Characterization of interleukin-2-initiated versus OKT3-initiated human tumor-infiltrating lymphocytes from glioblastoma multiforme: growth characteristics, cytolytic activity, and cell phenotype.

Outgrowth of tumor-infiltrating lymphocytes (TIL) from the human primary brain tumor glioblastoma multiforme was achieved by OKT3 initiation (10 ng/ml), followed by sustained expansion by interleukin-2 (IL-2; 200 U/ml). Tumor-infiltrating lymphocyte (TIL) initiation by this process was performed in parallel with the standard "IL-2-only" method. Of ten tumors, seven yielded TIL in response to OKT3/IL-2, whereas only three of these seven grew after initiation with IL-2 alone. On the basis of cell doubling times, at least 60 doublings, resulting in (hypothetically) up to 10(23) TIL from as few as 2 x 10(5) cells in tumor suspensions, could be achieved using OKT3/IL-2. OKT3-initiated TIL proliferated in culture for as long as 288 days, although senescence of some cultures occurred at as early as 73 days. Significant heterogeneity of lymphocytes infiltrating the fresh tumors and heterogeneity of resultant TIL phenotype and function were apparent, yet several common trends were noted. In all cases after OKT3 initiation, significant net growth was not apparent until approximately 14 days. In contrast, in the three samples that grew in response to IL-2 alone, log-phase growth was always observed earlier. During the early phase of the cultures, all TIL expressed some killing activity toward a broad spectrum of tumors, including the autologous tumor. No consistent preference of TIL for lysis of autologous tumor was observed. Glioblastoma multiforme TIL cultures contained a mixture of CD8+ and CD4+ cells, with few CD16+ or NKH-1+. Of the six TIL examined in detail for population phenotype in relationship to time in culture, four eventually became exclusively CD4+. Further analysis of these CD4+ TIL indicated that all were of the helper-inducer class, 4B4+ and 2H4-. Concurrent with the decline in CD8+ cells, a decline in the cytolytic activity of these TIL cultures occurred. Furthermore, in two TIL that remained CD8+, a decline in the cytolytic activity also occurred. Therefore, loss of killing activity was not merely a reflection of the major cell phenotype changes. These results indicate that the OKT3/IL-2 process provides an alternative to IL-2 alone for TIL initiation and growth, as well as providing a novel system for further analysis of tumor-derived lymphocyte and accessory cell functional potential.