Le Peng1,2, Yesilda Balavarca3, Tobias Niedermaier1,2, Korbinian Weigl1,4, Michael Hoffmeister1, Hermann Brenner1,3,4. 1. Division of Clinical Epidemiology and Aging Research, German Cancer Research Center (DKFZ), Heidelberg, Germany. 2. Medical Faculty Heidelberg, Heidelberg University, Heidelberg, Germany. 3. Division of Preventive Oncology, National Center for Tumor Diseases (NCT) and German Cancer Research Center (DKFZ), Heidelberg, Germany. 4. German Cancer Consortium (DKTK), German Cancer Research Center (DKFZ), Heidelberg, Germany.
Abstract
INTRODUCTION: Fecal immunochemical tests (FITs) for hemoglobin are increasingly used in colorectal cancer (CRC) screening. The use of uniform positivity thresholds (cutoffs) within screening populations is expected to imply lower positive predictive values (PPVs) and higher numbers of colonoscopies needed (numbers needed to scope [NNSs]) to detect advanced neoplasms among screening participants at lower risk compared with those at higher risk. We aimed to assess such variation and its potential implications in a large screening cohort. METHODS: A quantitative FIT (FOB Gold; Sentinel Diagnostics, Milan, Italy) was conducted in fecal samples collected by 4,332 participants of screening colonoscopy before bowel preparation. Participants were classified into 3 risk groups (low, medium, and high) by tertiles of a previously derived risk-factor-based risk score. We determined the variation of PPVs and NNSs for detecting advanced neoplasms (i.e., CRC or advanced adenoma) when using the same FIT cutoffs and variation of FIT cutoffs that would yield uniform PPVs across risk groups. RESULTS: When a fixed FIT cutoff of 10 μg/g was used, the PPV increased from 23.3% to 41.8% from the low- to the high-risk group, with NNS decreasing from 4.3 to 2.4 (P < 0.001). Similar variations of PPVs and NNSs across risk groups were observed at higher FIT cutoffs. When risk group-specific cutoffs were defined to achieve fixed PPVs of 25%, 30%, and 35% across all risk groups, cutoffs varied from 5.3 to 11.4, 6.5 to 18.7, and 7.5 to 31.0 μg hemoglobin/g feces, respectively, between high- and low-risk groups (P < 0.05 for all differences). DISCUSSION: Using risk-adapted cutoffs may help to achieve target levels of PPV and NNS and might be an option to consider for personalized FIT-based CRC screening.
INTRODUCTION: Fecal immunochemical tests (FITs) for hemoglobin are increasingly used in colorectal cancer (CRC) screening. The use of uniform positivity thresholds (cutoffs) within screening populations is expected to imply lower positive predictive values (PPVs) and higher numbers of colonoscopies needed (numbers needed to scope [NNSs]) to detect advanced neoplasms among screening participants at lower risk compared with those at higher risk. We aimed to assess such variation and its potential implications in a large screening cohort. METHODS: A quantitative FIT (FOB Gold; Sentinel Diagnostics, Milan, Italy) was conducted in fecal samples collected by 4,332 participants of screening colonoscopy before bowel preparation. Participants were classified into 3 risk groups (low, medium, and high) by tertiles of a previously derived risk-factor-based risk score. We determined the variation of PPVs and NNSs for detecting advanced neoplasms (i.e., CRC or advanced adenoma) when using the same FIT cutoffs and variation of FIT cutoffs that would yield uniform PPVs across risk groups. RESULTS: When a fixed FIT cutoff of 10 μg/g was used, the PPV increased from 23.3% to 41.8% from the low- to the high-risk group, with NNS decreasing from 4.3 to 2.4 (P < 0.001). Similar variations of PPVs and NNSs across risk groups were observed at higher FIT cutoffs. When risk group-specific cutoffs were defined to achieve fixed PPVs of 25%, 30%, and 35% across all risk groups, cutoffs varied from 5.3 to 11.4, 6.5 to 18.7, and 7.5 to 31.0 μg hemoglobin/g feces, respectively, between high- and low-risk groups (P < 0.05 for all differences). DISCUSSION: Using risk-adapted cutoffs may help to achieve target levels of PPV and NNS and might be an option to consider for personalized FIT-based CRC screening.
Authors: Berbel L M Ykema; Tanya M Bisseling; Manon C W Spaander; Leon M G Moons; Dorien van der Biessen-van Beek; Lisette Saveur; Martijn Kerst; Sasja F Mulder; Ronald de Wit; Danielle Zweers; Gerrit A Meijer; Jos H Beijnen; Iris Lansdorp-Vogelaar; Flora E van Leeuwen; Petur Snaebjornsson; Monique E van Leerdam Journal: BMC Gastroenterol Date: 2021-02-12 Impact factor: 3.067
Authors: Hanyue Ding; Jiaye Lin; Zijun Xu; Xiao Chen; Harry H X Wang; Liwen Huang; Junjie Huang; Zhijie Zheng; Martin C S Wong Journal: Cancers (Basel) Date: 2022-02-21 Impact factor: 6.639