Phase I study of α-galactosylceramide-pulsed antigen presenting cells administration to the nasal submucosa in unresectable or recurrent head and neck cancer

Abstract

Background

Human Vα24 natural killer T (NKT) cells are activated by the specific ligand, α-galactosylceramide (α-GalCer), in a CD1d-dependent manner. Potent anti-tumor activity of activated NKT cells has been previously demonstrated.

Methods

We conducted a phase I study with α-GalCer-pulsed antigen presenting cells (APCs) administered in the nasal submucosa of patients with head and neck cancer, and evaluated the safety and feasibility of such a treatment. Nine patients with unresectable or recurrent head and neck cancer received two treatments 1 week apart, of 1 × 10⁸ of α-GalCer-pulsed autologous APCs into the nasal submucosa.

Results

During the clinical study period, no serious adverse events (Common Terminology Criteria for Adverse Events version 3.0 greater than grade 3) were observed. After the first and the second administration of α-GalCer-pulsed APCs, an increased number of NKT cells was observed in four patients and enhanced natural killer activity was detected in the peripheral blood of eight patients.

Conclusion

The administration of α-GalCer-pulsed APCs into the nasal submucosa was found to be safe and induce anti-tumor activity in some patients.

Introduction

Head and neck cancer has been estimated to be the sixth most common malignancy worldwide, with about 500,000 patients diagnosed annually. The management of head and neck cancer has generally involved the combined-modalities of chemotherapy, radiation therapy and surgery. In recent years, cisplatin-based chemotherapy combined with radiation has been widely employed in the treatment of advanced head and neck cancer with the intent of preserving speech and swallowing or in the post-operative setting for patients at high risk for recurrence [1, 2]. However, the increased toxicity and extensive functional morbidity induced by these combined therapies can severely impair the quality of life (QOL), and, at the same time, the prognoses remains poor [3]. The improved survival rate by intensive combined therapy has only been shown to be 0∼10% more effective than conventional radiotherapy. In order to improve the prognosis and QOL of patients with head and neck cancer, the development of new treatment strategies is therefore of critical importance [4–6].

Natural killer T (NKT) cells represent a unique lymphocyte subpopulation that is characterized by the co-expression of T cell and natural killer (NK) receptors [7–10]. NKT cells exert strong anti-tumor activity both in vivo and in vitro settings [8, 11, 12]. NKT cells are activated by a specific glycolipid antigen; α-galactosylceramide (α-GalCer) in a CD1d-dependent manner [13, 14]. CD1d is a HLA class 1b antigen priming molecule that does not exhibit any type of allelic polymorphism. Therefore, the anti-tumor activity of NKT cells is not restricted to MHC antigens. Several clinical trials aimed at NKT cell activation using α-GalCer have been carried out [15–17]. Recently, we have demonstrated the safety and effectiveness of induction of NKT cell activation by the administration of 1 × 10⁹ antigen presenting cells (APCs) pulsed with α-GalCer, in patients with lung cancer [18].

The present studies were designed to evaluate the safety, immunological responses and possible anti-tumor effects of a smaller number (1 × 10⁸) of α-GalCer-pulsed APCs, administered in the nasal submucosa in patients with unresectable or recurrent head and neck cancer.

Patients and methods

Patient eligibility criteria

A group of patients between 20 and 80 years of age, with a histological diagnosis of squamous cell carcinoma of the head and neck, for which no standard treatment was available, were enrolled in the study. Other criteria for enrollment included a performance status (PS) of 0, 1, or 2; an expected survival of 6 months or more; and normal or near-normal renal, hepatic, and hematopoietic functions. The patients had not received chemotherapy or radiotherapy for at least 6 weeks prior to enrollment. The exclusion criteria for exclusion from these studies included positive antibody activity against HIV, hepatitis C virus, or human T-cell lymphotrophic virus; presence of hepatitis B surface antigen; active inflammatory disease or active autoimmune disease; a history of hepatitis; pregnancy or infections; concurrent corticosteroid therapy; and existence of another malignant neoplasm. The histological types and the anti-tumor effects of the treatment were classified according to the general rules for the clinical and pathologic recording of cancer as described by the Japanese Head and Neck Cancer Society.

