This article is a CME certified activity. To earn credit for this activity visit:
http://cme.medscape.com/viewarticle/720825

Introduction

Treating cancer by harnessing a patient’s immune system has several major advantages over other forms of cancer therapy (Table 1). For example, some cells of the immune system can respond to specific immunogenic proteins, or antigens, expressed by the tumor. This characteristic allows specificity of the immune response to cancer without excessive toxicity to normal tissues, as is seen with cytotoxic chemotherapy. Antigen-specific T lymphocytes, presumably the most important component of the immune system in mediating an antitumor response, have the capability of homing to any site of cancer even if disease deposits are located deep in tissues.^[1] Therefore, unlike other standard forms of cancer treatment, the immune response has the potential to eradicate cancer in any location. Cancer-specific T and B lymphocytes are cells that can directly induce tissue destruction and will continue to proliferate and function as long as there is antigen present to stimulate their activity. For this reason, once a robust immune response is elicited, no further immune-based treatments would be necessary. Finally, a key characteristic of an effective immune response is the generation of immunologic memory, which is the persistence of an antigen-specific immune response over many years. If the cancer antigen is sensed again, even decades after the initial diagnosis of disease, immune cells will rapidly respond, proliferate, and destroy antigen-expressing cancer cells before those cells have the chance to become re-established.

Unfortunately, the development of actual immune-based therapies for the treatment of cancer has been challenging. In large part, the challenges have been due to the nature of the immunogenic proteins that are expressed by human tumors. Over the last decade, a host of human tumor antigens have been identified as potential therapeutic targets. Some antigens expressed in tumors are viruses — hepatitis B virus (HBV) in hepatocellular carcinoma, Epstein-Barr virus (EBV) in lymphomas and nasopharyngeal carcinomas, and human papillomavirus (HPV) in cervical cancer are just a few examples — but most immunogenic cancer-associated proteins are normal cellular proteins (self proteins) that have become qualitatively or quantitatively altered in the malignant state as compared with their expression in normal tissues. Clinical responses to immunotherapies targeting either viral or self proteins have been reported.^[2] Immunologic targeting of a cancer-related self protein is hampered by the multiple mechanisms our bodies have of preventing autoimmunity. The inflammatory response that develops when cancer grows elicits immune system cells that are likely to turn off a destructive cancer-specific immune response because the antigens being recognized are perceived as “self.” Certain types of macrophage, termed M2, will secrete cytokines that prevent T-cell proliferation.^[3] Immature myeloid cells, myeloid-derived suppressive cells, are also present in the tumor bed and can inhibit the generation of a clinically productive immune response by preventing antigen-specific T cells from functioning correctly.^[4] T cells themselves can differentiate into regulatory cells when they sense self antigens and prevent further tumor recognition via secretion of interleukin (IL)-10 and transforming growth factor (TGF)-beta, which are immune-suppressant cytokines.^[5] These are just a few of the natural defense mechanisms in place for preventing the development of autoimmune disease; unfortunately, these same mechanisms limit the tumor-specific immune response. Effective cancer immunotherapy must generate a destructive immune response as well as control tolerizing mechanisms that are in place to limit self-specific immunity.

Despite the challenges, there are several immune-based therapies that are routinely used in the treatment of cancer patients. Cancer immunotherapy is generally classified as being “active” or “passive” (Table 2). Active immunotherapy is a treatment modality that functions by stimulating the patient’s own immune system to generate the cells needed to impart an antitumor effect. An example of an active immunotherapy would be a vaccine. The vaccine is administered to stimulate T or B lymphocytes to recognize and destroy the cancer. The use of nonspecific immunomodulators such as bacillus Calmette-Guerin (BCG) would also be considered active immunotherapy. After administration of BCG, it is assumed that cells of the innate immune system, present in the patients, would respond and cause inflammation that could result in the eradication of superficial bladder cancer. In the case of active immunotherapy, patients must have immune systems capable of competently responding to stimulation. For this reason, in general, active immunotherapy is not effective in patients with advanced-stage refractory disease who may have a depressed number of immune system cells able to adequately function. Passive immunotherapy provides the immune response to the patients. Monoclonal antibody therapy is considered a passive immunotherapy. Rather than stimulating a patient’s own antibody response, the infusion of monoclonal antibodies provides the antigen-specific antibodies to the patient. Similarly, rather than stimulating a patient’s own T cells via vaccination, adoptive T-cell therapy infuses high numbers of antigen-specific T cells into patients, thus providing immediate robust immunity to a specific target. Because patients do not have to generate their own endogenous immune response, passive immunotherapy is often used in the treatment of patients with well established and even refractory cancers. An example of adoptive T-cell therapy would be the use of donor lymphocyte infusions in the treatment of chronic myeloid leukemia (CML) that has relapsed after allogeneic hematopoietic stem cell transplant (HSCT).

