Gene Expression of Human Lymphocytes Used For ACT of Metastatic Melanoma

Recent reports have detailed the efficacy of the adoptive cell transfer (ACT) of autologous anti-tumor lymphocytes for the treatment of patients with melanoma refractory to standard therapy. Here we report the microarray analysis of tumor infiltrating lymphocytes (TIL) generated from twenty-eight melanoma patients enrolled in a treatment protocol that combines non-myeloablative lymphodepletion followed by adoptive cell transfer. The patients were divided into two groups Ÿ twelve who experienced an objective response according to RECIST criteria and sixteen who did not. Using several methods of data analysis, no consistent overt changes in gene expression, pathways or classical markers of lymphocyte development were detected between TIL administered to responding versus non-responding patients. Only one non Y chromosome gene, phosphoserine phophatase-like (PSPHL), was found to be over expressed primarily in lymphocytes that did not result in tumor regression and seemed to negatively correlate with in vivo survival of transferred TIL. As persistence of lymphocyte clonotypes in vivo was correlated with patient response, cyro-preserved treatment TIL cultures were bead-separated into in vivo persisting and non-persisting fractions. Subsequent microarray analysis of these persisting and non-persisting clonotypes also did not reveal any consistent differences in gene expression. Finally, to determine if the lymphocytes used in our adoptive cell transfer studies had similar gene expression profiles, the twenty-eight TIL samples were compared to 10 samples generated through limiting dilution cloning from an early ACT trial. This analysis revealed differences in the gene expression of lymphocytes that have been used in these two different ACT trials. However, the inability to find differential gene expression in tumor infiltrating lymphocytes that correlate to patient response or persistence suggests that differences affecting these clinical events lie in host or tumor factors rather than transferred cells. Keywords: cell type comparison design There were 28 patient samples with 12 responding and 16 non-responding patients. There were 6 persisting and 6 non-persisting clonotypes. There were 10 cloned lymphocyte samples. All 50 samples were arrayed using Universal Reference as a control in a two-color hybridization.