Comment on Powell et al, page
The clinical responses achieved by adoptive immunotherapy correlate with the in vivo persistence of infused melanoma-specific T cells that have up-regulated expression of IL-7Rα and have recapitulated an effector memory cell phenotype.
The premise and promise of T-cell therapy stems from the ability to selectively augment a recipient's immune response against a desired pathologic antigen. Adoptive immunotherapy for melanoma is an attractive approach, for it is well accepted that T cells can recognize melanoma-associated antigens. The translation of this concept into T-cell trials with therapeutic value for patients with metastatic tumor has recently been accomplished by Rosenberg and colleagues1 using heterogeneous populations of melanoma-specific T cells, ex vivo expansion and infusion of large numbers of T cells, co-infusion of CD4+ and CD8+ T cells, a pre-immunotherapy lymphodepletion regimen, and repeated administration of exogenous interleukin 2 (IL-2).1-3 In this issue of Blood, the authors follow the fate of adoptively transferred T cells in 6 responding melanoma patients and develop a rationale for their observed clinical responses.FIG1
Schematic of the adoptive transfer of a population of in vitro–expanded melanoma-specific CD8+ T cells in the lymphopenic recipient and, following contraction of the infused effector cells, emergence of clonal/oligoclonal effector memory T cells.
Schematic of the adoptive transfer of a population of in vitro–expanded melanoma-specific CD8+ T cells in the lymphopenic recipient and, following contraction of the infused effector cells, emergence of clonal/oligoclonal effector memory T cells.
To generate the numbers of tumor-specific T cells for adoptive immunotherapy trials, Powell and colleagues have developed a protocol to grow tumor-infiltrating lymphocytes (TILs) in vitro by stimulating with autologous resected tumor and high-dose IL-2, followed by rapid numerical expansion mediated by OKT3 (anti-CD3) and IL-2 that delivers a potent mitogenic signal. A potential drawback to this culturing methodology is that the expanded CD8+ T cells typically down-regulate cell-surface expression of CD27, CD28, IL-7Rα (CD127), as well as CD62L and CCR7. This phenotype is consistent with differentiated effector T cells, which are associated with a decreased ability to proliferate and a reduced potential to survive in vivo. However, it appears from their in vivo data that not all of the cultured T cells were terminally differentiated.
Adoptive transfer of these melanoma-specific T cells (identified by tetramer-binding and/or Vβ T-cell receptor expression) in some lymphopenic recipients resulted in rapid in vivo expansion followed by sustained homeostatic proliferation. Serial sampling from peripheral blood of patients demonstrated that the infused T cells had increased expression of the costimulatory molecules CD27, CD28, and IL-7Rα and this was associated with long-term persistence of effector memory T cells (figure). Extrapolating from these data, it appears that the infused T cells that survive are capable of responding to IL-7, a cytokine that is associated with transition of effector CD8+ T cells to memory cells and regulating the size of the T-cell pool by helping to compensate for T-cell depletion.4,5 However, other facets of the National Cancer Institute (NCI) T-cell trial may contribute to survival of the infused CD8+ T cells, such as co-infusion of CD4+ T-helper cells, as well as removal of regulatory T-cell subsets in the lymphopenic recipient. It remains to be seen if the effector memory T cells will continue to persist and whether these lymphocytes can function to control/prevent melanoma growth over the long term. Indeed, to help correlate the surrogate data obtained by immune monitoring with clinical outcome, these studies will need to evolve from sampling of peripheral blood to evaluating the immunobiology of adoptively transferred T cells residing in the tumor microenvironment, preferably using tools that also probe the functional activity of sampled T cells.
In this issue of Blood, Powell and colleagues highlight the power of the in vivo processes to select for a subpopulation of transferred T cells, such as memory-cell precursors, that can selectively survive in a lymphopenic host, and have important implications for the field of adoptive immunotherapy in general. Many adoptive immunotherapy trials numerically expand T cells ex vivo by providing a stimulatory signal through endogenous αβ T-cell receptor or by cross-linking CD3, in the presence of IL-2. The data in this paper will help reassure investigators that these in vitro growth conditions may not be an impediment to generating long-lived T cells in vivo. However, it is still to be determined whether these results from patients with melanoma can be translated to other T-cell therapy trials employing lymphodepleting-conditioning regimens, including clinical trials infusing T cells with specificity directed through chimeric immunoreceptors. Hopefully, such experiments will continue to delineate critical variables contributing to the in vivo survival of transferred T cells in the lymphopenic and lymphoreplete environments. Furthermore, additional studies will be needed to test whether adoptive immunotherapy can be improved by the a priori identification, or enforced expression, of receptors (such as IL-7Rα+) that endow T cells with improved in vivo survival. As these future adoptive immunotherapy trials are implemented, investigators and regulators will need to balance the relative risks and benefits obtained from transferring homogeneous welldefined clonal populations of antigen-specific T cells versus infusing heterogeneous bulk populations potentially containing antigen-reactive T cells capable of taking advantage of the recipient's lymphopenic milieu.