Preparation of APCs containing dendritic cells from peripheral blood

All procedures were conducted according to Good Manufacturing Practice standards. Peripheral blood mononuclear cells (PBMCs) were collected from peripheral blood samples from nine patients and separated by density gradient centrifugation (OptiPrep, Nycomed Amersham, Oslo, Norway). PBMCs were washed three times and resuspended in AIM-V (Invitrogen Corp., Carlsbad, CA, USA) with 800 units/ml of human granulocyte macrophage colony-stimulating factor (GeneTech Co., Ltd., China) and 100 Japanese reference units per milliliter of recombinant human IL-2 (Imunace, Shionogi, Osaka, Japan) as described [18]. After 7 days of cultivation, the cells were harvested and washed three times and then resuspended in 0.2 ml of 2.5% albumin in saline. The cultured cells were pulsed with 100 ng/ml of α-GalCer (KRN7000; Kirin Brewery, Gunma, Japan) on the day before administration. The cultured cells were then administered into the nasal submucosa of each patient. The criteria for α-GalCer-pulsed APCs administration included a negative bacterial culture 48 h prior to APCs administration, cell viability >60%, and an endotoxin test with <0.7 Ehrlich units/ml, 48 h before APCs administration.

Clinical protocol and study design

This study was conducted at the Department of Head and Neck Surgery, Chiba University Hospital, Japan. Written informed consent was obtained from each of the patients before undergoing a screening evaluation to determine their eligibility. The protocol was approved by the Institutional Ethics Committee. On days 0 and 7, 150 ml aliquots of peripheral blood were collected from each patient. The PMC were processed and pulsed with the antigen as described. The cells were resuspended to a final volume of 0.2 ml and injected into the nasal submucosa. Each patient received two injections of 1 × 10⁸ α-GalCer-pulsed APCs. The injection site was the anterior portion of the bilateral inferior turbinate. We used a 27 G needle and a 1-mL syringe for the nasal submucosal injection. Extensive clinical and laboratory assessments were conducted weekly and included a complete physical examination and determination of standard laboratory values for 5 weeks. All patients underwent hematological and tumor status assessments by computed tomography both at baseline, 4 and 8 weeks after the first administration ofα-GalCer-pulsed APCs (namely, 9 weeks after study entry).

APCs phenotype evaluation

The APC phenotypes were determined using a FacsCaliber flow cytometer (BD biosciences). The monoclonal antibodies (mAb) employed included FITC-labeled anti-HLA-DR, PE-labeled anti-CD86, and APC-labeled anti-CD11c (Becton Dickinson, San Diego, CA, USA). Isotype-matched mAbs were used as negative controls. The phenotypes of the cultured cells containing DCs were analyzed by flow cytometry prior to each administration. The DC-rich population (large, granular lymphocyte fraction) was electronically gated by forward and side scatter parameters (FSChigh SSChigh).

Immunological monitoring

The PBMC samples were obtained by extracting 30 ml of blood at least twice before the administration of APCs and weekly up to 3 weeks after the final treatment. The frequencies of Vα24⁺Vβ11⁺ NKT cells in PBMCs were assessed by flow cytometry. Mononuclear cells were stained with FITC-conjugated anti-T-cell receptor (TCR) Vα24 mAb (C15; Immunotech, Marseilles, France), PE-conjugated anti-TCR Vβ11 mAb (C21, Immunotech), and APC-conjugated anti-CD3ε mAb (UCTH1; PharMingen) as previously described [19]. The stained cells were subjected to flow cytometry and the percentages of Vα24⁺Vβ11⁺CD3⁺ cells among mononuclear cells were calculated. The peripheral NKT cell numbers (counts/ml) were subsequently estimated based on the PBMCs cell counts. The remaining PBMCs were suspended to 5 × 10⁵/ml in Cell Banker 2, and stored at −80°C.

ELISPOT assay of IFN-γ producing cells in PBMCs

Frozen PBMCs from each patient were thawed and cultured for 6 h in AIM-V. The cultured cells were washed and transferred to an ELISPOT assay kit (BD Pharmingen) in 96-well filtration plates coated with anti-IFN-γcapture antibody for 16 h [20]. The stimulation conditions were 100 ng/ml of α-GalCer in AIM-V. After extensive washing with PBS, biotinylated anti-human IFN-γ antibody was added. Two hours later, spots were detected by an avidin–biotin–peroxidase complex and aminoethyl carbazole solution. The mean values of the spots in three wells were utilized for the subsequent analysis. According to our ELISPOT assay protocol, the majority of the IFN-γ producing cells detected after α-GalCer stimulation for 16 h were determined to be CD56⁺ NK and NKT cells.

Evaluation of antitumor effect

The disease progression was defined as an increase of target lesions of malignancy by >20% and the improvement of the disease was defined as a decrease in the target lesions by 30% or more. Any adverse events and changes in laboratory values were graded according to the Common Terminology Criteria for Adverse Events version 3.0.