The potential mechanisms of action of many cancer immunotherapies are multifactorial and often not fully understood because the immune system is a complex organization of numerous components, pathways, and interdependent interactions. The cell types involved in mediating tumor-specific immunity will define the clinical efficacy as well as the toxicities associated with targeted immune-based treatments.

Components of the Immune System

Immunity is described as either “innate” or “adaptive.” Innate immune cells are those that are the first responders to abnormalities that are detected by the immune system. These cells include monocytes, macrophages, and other antigen-presenting cells (APCs), as well as neutrophils, eosinophils, mast cells, and natural killer (NK) cells. Innate immune cells do not recognize specific antigens; rather, they recognize substances expressed by pathogens via receptors, known as toll-like receptors (TLRs) on their cell surface. DNA, RNA, glycans, and other substances that are present in broad categories of pathogens such as bacteria and other infectious organisms will stimulate innate immunity via these receptors. Recognition of pathogens results in an immediate influx of a variety of phagocytic cells that can eliminate the pathogen. Moreover, in certain conditions, some cells of the innate immune system play an important role in processing and presenting the pathogens they enveloped to adaptive immune system cells, resulting in long-lasting immunity. Cells of the innate immune system do not proliferate in response to specific antigens, nor do they maintain immunologic memory. Innate immune system cells are the major component of most inflammatory responses.

B and T lymphocytes, or B and T cells, are the main constituents of the adaptive immune response. These cells display all the characteristics shown in Table 1 and respond to specific antigens. B cells produce proteins called antibodies that bind to soluble antigens. Among other functions, antibody binding to antigen can result in:

• Internalization of the antigen into the B cell, where it is processed and presented to T cells;
• Fixation of complement to the antibody and the initiation of an enzymatic cascade that will result in the destruction of the cell expressing the antigen; and/or
• Fixation of APCs to receptors on the antibody that will result in further phagocytosis of the antigen and death to antigen-expressing cells.

T cells have the capability of directly killing abnormal cells. The cytotoxic CD8-positive T cell (CTL) is thought to be the most important effector cell in the immune eradication of cancer. CTLs can kill cells directly via enzyme-mediated mechanisms or can induce senescence in cells expressing antigen. T cells recognize antigen that has been processed and presented in immune receptor molecules, major histocompatibility molecules (MHCs), on the surface of APCs, or other cells. After processing, the protein antigen is cleaved into amino acid fragments called peptides. The placement of antigenic peptides in the MHC is specific, and either CTL or CD4-positive T-helper (Th) cells will recognize the peptide-MHC complex, depending on the amino acid sequence of the peptide and the type of MHC molecule. CTLs recognize peptide fragments in the context of MHC class I, and Th cells recognize peptides presented in MHC class II. Th cells are critical to the overall function of the adaptive immune response. Th cells secrete cytokines in response to antigen recognition. These cytokines can either stimulate further proliferation of CTLs (interferon [IFN]-gamma, TNF-alpha, and IL-2) or the proliferation of B cells (IL-4, IL-5, IL-10), which often results in a dampening of the CTL response. Furthermore, Th cells can differentiate into T regulatory cells and secrete cytokines (IL-10 and TGF-beta), which will turn off tissue-destructive CTL. Therefore, Th cells can either generate destructive inflammation or immune suppression, depending on the antigen and the conditions in which the antigen is presented to the T cells.

Because the therapeutic function of B and T cells is linked with the effective processing function of APCs, immune-based therapies must take both the innate and adaptive immune systems into account to be clinically effective.

Cancer Immune-Based Therapies

There are many types of immune-based therapies routinely used in the treatment and prevention of both solid tumors as well as hematopoietic malignancies. Both active and passive immune therapies have been approved for clinical use, and treatments can target both the innate and adaptive immune systems. The majority of immunotherapeutic approaches currently used and/or under development are designed to elicit an adaptive immune response due to the unique characteristics of such a response in the treatment of cancer and prevention of relapse (Table 1). Immune-based cancer therapies either aim to directly stimulate B or T cells (monoclonal antibodies, vaccines, adoptive T-cell therapy) or indirectly generate an adaptive immune response via providing appropriate growth factors and a cytokine milieu to support adaptive immunity (cytokine therapy) or enhance APC function so that T cells can be stimulated via cross priming (immunomodulators). Cross priming, the presentation of tumor antigens to T cells indirectly via APCs, is the primary mode by which T cells recognize tumors.

