A253 measurement of natural killer cell activity (nka) in subjects undergoing colonoscopy:test performance of a new blood test at different cut-offs for the detection of colorectal cancer (crc)

Low NKA has been linked to a higher risk of cancer and has been reported in CRC patients.A novel simple blood test in vitro diagnostic device (IVDD) which measures NKA in a small volume of whole blood is now available. Tha aim of the study is to evaluate the test performance metrics of the IVDD at different cut-offs in subjects with CRC and adenomatous polyps (AP). This study measured NKA in 1081 subjects presenting for screening or prescribed colonoscopies using a biological assay performed as per the manufacturer’s directions. In the 872 evaluable subjects,statistically significant differences were found between the NKA of subjects positive for CRC (n=23), and that of subjects negative for CRC (n=849) [CRC mean 317.1 pg/ml (DS:845.5),CRC-negative mean 745.7 pg/ml (SD:1028.5),p=0.001; CRC median 86.0 pg/ml (IQR:43.3–151.0),CRC-negative median 298.1 pg/ml (IQR:100.4–920.2),p<0.001]. The prevalence of CRC was 2.6% and of AP >10 mm was 14.6%. Receiver Operator Characterisitcs (ROC) analysis show an optimum cut-off for detection of CRC at 181 pg/ml,with an area under the curve of 73% (p<0.0001). At cut-offs of 200 pg/ml, high sensitivity and negative predictive values for detection of CRC were seen. At cut-offs of 300 and 500 pg/ml, a slight improvement in the test performance was seen in the sensitivity for AP>10 mm. The odds ratio for the NKA IVDD for the detection of cancer at a cut-off of 200 pg/ml was 10.3 (95% CI 3.03–34.9). The clinical results in the present study using new simple blood test for measurement of NKA confirm that there are limited benefits of using different cut-offs other than the optimal cut-off (220 pg/ml) for this new IVDD as determined by ROC analysis for detection of CRC None