Results

Patient characteristics

In accordance with the protocol, a total of nine patients who satisfied all entry criteria were enrolled into this study from January 2005 to March 2006. All patients completed this study without dropping out. The patient characteristics are summarized in Table 1. The median age was 59.2 years, and PS0/1 was 6/3, respectively. Five patients had previously undergone a surgical resection, while four patients had only undergone chemo-radiotherapy. Cases 001, 006, and 007 still had unresolved cancer lesions despite prior treatments, whereas the remaining cases had recurred after initial remission after previous treatment. As for the cancer lesions themselves, five cases had evidence of disease limited to the local site, and four cases exhibited distal metastasis.

Table 1.

Patient profiles

Case	Age/gender	Primary lesion	Present tumor lesion	Pretreatment	History of bilateral neck dissection
Case 001	54/male	Hypopharynx	Primary lesion	CT, RT	–
Case 002	48/female	Maxillary sinus	Primary lesion	CT, RT	–
Case 003	65/male	Hypopharynx	Cervical lymphnode	CT, RT, OP	–a
Case 004	57/male	Maxillary sinus	Lung	CT, RT, OP	–a
Case 005	60/male	Hypopharynx	Lung	CT, RT, OP	+
Case 006	71/male	Hypopharynx	Lung	CT, RT	–
Case 007	63/male	Maxillary sinus	Primary lesion	CT, RT	–
Case 008	61/male	Mesopharynx	Primary lesion	CT, RT, OP	–a
Case 009	54/male	Larynx	Skin	CT, RT, OP	+

CT chemotherapy, RT radiation therapy, OP operation

^aUnilateral neck dissection

Adverse events

No major toxicity (above CTCAE-grade 2) or severe side effects were observed in any of the patients (Table 2). One patient (case 003) exhibited temporary anemia (CTCAE-grade 2) 1 week after the first injection, but recovered the following week without any treatment.

Table 2.

Clinical effects

Case	Clinical outcome	Adverse event
Case 001	SD	–
Case 002	SD	–
Case 003	PR	Anemia (grade 2)
Case 004	PD	–
Case 005	PD	–
Case 006	SD	–
Case 007	SD	–
Case 008	SD	–
Case 009	PD	–

Clinical outcome

All nine cases were evaluated at the end of the clinical trial period. According to the findings of computed tomographic examinations with enhancement, one patient demonstrated a partial response with some clinical improvement (case 003). In this patient, the tumor diameter decreased from 22 to 7 mm. The computed tomography findings and the photographic view of the neck tumor mass before and after cell therapy in this patient (case 3) are shown in Fig. 1.

Fig. 1.

Computed tomography findings and photographs of case 003 before and after treatment. The arrows indicate the tumor lesion

In other five cases no change was noticed in the existing diseases status, and three remaining patients cases exhibited continued progression of the disease (Table 2).

Phenotypes of APCs containing dendritic cells

The APC profiles of three representative cases are shown in Fig. 2. In all preparations, the FSChigh, SSChigh cells exhibited a mature monocyte-derived DC phenotype, expressing HLA-DR, CD11c, CD80, and CD86 (Fig. 2a). The proportion of DCs in the APCs were observed to vary from case to case (Fig. 2b).

Fig. 2.

A flow cytometric analysis of APCs (GM/IL-2 cultured) in all cases. a The expression levels of HLA-DR, CD11c, CD86 assessed by flow cytometry. The shaded areas: background staining with an isotype control. The solid lines: staining profiles of the indicated molecules. The values represent the percentages of the positive cells. b A summary of the contents of the APCs administered for the first time. The percentages of the indicated fraction in whole APCs are shown. The number of cells with a DC phenotype (CD11c⁺HLA-DR⁺CD86⁺) in the APCs (1 × 10⁸) is shown

Immunological monitoring

Specific immunological testing was conducted in each of the nine patients who completed this study (summarized in Table 3). As shown in Fig. 3, one patient (case 001) exhibited a dramatic increase in the number of circulating NKT cells after the first administration of α-GalCer-pulsed APCs. The absolute number of Vα24 NKT cells increased in four patients (cases 001, 002, 007 and 008). The increased levels were sustained for at least 1 week. However, no such increase was detected after the second administration of α-GalCer-pulsed APCs in cases 001, 002, and 007. The NKT cell numbers decreased slightly in five patients (cases 003, 004, 005, 006 and 009). In two patients who exhibited decreased NKT cells (cases 004 and 009), the number of NKT cells returned to pretreatment baseline values, while the cell numbers in the remaining cases continued to remain low. Only case 003 exhibited increased peripheral blood NK counts (data not shown), however, the remaining cases did not show any significant change in the number of NK cells in PBMCs.