Monoclonal Antibodies

Monoclonal antibodies directed against growth factor receptors on the surface of tumors have become integrated into cancer therapy for multiple human malignancies and have been shown to improve overall survival when used as part of standard therapy. A few examples of clinically approved monoclonal antibodies for cancer therapy include:

• HER2/neu-specific monoclonal antibodies that are used in the treatment of breast cancer;
• Epidermal growth factor receptor (EGFR)- and vascular endothelial growth factor (VEGF)-specific monoclonal antibodies that are part of the treatment regimen for colon, non-small-cell lung, head and neck, and breast cancer; and
• CD20-specific monoclonal antibodies, a mainstay of lymphoma treatment.

There are several other different monoclonal antibodies approved for clinical use in the treatment of both solid tumors and hematopoietic cancers (Table 3).

A few selected agents clinically approved for treating solid tumors are described below as examples of the mechanisms of action, response rates, and toxicities that can be expected with this class of agents. Moreover, although many of these agents are used in multiple tumor types, only a few examples are reviewed for each monoclonal antibody.

Trastuzumab is an immunoglobulin G(1) kappa monoclonal antibody that is directed against the extracellular domain of the HER2/neu growth factor receptor. HER2/neu is upregulated in approximately 20% of all breast cancers.^[6] The antitumor functions of trastuzumab are associated with its ability to modulate signaling through the HER-2/neu receptor as well as initiate antibody-dependent cell-mediated cytotoxicity (ADCC). Recent studies suggest that trastuzumab disrupts HER2/HER3 interactions, leading to downregulation of AKT signaling, which results in decreased cell proliferation.^[7] Additional biologic mechanisms of action also include inhibition of extracellular domain cleavage, antiangiogenic effects, and decreased DNA repair.^[8] Several preclinical models have demonstrated the ability of trastuzumab to mediate direct cell killing via ADCC.^[9,10] Moreover, specific single nucleotide polymorphisms in Fc receptors, which are involved in the binding of APCs during ADCC, are associated with improved objective response rates and progression-free survival (PFS) in HER2/neu-positive metastatic breast cancer patients who are receiving trastuzumab therapy.^[11] This finding further underscores the importance of ADCC as a mechanism of action of the agent. Trastuzumab is an integral part of the therapy of HER2/neu-positive breast cancer from early stage tumors to advanced stage disease. In the adjuvant setting, at any stage, the use of trastuzumab reduces recurrence by about 50% and increases overall survival by about 30%.^[12] In women with metastatic HER2/neu-positive breast cancer, the use of trastuzumab may result in an improved prognosis compared with women with metastatic HER2/neu-negative breast cancer. A single institution study^[13] evaluating more than 2000 women with metastatic breast cancer demonstrated a 44% reduction in the risk for death among women with HER2/neu-positive disease who received trastuzumab compared with women with HER2/neu-negative breast cancer (P < .0001, hazard ratio [HR] = 0.56). The major toxicity associated with trastuzumab therapy is cardiac dysfunction. Most of the cardiac toxicity seen with treatment is limited to asymptomatic decreases in the left ventricular ejection fraction (LVEF); however, severe congestive heart failure will occur in approximately 4% of patients.^[14] In cases where left ventricular dysfunction is asymptomatic, most studies have withheld treatment if the LVEF decreased by more than 15% of baseline or fell below 40%-45%.^[15] It should be noted that observed cardiac toxicity is almost always reversible with discontinuation of the drug, and the drug can often be restarted once left ventricular function has been restored.

