Increased vascular permeability in organs mediated by the systemic administration of lymphokine-activated killer cells and recombinant interleukin-2 in mice.

Significant tumor regressions in mice with established carcinomas, sarcomas, and melanomas and in humans with advanced cancers have been observed following immunotherapy with lymphokine-activated killer (LAK) cells and recombinant interleukin-2 (IL-2). However, dose escalations of LAK cells and IL-2 have been prevented by the development of a vascular leak syndrome (VLS). Although IL-2 alone can produce this VLS, we investigated the role of transferred LAK cells, generated by the incubation of syngeneic splenocytes in IL-2, in mediating this phenomenon. We used a murine model to quantitate the vascular leak by measuring the extravasation of iv injected 125I-bovine serum albumin. A permeability index (PI) was calculated by dividing the mean cpm of tissues from treated mice by those from control animals. The systemic transfer of LAK cells and IL-2 produced a significantly greater extravasation of albumin in the lungs, liver, and kidneys than after Hanks' balanced salt solution, IL-2, or LAK cells alone (in the lungs, for example, PI = 4.7, 1.4, and 1.6 after LAK cells and IL-2, LAK cells alone, and IL-2 alone, respectively). To eliminate the contribution to the leak by host lymphocytes, we irradiated mice before cell transfer. As compared to controls, LAK cells and IL-2 resulted in higher extravasation in the lungs, liver, kidneys, and spleen. However, a similar vascular leak was not observed in the lungs, liver, and kidneys after treatment with IL-2 plus fresh or cultured (without IL-2) splenocytes. Moreover, the combination of IL-2 excipient and LAK cells or IL-2 and irradiated LAK cells did not produce a fluid leak. The development of the VLS by LAK cells was directly related to the dose of concurrently administered IL-2 and the number of injected cells. Depletion of Thy 1.2-positive lymphocytes using antibody and complement prior to generating the LAK cells used in adoptive transfer did not abrogate the VLS in any of the organs tested. Similarly, depletion of L3T4 and Lyt-2 positive cells in vivo using monoclonal antibodies prior to harvesting spleens for generation of LAK cells also had no impact on the VLS. In contrast, in vitro treatment of LAK precursor cells with antibody to asialo-GM-1 plus complement completely eliminated the VLS when the depleted cells were cultured in IL-2 and subsequently transferred.(ABSTRACT TRUNCATED AT 400 WORDS)