目的：糞便潛血檢驗可做為篩檢大腸癌或瘜肉的指標之一。新一代糞便免疫化學檢查 (fecal immunochemical test；FIT)，主要以人類血紅素特異抗體作為反應方法學，因不受其 他動物的肉和血及蔬菜和藥物等影響，故特異性高。本研究評估新研發OC-Light S FIT(EIKEN)的糞便潛血免疫法之效能。 材料與方法：比較OC-SENSOR IO (EIKEN)，EASEmed ，OC-Light S FIT三種方法學。 (1) 以OC-SENSOR IO所定量的50ng/mL濃度，取50ng/mL以上及以下的檢體各20支，分別評 估EASEmed及OC-Light S FIT的準確度、敏感度、特異性，及一致性。並以kappa coefficient 統計比較一致性。(2) 驗證Cut-off值：依CLSI EP12-A2指引，取濃度60及40 ng/mL之病人檢 體各1支，以EASEmed及OC-Light S FIT重複分析20次。(3)干擾測試：分別以牛、豬、雞的 血液稀釋至濃度100 ng/mL，並分為兩組:第一組配製成含有EIKEN原廠校正品50 ng/mL，第 二組則不含校正品，再以EASEmed及OC-Light S FIT分別測定。 結果：(1)準確度：OC-Light S FIT與EASEmed的敏感度分別為89％及94％，特異性分別 為80％及50％。(2) 比較一致性：OC-Light S FIT與EASEmed 分別為0.68 及0.43。(3) 驗證 Cut-off值：OC-Light S FIT與EASEmed在40ng/ml皆出現反應線。(4)干擾測試：OC-Light S FIT與EASEmed都不受其他動物血液之干擾。 結論：OC-Light S FIT與EASEmed原廠宣告敏感度為50ng/ml，實際在濃度40ng/ml都出 現陽性反應。EASEmed在濃度30 ng/ml以上即出現明顯陽性反應，產生過高的偽陽性，導致 敏感度高而特異性較差。OC-Light S FIT的敏感度表現是較符合原廠之宣告，且與定量法之 一致性較佳。研究結果OC-Light S FIT可較準確作為糞便潛血篩檢用。目前國內糞便潛血檢 驗定性方法的敏感度為 50 ng/ml，皆高於衛福部國健署在大腸直腸癌防治篩檢使用免疫定 量法所採用的陽性閥值(100 ng/ml )，因此大腸癌篩檢定量結果為陰性時，定性結果卻可能 呈現陽性。建議檢驗室需通報臨床定性與定量分析方法之差異，以提升醫療品質及病人安 全。

Purpose: Fecal occult blood test can be used as one of the indicators for screening colorectal cancer or colorectal polyp. A new generation of fecal immunochemical test (FIT), mainly using human hemoglobin-specific antibodies as a chromatographic immunoassay (CIA) method, is highly specificity because it is not affected by meat, blood, vegetables and drugs of other animals. This study evaluated the efficacy of the newly developed OC-Light S FIT (EIKEN) fecal occult blood immunoassay. Material and Method: Compare OC-SENSOR IO (EIKEN), EASEmed and OC-Light S FIT three methodologies. (1) Taking 50 ng/mL concentration of OC-SENSOR IO, take 20 samples of above 50 ng/ mL and under 50 ng/ mL respectively, to evaluate the accuracy, sensitivity, specificity and of consistency EASEmed and OC-Light S FIT, respectively. And compare the consistency with kappa coefficient. (2) Verify Cut-off value: According to CLSI EP12-A2 guidelines, take one sample of each patient at a concentration of 60 and 40 ng/ml, and repeat analysis 20 times with EASEmed and OC-Light S FIT. (3) Interference test: diluted to 100 ng/mL in blood of cattle, pigs and chickens, and divided into two groups: the first group was prepared to contain 50 ng/mL of the original EIKEN calibrator, and the second group contained no calibrator. Determined by EASEmed and OC-Light S FIT. Results: (1)Accuracy: The sensitivity of OC-Light S FIT and EASEmed is 89% and 94%, respectively, and the specificities are 80% and 50%, respectively. (2) Consistency: OC-Light S FIT and EASEmed are 0.68 and 0.43 respectively. (3) Cut-off value: Both of OC-Light S FIT and EASEmed have reaction line at 40 ng/mL. (4) Interference: OC-Light S FIT and EASEmed are not interfered with by other animals' blood. Conclusions: Both of OC-Light S FIT and EASEmed manufacturers declaration sensitivity is 50ng/mL. Actually, there is a positive reaction at a concentration of 40 ng/mL. EASEmed showed a significant positive reaction at a concentration above 30 ng/mL, resulting in an excessively high false positive, so the sensitivity is higher but the specificity is poor. The Sensitivity of OC-Light S FIT meets manufacturer declaration, and has a better consistency with quantitative methods. OC-Light S FIT is more accurate as a fecal occult blood screening test. At present, the sensitivity of the qualitative method for fecal occult blood test is 50 ng/mL, which is higher than the positive threshold (100 ng/ml) adopted by the Ministry of Health and Welfare in the screening and detection of colorectal cancer. Therefore, when the quantitative result is negative, the qualitative result may be positive. It is recommended that the laboratory should inform the differences between qualitative and quantitative methods to the clinical, for improving medical quality and patient safety.