Cetuximab is a humanized monoclonal antibody that binds to the extracellular domain of EGFR. Similar to trastuzumab, the antitumor effects of cetuximab include blocking EGFR phosphorylation and activation, inhibition of angiogenesis and invasion, activation of proapoptotic proteins, and ADCC.^[16] Cetuximab is approved for use in the treatment of metastatic colorectal cancer as well as head and neck squamous cell carcinoma. Cetuximab combined with standard chemotherapy has been shown to improve PFS among patients with metastatic colon cancer. In a study of almost 2000 patients randomized to receive either cetuximab and FOLFIRI (irinotecan, fluorouracil, and leucovorin) or FOLFIRI alone as primary therapy,^[17] the HR for PFS for the cetuximab-treated group was 0.85 (P = .048). The benefit was more pronounced among patients with wild-type KRAS mutation. Indeed, multiple studies have shown that patients who have a KRAS mutation detected in codon 12 or 13 do not benefit from anti-EGFR monoclonal antibody therapy.^[18] For this reason, KRAS mutation testing is recommended as part of the evaluation of patients under consideration for anti-EGFR monoclonal antibody therapy.^[18] For the treatment of locally advanced head and neck cancer, the addition of cetuximab to radiation therapy has improved both local-regional control and overall survival.^[19] The major toxicity associated with cetuximab therapy is a severe acneiform rash — although research has shown that the development of a rash is associated with improved survival with the use of cetuximab.^[20] A follicular eruption will occur in about 90% of patients treated with cetuximab, and the reaction will be severe (Grade 3-4) in about 15% of cases.^[21] Low-grade skin toxicity is effectively treated with topical therapy, while more significant skin reaction may require drug discontinuation until it resolves.^[22] What type of impact intermittent discontinuation of the drug may have on clinical outcome is not known at this time.

Bevacizumab is a humanized monoclonal antibody that binds VEGF and inhibits the development of angiogenesis. Bevacizumab has been studied extensively in and is approved for the treatment of non-small-cell lung cancer (NSCLC), colon cancer, and breast cancer. In 2006, a study of almost 900 patients was reported evaluating bevacizumab with or without paclitaxel for the treatment of Stage IIIB or IV NSCLC.^[23] Both median survival and PFS were improved (P = .003 and P < .001, respectively) in the arm that received bevacizumab. A recent meta-analysis^[24] of over 2000 patients evaluating the use of bevacizumab in unresectable NSCLC found that high doses of the agent were associated with improved 2-year overall survival and PFS, though at an increased risk for treatment-related death. Low-dose bevacizumab was associated with improved PFS and no greater risk for treatment-related death. Bevacizumab, when combined with irinotecan, fluorouracil, and leucovorin, has also demonstrated a survival benefit in patients with metastatic colorectal cancer.^[25] Finally, a randomized study of over 700 patients with metastatic breast cancer demonstrated that bevacizumab, given in combination with paclitaxel, improved PFS, but not overall survival, compared with paclitaxel alone.^[26] Grade 3-4 hypertension, seen in almost 15% of patients, was the most common serious side effect in the study — and, in fact, the major toxicity associated with bevacizumab treatment is the development of hypertension. It is presumably due to the effect of the drug on the vasculature. Bevacizumab should be used with caution in patients with a history of hypertension and not used in patients with uncontrolled hypertension. The drug is also associated with an increased risk for thromboembolic events. A recent study^[27] evaluated the impact of the development of hypertension on clinical outcome in patients receiving bevacizumab with chemotherapy in the context of a large randomized trial. Both overall survival and PFS were improved in those patients who developed hypertension after starting treatment.

It is unclear how much of the success of monoclonal antibody therapy is due to the initiation of ADCC and the subsequent generation of an adaptive immune response. It is presumed that the toxicities that predict improved outcomes that have been seen with the use of specific monoclonal antibodies are a surrogate for successful tissue targeting of these agents.

Vaccines

Cancer vaccines are a form of active immunotherapy. Most vaccines are constructed to stimulate APCs to present specific antigens to T cells, which will generate either a robust endogenous T-cell response and CTLs or a B-cell response with resultant neutralizing antibodies. In the case of vaccines, the patient’s immune system must have the capability of responding to antigen-specific stimulation. A variety of cancer vaccines have been tested for both prophylactic and therapeutic indications. Multiple studies, enrolling hundreds of patients, have demonstrated that vaccines targeting cancer — even those targeting self proteins — are associated with rare toxicity; the most common are injection site reactions due to the immunologic adjuvants that are often included in the vaccine to assist in boosting the immune response.

