This article reviews up-to-date clinical studies combining chemotherapy with immunotherapy; it also explains the rationale for this combination in treatment of pancreatic cancer.
Pancreatic cancer has one of the poorest prognoses out of all human malignancies. Only 20% of patients who present with this disease have surgically resectable tumors at the time of diagnosis, and of these 20% of patients, the vast majority will experience recurrence within the first 2 years. Chemotherapy is the mainstay of treatment. Unfortunately, little improvement has been made in the past few decades with regard to overall survival (OS) with chemotherapy alone. Of note, the immune system’s involvement in cancer development and progression has sparked much interest in recent years. The model of the cancer-immunity cycle suggests an interplay of immune-suppression and immune-stimulation. In normal individuals, a state of immunosurveillance is in place. However, within the tumor microenvironment, inhibitory signals and immunosuppressive cells are present and tip the scale in favor of immune suppression. This enables cancer cells to evade the immune system and allows tumors to grow. Studies in a variety of solid tumors have shown that many chemotherapies, including those used as standards of care in pancreatic cancer, have immune-modulating functions, including inhibition of immune suppression and stimulation of immune function. This article reviews up-to-date clinical studies combining chemotherapy with immunotherapy; it also explains the rationale for this combination in treatment of pancreatic cancer.
Pancreatic ductal adenocarcinoma (PDA) has one of the most dismal prognoses out of all human malignancies, currently ranking as the fourth most common cause of cancer-related death in the United States.1 Although advances have been made in both understanding the molecular biology of the disease as well as its imaging, the current 5-year survival rate for all stages combined is only 6%, which is one of the lowest out of all cancer diagnoses.1-2 Currently, the only hope of cure for PDA is surgical resection.
However, even after an R0 resection with node-negative disease, the long-term survival rate is lower than 10%.2 This is due to a high recurrence rate, with 80% of those patients experiencing recurrence within 2 years of surgical resection. Over the last few decades, new chemotherapies have been discovered; still however, the effect on OS in patients with metastatic tumors has been minimal. For example, in 1997 gemcitabine was shown to improve the median OS by 1.2 months when compared with fluorouracil.3 More recently, 2 sets of chemotherapy combinations have also been shown to improve OS in this disease. The chemotherapy cocktail of fluorouracil, oxaliplatin, irinotecan and the vitamin leucovorin (FOLFIRINOX) used in the first-line setting was shown to improve OS to 11.1 months when compared with gemcitabine alone.4
In October 2013, nab-paclitaxel was shown to achieve an OS of 8.5 months compared with 6.7 months achieved with gemcitabine alone.5 The fact remains that, although we have seen advances in the treatment of this disease, the clinical effect on the overall prognosis has been slight, with 5-year survival essentially unaffected. In addition, the trial that achieved the greatest improvement in OS with FOLFIRINOX was accompanied by substantial toxicities and, therefore, is only recommended for robust patients with an Eastern Cooperative Oncology Group (ECOG) performance status of 0 to 1.4 The meager improvement in OS is thought to be due to the disease’s resistance to both chemotherapy and targeted therapy.6
Today, many studies have focused on the immune system’s role in cancer and how it can recognize both cancer cells and tumor-associated antigens. Therefore, many investigators have turned their focus to immunotherapy as a potential treatment for PDA.Chen and Mellman have delineated the cancer-immunity cycle, which depicts the immune system’s role in controlling tumor growth in normal individuals. Understanding this cycle provides insight into how tumors can evade it. As cancer develops, oncogenic proteins are produced. In the normal individual, these cancer-associated antigens are released into the microenvironment during necrotic or immunogenic cell death. The antigens are captured by antigen-presenting cells (APCs), such as dendritic cells, processed into peptides, and presented on the cell surface bound to major histocompatibility complex I (MHCI) molecules.