Table 3.

Immunological profiles of PBMCs

Case	Primary NKT cells frequency (%)	Primary NKT cell counts/ml	NKT cell increase	Enhanced NK activity
Case 001	0.14	1,021	+	+
Case 002	0.014	150	+	+
Case 003	0.0025	20	–	+
Case 004	0.04	664	–	+
Case 005	0.05	767	–	–
Case 006	0.007	67	–	+
Case 007	0.004	15	+	+
Case 008	0.022	150	+	+
Case 009	0.02	231	–	+

The frequencies and numbers of NKT cells in PBMCs at primary points were shown

Fig. 3.

The number of Vα24 NKT cells and NK cells in the peripheral blood during the course of treatment in three representative cases. The number of peripheral blood NKT cells (Vα24⁺Vβ11⁺ cells) and NK cells (CD3-CD56⁺ cells) are shown. Peripheral blood samples were taken before intranasal injection on days 7 and 14. The primary frequencies of the NKT cells in PBMCs of each case were as follows; case 001, 0.14%; case 002, 0.014%; case 003, 0.0025%

ELISPOT assays revealed IFN-γ spot forming cells, thus reflecting the NK and NKT activity, and which increased substantially in eight cases (except for case 005) (Fig. 4, three representative cases are shown). Even after the second administration of α-GalCer-pulsed APCs, three patients (case 003,004, and 008) exhibited a further increase in the number of IFN-γ spot-forming cells. Case 005 exhibited a decrease in the number of IFN-γ spots after the second administration.

Fig. 4.

The increased number of IFN-γ spot forming cells induced by in vitro stimulation with α-GalCer during the course of treatment in three representative cases. Peripheral blood samples were taken before the intranasal injection on days 7 and 14. Freeze-stocked PBMCs were thawed, and cultured for 6 h in AIM-V at 37°C. Next, the cells were stimulated with either α-GalCer (100 ng/ml) or vehicle in 5 × 10⁵/well in a 96-well flat bottom plate for 16 h. All spots were counted, and the mean values are shown with SDs

Discussion

The relative frequency of NKT cells in the peripheral blood are generally quite low (usually less than 0.1% of PBMCs), these cells play a very important roles in the development of both innate and acquired immune responses [8, 9]. The anti-tumor activity of NKT cells has been well documented in several recent publications [11, 12]. NKT cells exhibit cytotoxic effects through direct killing. These cells have also been shown to activate NK and specific cytotoxic T cells. [21–24]. In our previous studies employing intravenous administration of α-GalCer-pulsed APCs [18] or activated NKT cells [20] in patients with lung cancer, an increased number of NKT cells and a significant enhancement of NK cell activity was observed in 30 and 80% of the patients, respectively.

Most head and neck cancers arise from the upper respiratory or digestive mucosa. In a study with mice, the proliferation of NKT cells was observed in cervical lymph nodes following the intranasal administration of α-GalCer [25], and when Salmonella toxin was administrated orally to sheep, the migration of antigen presenting DCs was observed from the oral mucosa to the regional lymph nodes [26]. In our previous study, when autologous indium-labeled DCs that had been pulsed with specific tumor antigen, were administered into the submucosa, the labeled DCs migrated quickly and selectively to the regional lymph nodes where they then remained for more than 1 week (unpublished observation). We therefore chose the submucosal administration of α-GalCer-pulsed APCs, which is expected to effectively induce mucosal anti-tumor responses. In fact, in some patients, we detected an increased number of NKT cells in the PBMCs and an increase in the number of IFN-γ producing NK cells in PBMCs (Table 3, Figs. 3 ,4). This is a striking observation because we injected only 1 × 10⁸ APCs into the nasal submucosa, however, substantial changes in the systemic NKT/NK activity were observed. These results indicate that the regional immune system in the upper respiratory and digestive organs is quite unique in its ability to initiate and modulate local immunologic functions. The observations reported here suggest that submucosal administration of antigen or antigen-pulsed APC is an effective and useful route for the modulation of immune system.

In this study, we prepared APCs by culturing whole PBMCs with IL-2 and GM-CSF, but not by the use of either adherent or CD14-positive cells. When the proliferative activity of NKT cells was examined, a stronger response was observed with APCs derived from whole PBMCs than with those derived from adherent cells [27], thus suggesting the existence of a difference in the effective antigen presentation between glycolipid and protein antigens.

