Purpose: Dendritic cell (DC) vaccines are anticipated as a treatment regimen for the future in the management of solid tumors. Previously we have shown that DC-based vaccines show a consistent immune response, safety, and long-term survival with neoadjuvant radiation.

Methods: Patients (N = 14) were assigned to neoadjuvant 50 Gy of external beam radiation (EBRT), given in 25 fractions delivered 5 days/week, combined with 4 intratumoral injections of DCs, followed by complete resection. The control arm consisted of 6 patients treated with just neoadjuvant radiation followed by surgery. The primary endpoint was to establish safety and efficacy of the vaccine with immunological correlation.

Results: Median follow-up was 30 months. The primary endpoint of demonstrating the safety of the DC vaccine was achieved, but clinical efficacy with immunological correlation was not statistically significant. Eleven out of fourteen (66.6%) patients on the DC vaccine arm were alive, of which 8/14(57%) had no systemic recurrences over a period of 4 years. Favorable immunological responses correlated with clinical responses in some cases.

Conclusions: Neoadjuvant radiation with dendritic cell injections has the potential to be safe, effective, and immunogenic in high-grade soft tissue sarcoma and needs to be confirmed in randomized larger clinical trials.

Tumor heterogeneity, high grade, and larger tumors portend a poor prognosis in soft tissue sarcomas.

 In particular chemotherapy, radiation, and surgery, have not succeeded in establishing remission in these patients where comorbidities and older age may be risk factors as well.

 Therefore, tumor vaccines may be ideal in this population due to their favorable toxicity profile.

Investigations in immunology have developed several tumor-vaccine, antigen vaccine, and dendritic cell (DC)-vaccine strategies.

 In preliminary studies using animal tumor models, we tested the hypothesis that a combination of apoptosis-inducing therapy with intratumoral administration of DCs can result in a potent antitumor response. We used mouse models to prove the concepts that, after tumor irradiation, injected DCs migrate to the tumor site, induce an antitumor immune response after engulfing tumor cells.

Our group had also studied the systemic immune response of hypo-fractionated high-dose RT in metastatic breast cancer subjects and evaluated the tumor-specific RT-induced autoantibody (RIAA) formation. The 3-year overall survival in IgM or IgG positive patients were significantly superior to those in the negative group, and there were no significant disparities in subject and treatment characteristics between these two groups to account for differences in outcome. Greater than half of our subjects demonstrated the formation of RIAA, considering that those who were included had previously received multiple immunosuppressive chemotherapeutic regimens. Importantly, those patients who exhibited RIAA had a prolonged interval before developing new metastatic lesions, and RIAA formation conferred superior survival outcomes. Radioimmunoassay experiments defined potential immunomodulatory mechanisms leading to RIAA formation to detect differences in cytokine levels in those patients who mounted an antibody response. All patients with GM-CSF, IFN-γ or TNF-α, enjoyed an immune response suggesting an interaction between radiation therapy (RT) immunologic signaling and these cytokines. Our observations were compelling for the assertion that hypofractionated RT can induce tumor-specific autoimmunity. This autoimmunity works synergistically with the local tumor-killing effects of RT, and ultimately, translates into an improvement in survival.

Survivin, a member of the inhibitor-of-apoptosis IAP gene family, is correlated with tumor progression and poor prognosis.

 It has a strong potential as a tumor marker for solid tumors of mesodermal origin.

 Survivin is also correlated with poor survival in tumor types, like breast, glioma, lymphoma, and colorectal cancers. Our group had previously developed an experimental system to measure survivin-specific immune responses in healthy individuals and prostate-cancer patients in which it demonstrated that survivin-specific immune responses can be rated and detected in vitro.

 Previously, studies have also shown a highly increased risk of tumor-related death in a group of soft tissue sarcomas (STS)patients who demonstrated a significant correlation between the co-expression of the survivin mRNA and the telomerase reverse transcriptase mRNA. Prognostic information of STS can be obtained from ELISA that is adapted for survivin content measurement in cell lysates and tissue extracts.

