OUT OF THE Lab

 

The human immune system is made up of many different kinds of cells that are responsible for eliminating harmful “invaders” including microbial pathogens such as bacteria, viruses, as well as altered or sick or cancerous cells, from the body. T lympho­cytes, or T cells, play a central role in orchestrating most immune responses. In general, microbial pathogens are composed of foreign proteins or antigens. These antigens, when present in the host, are picked up and digested into peptides by specialized immune cells called antigen pre­senting cells (APCs). APCs subsequently present these foreign peptides to the rest of the immune system by placing them into surface exposed pre­senting proteins collectively known as major histo-compatibility complex (MHC) molecules. The MHC/peptide complexes on the surface of the APC are subsequently screened for “foreignness” through their ability to bind to pathogen-specific T cell receptors present on the T cell. When a T cell recognizes such antigenderived peptides as foreign, it becomes activated, it divides, and it attacks and kills cells which express the same peptide.Each T cell only expresses a single T cell receptor capable of recog­nizing only one specific antigen-derived peptide.

While this system is highly efficient in protecting the host from truly foreign microbes, most tumor cells unfortunately fail to express proteins or anti­gens deemed foreign by the host T cell immune repertoire. However, while tumor antigens are self-antigens, they may elicitimmune detection either through ectopic expression or over expression relative to the levels found in normal cells. Tumor cells which aberrantly express self-antigens in this manner may potentially be recognized and killed by host T cells. However, if tumor antigen recognition is an initial prereq­uisite for establishing immune response, it alone is insufficient for tumor eradication. Tumors cells have generated a wide range of mechanisms to eludeimmune responses including loss of tumor antigen expression, decrease in MHC expression, induction of premature tumor-specific T cell death (apoptosis) or unresponsiveness (anergy), as well as generalized T cell sup­pression through the secretion of specialized molecules (cytokines).

Therefore, for the successful application of T cell-based ther­apy of cancers, not only must a population of T cells exist which recognize the tumor, but these T cells must furthermore have the ability(i) to expand to clini­cally sufficient numbers,(ii) tomigrate specifically to the tumor site, and (iii) to mature into effector cells capable of killing the target tumor cell. In other words, for T cell-based anti-tumor therapy to succeed, one has to devise means of overcom­ing the multiple mechanisms whereby tumor cells naturally escape the detection of the host’s own immune system.

Investigators in Dr. Sadelain’s laboratory at Memorial Sloan Kettering Cancer Center (MSKCC) have proposed to gen-

erate tumor-specific T cells through the transfer of genes that encode tumor-targeted chimeric antigen receptors (CARs). CARs consist of a tumor antigen-binding domain, derived from a mouse mono-clonal antibody, fused to an intracellular signaling domain, derived from the T cell receptor, capable of activating T cells. Such CARs fuse antigen recogni­tion with signal transduction, two functions that are borne byseparate molecules in the physi­ological T cell receptor. This approach is advantageous sinceit allows investigators to easilygenerate CARs to a wide variety of tumor-associated antigens. Furthermore, the same CAR gene can be used to modify T cell from any patient bearingtumors which express the target­ed antigen. Finally, since these artificial receptors function in both the CD4 + (helper) and CD8 + (cytotoxic) T cell sub-sets, transduction of patient T cells with CARs could gener­ate both helper and cytotoxictumor-specific T cells which, in theory, would result in a moresustained anti-tumor T cell response.

non-Hodgkins lymphomas(NHL). Engineered peripheralblood T cells can be targeted tothese B cell malignancies through the expression of a CD19-specific CAR (termed 19z1). These engineered T cellsmay subsequently be expandedby co-culture on artificial anti­gen presenting cells (AAPCs)derived from mouse NIH 3T3fibroblasts engineered to expressboth the CD19 and CD80 anti­gens with the addition of the cytokine IL- 15 to the culturemedium. This approach has allowed investigators to generateclinically significant numbers of

tumor-targeted T cells which retain the ability to kill CD19-expressing tumor cells. When injected into immune-sup­pressed (SCID-Beige) mice, these engineered T cells are able to eradicate established Bcell tumors as demonstrated bypositron emission tomography (PET) scanning (Fig 1. PET scans of SCID mouse: a. tumorfree mouse showing uptake in the brain (Br), heart (H), and excretion through the kidney (K); b. mouse 21 days after injection with tumor cells, showing uptake of tumor cellsin the bone marrow of the verte­bral column (V) and Calvarium

(C); c. mouse 6 months after 19z1 T cells injection showingno detectable uptake in the bonemarrow). Overall, 50% of mice treated with a single dose of modified T cells were cured (Fig. 2a), while treatment for 2 consecutive days with these T cells resulted in a 75% long-term(>300days), tumor free survival (Fig 2b).

Dr. Sadelain’s work is signifi­cant in that it defines the requirements for activation andexpansion of genetically modi­fied tumor-specific T cells to eradicate tumor in SCID mice,thereby expanding the potentialscope of adoptive T cell therapyin clinical setting. Dr. Sadelainand his team are currently in the process of planning a clinical

trial to test the safety and efficacy of this approach in patients with CLL.

Viviane Martin, PhD

Office of Industrial Affairs Memorial Sloan Kettering Cancer Center

martinv@mskc c.org