The patients tolerated well the treatment with 0.2 ml saline containing 1 × 10⁸ of α-GalCer-pulsed APCs twice with a 1-week interval. The treatment did not induce any pain and could be administered to outpatients easily without any anodyne. No severe adverse events were observed in the present investigation. Temporally associated anemia (CTCAE-grade2) was observed in one case (case 003). This patient had gastric and esophageal cancer, and had already undergone a total gastrectomy and irradiation therapy 5 years previously, and hypopharyngeal cancer had later developed as a new malignancy. Despite surgery and chemotherapy, the patient experienced local recurrence, and therefore he was registered for this study. Even before entry into the study, an aggravation of the anemia was occasionally observed in this patient. Therefore, the causality of anemia as a result of immunotherapy employed in this study could not be clearly established.

In an animal model, some hepatic disorders were observed after α-GalCer administration [28–30]. In our previous clinical study with lung cancer patients, some mild adverse events, including hot flashes, headache, hyperkalemia, and general fatigue (CTCAE-grade 1 and 2) were observed [18]. The reason why fewer adverse events were observed in the current study may be related to the fact that a smaller number of α-GalCer-pulsed APCs were administered.

The observed immunological responses were comparable with those obtained in the previous study with lung cancer patients [18], into whom 1 × 10⁹/m² of α-GalCer-pulsed APCs were administered intravenously. Even the smaller number of α-GalCer-pulsed APCs was sufficient to induce a significant activation of NKT cells and NK cells when administered directly into the sub-mucosa. The four patients who exhibited increased NKT cells (Table 3) had not undergone a bilateral neck dissection (cases 001, 002, 007, and 008). The presence of regional lymph nodes might therefore be indispensable for the effective increase of NKT cells. Nasal mucosal administration is easy for patients, as only 150 ml of peripheral blood extract is sufficient to prepare 1 × 10⁸ APCs and no apheresis is necessary.

Following the intranasal mucosal administration of α-GalCer-pulsed APCs, the IFN-γ producing cells, which reflected the NK and NKT activities, were significantly enhanced in 89% (8/9 patients). It is difficult to determine the frequency of IFN-γ producing cells are NKT cells, because the expression of Vα24 TCR tends to be very quickly down-modulated after activation [31, 32] and the specific activation of NKT cells has been reported to lead to a rapid induction of extensive NK cell proliferation and cytotoxcity, partially depending on the IFN-γ production by NKT cells [21, 22, 33]. Case 005 alone did not exhibit any augmentation of the activities. This case had previously undergone an extensive bilateral neck dissection, which may therefore have impaired the development of increased NK and NKT activities. After the activation of NKT cells, the anti-tumor activities are considered to be mediated by NK cells as well as NKT cells, as the absolute number of NKT cells is quite few in comparison to that of NK cells. Therefore, the enhanced NK activities observed in most of the patients enrolled in this study are thus considered to support the effectiveness of this treatment. On the other hand, in five of nine patients who received intranasal α-GalCer-pulsed APCs to the submucosa, the number of NKT cells did not increase in the peripheral blood. The reason for this was not clear. The increased number of NKT cells might accumulate in the cancer lesions from the peripheral blood. Chang et al. reported the sustained expansion of NKT cells after two intravenous injections of α-GalCer-pulsed mature dendritic cells [34], however, the number of local infiltrating NKT cells around cancer lesions still needs to be determined.

Regarding the clinical outcome, one patient, five patients and three patients exhibited PR, SD, and PD, respectively (Table 2). This outcome does not rule out the clinical efficacy of the treatment, because the status of the patients enrolled in this study, who had either unresectable or recurrent cancer, must be categorized as very serious. The results therefore most likely reflect the poor condition of these patients. Case 003 exhibited a PR response. This patient had developed three separate cancers and had undergone combined therapies as described above. The recurrent local tumor reduced in size, from 22 to 7 mm in diameter, after two treatments with α-GalCer-pulsed APCs and the pain was also alleviated for several weeks.

In summary, the submucosal administration of α-GalCer-pulsed APCs for patients with head and neck cancer was safe and a small number of these APCs without inconvenience of apheresis with no adverse effects could exhibit significant immune responses and some positive clinical effects. Further studies should therefore be conducted to examine NKT cells in the cancer tissue, as well as the regional neck lymph nodes, following submucosal administration.

Abbreviations

α-GalCer

α-Galactosylceramide

APCs

Antigen presenting cells

CT

Chemotherapy

CTCAE

Common Terminology Criteria for Adverse Events

DCs

Dendritic cells

NK

Natural killer

NKT

Natural killer T

OP

Operation

PBMCs

Peripheral blood mononuclear cells

PD

Progressive disease

PR

Partial response

PS

Performance status

QOL

Quality of life

RT

Radiation therapy

SD

Stable disease