Overexpression of this protein was documented with high frequency in patients with STS. Therefore, we formally evaluated the survivin-specific response in eighteen subjects enrolled in phase I high-risk STS study involving preoperative RT and DC. Two sources of TAAs were used to evaluate comparatively the immune response in patients: whole tumor cell lysate (TCL) and survivin. The immunological response to both TCL and survivin were present and correlated to some extent with the clinical responses.

 Long term progression-free survival was 53 months, and overall survival was 47 months.

 Patients were divided into two groups based on the level of survivin expression in tumor samples: subjects with low level of survivin expression in tumor cells had a detectable response to survivin, whereas subjects with a high level of expression have developed survivin specific response to the treatment.

There is a clear correlation between immune response to survivin and its expression in tumors. The fact that such connection is not present for the response to whole tumor lysates indicates the specificity of the responses and argues for survivin as a good surrogate marker for future STS trials.

Treatment with radiation and DCs also resulted in dramatic increase in the presence of T-lymphocytes in tumor tissues (

< .01). Three-fourths of the patients had significant anti-autologous tumor cell immune responses, as seen in peripheral blood mononuclear cells before and after treatment. Responses showed a four-fold increase over controls after the last DC injection. Interestingly, these responses persisted even 30 weeks after starting treatment

. Significant differences were seen in the levels of two cytokines: IL-1 receptor antagonist (IL-1Ra), and interferon-inducible protein 10 (IP-10). Before treatment, the plasma concentrations of these molecules were similar between responder and nonresponders. However, during and after the treatment there was a significant increase in the IL-R1a level in the responders (

<0.03) while IP-10 levels are higher in the non-responders(

This raises the possibility of IL-1Ra and IP-10 as surrogate markers of immune response. The multifaceted immunologic assessment suggested significant anti-autologous tumor cell immune responses and thus provides the foundation for future trials involving radiotherapy and DCs in STS.

We first designed a phase I pilot study that looked at the combination of external beam radiation with intratumoral injection of dendritic cells as neo-adjuvant treatment of high-risk soft tissue sarcoma patients.

 We found that only after EBRT-DC therapy tumor-specific immune responses were observed in 52.9% of patients.

Our data also suggested that change in fractionation of EBRT may be warranted to provide more time for DC migration out of tumors. It was demonstrated that combination of EBRT and intratumoral DC administration could be a promising method of the treatment of patients with high-risk STS and provide rationale for more extensive clinical studies in this and other types of cancer.

Hence, we designed to execute the change in fractionation of the EBRT in a phase II trial with one arm enrolling patients to receive external beam radiation (EBRT) 50 Gy in 25 equal fractions; delivered 5 days per week (Monday through Friday) combined with intratumoral injections of DCs 3 times on the second, third, fourth Friday during the radiation. The primary endpoint was to investigate the safety and efficacy and the ability of an intensified RT regimen (namely, conventional RT with a high-dose hypo-fractionated boost) and DC administration, to induce an enhanced T lymphocyte immune response specific for soft tissue sarcoma-tumor associated antigens (STS-TAAs).

The Institutional review board approved the study at the University of South Florida, Tampa and University of Florida at Gainsville. Written informed consent was obtained from study participants or a legally authorized representative before enrollment.

All patients had clinically confirmed stage III, histologically proven soft tissue sarcomas over 5 cm, with a significant risk of progressing to distant metastases. Selection criteria included an ECOG performance status of 0 or 1, at least 18 years of age, no previous therapy, no autoimmune disease, and primary tumor of over 5 cm in size by RECIST criteria. Adequate function of the bone marrow (white count >3000, platelets >100, 000, hematocrit > 25), liver (serum bilirubin < 2 mg/dl), and kidney (creatinine < 2) were required for eligibility. Patients were excluded if they had gastrointestinal stromal tumors or tumors in the retroperitoneum or head and neck location, ongoing immunosuppressive therapy like steroids, azathioprine, cyclosporine or methotrexate, ongoing infection, prior radiation, demonstrated metastatic disease and HIV or a primary immunodeficiency disorder.

