15353-j-veluchamy

128 | Chapter 5 Discussion In order to test the cytotoxic potential of NK cells for treating advanced colorectal cancer patients, we compared their functional status before and after chemotherapy. We observed that peripheral blood NK cell numbers were reduced in mCRC patients and that residual NK cells were dysfunctional and unable to mount a strong effector response when stimulated with an NK cell sensitive tumor target. Though an increase in NK cell cytotoxicity was observed when tumor target cells were coated with the anti-EGFR mAb cetuximab, reflecting a capacity for ADCC, cytotoxicity was still significantly lower (both before and after chemotherapy) than that observed in healthy controls. These data indicate a decreased functional state of NK cells in patients with mCRC, which is in line with studies in mice where the cytokine production and anti-tumor activity of adoptively transferred NK cells were highly affected following long-term exposure to tumors 30 . Through recognition of MHC class I molecules KIRs prevent NK cells from targeting healthy cells while allowing them to detect tumor or infected cells with low or downregulated expression of MHC class I in a process known as “missing self” 31 . Severely diminished or aberrant expression of MHC class I has been reported in the majority of colorectal adenocarcinomas 32,33 , which makes them an ideal target for NK cell-mediated killing. Although NK cells are infrequent in colorectal tissues 18 , several independent studies investigated the clinical impact of NK cell infiltration on the prognosis of CRC, as well as in other types of carcinoma. These clinical studies, including a recent tissue microarray of 1414 CRC biopsies, led to the conclusion that NK cell infiltration in tumors correlated with better overall response rates and progression-free survival in CRC patients 34-37 , suggesting that therapies aimed at boosting NK cell functions could be beneficial in mCRC and possibly also in other types of cancer. We evaluated and compared the cytotoxic efficacy of two different sources of feeder cell free allogeneic NK cells, i.e. A-PBNK cells and in vitro expanded and differentiated UCB-NK cells. In vitro NK cell cytotoxicity experiments revealed that the cytotoxic activity of UCB- NK cells against CRC cells was significantly higher than that of A-PBNK cells and in addition demonstrated that, while an increase in cytotoxicity through ADCC was not evident with UCB-NK cells, their cytotoxic potential was still comparable to that observed with A-PBNK potentiated by cetuximab mediated ADCC. It is possible that the stronger cytotoxic effects of UCB-NK cells result from a more intense stimulation with cytokines in comparison to A-PBNK cells. The failure to observe ADCC-enhanced cytotoxicity with UCB-NK cells in vitro can be explained by their low expression levels of CD16 29 . As we previously observed in vivo up-regulation of CD16 on UCB-NK cells upon their transfer to NOD/SCID/IL2Rgnull (NSG) mice 38 , we decided to also test the efficacy of cetuximab treatment in combination with UCB-NK cells in an in vivo model. Treatment of SW480 RAS mut tumors in BRGS mice with UCB-NK cells, resulted in control of disease progression and translated into a significantly longer survival. As expected, cetuximab monotherapy did not result in a decreased SW480