involved in modulating effector cells of the innate immune system, together with recent understanding of NK cell recognition and killing of target cells, have provided a basis for the rational investigation of immunoregulatory cytokine combinations in the treatment of specific cancers. Cytokines such as IL-281 and IFNγ82 have been shown, both in murine models and clinical trials, to enhance ADCC by stimulating or expanding NK cells, macrophages and monocytes in vivo. In a phase II clinical trial of combined therapy with IL-2 and rituximab, IL-2 expanded the number of FcR-bearing cells in vivo and enhanced in vitro ADCC by rituximab81. The effects of granulocyte colony-stimulating factor (G-CSF)-mediated neutrophil stimulation on rituximab activity have been studied in a phase I/II trial in patients with low-grade lymphoma83. Although the overall response rate was comparable to that reported for rituximab monotherapy, remission duration in this pilot phase II study was remarkably long. A recent phase 1 study demonstrated the feasibility of integrating IL-2 into a regimen of ch14.18(an antibody targeting ganglioside GD2, a cell surface antigen highly expressed on human neuroblastomas) plus GM-CSF in an effort to boost effector immune responses84.
Combination therapy employing monoclonal antibodies and small molecule tyrosine kinase inhibitors (TKI), which act to block activation of intracellular signalling in tumour cells or stromal cells following receptor engagement, has been proposed as an approach to enhance clinical efficacy through more effective cancer-specific signalling perturbation with concurrent activation of immune effector mechanisms. The effect of combining SU6668 (a VEGFR2-, PDGFRβ– and FGFR1 TKI) with a mouse CD86–IgG fusion protein, which provides a stimulatory signal to CD86, was tested in pre-clinical mouse models. These studies demonstrated enhanced anti-tumour and anti-metastatic responses, as well as an increase in tumour-infiltrating CD8+ T cells, compared with monotherapy85. Similarly, when the effect of combining cetuximab with gefitinib (an EGFR TKI) was investigated in preclinical models, combination treatment resulted in permanent regression of large tumours at suboptimal doses, whereas cetuximab monotherapy induced transient regression only at the highest doses86. Combination therapy provided superior inhibition of EGFR-signalling, greater inhibition of cell proliferation and vascularization, and enhanced apoptosis86. A phase I clinical trial demonstrated the feasibility of gefitinib and cetuximab combination therapy in patients with refractory non-small cell lung cancer87. Recent work has shown enhanced efficacy of combination therapy using lapatinib, a small molecule HER2-specific TKI, and trastuzumab compared to monotherapy88. In addition to its TKI activity, lapatinib promoted the formation of inactive HER2 homodimers at the plasma membrane, resulting in enhanced trastuzumab-mediated ADCC88. A recent phase III clinical trial demonstrated that trastuzumab and lapatinib combination therapy improved progression free survival compared to lapatinib monotherapy in patients with HER2 positive, trastuzumab refractory metastatic breast cancer89.
Given the evidence suggesting that monoclonal antibody therapy can have both direct and indirect effects on the adaptive immune response, combining antibody therapy with vaccination strategies may prove to be efficacious. Using irradiated GMCSF-producing B16 melanoma cells as a vaccination strategy, it was shown that the addition of anti-CTLA4 antibody was more effective at eradicating established B16 tumours compared to vaccination alone90. This enhanced anti-tumour effect was dependent on CD8+ T cells and NK1.1+ cells90. Similarly, in a mouse tumour model induced by the subcutaneous injection of a colon carcinoma, the addition of the TGFβ-specific antibody ID11 to an irradiated tumour cell vaccine significantly enhanced anti-tumour activity in a CD8+ T cell-dependent manner91. Addition of TGFβ-specific antibody to a peptide vaccine was found to increase the number and lytic activity of tumour antigen-specific CTLs, but had no effect on the numbers of TREG or TH17 cells92. Therefore pre-clinical data support the idea of combining antibodies that target the immunosuppressive tumour microenvironment with vaccination. Similarly, studies combining antibodies specific for tumour antigens, such as trastuzumab, with peptide-based vaccines that stimulate HER2-specific T cell responses are underway and have shown initial clinical promise93.
Advances in antibody engineering have generated novel antibody constructs that permit the testing of new antibody based therapy strategies. Bispecific antibodies simultaneously targeting epitopes on tumour as well as immune effector cells have long been shown to have equivalent or superior potency as compared to their IgG counterparts. However, they showed limited clinical efficacy largely attributed to host toxicities caused by concurrent T cell costimulation and short half lives. BiTE molecules, which employ a new format of bispecific anti-CD3 antibody, are formed by linking two Fv fragments using flexible linkers. BiTE molecules show enhanced tumour cell lysis, high protein stability and efficacy at low T cell to target cell ratios94. MT110, a BiTE molecule specific for human epithelial cell adhesion molecule (Ep-CAM) and CD3, has shown anti-tumour efficacy in a mouse xenograft model95 and is currently being studied in a phase I clinical trial (clinicaltrials.gov, NCT00635596). Additional advances in protein engineering have focused on equipping non-Ig family proteins with novel binding sites and are collectively called engineered protein scaffolds. In contrast to antibodies, scaffolds are usually comprised of monomeric proteins or a stably folded extramembrane domain of a surface receptor that is highly stable and compatible with high yield expression systems96. Examples of engineered protein scaffolds include designed ankyrin repeat domains (DARPins) and Adnectins, which are composed of the 10th extracellular domain of fibronectin III that folds to adopt an Ig-like structure. A VEGFR-2 adnectin was recently reported to possess anti-tumour activity against human tumour xenografts97.