The primary objective of the study was to investigate safety and tolerability and the ability of an intensified radiation regimen (namely conventional RT with a high dose hypo-fractionated boost) and DC vaccine to induce an enhanced T-lymphocyte immune response specific for soft tissue sarcoma-tumor associated antigens. Secondary objectives were to determine the efficacy, disease-free survival at 1, 2 and 3 years after treatment, response rate, time to progression to lung metastases, and local recurrence rate. The planned accrual was 20 patients in each arm, but for funding reasons a total of 14 patients in the DC arm A and 6 patients on control Arm B were all who could be enrolled.

There was no randomization and patients were all accrued during the same period, using the same eligibility criteria followed the same RT protocol, had the coincidental timing of blood draws and this analysis was performed at only one laboratory at Moffitt Cancer Center. We had enrolled 14 patients on the DC arm (arm A) to receive neoadjuvant radiation and intratumoral DC vaccine whereas 6 patients on the control arm B were treated with neoadjuvant radiation alone.

Patients on both DC arm A and control arm B were treated with external beam radiation (EBRT) 50 Gy in 25 equal fractions, delivered 5 days per week (Monday through Friday). Midway through the second week of treatment, a 10 Gy dose core tumor volume boost of RT were delivered to the core of the tumor in a single fraction. This is in addition to the conventional 2 Gy standard dose the patient receives the same day. The boost will take place after 8 fractions of conventional RT (on day 10) (see Figure 1). Tumors were surgically removed 3 to 6 weeks after the completion of radiation.

On the treatment arm A, patients received neoadjuvant radiation combined with intratumoral injections of DC at a dose level of 1 ml containing 107 cells injected , using 22 to 25 gauge, 1.5-inch needles, on Fridays 2 to 4 hours after the 10th, 15th, 20th, and 25th RT dose. In some situations the dates of DC injection may be altered at the discretion of the treating physician with the following stipulations: 1) a DC injection must be given on day 1 or day 2 following the core boost RT dose and 2) no RT will be delivered in the 48 hours following the DC injections. The plan was to inject the entire DC product evenly throughout the tumor and tattoo the site so that the needle would be inserted at the same site each time.

To maximize the immunological response, the radiation therapy was coordinated to ensure that adequate apoptosis (forming apoptotic bodies for uptake into DCs with subsequent tumor antigen presentation) was ongoing within the soft tissue sarcoma at the time of intratumoral dendritic cell injection. No studies of serial sarcoma tumor biopsies after external-beam radiation to establish the time course of apoptosis had been published at the time of the design of this study. In our protocol, dendritic cell injections were given after 18, 27, 36, and 50.40 Gy of radiation DCs were injected on Friday morning after completing the radiation fraction for that day as this ensured that there was a period of 48 hours before the next radiation treatment thus minimizing the risk of inactivation of DCs.

Safety and tolerability were evaluated according to the NCI Common Terminology Criteria for Adverse Events (AEs), version 4.0.

Peripheral blood mononuclear cells (PBMC) cultures with granulocyte-macrophage colony stimulating factor (GM-CSF) and interleukin-4 were used to prepare DCs as described in previous studies 10

 cells of dendritic cells were injected in a total volume of 1 ml.

 cells were injected intratumorally before surgery. DC labeling and imaging were published previously.

Peripheral blood mononuclear cells (PBMC) were collected and frozen at -180° C from patients before the start of radiation after initial diagnosis, at the time of surgery and postsurgical resection at 2 and 3 months. Tumor cells were harvested by core biopsies of patients’ tumors before treatment and tumor cell lysates (TCL) were prepared by repeated snap freeze-thawing cycles and stored in liquid nitrogen, as described.

