Tumor-infiltrating lymphocytes (TILs) can be grown in vitro in medium containing interleukin-2 (IL-2). In clinical trials at the Surgery Branch of the National Cancer Institute, patients with metastatic malignant melanomas were treated with IL-2 plus the adoptive transfer of autologous TILs. At the time of treatment, TILs were assayed for in vitro lysis of fresh autologous and allogeneic melanoma cells and Daudi cells. Patients were evaluated for clinical response 4-8 weeks later. Lysis of autologous tumor cells by TILs was significantly higher for responding than for nonresponding patients. Tumor cells from responding and nonresponding patients were equally sensitive to lysis by allogeneic lymphokine-activated killer (LAK) cells. There was no difference between TILs from responding and nonresponding patients for lysis of LAK-sensitive Daudi cells, which was low in most cases and demonstrated that TIL lysis of autologous tumor cells was not due to LAK cells. The observed association of autologous tumor cell lysis by TILs with clinical response suggests that the development of culture methods to optimize lysis of autologous tumors may lead to increased response rates using this TIL treatment regimen.

Between 1989 and 1993, 255 tumor biopsies representing 4 tumor histologies (melanoma, breast cancer, colon cancer and renal cell cancer) were received by the Surgery Branch of the National Cancer Institute. Tumor-infiltrating lymphocytes (TIL) were grown from single-cell suspensions of tumor biopsies over the course of 30-45 days. The TIL were grown in medium containing IL-2. To obtain numbers suitable for therapy (>10(11)), TIL were expanded using a large-scale system of cell culture and harvesting. While the largest number of biopsies was obtained from melanoma patients, TIL were successfully grown from 160 of 255 tumor biopsies representing all 4 histologies. Under the culture conditions employed, several characteristics of TIL expansion were observed. The cell surface phenotype of TIL which grew out from the tumor biopsies was generally a mix of CD3+/CD4+ or CD3+/CD8+ lymphocytes. Only TIL from melanoma biopsies were found to be consistently cytolytic and, in many cases, lysed autologous tumor cells preferentially. Interestingly, TIL derived from extra-nodal sites of metastatic melanoma biopsies (subcutaneous, lung, bowel; 36 of 67, 54%) were more likely to have these cytolytic characteristics than TIL derived from tumor-involved lymph node biopsies (7 of 39, 18%). The present study summarizes 5 years of laboratory effort and validates the technologies developed for the large-scale growth and harvesting of TIL. In addition, it summarizes the laboratory effort supporting previously published clinical reports on TIL from our group.

No abstract

No abstract

The immune adherence (IA) assay was used to measure serum reactivity of patients with melanoma and osteosarcoma against paired cell lines (tumor cells and normal skin fibroblasts obtained from the same individuals) grown in tissue culture. Sera from 224 patients with various stages of melanoma were compared with sera from 100 normal age-and sex-matched donors. None oft.he 18 stage I sera (0%), 23 of 166 (14%) stage II sera, 3 of 40 (7%) stage III sera, and 3 of 100 (3%) normal sera were highly reactive to a standard allogeneic melanoma-fibroblast pair. None of the sera exhibited unique activity against melanoma. There was no correlation between stage of melanoma and high serum reactivity, nor was this reactivity predictive of recurrence. Sera from 39 tumor-bearing osteosarcoma patients prior to amputation were compared with sera from 50 normal ageand sex-matched donors. Eight of 39 (21%) patient sera and 1 of 50 (2%) normal sera were highly reactive to an osteosarcoma-fibroblast pair. No sera had reactivity uniquely directed against osteosarcoma. Eight osteosarcoma and two melanoma patients were tested simultaneously against their autologous cultured tumor and skin cells. Only one of these patients exhibited high reactivity towards autologous cells, and this reactivity was equal against both osteosarcoma and normal cells. None of seven highly reactive osteosarcoma or six highly reactive melanoma sera had residual tumor-specific reactivity against allogeneic osteosarcoma or melanoma after absorption with cultured fibroblasts, cultured fetal fibroblasts, or fetal calf serum.

No abstract

No abstract

No abstract