FcγRs provide the key link between therapeutic antibodies and the cellular immune system, and enable monoclonal antibodies to induce adaptive immune responses. The magnitude and quality of the innate immune responses induced by ADCC might have an impact on the ensuing adaptive immune response. ADCC-inducing approaches therefore offer a promising basis to improve antibody efficacy. New insights from animal models and clinical trials suggest a rationale for ADCC-based combination therapy, approaches that promote antigen presentation (for example, Toll-like receptor agonists98), co-stimulation (for example, CTLA-4 antibody) and T cell activation or expansion (for example, IL-281). Furthermore, antibody structures can be modified to selectively engage activating rather than inhibitory FcγRs. Fusion antibodies with immunostimulatory motifs that induce and amplify antigen presentation and costimulation have also shown promise99. However, new methods and more investigation are still needed to more accurately detect and monitor immune responses directed against tumour antigens in vivo.
The past century has witnessed the evolution of the ‘magic bullet’ from concept to clinical reality. The attributes of target specificity, relatively low toxicity as well as the ability to activate the immune system suggest the continuing promise of therapeutic antibodies. Therapeutic antibodies currently provide clinical benefit to patients with cancer and have been established as ‘standard of care’ agents for several highly prevalent human cancers. The next generation of unconjugated antibody therapies will undoubtedly yield many effective new treatments for cancer over the next decade. These advances will arise from the identification and validation of new targets, the manipulation of tumour–host microenvironment interactions, and the optimization of antibody structure to promote the amplification of anti-tumour immune responses.
Some solid tumours exhibit significant CD4+ and CD8+ T cell infiltrates. CD8+ cytotoxic T lymphocytes (CTLs) are critical for anti-tumour immunity. They recognize specific peptides presented by tumour cells on MHC class I molecules and release perforin and granzymes to initiate tumour cell apoptosis. Tumour antigen-specific CTLs can be detected in the serum of cancer patients and the influx of antigen-specific CTLs in the tumour microenvironment is associated with favorable clinical outcomes100. CD4+ helper T (TH) cells can be subdivided into TH1, TH2, and TH17 cell subsets based on their cytokine expression profiles. TH1 cells promote the differentiation of naïve CD8+ T cells into activated CTLs by increasing the expression of stimulatory B7 family members on antigen presenting cells (APCs) as well as by producing IL-2, which stimulates T cell expansion. However, tumour-infiltrating CD4+ T cells are often TH2-polarized and secrete suppressive factors, such as IL-4 and IL-10, that can inhibit TH1 responses101. The roles of TH17 cells, which produce IL-17A, IL-22 and express IL-23R, in anti-tumour responses are currently being elucidated; reports show that TH17 cells can be involved in both the promotion of anti-tumour immunity102 and the enhancement of tumour growth103.
Regulatory T (TREG) cells derived from the thymus comprise another important subset of CD4+ T cells. They are defined by the coexpression of CD25 (IL-2Rα chain) and the transcription factor forkhead box P3 (FOXP3), and function to suppress immune responses. Many solid tumours accumulate TREG cells in the tumour microenvironment and this accumulation is associated with poor clinical outcomes in some cancers45. TREG cells can exert their suppressive effects by (i) inducing expression of inhibitory B7 family members on APCs, (ii) directly killing APCs and effector T cells through perforin and granzyme release (iii) engagement of CD80 and CD86 on APC through CTLA4, which leads to T cell anergy and death and production of suppressive mediators such as IL-10 and TGFβ104. Therefore, effective anti-tumour immunity is critically dependent on optimizing the balance between tumour antigen-specific CTLs and immune-suppressive T cells in the tumour microenvironment.
L.M.W. and S.W. are supported by the US National Institutes of Health (NIH) (grants CA51008, CA50633, CA121033.)
Louis M. Weiner received his MD from Mount Sinai School of Medicine, trained in internal medicine at the University of Vermont and completed his hematology/oncology fellowship at Tufts-New England Medical Center. He was subsequently based at Fox Chase Cancer Center, where he served as Chairman of the Department of Medical Oncology and Vice President for Translational Research. He is currently the Director of the Lombardi Comprehensive Cancer Center at Georgetown University Medical Center, and Chair of the Department of Oncology. Dr. Weiner’s research focuses on monoclonal antibody engineering and immunotherapy.
Rishi Surana is a M.D./Ph.D. student in the tumour biology training program at the Georgetown University Medical Center
Shangzi Wang received her Ph.D. from the Ocean University of China, Qingdao (China), and conducted her postdoctoral training with Dr. Weiner at the Lombardi Comprehensive Cancer Center at Georgetown University Medical Center.
Competing financial interests
The authors declare no competing financial interests.