Next, the APCs recruit T cells by the release of proinflammatory cytokines such as TNF-alpha, IL-1, and IFN-alpha. The binding of the T cells to the MHCI/ antigen complex on the APC is the first step of T-cell priming and activation. Following activation, T-effector cells travel to the tumor site, infiltrate the tumor and, ultimately, the T-cell receptor binds to a tumor cell and kills tumor cells via the cytotoxic T-cell response (CTR). At this point, the cycle restarts as the immunogenic death releases more cancer cell antigens; therefore, the cycle propagates itself and potentiates its own response (Figure 1).7
The idea of the cancer-immunity cycle proposes that, for a cancer immune response to be generated, the net balance between immune stimulation versus immune suppression must be tipped in favor of the former. Studies in various cancers have suggested that tumors evade the immunogenic process mostly by factors that promote immunosuppression. In pancreatic cancer, these evasion mechanisms include: 1) inhibition of antigen presentation by downregulation of HLA/ tumor antigen complexes on tumor cells8;2) upregulation of immune cell inhibitors, such as programmed death ligand (PD-L1), upon which binding to the programmed death (PD- 1) receptor on T cells results in T cell apoptosis and immunosuppressive cytokine release9; 3) receptor inhibition of costimulatory interactions between CD28 and B7 proteins by CTLA-4, which is further suggested by CTLA-4 blockade slowing tumor growth10; and 4) infiltration of tumors by leukocytes that promote immune suppression and immune tolerance such as regulator T cells (Tregs) and myeloid-derived suppressor cells (MDSCs) (Figure 2).6 Of particular interest, from a medical oncology standpoint, are the effects of chemotherapy on the cancer-immunity cycle and, ultimately, how chemotherapies may be used in treatment to synergize with immunotherapies.Two general ways chemotherapy may affect the cancer-immunity cycle are improving immune stimulation and inhibiting immune suppression. On the immune stimulation side of the equation, some chemotherapies may improve the immune system’s recognition of malignant cells. For example, gemcitabine was shown to increase the spectrum of epitopes that are presented on the surfaces of mesothelioma cells.11
Figure 1. Neoplastic growth control via normal immunosurveillance and the effects of various chemotherapies. The cancer immunity cycle starting at cancer cell death with release of antigens into the surrounding microenvironment. The antigens are captured by APCs, processed and presented on the APC surface. The APCs, with antigens on their surface, then prime and activate T cells. T cells then move to the site of the tumor and are now “programmed” to bind to and kill tumor cells. Certain chemotherapies potentiate the immune response (green arrows).
Furthermore, gemcitabine and oxaliplatin have been shown to upregulate HLA molecules on tumor cells.12 These molecular mechanisms increase the chances of activated T cells recognizing malignant cells. In addition to this mechanism, some chemotherapies have been shown to improve the ability of the tumor cells to induce an immune response, essentially improving the immunogenicity. For example, oxaliplatin has been shown to upregulate cancer death associated molecules (CDAMs) on the surfaces of colon cancer cells.13-14
CDAMs essentially tag dying tumor cells for phagocytosis. Additionally, some chemotherapies can also induce tumor cells to secrete chemical signals that promote dendritic cell maturation. For example, when lung cancer cells are treated with gemcitabine or oxaliplatin in vitro, the resulting supernatant has the capacity to induce dendritic cell maturation.12 Cyclophosphamide was also shown to induce dendritic cell differentiation as well as release CDAMs in various cancer cell lines in vitro.15 Furthermore, some chemotherapies upregulate the expression of costimulatory molecules on malignant cells, such as CD80/B7-1, which is required for generating an immune response.16
Additional examples of chemotherapy improving the immune response include increasing the permeability of the cell to cytotoxic granzyme B and upregulating antigen presenting complexes on dendritic cells, both of which are promoted by paclitaxel.17-18 Finally, as a class, taxanes can increase the production and local release of cytokines that stimulate T-cell priming and activation, such as IL-1 and TNFalpha.19 All of these are examples of how chemotherapies can augment the immunogenicity of malignant cells.20 The other side of the equation addresses immune suppression. Studies in a variety of cancers have shown that many tumors are infiltrated with immune-suppressive cells. In a pancreatic cancer mouse model, for example, it has been shown that these tumors are infiltrated with suppressive regulatory T cells (Tregs) and MDSCs (Figure 2).
Treatment of these mice with single agent gemcitabine decreased immunosuppressive MDSCs. Additionally, tumor regression and improvement in OS were seen in mice treated with gemcitabine in combination with a dendritic cell vaccine. These results showed that hindering the tumor’s endogenous immunosuppressive function, in turn, allowed the dendritic cell vaccine to propagate its immunogenic function. The same study showed that pancreatic cancer patients had higher levels of MDSCs in their peripheral blood compared with healthy donors and these levels were significantly decreased in the majority of these patients after 1 cycle of single agent gemcitabine. In light of these findings, further studies are being pursued in human clinical trials.Clinical trials testing pancreatic cancer vaccines started in the late 1990s (Table).22 GVAX was the first pancreatic cancer whole cell vaccine that was combined with chemotherapy (cyclophosphamide).23 The GVAX vaccine consists of pancreatic cancer cells genetically modified to secrete granulocyte macrophage colony stimulating factor (GM-CSF). Cyclophosphamide is thought to promote immune stimulation by promoting dendritic cell maturation and upregulating HLA molecules on tumor cells. Cyclophosphamide is also thought to negatively regulate the immune suppressive functions of Tregs. Therefore, the rationale of this combination is quite clear as pancreatic cancer cells secreting GM-CSF is immune stimulating and cyclophosphamide both stimulates immunogenicity and relieves immune suppression. It should be noted that this study was the first vaccine study in patients with metastatic pancreatic cancer. Importantly, the median survival more than doubled when the vaccine was combined with cyclophosphamide compared with the vaccine arm alone (4.7 months vs 2.3 months).
