Search

Natural killer (NK) cells have long been considered as potential agents for adoptive cell therapy for solid cancer patients. Until today most studies utilized autologous NK cells and yielded disappointing results. Here we analyze various modular strategies to employ allogeneic NK cells for adoptive cell transfer, including donor-recipient HLA-C mismatching, selective activation and induction of melanoma-recognizing lysis receptors, and co-administration of antibodies to elicit antibody-dependent cell cytotoxicity (ADCC). We show that NK cell activation and induction of the relevant lysis receptors, as well as co-administration of antibodies yield substantial anti-cancer effects, which are functionally superior to HLA-C mismatching. Combination of the various strategies yielded improved effects. In addition, we developed various clinically-compatible ex vivo expansion protocols that were optimized according to fold expansion, purity and expression of lysis receptors. The main advantages of employing allogeneic NK cells are accessibility, the ability to use a single donor for many patients, combination with various strategies associated with the mechanism of action, e.g. antibodies and specific activation, as well as donor selection according to HLA or CD16 genotypes. This study rationalizes a clinical trial that combines adoptive transfer of highly potent allogeneic NK cells and antibody therapy.

Prostate cancer is the most common non-dermatologic malignancy in men in the Western world. Recently, a frequent chromosomal aberration fusing androgen regulated TMPRSS2 promoter and the ERG gene (TMPRSS2/ERG) was discovered in prostate cancer. Several studies demonstrated cooperation between TMPRSS2/ERG and other defective pathways in cancer progression. However, the unveiling of more specific pathways in which TMPRSS2/ERG takes part, requires further investigation. Using immortalized prostate epithelial cells we were able to show that TMPRSS2/ERG over-expressing cells undergo an Epithelial to Mesenchymal Transition (EMT), manifested by acquisition of mesenchymal morphology and markers as well as migration and invasion capabilities. These findings were corroborated in vivo, where the control cells gave rise to discrete nodules while the TMPRSS2/ERG-expressing cells formed malignant tumors, which expressed EMT markers. To further investigate the general transcription scheme induced by TMPRSS2/ERG, cells were subjected to a microarray analysis that revealed a distinct EMT expression program, including up-regulation of the EMT facilitators, ZEB1 and ZEB2, and down-regulation of the epithelial marker CDH1(E-Cadherin). A chromatin immunoprecipitation assay revealed direct binding of TMPRSS2/ERG to the promoter of ZEB1 but not ZEB2. However, TMPRSS2/ERG was able to bind the promoters of the ZEB2 modulators, IL1R2 and SPINT1. This set of experiments further illuminates the mechanism by which the TMPRSS2/ERG fusion affects prostate cancer progression and might assist in targeting TMPRSS2/ERG and its downstream targets in future drug design efforts.

NK cells are key players in anti tumor immune response, which can be employed in cell-based therapeutic modalities. One of the suggested ways to amplify their anti tumor effect, especially in the field of stem cell transplantation, is by selecting donor/recipient mismatches in specific HLA, to reduce the inhibitory effect of killer Ig-like receptors (KIRs). Here we suggest an alternative approach for augmentation of anti tumor effect of allogeneic NK cells, which is founded on profile matching of donor NK lysis receptors (NKLR) phenotype with tumor lysis-ligands.We show that an NKLR-mediated killing directly correlates with the NKLR expression intensity on NK cells. Considerable donor variability in the expression of CD16, NKp46, NKG2D and NKp30 on circulating NK cells, combined with the stability of phenotype in several independently performed tests over two months, indicates that NKLR-guided selection of donors is feasible. As a proof of concept, we show that melanoma cells are dominantly recognized by three NKLRs: NKG2D, NKp30 and NKp44. Notably, the expression of NKp30 on circulating NK cells among metastatic melanoma patients was significantly decreased, which diminishes their ability to kill melanoma cells. Ex vivo expansion of NK cells results not only in increased amount of cells but also in a consistently superior and predictable expression of NKG2D, NKp30 and NKp44. Moreover, expanded NK cultures with high expression of NKG2D or NKp30 were mostly derived from the corresponding NKG2D(high) or NK30(high) donors. These NK cultures subsequently displayed an improved cytotoxic activity against melanoma in a HLA/KIR-ligand mismatched setup, which was NKLR-dependent, as demonstrated with blocking anti-NKG2D antibodies.NKLR/NKLR-ligand matching reproducibly elicits enhanced NK anti-tumor response. Common NKLR recognition patterns of tumors, as demonstrated here in melanoma, would allow implementation of this approach in solid malignancies and potentially in hematological malignancies, either independently or in adjunction to other modalities.