IL-15 administered by continuous intravenous infusion (CIV) results in 100-fold expansion of the CD8⁺ T effector memory population. IL-15 was administered for 10 days using 4 different dosing strategies (see panel A description below). The control group received 5% dextrose by CIV for 10 days. (A) Effect of IL-15 administration on total lymphocyte count, monocyte count, and lymphocyte subsets. Each point corresponds to the mean ± SEM and indicates the fold change in the absolute count relative to pretreatment (day −4). Gray bars above x-axis represent duration of IL-15 administration. (B) Representative example of CD8⁺ memory subset expansion in a RMs receiving 20 μg/kg/d of IL-15 by CIV. The left panel shows the pretreatment distribution of CD3⁺CD8⁺ naive (T_N, CD28⁺ CD95⁻), central memory (T_CM, CD28⁺ CD95⁺), and effector memory (T_EM, CD28⁻ CD 95⁺) populations. The right panel shows the distribution of CD3⁺CD8⁺ T_N, T_CM, and T_EM populations after 10 days of IL-15 treatment. The center panels show pre- and day-10 expression of Ki67 and CCR7 on CD3⁺CD8⁺ T_N, T_CM, and T_EM gated populations. (C) Dynamics of CD8⁺ memory subset expansion by different IL-15 dosing strategies. Data from the different dose groups are shown as in panel A.

Pharmacokinetics of IL-15 when administered by continuous intravenous infusion versus subcutaneous dosing. (A) Serum levels of IL-15 were obtained at 0.25, 0.5, 1, 2, 4, 8, and 24 hours after dose on treatment days 1 and 10 in animals receiving 20 μg/kg/ (♦) or 40 μg/kg/d (■) of IL-15 by SC injection. Each point corresponds to the mean ± SD for 3 animals. (B) Serum IL-15 levels were measured on days 2 and 10 in RMs receiving 20 μg/kg/d of IL-15 given by CIV. Levels were measured during 2 treatment cycles given 8 weeks apart. Each point corresponds to the mean ± SD. For day 2 versus day 10 in cycle 1, P = .06 by Wilcoxon rank test.