Fludarabine modulates immune response and extends in vivo survival of adoptively transferred CD8 T cells in patients with metastatic melanoma

Abstract

Background: Adoptive T cell therapy involving the use of ex vivo generated antigen-specific cytotoxic T lymphocytes provides a promising approach to immunotherapy. It has become increasingly apparent that anti-tumor efficacy using adoptively transferred T cells is linked to their duration of in vivo persistence and can only be achieved when combined with some form of pre-infusion patient conditioning regimen. An optimal conditioning regimen that provides a positive benefit without serious toxicities has yet to be defined. We have established a unique clinical model that allows for evaluation of a given conditioning regimen on adoptively transferred T cells in humans. In this first-in-human study (FHCRC #1796), we evaluate the use of fludarabine, an FDA-approved reagent with predictable lymphodepleting kinetics and duration of action, as a conditioning regimen that promotes homeostatic upregulation of cytokines and growth signals contributing to in vivo T cell persistence.

Methods/findings: We conducted a phase I study in patients with refractory metastatic melanoma. Patients received two infusions of a single tumor-reactive antigen-specific CTL clone expanded to 10(10)/m(2); the first infusion was given without fludarabine conditioning, and the second CTL infusion was given after a course of fludarabine (25 mg/m(2)/dayx5 days). This design permits intra-patient comparison of in vivo T cell persistence pre- and post-fludarabine. Nineteen CTL infusions were administered to ten patients. No serious toxicities were observed. Three of nine evaluable patients experienced minor response or stable disease for periods of 5.8-11.0 months with two additional patients demonstrating delayed disease stabilization. The median overall survival in this heavily pre-treated population was 9.7 months. Fludarabine led to a 2.9 fold improvement in the in vivo persistence of transferred CTL clones from a median of 4.5 days (range 0-38+) to 13.0 days (range 2-63+) (p<0.05). Fludarabine lymphodepletion increased plasma levels of the homeostatic cytokines IL-7 and IL-15. Surprisingly, fludarabine also increased the relative percentage of CD4+ T cells expressing the regulatory protein Foxp3.

Conclusions/significance: Lymphodepletion with fludarabine enhances transferred T cell persistence but suggest that additional improvements to optimize T cell survival and address regulatory T cells are critical in providing anti-tumor efficacy.

Trial registration: ClinicalTrials.gov NCT00317759.

Conflict of interest statement

Competing Interests: The authors have declared that no competing interests exist.

Figures

Figure 1. Treatment schema.

Patients #1–4 received gp100, MART-1, or tyrosinase specific CD8+ T cell (CTL) infusions and fludarabine conditioning without IL-2. Patients #5–10 received very low dose IL-2 at 2.5×105 s.c. bid for 14 days following each T cell infusion. Patients were evaluated for response four weeks after the CTL infusion #2 and every 2 months thereafter.

Figure 2. Plasma levels of IL-7 and IL-15 were measured at serial timepoints pre- and post fludarabine conditioning.

CTL infusion #2 (INF#2) is administered two days after the final dose of fludarabine. *p = 0.053 comparing pre-fludarabine (pre) versus day of INF#2 (for IL-7) and **p

Figure 3

A. In vivo persistence of CTL clones as measured by tetramer+ CD8+ T cells from post-infusion, cryopreserved PBMCs. Superimposed are the correlative IFN-γ ELISPOT data. ELISPOT results are presented as the mean number of spot forming cells/105 PBMCs. Shown is patient #8 who demonstrated prolonged persistence of Mart-127–35 specific CTL. B. In vivo persistence and ELIspot data for patient #7 who received only one infusion of Gp100154–162 specific CTL clones. The long arrow indicates initiation of high dose dexamethasone (dex).

Figure 4. Fludarabine conditioning increases regulatory CD4+Foxp3+ T cells.

A. Representative example (patient #5) of flow cytometric analysis for CD4+Foxp3+ T cells from pre and post-CTL infusion PBMC samples. Infusion number and post-infusion day is designated along with percentage of CD4+Foxp3+/CD4+ cells. B. The relative percentage and absolute CD4+Foxp3+ population is reported for the nine patients who received fludarabine. Samples are tested in duplicate with the average values reported. Comparisons between baseline Foxp3 levels vs. day 0 of INF#2 (**p

Figure 5. CD4+Foxp3+ levels are sustained in the presence of IL-2 (n = 5) compared to the absence of IL-2 (n = 4) at day 21 after CTL Infusion#2 (trend at p = 0.093, unpaired t-test).