Trial Information

Phase I/II Study Using a Non-Myeloablative Lymphocyte Depleting Regimen of Chemotherapy Followed by Infusion of Allogeneic Tumor-Reactive Lymphocyte Cell Line DMF5 in Metastatic Melanoma

Background:

In previous trials in the Surgery Branch, a 51 percent objective response rate has been
observed in heavily pre-treated patients with metastatic melanoma undergoing adoptive cell
transfer therapy utilizing a non-myeloablative preparative regimen followed by
administration of autologous tumor-reactive lymphocytes and subsequent treatment with
high-dose aldesleukin.

However, in patients with metastatic melanoma undergoing metastasectomy, recovery of
adequate numbers of tumor specific T lymphocytes from surgical specimens is possible in
approximately half of all patients, thus limiting the application of adoptive cell transfer
therapy.

Murine models performed in the Surgery Branch have demonstrated solid tumor regression in
mice treated with allogeneic tumor specific T cells combined with a preinfusion
lymphodepleting regimen.

We have identified a tumor specific lymphocyte cell line (DMF5) used previously in an
autologous adoptive cell transfer protocol that was associated with an objective clinical
response in that patient.

In subsequent preclinical testing of this lymphocyte population, we have demonstrated high
specificity against HLA-A 0201 positive melanoma cell lines as well as the common shared
melanocyte differentiation antigen MART-1:27-35. We have expanded this lymphocyte population
to provide up to 30 individual allogeneic cell transfers to HLAA 0201 positive patients with
metastatic melanoma.

In this trial we want to test our hypothesis that objective tumor regression can be achieved
with the DMF5 allogeneic T-cell product using a non-myeloablative regimen followed by cell
transfer and high-dose aldesleukin.

It should be emphasized that this protocol is designed to test whether highly melanoma
reactive allogeneic lymphocytes can mediate cancer regression. The DMF5 cell line is a
limited reagent only available for the treatment of up to 30 patients. However, if this
treatment results in cancer regression, it will represent an important step in our
development of an allogeneic T-cell receptor engineered universal effector cell line for the
treatment of patients with cancer.

Objectives:

To evaluate the safety of the administration of the DMF5 allogeneic T-cell product in
patients receiving the non- myeloablative conditioning regimen, and aldesleukin.

To determine whether this allogeneic tumor-specific lymphocyte cell line, hereafter referred
to as DMF5, infused in conjunction with the administration of high-dose aldesleukin may
result in objective clinical tumor regression in eligible HLA-A 0201 positive patients with
metastatic melanoma receiving a non-myeloablative lymphoid depleting preparative regimen.

To determine the in vivo survival of the infused cells following the non-myeloablative
regimen, via analysis of the sequence of the variable region of the T cell receptor or flow
cytometry (FACS).

Eligibility:

Patients with metastatic melanoma who are greater than or equal to 18 years of age, HLA-A
0201 positive, do not have suitable autologous tumor reactive TIL cells available, and are
able to tolerate high-dose aldesleukin.

Design:

Patients will receive a non-myeloablative lymphocyte depleting preparative regiment
consisting of cyclophosphamide (60 mg/kg/day times 2 days intravenous (IV)) and fludarabine
(25 mg/m2/day IV times 5 days).

Patients will receive intravenous adoptive transfer of the tumor reactive lymphocyte cell
line DMF5 (after its expansion in interleukin-2 and OKT3) followed by high-dose intravenous
(IV) aldesleukin (720,000 IU/kg/dose every 8 hours for up to 15 doses).

Patients will undergo complete evaluation of tumor with physical examination, CT of the
chest, abdomen, and pelvis and clinical laboratory evaluation four to six weeks after
treatment and then monthly for approximately 3 to 4 months or until off study criteria are
met. The study will be conducted using a Phase I/II optimal design. The protocol will
proceed in a phase 1 dose escalation design, with three cohorts. Should a single patient
experience a dose limiting toxicity at a particular dose level, three more patients would be
treated at that dose to confirm that no greater than 1/6 patients have a DLT prior to
proceeding to the next higher level. If a level with 2 or more DLTs in 3-6 patients has been
identified, three additional patients will be accrued at the next lowest dose, for a total
of 6, in order to further characterize the safety of the maximum tolerated dose prior to
starting the phase II portion. If a dose limiting toxicity occurs in the first cohort, that
cohort will be expanded to 6 patients. If 2 DLTs are encountered in this cohort, the study
will be terminated.

Once the MTD has been determined, the study then would proceed to the phase II portion, and
initially, 9 total patients will be administered the therapy at the maximum tolerated dose.
The plan will utilize a Simon two-stage optimal phase II design. If 0 of the 9 patients
experiences a clinical response, then no further patients will be enrolled but if 1 or more
of the first 9 evaluable patients enrolled have a clinical response, then accrual will
continue until a total of 30 evaluable patients have been enrolled.

This design has the ability to distinguish a 5% response rate (p0=0.05) from a 25% response
rate (p1=0.25), with 10% probability of falsely "accepting" the DMF5 cell therapy approach
(alpha=0.10), and 10% probability of incorrectly discarding this strategy as if it were
unacceptably poor (beta=0.10). This design also has 63% probability of stopping early (at 9
patients) if the true response rate is 5%.