Defects in CB cellular immunity (Cairo, et al, Blood 82:2269, 1993) and cytokine production (Cairo, et al, Pediatr Res 30:326, 1991; 31:574, 1992; Lee & Cairo, et al, Blood 86:163a, 1995) may contribute to the decreased incidence of GvHD post cord blood transplantation, and potentially increase the relapse rate post transplant. Interleukin-15 is a novel cytokine that has no sequence homology to IL-2, but requires the use ofβ- and γ-chains of the IL-2 receptor complex for signaling. IL-15 shares many of the T cell stimulatory activities described for IL-2 (Armitage, et al, J Immunol 154:483, 1995). Recent studies have demonstrated induction of LAK cell activity in adult peripheral blood (aPBL) by IL-15 against known tumor targets (Grabstein, et al, Science 264:965,1994). Sweetman, et al, have demonstrated decreased IL-15 expression in cord MNC vs. adult MNC (Sweetman, et al, SPR abstract, 1996). In this study we sought to determime the effects of IL-15 on LAK cells generated from CB by comparing with IL-12 and IL-2 and aPBL. Mononuclear cells were isolated by density gradient separation, and monocytes were depleted by plastic adherence. LAK cells were generated by stimulating the non-adherent cells with IL-15 for 72 hours followed by a standard 3-hr 51Cr-release cytotoxicity assay against Daudi, a Burkitt's lymphoma cell line. LAK cell cytotoxicity against Daudi cells was shown to be dose-dependent, 93 ± 31 Lytic Unit (LU) for 0.1ng/ml, 343 ± 76 for 1.0, 774 ± 116 for 10, 1036 ± 195 for 100, and 1302 ± 311 for 500 ng/ml (n=5). The peak cytotoxic activity was reached with 100 ng/ml IL-15, similar to the activity generated with 1000 u/ml IL-2 (1090 ± 245 LU, n=3, p=0.886), and higher than 100 u/ml IL-12 (381 ± 89 LU, n=3, p=0.035). When compared to aPBL, CB cells show significantly increased response to IL-15 stimulation at the lower doses of 1 ng/ml (343 vs 90 LU, p=0.033) and 10 ng/ml(774 vs 357, p=0.036), but no significant difference at 100 ng/ml (p=0.324). The LAK cells generated from CB with 10 ng/ml IL-15 produced a wide tumorcidal spectrum against LAK-nonsensitive tumor cells, including acute lymphoblastic leukemia cell lines CCRF-CEM (T cells) (318 ± 3 LU vs 110 ± 39 for media control) and CCRF-SB (B cells) (436 ± 114 vs 36 ± 12), and neuroblastoma cell lines NB-100 (330 ± 40 vs 67 ± 31) and SK-N-MC (358 ± 13 vs 100 ± 30). Our data suggests that IL-15 is able to induce a LAK response from CB cells comparable to aPBL and at low doses generating a higher response in the cord vs. the adult. This effect with IL-15 was equal to high doses of IL-2 and higher than levels seen using IL-12, thus suggesting a possible use for cancer immunotherapy following cord blood transplantation.