 Two sources of tumor-associated antigens (TAAs) were used to evaluate comparatively the immune response in patients: whole tumor cell lysate (TCL) and survivin. Lysates from patient PBMCs were used as controls. T-cell responses to TCL were assessed using IFN-γ ELISPOT

 and proliferation assays. To evaluate T-cell response to survivin, DCs were infected with adenovirus-survivin (Ad-surv), as described previously, to serve as stimulator cells.

 As a control, we used an adenoviral construct containing the empty vector.

Histological and Immunohistochemical Evaluation of Tumor Tissues

A pretreatment biopsy confirmed every tumor, utilizing histology, immunohistochemistry (IHC), and cytogenetic or molecular testing, when necessary. Therapy response assessments included tumor necrosis, fibrosis, and hyalinization as well as cytological changes. The viable tumor cells were estimated as 100% minus the percentage of tumor necrosis/fibrosis/hyalinization. Antibodies used for the studies included survivin (Novus Biological, Littleton, CO), CD3 (Ventana, Tucson, AZ), CD4 (Cell Marque, Rocklin, CA), and CD8 (Ventana).

We used Graph Pad Prism Software for statistical analysis. Kaplan-Meier and log-rank tests were employed for determining survival outcomes. Categorical data were summarized by frequency and percentage. Continuous data were summarized by number, mean, standard deviation, median, minimum and maximum values.

From May2011 to March 2013, we enrolled a total of 14 patients on DC arm A and 6 patients on control arm B (NC00012334). The demographics and patient characteristics are described in Table 1. The majority of the patients were male 13/20 (65%) and white 16/20 (80%). All twenty patients were included in the full analysis set and used to assess treatment effects and outcomes.

Dendritic cell vaccine -related adverse events (all grades) were fatigue (11.7%), hyperglycemia (8.0%), lymphopenia (7.8%) and anemia (7.8%) followed by pain (6.8%) and hyperpigmentation (6.8%). In the radiation only group, the most frequently reported adverse events were erythema multiforme (17.9%) and pain (5.7%), as well as fatigue (11.8%) (Table 2).

All treatment-related adverse events were classified into grades as seen in Table 2. Some of the adverse events on the dendritic cell vaccine arm may be from surgery, post-operative complication or both. Overall; the dendritic cell vaccine was very well tolerated and its safety profile was consistent with the previously described studies.

The proportion and type of all adverse events were not different between treatment groups. Four events from DC vaccine arm consisting of rash, injection site reaction, lymphopenia, and allergic/hypersensitivity reactions were classified as related to the investigational DC vaccine. No patient in the vaccine group or radiation only group was withdrawn from treatment because of an adverse event. No deaths occurred during the treatment.

The median OS was 29.5 months (95% CI, 23-46 months ) for the DC vaccine arm A versus 26.5 months (95% CI, 12-28 months) for the RT only arm. This difference was statistically not significant (HR, 0.44; 95% CI, 0.04448-2.912; 

= .4; Figure 2B.). Eleven patients from the DC vaccine arm and 4 patients from the RT cohort (arm B) were alive at the time of the analysis and were consequently censored. The 1-, 2-year and 3-year survival rates were 93.8%, 77.4%, and 58.4%, respectively, for the DC vaccine arm A compared with 75%, 60%, and 40%, respectively, for the RT-only arm B (Figure 2A).

Median PFS was 29.5 months (95% CI, 3-46 months) in DC vaccine arm A versus 20.5 months (95% CI, 12-35 months) in the RT arm only arm B. Vaccinated patients did not show a statistically significant advantage in PFS (HR, 0.39; 95% CI, 0.05-3.19; 

Eight patients from the DC vaccine arm A and 3 from RT-only arm B have not progressed at the time of this report. Of the 8 that have progressed on the DC vaccine arm, 2 have had local recurrence and 6 have had metastatic disease to the lung. Three have died following systemic progression to the lung, but the remaining 5 patients are currently receiving other treatments including surgery, radiation, and chemotherapy. Eleven out of 14 patients (79%) were alive on the DC combined with radiation (arm A) compared with 2 of the 6 patients (66%) alive on the control arm treated who were treated with radiation only (arm B).

