In Vitro Preclinical Study Results for a Phase I Adoptive Immunotherapy Trial Using Cetuximab in Association with Allogeneic NK Cell Infusion in KRAS Oncogene Mutated Metastatic Colorectal Cancer Patients.

Prognosis for metastatic colorectal-cancer (mCRC) patients is unfavorable, median survival from diagnosis varying between sixteen and twenty-four months. Numerous clinical trials have demonstrated a clinical advantage for patients treated with a combination of chemotherapy and cetuximab. However recent retrospective evidence from several randomized studies, has established that mCRC patients with tumors harboring a mutation in the KRAS gene do not derive benefit from the administration of cetuximab. We hypothesized that KRAS mutated CRC cells may be susceptible to cetuximab-induced ADCC mediated by healthy donor NK cells.

7 different CRC cell lines (DLD 1, HCT-116, HT-29, LOVO, SW620, SW48, SW480) were analyzed for EGFR membrane expression levels, tested for KRAS mutational status and MHC class I-II expression. CD56+CD3- NK cells were purified from volunteer healthy donors peripheral blood mononuclear cells, by a two-steps CD3+ cells negative selection followed by CD56+ cells positive selection by MACS cell separation columns (Miltenyi biotec). Purified NK cells were incubated overnight with medium alone or stimulated with IL-2 (100U/ml) or IL-15 (20U/ml). CRC cell lines, were labeled with 51Cr overnight, and used as target cells for NK assay or incubated at 4°C for 30 minutes with Cetuximab 100 ug/ml for ADCC assay. Cytotoxic activity was evaluated in a 4-hour cytotoxicity assay, at an effector/target (E/T) ratio ranging between 100:1 and 2:1.

all but one (SW620) CRC cell lines express EGFR (range 40-90%) and MHC class I-II molecules; 4/7 harbor KRAS mutation (DLD 1, HCT-116, SW620, SW480). All CRC cell lines were susceptible to unstimulated allogeneic NK cells lysis (mean 20% ± 4,57, at ratio 25:1; mean 15% ± 3,65, at ratio 12:1) and lytic activity is significantly enhanced (p<0.05) by either IL-2 or IL-15 pre-activation (mean 48% ± 5,8, at ratio 25:1; mean 38% ± 4,3, at ratio 12:1, respectively). No difference in lytic activity was observed between IL-2 vs IL-15 pre-activation and between thawed vs fresh NK cells. SW48, KRAS wilde-type (EGFR expression 74%) and HCT-116 KRAS mutated (EGFR expression 61%) cell lines were chosen as targets for ADCC assay. When unstimulated NK cells were used as effectors, SW48 lysis increased from 38% without cetuximab pre-coating to 50% with cetuximab pre-coating (E/T ratio of 50:1, p<0.05). Similarly, HCT-116 lysis increased from 20% without cetuximab pre-coating to 36% with cetuximab pre-coating (E/T ratio of 50:1, p<0.05). When IL-2 or IL-15 pre-activated NK cells were used as effectors, both targets lysis increased from 50% without cetuximab pre-coating to 71% with cetuximab pre-coating (E/T ratio of 100:1, p<0.05). Pre-coating with cetuximab did not enhanced unstimulated or pre-activated NK lysis against SW620 (EGFR expression 0%).

allogeneic NK cells from different healthy donors are able to lyse all tested CRC lines (4 KRAS mutated). IL-2 and IL-15 activation significantly enhances NK cytotoxicity; moreover lytic activity of thawed and fresh NK cells is similar allowing the use of both populations in clinical trials. In ADCC assay NK cell lytic activity is significantly enhanced by cetuximab; surprisingly cetuximab mediated ADCC activity is independent from CRC cell lines KRAS oncogene mutational status and require EGFR membrane expression. Pre-coating with cetuximab also increase lytic activity of IL-2 and IL-15 activated NK cells. Further experiments are in progress to evaluate the susceptibility to the lysis of primary tumor cells derived from KRAS mutated and wild type CRC patients, in NK and ADCC assays in order to confirm data obtained against established CRC cell lines and to correlate cetuximab-mediated ADCC with EGFR membrane expression. These findings support a phase I study with allogeneic NK cells infusions in association with cetuximab and IL-2 in mCRC patients harboring a mutation in the KRAS gene.

No relevant conflicts of interest to declare.