In this study NPC patient will receive 4 days of treatment with CD45 antibody followed by one dose of LMP1- and LMP2-CTL. From this, we can learn if treating the patient first with the CD45 antibody will also let LMP1- and LMP2-CTL we give grow better. In addition, we will find out, if LMP1- and LMP2-CTL are safe and have enhanced anti-tumor activity in comparison to standard EBV-CTL.

This study aims to determine the safety of autologous LMP1- and LMP2- specific cytotoxic T-lymphocytes (CTL) in combination with CD45 monoclonal antibody (MAb) in patients with EBV-positive nasopharyngeal carcinoma (NPC).

And to obtain information on the expansion, persistence and anti-tumor effects of autologous LMP1- and LMP-2 specific CTL given after lymphodepletion with CD45 MAb in patients with EBV-positive NPC.

While patients with nasopharyngeal carcinoma (NPC) may be cured by chemotherapy and radiotherapy, the outlook for patients who are resistant to this treatment or who relapse is poor. Almost all patients with undifferentiated nasopharyngeal carcinoma have the EBV virus in their tumors which may be a target for immunotherapy approaches. We have successfully used specialized immune system cells grown in the laboratory and trained to recognize and kill EBV infected cells (EBV-specific cytotoxic T-lymphocytes [EBV-CTL]) to prevent and treat another type of cancer called post transplant lymphoma that occurs after bone marrow transplant. In post transplant lymphoma, the tumor cells have 9 proteins made by EBV on their surface. However in nasopharyngeal carcinoma that develops in patients with a normal immune system, the tumor cells only express 2 EBV proteins that are much harder for the immune system to recognize. In a previous study we made EBV-CTL that recognized all 9 proteins and gave them to patients with NPC. For patients without evidence of active disease at the time of therapy, there disease remains in remission. For those patients with active disease at the time they received CTL, some patients had a partial response to this therapy, and only three patients had a complete response. We think the main reason for this is that many of the T cells reacted with EBV proteins that were not on the tumor cells, and the other is that the infused T cells have not enough space to grow.

The two EBV proteins present on NPC tumor cells that are good targets for T-cell therapies are called LMP1 and LMP2. We are therefore planing to generate T cells specific for LMP1 and LMP2 and infuse these cells into NPC patients. To make LMP1- and LMP2-CTL, we have obtained blood from the patients and grown special type of cell called a dendritic cell (DC) and EBV infected lymphoblastoid cells (LCL). We have then transferred an adenovirus vector that carries the LMP1 and LMP2 gene into the DC and the LCL. These DC and LCL are then treated with radiation so they cannot grow and are used to stimulate and expand LMP1- and LMP2-CTL. This stimulation trains the T cells to kill cancer cells with LMP1 and LMP2 on their surface.

To 'create space' for EBV-CTL growth after infusion in NPC patients we have already used a special protein called a CD45 antibody, which removes for a short period of time most of the patient's T cells. The preliminary results of this study is encouraging: the use of the CD45 antibody is safe and we observed enhanced EBV-CTL growth after infusion. In addition, all patients who has EBV-CTL growth had clinical responses.

We and others have demonstrated the feasibility of CTL therapy for EBV-positive NPC in immunocompetent patients, providing preliminary evidence of anti-tumor activity of EBV-CTL in this patient population. Not all patients responded, however, suggesting the need for further improvement. We propose that CTL failure can be overcome by increasing the specificity of the infused CTL product. That is, infusion of CTL specific for LMP1 and LMP2 will produce greater clinical benefit than EBV-specific CTL. The rationale for this approach is straight forward: EBV-specific CTL lines generated by standard methods are dominated by T-cell clones not reactive to the subdominant EBV proteins LMP1 and LMP2 expressed in NPC. We also propose that the failure of adoptively transferred CTL to measurably expand in the peripheral blood of NPC patients is a consequence both of lymphoid homeostasis in these lympho-replete patients and of the inhibitory T-cell infiltrate at the sites of disease. We will therefore use monoclonal antibodies targeting the CD45 antigen (CD45 MAbs), to lymphodeplete NPC patients prior to the infusion of EBV-specific CTL. Preliminary results indicate that CD45 MAb depletion can augment CTL expansion, and that such expansion is associated with a higher disease response rate. We will confirm and extend these promising new data in this Phase I clinical trial.

Inclusion Criteria:

All patients with NPC in first or subsequent relapse or with primary refractory disease or high risk (T3 or T4, or node positive) in whom the EBV genome or antigens have been demonstrated in tissue biopsies will be eligible for this trial.

Any patient with EBV positive NPC, in relapse or with primary resistant disease

Patients with a life expectancy 6 weeks or greater.

Patients with a Karnofsky score (age at least 16; for Karnofsky scale see full protocol) or Lansky score (less than 16; for Lansky scale see full protocol) of 50 or greater as described below:

Patients with bilirubin <2x normal, SGOT <3x normal, and Hgb greater than 8.0.

Patients with a creatinine 2x normal or less for age.

Patients should have been off other investigational therapy for one month prior to entry in this study.

Patient, parent/guardian able to give informed consent.

Exclusion Criteria:

Severe intercurrent infection.

Due to unknown effects of this therapy on a fetus, pregnant women are excluded from this research. The male partner should use a condom.

Note: Patients who would be excluded from the protocol strictly for laboratory abnormalities can be included at the investigator¡-s discretion after approval by the CCGT Protocol Review Committee and the FDA reviewer.