Development and external validation of a faecal immunochemical test-based prediction model for colorectal cancer detection in symptomatic patients

Abstract

BackgroundRisk prediction models for colorectal cancer (CRC) detection in symptomatic patients based on available biomarkers may improve CRC diagnosis. Our aim was to develop, compare with the NICE referral criteria and externally validate a CRC prediction model, COLONPREDICT, based on clinical and laboratory variables.MethodsThis prospective cross-sectional study included consecutive patients with gastrointestinal symptoms referred for colonoscopy between March 2012 and September 2013 in a derivation cohort and between March 2014 and March 2015 in a validation cohort. In the derivation cohort, we assessed symptoms and the NICE referral criteria, and determined levels of faecal haemoglobin and calprotectin, blood haemoglobin, and serum carcinoembryonic antigen before performing an anorectal examination and a colonoscopy. A multivariate logistic regression analysis was used to develop the model with diagnostic accuracy with CRC detection as the main outcome.ResultsWe included 1572 patients in the derivation cohort and 1481 in the validation cohorts, with a 13.6 % and 9.1 % CRC prevalence respectively. The final prediction model included 11 variables: age (years) (odds ratio [OR] 1.04, 95 % confidence interval [CI] 1.02–1.06), male gender (OR 2.2, 95 % CI 1.5–3.4), faecal haemoglobin ≥20 μg/g (OR 17.0, 95 % CI 10.0–28.6), blood haemoglobin <10 g/dL (OR 4.8, 95 % CI 2.2–10.3), blood haemoglobin 10–12 g/dL (OR 1.8, 95 % CI 1.1–3.0), carcinoembryonic antigen ≥3 ng/mL (OR 4.5, 95 % CI 3.0–6.8), acetylsalicylic acid treatment (OR 0.4, 95 % CI 0.2–0.7), previous colonoscopy (OR 0.1, 95 % CI 0.06–0.2), rectal mass (OR 14.8, 95 % CI 5.3–41.0), benign anorectal lesion (OR 0.3, 95 % CI 0.2–0.4), rectal bleeding (OR 2.2, 95 % CI 1.4–3.4) and change in bowel habit (OR 1.7, 95 % CI 1.1–2.5). The area under the curve (AUC) was 0.92 (95 % CI 0.91–0.94), higher than the NICE referral criteria (AUC 0.59, 95 % CI 0.55–0.63; p < 0.001). On the basis of the thresholds with 90 % (5.6) and 99 % (3.5) sensitivity, we divided the derivation cohort into three risk groups for CRC detection: high (30.9 % of the cohort, positive predictive value [PPV] 40.7 %, 95 % CI 36.7–45.9 %), intermediate (29.5 %, PPV 4.4 %, 95 % CI 2.8–6.8 %) and low (39.5 %, PPV 0.2 %, 95 % CI 0.0–1.1 %). The discriminatory ability was equivalent in the validation cohort (AUC 0.92, 95 % CI 0.90–0.94; p = 0.7).ConclusionsCOLONPREDICT is a highly accurate prediction model for CRC detection.Electronic supplementary materialThe online version of this article (doi:10.1186/s12916-016-0668-5) contains supplementary material, which is available to authorized users.

The fecal hemoglobin concentration, age and sex test score: Development and external validation of a simple prediction tool for colorectal cancer detection in symptomatic patients

“…The development and external validation of the multivariable prediction model was designed according to the TRIPOD guidelines . All diagnostic accuracy studies from which data were derived for this work were performed according to the STARD guidelines, as documented in the relevant publications …”

“…Quantitative FIT allow for assessment of fecal hemoglobin concentration (f‐Hb). There are several prediction models in asymptomatic individuals for CRC screening based on f‐Hb, and a CRC prediction model in symptomatic patients including f‐Hb has been recently developed and externally validated . This prediction model was shown to have high diagnostic accuracy and discriminated between three risk groups, one of them with a negligible risk for CRC.…”

“…In consequence, we carried out a derivation and external validation multivariable prediction model study, based on the hypothesis that a simple to calculate risk score, based on only three readily available variables—f‐Hb, age and sex—could simplify the evaluation of symptomatic patients and provide a easy to use tool for the risk stratification of these to facilitate prioritisation of referrals for endoscopy . In order to perform this study on the FAST Score, detailed data obtained in five previously peer‐reviewed and published studies evaluating the diagnostic accuracy of FIT in 5,548 symptomatic patients referred for colonoscopy were used …”

Setting up a service for a faecal immunochemical test for haemoglobin (FIT): a review of considerations, challenges and constraints

“…Moreover, f-Hb increases as deprivation increases 13 14. One potential solution that takes these and other relevant variables into consideration in combination with the f-Hb concentration are risk-scoring models such as the very simple Score Card,15 the very complex COLONPREDICT model16 and the FAST score, which simply takes f-Hb, age and sex into account 17. However, their use in routine practice has yet to be evaluated in depth, but this sort of approach might simplify the interpretation of FIT results among GP considering making a referral to secondary care, particularly since the scores could be reported using the functionality available in most laboratory information systems, perhaps along with interpretative guidance.…”

Reduction of faecal immunochemical test false‐positive results using a signature based on faecal bacterial markers

“…FIT analysis was performed at Complexo Universitario de Ourense employing the same sample used in the CRC‐specific biomarkers analysis. Stool samples for faecal haemoglobin determination were analysed using the OC‐Sensor tube collector and the assay was performed using the automated OC‐Sensor, which detects gastrointestinal bleeding associated with disorders such as CRC, polyps and diverticulitis (Eiken Chemical Co., Tokyo, Japan) . Positive tests were those with a concentration of faecal haemoglobin equal or higher than 100 ng/mL (20 µg Hb/g of faeces; FIT100).…”