There are several vaccines that are used today to prevent the development of cancer. These vaccines target viral antigens. The chronic infections induced by these pathogens are associated with the development of malignancy. Vaccines constructed for the prevention of viral infections, such as HBV and HPV, are designed to generate neutralizing antibodies that will clear the infectious pathogen. HBV vaccines have been routinely administered since the 1980s, and recent reports suggest that the prevention of the development of HBV chronic carriers via the prevention of infection is having an impact on the incidence of hepatocellular carcinoma (HCC) in endemic areas. In a 20-year follow-up study among individuals in Taiwan, the rate of HCC was significantly lower (relative risk of 0.31, P < .001) among patients who had been vaccinated compared with those who were not.^[28] Similarly, a recently approved vaccine for the HPV virus has shown great efficacy in preventing the development of cervical cancer. A double-blind study^[29] of an HPV type-16 vaccine was conducted in over 2000 young women. There was 100% efficacy in preventing the development of HPV infection, while all cases of cervical intraepithelial neoplasia (CIN) occurred in the control arm. HPV vaccines have shown efficacy in preventing other HPV-associated cancers as well. A quadrivalent (HPV 6, 11, 16, and 18) vaccine was effective in preventing vulvar intraepithelial neoplasia (VIN) in most patients enrolled in a large randomized clinical trial.^[30]

Unfortunately, once patients have already developed infection and/or preinvasive lesions, the ability of a vaccine to induce disease resolution is limited.^[31] The inability to clear disease is most likely due to the type of immune response generated with these vaccines. B cell-based antibody production is most commonly induced with immunization. Once malignant disease has developed, a CTL/Th cell immune response is needed for eradication. These data suggest that prophylactic cancer vaccines have and will play an important role in the prevention of human malignancy. Although the antigens to target with vaccination for infectious disease are well established, antigens that would give broad coverage for prevention of malignancies such as colon and breast cancer have not yet been identified.

No cancer vaccine is currently approved to either treat cancer or to prevent disease recurrence, although vaccines have been tested in patients with advanced stage malignancy for decades. A review of over 400 patients with melanoma, immunized with a variety of vaccines targeting the disease, estimated that overall response rates range from 2% to 9%.^[32] Still, as a better understanding of T-cell stimulation and function has evolved over the last 10 years, newer vaccine technologies have been developed that are much more successful in eliciting CTL/Th cell immunity than previous platforms. For this reason, recent studies are beginning to suggest that cancer vaccines may have an impact on disease outcome. Clinical response appears to be related to the level of immunity achieved. In a study of 20 women with grade 3 VIN,^[33] the use of a long synthetic peptide vaccine targeting HPV-16 was correlated with a clinical response in 79% of patients (47% complete responses) at 12 months of follow-up. Moreover, patients with complete responses had significantly higher levels of HPV-specific IFN-gamma producing T cells than those who did not achieve a response. Phase 2 clinical trials of vaccines targeting breast (HER2/neu peptide vaccine) and prostate (prostate-specific antigen vaccine) cancer as well as melanoma (gp100/MART1/tyrosinase peptide vaccine) have suggested that a survival benefit can be achieved with active immunization and that the magnitude of the IFN-gamma antigen-specific T-cell response obtained is associated with survival.^[34-37]

In the last year, phase 3 studies of cancer vaccines targeting a variety of malignancies have reported positive results. An integrated analysis of 2 randomized phase 3 studies evaluating the therapeutic efficacy of a vaccine targeting prostatic acid phosphatase (PAP) in 225 patients with hormone-refractory prostate cancer demonstrated a survival benefit with vaccination.^[38] Patients randomized to the vaccine arm experienced a 33% reduction in the risk for death (P = .011, HR = 1.5) compared with those receiving a placebo. The vaccine, a cell-based vaccine consisting of autologous APCs loaded with a PAP-GM-CSF fusion protein, was well tolerated with few side effects. An additional randomized phase 3 study of high-dose IL-2 with or without a peptide-based vaccine targeting the gp100 melanoma tumor antigen demonstrated survival benefit in 185 patients with advanced stage melanoma after vaccination.^[39] PFS in the vaccinated arm was 2.9 months compared with 1.6 months in the control arm (P = .010), while overall survival had a trend toward significance (17.6 months vs 12.8 in the control; P = .096). It is notable that both of these studies were performed in patients with end-stage metastatic disease that was no longer responding to standard therapy and that clinical benefit was still observed, considering that active immunization has proven to have the greatest benefit in the prophylactic setting.