Figure 2. Tumor microenvironment. Many solid tumors are infiltrated by immunosuppressive cells, such as MDSCs and Tregs. Certain chemotherapies inhibit these suppressive functions.
As previously mentioned, single-agent gemcitabine has long since been the first-line standard of care for pancreatic cancer across all stages. Gemcitabine has been found to have multiple effects on the cancer-immunity cycle. It improves immune stimulation by inducing dendritic cell maturation, increasing epitope presentation on tumor cells, and decreasing tumor infiltrative MDSCs.11,12,21 These studies show that gemcitabine also augments the immune response in favor of generating immunity. Therefore, the data lay the foundation for investigating combinations of this chemotherapy with immunotherapy.
The first example of such a combination is a phase I study in Japan investigating the combination of single-agent gemcitabine with a vascular endothelial growth factor receptor 2 (VEGFR2) peptide vaccine. VEGFR2 is highly expressed in the new vasculature of tumors, but not in normal blood vessels. The rationale for this study was to utilize the immune modulating characteristics of gemcitabine to generate an immune response to the VEGFR2 peptide. Twelve of the 18 patients enrolled in the study had either a partial response or stable disease and the median OS was 7.7 months, which was greater than treatment with gemcitabine alone.24 Two other studies have also investigated peptide vaccines, 1 using a personalized peptide and the other using the Wilms tumor peptide. The personalized peptide showed a median survival of 9 months when combined with gemcitabine, which is an improvement over chemotherapy alone.25-26
The algenpantucel-L hyperacute pancreas vaccine trials are currently investigating the benefit of adding a pancreatic cancer vaccine to the standard adjuvant treatment chemotherapy and chemoradiotherapy (known as the RTOG-9704 standard) for resected pancreatic cancer. The algenpanctucel-L vaccine consists of human pancreatic cancer cells that express α-gal epitopes, which are not present on human cells; however, they are present on cells of most other mammalian species. Bacterial gut flora within the human gastrointestinal tract continuously stimulate production of antibodies to α-gal, as the epitopes are frequently introduced through the human diet. Upon administration of the vaccine, the human immune system already has humoral immunity in place and antibodies recognize the α-gal epitopes on the vaccinated pancreatic cancer cells. These antibodies then fix complement and activate complement mediated lysis, starting the cancer-immunity cycle.27 Phase II data from this study showed an added benefit when combining algenpantucel-L with the RTOG-9704 standard adjuvant chemotherapy; 12-month disease-free survival was 62% with the vaccine compared with less than 50% in the initial study of the adjuvant standard of care (RTOG-9704).
Laheru et al, 2008
GVAX alone, MS = 2.3 mo
GVAX + Cy, MS = 4.3 mo
12 mo DFS = 62%
Median DFS = 14.1%
Peptide Miyazawa, 2010
1/18 = PR; 6/18 = PD
6/18 = SD
Median OS = 9 mo
Median TTP = 7 mo
PR = 33%
SD = 43%
PD = 24%
Wilms Tumor Peptide
2 mo DCR = 89%
MS = 259 days
FOLFIRINOX Gemcitabine + nab-paclitaxel
Cy indicates cyclophosphamide; DCR, disease control rate; DFS, disease-free survival; MS, median survival; OS, overall survival; PD, progressive disease; PFS, progression-free survival; PR, partial response; SD, stable disease; TTP, time to progression.