Accumulation of Myeloid-derived suppressor cells and Regulatory T Lymphocytes

We assessed the presence of myeloid-derived suppressor cells (MDSC’s) and T lymphocytes in a few cases and controls (Figure 5).

Six out of 14 individuals (43%) demonstrated evidence of an immune response to survivin, at least at 1 point after the start of the treatment (Figure 4 A-C). Tumor necrosis did not correlate with survival (

 = .6) (Table 3, Figure 3). We found that there were some patients whose immune responses correlated with survival (Table 3). Due to their small numbers, this was not statistically significant and warrants further evaluation through prospective randomized controlled clinical trials.

This phase II study showed that dendritic cell vaccine, when given in combination with neoadjuvant radiation, was safe and well tolerated in high grade, larger than 5 cm soft tissue sarcomas. This is an extension from the pilot study of combining EBRT with DC infusions with change in fractionation of EBRT that was expected to provide more time for DC migration out of tumors.

The median OS and PFS were 29.5 months for the DC arm A and 26.5 and 20.5 months, respectively, for RT-only control arm B.

The 1-, 2-year, and 3-year survival rates were 93.8%, 77.4%, and 58.4%, respectively, for the DC vaccine arm A compared with 75%, 60%, and 40%, respectively, for the RT only arm B. The DFS of the control arm reflects data from soft tissue sarcoma treatment of previously published literature.

Furthermore, immune responses to survivin were induced in 6 patients (43%) following DC vaccination who received at least 4 injections of DCs. The immune response to survivin that is expressed in soft tissue sarcomas is encouraging enough to be able to use it as a biomarker in future trials. However, we were unable to check the magnitude of these responses in this protocol.

The majority of patients recruited in this trial had different types of soft tissue sarcomas. We were unable to correlate specific histologies to positive immunological responses to be able to comment on better survival patterns with certain histological subtypes.

Survivin expression in tumor cells varied among the 20 patients. We found that 6 out of 14 patients in the DC treatment arm A and 2 out of 6 in the RT-only control arm B expressed survivin. Though the expression of survivin could be correlated with the immune response to the protein, this was not statistically significant. The fact that such a connection was not present for the response to tumor lysates indicates the specificity of the responses and argues for survivin as a good surrogate marker for future STS trials.

Although this was a small non-randomized phase II study in a heterogeneous patient group, the increased PFS, and OS were encouraging, but not conclusive. We could attribute several reasons for the clinical responses of this study. Immune competence and performance status of these patients, adaptive responses induced by lower doses of ionizing radiation,

 with increased expression of survivin, may all affect the clinical response to the vaccine. Therefore, the effect of DC vaccine may have to be proven by characterizing the association of survivin or other proteins and specific tumor antigen responses induced by the DC vaccine.

The boost given midway through treatment was to investigate the ability of an intensified RT regimen (namely, conventional RT with a high-dose hypofractionated boost) and DC administration to induce an enhanced T-lymphocyte immune response specific for STS-TAAs. However, we could only detect an association, but to prove causation we will conduct a larger prospective randomized trial.

It is believed that a clear perspective on the action of vaccines is difficult to achieve if characteristics that are uncontrolled such as longer-than-expected survival and symptom improvement are analyzed. Also, objective criteria like tumor necrosis or lymphocyte infiltration may be seen in the natural course of the malignancy, even without treatment.

Understanding the interaction between innate and adaptive immunity, defining ligands for Toll-like receptors and inhibition provided by regulatory T cells and CTLA-4 like negative regulatory signals and use of adoptive T-cell therapies in combination with DC vaccines may help achieve sustained remissions.

In conclusion, in the current study we have shown that DC infusions are safe and well tolerated when administered in combination with radiation. Using tumor associated antigens and survivin, we will move to the next step of a randomized larger clinical trial powered to substantiate stronger immune induction and sustained clinical outcomes.