Ultimately, therapeutic vaccination may be most effective when used in the adjuvant setting to prevent disease recurrence. This was the rationale behind a phase 3 study designed to evaluate the benefit of immunization against idiotype antigens expressed on follicular lymphoma.^[40] Follicular lymphoma expresses unique surface immunoglobulin, and the idiotype determinants of those immunoglobulins have been shown to be immunogenic.^[41] In the phase 3 study,^[40] patients with follicular lymphoma, who had achieved a complete clinical response with standard therapy, were randomized to receive an idiotype vaccine vs control immunization. At a median follow-up of almost 57 months, the median disease-free survival was 44.2 months in the vaccine arm compared with 30.6 months in the control arm (P = .045, HR = 1.6). Over the next few years, many more vaccines will be progressing to phase 3 clinical trials, and the likelihood of the development of a variety of vaccines to aid in the prevention of cancer relapse is high.

Adoptive T-Cell Therapy

The infusion of tumor competent T cells into a patient is a form of passive immunotherapy. Once the T cells are stimulated by antigen in vivo, they will proliferate and induce tumor eradication. If both Th cells and CTLs are infused, there is the potential for the infused cells to persist as memory T cells, which could result in a prolonged antitumor effect related to the initial infusion. The mechanism of action of T-cell transfer depends on the infused cancer-specific CTL-killing tumor cells expressing antigen and the Th cells supporting sustained proliferation and function of the CTLs. Adoptive T-cell therapy is most widely used in the HSCT setting for relapsed hematologic malignancy as a donor lymphocyte infusion (DLI), particularly for CML. DLI is the collection and infusion of peripheral blood mononuclear cells from the transplant donor into the transplant recipient after the transplant has failed. The treatment has been shown to reinduce a remission in some patients, most likely due to the infused T cells recognizing and responding to minor histocompatibility antigens (MHAs) expressed on the surface of the recipient’s relapsed tumor cells, a phenomenon called graft vs tumor effect. MHAs are recognized as foreign to the donor T cells and stimulate a destructive immune response. Unfortunately, MHAs are also expressed on normal tissues in the recipient, resulting in a higher risk for graft vs host disease (GVHD) in patients who receive DLI as part of their treatment.

DLI has shown the greatest therapeutic efficacy in the treatment of CML that has relapsed after HSCT. After DLI in this setting, a 3-year survival in more than 60% of patients can be obtained.^[42] The results with DLI are less impressive in other hematologic malignancies. In acute myeloid leukemia (AML), multiple myeloma, and lymphoma, survival associated with DLI after HSCT relapse is well below 50%.^[42] In a recent evaluation of over 500 AML patients in first hematologic relapse after HSCT, those who received DLI as part of their treatment for relapse had an estimated survival at 2 years of 21% compared with 9% for those who did not receive DLI.^[43] Of note, the time to achieve remission after DLI is quite prolonged compared with the time to response observed after administration of a cytotoxic chemotherapy. A study of CD4-positive DLI in 19 patients with relapsed CML demonstrated a 13-week (range, 9-30 weeks) time to cytogenetic response from the time of the first DLI and a 34-week (range, 11-56 weeks) time to molecular remission.^[44] Presumably, the delayed time to achieve maximal response or remission is due to the need for in vivo proliferation of T cells to achieve a level of magnitude that is therapeutically effective. Most likely, this is the reason that DLI is much more effective in cases with minimal residual disease rather than overt relapse: the immune response has time to overwhelm the cancer, whereas in frank relapse there is not time for T-cell expansion before the leukemia has grown out of control.

As stated, the major toxicity after DLI is the development of GVHD. In a study of over 300 patients who received DLI for CML that relapsed after HSCT, 38% of patients developed either acute or chronic GVHD.^[45] Risk factors for developing GVHD included an initial cell dose greater than 10⁷, DLI administered within 2 years of the initial transplant, and DLI derived from unrelated donors. Indeed, the risk for death was 2.3-fold greater in patients who underwent DLI and developed GVHD than in patients who did not develop GVHD after DLI. There are several approaches — including the use of low-dose immunosuppression prior to DLI or the purification and infusion of T cells specific for MHA — being studied to minimize the development of GVHD after DLI in an effort to improve overall survival rates.^[46,47]

In large part due to the success seen with DLI in inducing remissions in refractory hematologic malignancy, adoptive T-cell therapy is being studied as a treatment in several tumor types. EBV-mediated malignancies have been particularly amenable to the evaluation of T-cell transfer. Patients have a vigorous endogenous immune response to EBV because it is a foreign virus, and T cells are more easily collected and expanded ex vivo. Further, EBV-specific T cells can be of high avidity and easily demonstrate lytic activity in culture. Finally, EBV is directly implicated in the malignant transformation. A phase 2 study of the infusion of allogeneic EBV-specific T cells for the treatment of EBV-positive posttransplant lymphoproliferative disease demonstrated a 52% response rate at 6 months with minimal toxicity in 33 patients.^[48] The infusion of autologous EBV-specific CTLs was associated with a 20% response rate in patients with stage IV nasopharyngeal carcinoma (n=10).^[49] Infusions were well tolerated; notably, 2 patients developed inflammatory responses at the site of the tumor after T-cell infusion, indicating the cells were capable of homing and inducing the generation of significant inflammation.