The median disease-free survival was 14.1 months with the vaccine compared with 11.4 months with RTOG-9704. The 12-month OS was 86% with the vaccine compared with 69% in the standard adjuvant study. Phase III study of algenpantucel- L is still ongoing and will enable appropriate comparison to the standard of care. In the algenpantucel-L phase II study, 3 patients who had recurrence after immunotherapy had a resultant complete radiographic response to subsequent chemotherapy. This raises the possibility of increased chemosensitivity following immunotherapy.28
As previously mentioned, the recent advances in pancreatic cancer include the combination of gemcitabine and nab-paclitaxel (Abraxane) or the chemotherapy cocktail FOLFIRINOX as first-line treatment, depending on performance status, in stage 4 disease. Either combination is used at various cancer centers in the setting of locally advanced disease. Pertaining to the gemcitabine/nab-paclitaxel combination, we have already discussed the effects of gemcitabine on immunogenicity and immune suppression. Again, taxanes augment the cancerimmunity cycle by improving immunogenicity. Furthermore, paclitaxel reduces immune suppression by antagonizing Treg functions.20
Similarly, the chemotherapies that make up FOLFIRINOX augment the immune system both by improving immunogenicity as well as prohibiting immunosuppression. With these immune-modulating functions, another hyperacute pancreas vaccine study with algenpantucel-L is presently under way for locally advanced pancreatic cancer. This study has recently opened and has several arms including gemcitabine and nab-paclitaxel alone, FOLFIRINOX alone, and each standard chemotherapy in combination with the vaccine. Another immunotherapy option that has been explored is agonist monoclonal antibodies to CD40.
CD40 is a receptor of the tumor necrosis factor (TNF) family and is usually expressed on B cells, dendritic cells, and monocytes. Its ligand, CD40 ligand, is mainly expressed on activated T cells. Binding of CD40 ligand to the CD40 receptor is essentially immune-stimulating, resulting in the upregulation of MHC and costimulatory molecules, the release of proinflammatory cytokines and the further enhancement of T-cell activation. Vonderheide and colleagues29 investigated the stimulation of the CD40 receptor using an agonist monoclonal antibody in combination with gemcitabine.
They found that stimulating CD40 in combination with gemcitabine induced tumor regression in 80% of the mice, whereas none of the mice treated with gemcitabine alone saw tumor regression. Further studies showed the CD40 stimulated tumor regression was controlled by both T-cell dependent and T-cell independent mechanisms.29 Beatty and colleagues then went on to test this phenomenon in humans in a phase I clinical trial investigating agonist CD40 monoclonal antibody, CP-870,893, combined with gemcitabine in pancreatic cancer patients. In this study 10 patients were in the mean tolerated dose expansion cohort. Eight of these 10 received at least 2 cycles of chemotherapy.
Two of the 8 achieved a partial metabolic response as evidenced by >25% reduction of standard uptake value (SUV) on positron emission tomography (PET) scan. They also found an average SUV reduction of 26.9% within the primary lesion and all patients did have some reduction.30The cancer-immunity cycle is an ideal model to envision how tumor cells evade immunosurveillance as well as where future modalities may intervene with hopes of potentiating tumor cell death. The cancer-immunity cycle together with the immunemodulating functions of chemotherapies that are used in pancreatic cancer creates a rationale for investigating vaccinechemotherapy combinations.
Studies to date have suggested benefits of adding immunotherapies to standard chemotherapy regimens. Additional benefits are also suggested by the indication that immunotherapy may render improved chemosensitivity at later dates. In addition, vaccines are often well tolerated with minimal toxicities, which make them a favorable approach. The hope is that we can identify the appropriate combination of vaccine and immune-modulating chemotherapy that will eradicate the disease. There is also likely to be a role for immune checkpoint therapy with inhibitors of PD-1 and PD-L1. Such phase I single-agent studies are currently in progress for pancreatic cancer. The results of studies so far create hope that the combination of chemotherapy with immunotherapy may be a game changer in the treatment of pancreatic cancer.
The path to integrating immunotherapy with chemotherapy will likely be bumpy and afflicted by setbacks, many of which we have already seen. The only hope for changing the paradigm of treating pancreatic cancer is to maintain persistent clinical trials. Any pancreatic cancer patient should be presented with an opportunity to enroll in a trial, or should be referred to a center that offers them. These patients should be given a chance at tomorrow’s drugs as well as an opportunity to contribute to advancing the treatment of this disease.
ABOUT THE AUTHORS
Affiliations: Kara D. Forinash, MS, PA-C, is a physician assistant, Shari Pilon-Thomas, PhD, is a professor of medicine, and Gregory M. Springett, MD, PhD, is a professor of medicine at Moffitt Cancer Center, Tampa, FL.
Disclosures: Kara Forinash, and Drs Pilon-Thomas and Springett report no relevant conflicts of interest to disclose.
Address correspondence to: Kara Forinash, PA-C; 12902 Magnolia Dr. Tampa, FL 33612; phone: 813-745-3346; fax: 813-745-7229.