Shailaja Raj, Marilyn Bui, Xiuhua Zhao, Dungsa Chen, Randy Haysek, Ricardo Gonzalez, Douglas Letson, Alberto Chiappori, and Scott J. Antonia are with the H.Lee.Moffitt Cancer Center, Tampa, Florida; Anthony Conley is with MD Anderson Cancer Institute, Houston, Texas; Steven Eric Finkelstein is with Cancer Centers of America ,Sergio Lavilla-Alonso and Dimitry Gabrilovitch are with The Wistar Institute, Philadelphia, Pennsylvania; and Daniel Indelicato is with the University of Florida, Gainesville, Florida.

Mandard AM, Petiot JF, Marnay J, et al. Prognostic factors in soft tissue sarcomas. A multivariate analysis of 109 cases. 

Trojani M, Contesso G, Coindre JM, et al. Soft-tissue sarcomas of adults; study of pathological prognostic variables and definition of a histopathological grading system.

Tsujimoto M, Aozasa K, Ueda T, Morimura Y, Komatsubara Y, Doi T. Multivariate analysis for histologic prognostic factors in soft tissue sarcomas. 

Rööser B. Prognosis in soft tissue sarcoma.

el-Jabbour JN, Akhtar SS, Kerr GR, et al. Prognostic factors for survival in soft tissue sarcoma. 

Frustaci S, Gherlinzoni F, De Paoli A, et al. Adjuvant chemotherapy for adult soft tissue sarcomas of the extremities and girdles: results of the Italian randomized cooperative trial.

Lack EE, Steinberg SM, White DE, et al. Extremity soft tissue sarcomas: analysis of prognostic variables in 300 cases and evaluation of tumor necrosis as a factor in stratifying higher-grade sarcomas. 

Spira AI, Ettinger DS. The use of chemotherapy in soft-tissue sarcomas. 

Finkelstein SE, Fishman M, Conley AP, Gabrilovich D, Antonia S, Chiappori A. Cellular immunotherapy for soft tissue sarcomas. 

. 2012;4(3):283-290. doi: 10.2217/imt.12.3.

Nikitina EY, Gabrilovich DI. Combination of gamma-irradiation and dendritic cell administration induces a potent antitumor response in tumor-bearing mice: approach to treatment of advanced stage cancer. 

Altieri DC. Survivin and IAP proteins in cell-death mechanisms. 

2010;430(2):199-205. doi: 10.1042/BJ20100814.

Altieri DC. Validating survivin as a cancer therapeutic target. 

Mehrotra S, Languino LR, Raskett CM, Mercurio AM, Dohi T, Altieri DC. IAP regulation of metastasis. 

 2010;17(1):53-64. doi: 10.1016/j.ccr.2009.11.021.

Kappler M, Kotzsch M, Bartel F, et al. Elevated expression level of survivin protein in soft-tissue sarcomas is a strong independent predictor of survival. 

Kappler M1, Köhler T, Kampf C, et al. Increased survivin transcript levels: an independent negative predictor of survival in soft tissue sarcoma patients. 

Adida C, Haioun C, Gaulard P, et al. Prognostic significance of survivin expression in diffuse large B-cell lymphomas. 

Dohi T, Beltrami E, Wall NR, Plescia J, Altieri DC. Mitochondrial survivin inhibits apoptosis and promotes tumorigenesis. 

Reed JC1, Wilson DB. Cancer immunotherapy targeting survivin: commentary re: V. Pisarev et al., full-length dominant-negative survivin for cancer immunotherapy.

9:6523-6533, 2003. Clin Cancer Res. 2003;9(17):6310-6315.

Articles

Neoadjuvant radiation with dendritic cell injections has the potential to be safe, effective, and immunogenic in high-grade soft tissue sarcoma and needs to be confirmed in randomized larger clinical trials.

Dimitry Gabrilovitch MD PhD

Articles

Phase II Clinical Trial of Intratumoral Dendritic Cells Injections and Neoadjuvant Radiotherapy in Resectable Soft Tissue Sarcomas