In solid tumors, adoptive T-cell therapy has been most widely studied for the treatment of advanced stage melanoma. Early trials that infused nonspecifically expanded T cells derived from the tumor (ie, tumor infiltrating T cells or TIL) demonstrated minimal response rates.^[50] Lymphodepletion prior to T-cell infusion, however, has increased responses dramatically. Studies have shown that pretreatment with lymphodepletion regimens such as radiation, cyclophosphamide, and/or fludarabine prior to T-cell infusion removes cells that consume essential cytokines needed for in vivo T-cell expansion such as IL-7 and IL-15.^[51] The addition of myeloablative preparative regimens has increased the response rate of adoptive T-cell therapy in metastatic melanoma patients to as high as 70% but with significant added toxicity.^[52] The durability of the clinical response is variable with few long-term survivors. To optimize response rates and the number of durable responders, investigators are evaluating the use of:

• Highly purified T-cell populations, especially those T cells responsible for maintaining immunologic memory;
• Novel targeted pretreatment methods that may improve T-cell proliferation in vivo with decreased toxicity; and
• The development of engineered T cells that demonstrate high avidity for antigens expressed by the tumor and may, therefore, be more efficient killers.^[53-55]

Similar to DLI, adoptive T-cell transfer for solid tumors may have more sustained efficacy if used in minimal residual disease states rather than in refractory established tumors.

Immunomodulators

There are several categories of immunomodulators: cytokines, modulators of innate immunity, and, identified more recently, agents targeting specific immunosuppressive pathways. Cytokines are biologic proteins that are naturally secreted by immune system cells to initiate and sustain an immune response. Recombinant cytokines have been manufactured, evaluated in clinical trials, and have shown some benefit in the treatment of diseases such as melanoma, renal cell carcinoma, and CML, where they are approved for clinical use. Over the last few years, several novel targeted agents such as kinase inhibitors and mammalian target of rapamycin (mTor) inhibitors have shown much greater therapeutic efficacy in the treatment of these diseases than cytokines such as IL-2 and IFN-alfa 2b. For this reason, cytokines are much less commonly used for the treatment of cancer now than they were a decade ago.

IL-2 is approved for use in the treatment of metastatic renal cell carcinoma and melanoma, where the drug is associated with about a 10% response rate and less than 2% durable responses. The mechanism of action of IL-2 is unknown, but the cytokine significantly stimulates the proliferation and activation of both T cells and NK cells which are thought to have lytic activity against tumors. In renal cell carcinoma, response to IL-2 is most likely to be associated with a good performance status, minimal sites of metastatic disease and no liver metastasis, as well as normal neutrophil counts.^[56] In melanoma, high-dose IL-2 results in a response in a small minority of patients; the response rate improves when it is combined with chemotherapy, an approach called “biochemotherapy,” but there is no benefit to overall survival.^[57] Clinical response to IL-2 in melanoma has been associated with having only cutaneous metastasis as a site of metastatic disease and the development of lymphocytosis after treatment.^[58] The approved doses for IL-2 for treatment of advanced stage melanoma are quite high, 600,000 IU/kg; lower doses have not shown as great a clinical benefit as high-dose therapy. However, side effects are significant at this dose and include vascular leak syndrome often requiring systemic blood pressure support, diuresis, and, at times, intubation for respiratory support after the development of pulmonary edema. Along with expanding effector T cells, IL-2 has also recently been shown to expand the regulatory T-cell populations that inhibit the function of effector T cells such as CTLs.^[59] It may be that the minimal efficacy of IL-2 in inducing remissions is related to the lack of specificity of the cytokine to antitumor effector cells.

