In Vivo Administration of Interleukin 15 (IL-15) Safely Promotes the Survival and Proliferation of Adoptively Transferred Antigen-Specific T Cell Clones

Abstract

The adoptive transfer of antigen-specific CD8+ cytotoxic T lymphocytes (CTL) that have been expanded in vitro is a promising treatment for human malignancies and infections. Interleukin (IL)-2 is frequently administered to support the in vivo survival of transferred T cells, but causes systemic toxicity when given in high doses and promotes the expansion of CD4+ regulatory T cells, which can inhibit antitumor immunity. IL-15, like IL-2, belongs to the four α-helix bundle family of cytokines and shares functional activities with IL-2, including binding to the IL-2 receptor (R) β and γc signaling components and promoting the proliferation of activated T cells in vitro. Despite the similar structure and in vitro function of IL-2 and IL-15, mice deficient in IL-15 or IL-15Rα have a marked reduction in natural killer (NK) cells, NKT cells, and CD8+ memory cells, whereas mice deficient in IL-2 or IL-2Rα have lymphoid hyperplasia and autoimmunity. Because of its critical role in the maintenance of T cell memory, IL-15 is an attractive alternative to IL-2 for augmenting adoptively transferred T cell immunity in humans. We administered IL-15 subcutaneously to nonhuman primates and evaluated toxicity, immunological effects, and peak and trough plasma levels. After establishing a safe regimen of IL-15 dosing, we evaluated the ability of IL-15 to support the survival of adoptively transferred CD8+ effector T cell (TE) clones in vivo.Results: IL-15 was administered subcutaneously to five macaques at doses ranging from 2.5 – 15 μg/kg, given either daily or every 3 days, respectively. The animals were monitored for clinical toxicity and plasma levels. Peripheral blood T cell subsets were enumerated at intervals and evaluated for phenotype and expression of Ki-67, a nuclear antigen expressed by cells undergoing proliferation. Daily administration of high-dose IL-15 resulted in a pronounced increase in the absolute numbers and Ki-67-expression of CD8+ T cells and NK cells, respectively, and preferentially expanded CD8+CD95+CCR7− effector memory (TEM) and CD8+CD95+CCR7+ central memory T cells (TCM). However, daily IL-15 in doses of 5 – 15 μg/kg was associated with accumulation of IL-15 in serum, and caused toxicities that were reversible when IL-15 was discontinued. By contrast, intermittent IL-15 treatment every 3 days was safe and induced only a moderate increase in NK cells, CD8+ TEM and TCM, and enhanced expression of Ki-67 in these cell subsets. This coincided with an increase of the absolute number of cytomegalovirus (CMV)-specific CD8+ T cells in the peripheral blood, but total numbers of CD4+ FoxP3+ T cells were not increased with IL-15. We then examined the ability of IL-15 administered every 3 – 4 days for 3 weeks to support the in vivo persistence of TCM-derived CMV-specific CD8+ TE clones that were marked to express a truncated macaque CD19 surface molecule and transferred to the animals without prior lymphodepletion. As previously reported, CD8+ TE clones derived from TCM precursors survive in vitro in low-doses of IL-15 in the absence of T cell receptor stimulation, persist long term in vivo after transfer and revert to the memory pool (Berger et al, JCI 2008, 118:294). In comparison with animals that received CD8+ T cells alone in which transferred T cells persisted at a stable level of 0.2 – 0.8% of circulating CD8+ cells, the administration of IL-15 after T cell transfer resulted in the establishment of a high-level T cell response (10 – 15% of CD8+ T cells; >100 cells/μL) that persisted for >6 months after IL-15 was discontinued. The CD19+CD8+ TE clones re-acquired a memory T cell phenotype in vivo and expressed bcl-2, bcl-xL and Ki-67 comparable to endogenous CD8+ T cells. The transferred cells were present in large numbers in bone marrow and lymph node samples obtained on day 14 and day 56 after infusion suggesting that they efficiently occupied niches of T cell memory. This data in a large animal model predictive of clinical translation demonstrates that IL-15 can be safely administered, exerts a profound immunologic effect, and dramatically augments the long-term survival of ex vivo expanded antigen-specific CD8+ CTL clones after adoptive transfer without promoting in vivo expansion of CD4+ Foxp3+ regulatory cells. Thus, IL-15 may be a safer and more effective alternative to IL-2 and/or lymphodepletion to support the in vivo persistence of adoptively transferred tumor or virus-specific T cells in human immunotherapy.