IFN-alfa 2b is the only approved treatment for the adjuvant therapy of melanoma associated with a high risk for relapse. The drug is also used in the treatment of renal cell carcinoma, CML, and other myeloproliferative disorders. The mechanisms of action of IFN in mediating tumor regressions are multifactorial. IFN has been shown to be antiproliferative and antiangiogenic, and to induce the apoptosis of tumor cells and regulate the function of T cells.^[60] The use of IFN-alfa 2b is controversial in melanoma: some studies have shown survival benefit in the adjuvant setting whereas others have not duplicated this result. An Eastern Cooperative Oncology Group study randomizing almost 300 patients to receive IFN vs observation demonstrated an improvement in both relapse-free survival (P = .0023) and overall survival (P = .0237), but a pooled analysis of multiple studies by the same group showed only an improvement in relapse-free survival (P = .006), not overall survival.^[61,62] IFN treatment is toxic, with common side effects including flu-like symptoms that can be debilitating, gastrointestinal disorders including hepatotoxicity, arthralgias, and neuropsychiatric symptoms. Because the success of the therapy depends on patient compliance to the treatment, symptom management is key; guidelines have been published on the medical management of IFN toxicity.^[63] Lower doses of IFN are less toxic but do not improve survival rates.^[64]

Immunomodulation with intravesicular BCG has been a mainstay of the treatment of early bladder cancers. BCG is a mycobacterial cell wall preparation that potentially functions by stimulating cells of the innate immune system via TLRs. Studies have shown that BCG will activate APCs such as dendritic cells and induce these cells to secrete cytokines that will elicit an inflammatory adaptive T-cell response.^[65] A recent systemic review of randomized trials and meta-analyses demonstrated that intravesicular BCG given after transurethral resection for superficial bladder cancer reduced the risk for recurrence by almost 70% compared with surgery alone.^[66] Complications associated with the use of BCG include urinary frequency, cystitis, fever, and hematuria.

A newer class of immunomodulators are antibodies that have been designed to block proteins that inhibit the generation of an immune response or engage receptors that stimulate immunity. One of the first in class is anticytotoxic T lymphocyte antigen 4 antibody (CTLA-4). CTLA-4 is a protein expressed by T cells that will bind receptors on APCs, turning off a T-cell response. The CTLA-4 signaling pathway is a major mechanism of self-regulation of T cells whereby immunologic tolerance is maintained. Recent studies in murine models demonstrate that anti-CTLA-4 can both directly enhance T-cell effector function as well as block T regulatory cell activity.^[67] A published phase 2 study of 56 patients with progressive metastatic melanoma recorded an overall response rate to anti-CTLA 4 of 13%.^[68] Treatment was associated with significant toxicity, mostly autoimmune phenomena such as colitis, vitiligo, and hypophysitis. Investigators demonstrated that grade 3 and 4 autoimmune toxicity was associated with a higher clinical response (P = .008).^[68] Another antibody in clinical trials is anti-CD137. CD137 is expressed on T cells; when the antibody binds its ligand, T cells will activate and proliferate. Because these approaches are not specific to an antigen and have been shown to be general immune stimulators, the issue of toxicity is a concern.

Conclusions

There are several classes of immune-based cancer treatment that are currently used in the clinic. As technology and our understanding of how the immune system recognizes and responds to cancer have advanced, so has the development of new immune-based cancer therapeutics. We should expect to see immunotherapy further integrated at all stages of cancer prevention and treatment within the next decade.

Supported by an independent educational grant from Bristol-Myers Squibb.

This article is a CME certified activity. To earn credit for this activity visit:
http://cme.medscape.com/viewarticle/720825

Authors and Disclosures

As an organization accredited by the ACCME, Medscape, LLC, requires everyone who is in a position to control the content of an education activity to disclose all relevant financial relationships with any commercial interest. The ACCME defines “relevant financial relationships” as financial relationships in any amount, occurring within the past 12 months, including financial relationships of a spouse or life partner, that could create a conflict of interest.

Medscape, LLC, encourages Authors to identify investigational products or off-label uses of products regulated by the US Food and Drug Administration, at first mention and where appropriate in the content.

Author(s)

Mary L. Disis, MD

Associate Professor of Medicine, University of Washington, Seattle, Washington Disclosure: Mary L. “Nora” Disis, MD, has disclosed the following relevant financial relationships: Received grants for clinical research from: Hemispherx Biopharma; GlaxoSmithKline Served as an advisor or consultant for: VentiRx Pharmaceuticals, Inc. Dr. Disis does not intend to discuss off-label uses of drugs, mechanical devices, biologics, or diagnostics not approved by the US Food and Drug Administration (FDA) for use in the United States. Dr. Disis does intend to discuss investigational drugs, mechanical devices, biologics, or diagnostics not approved by the FDA for use in the United States.