COLOFIT and colorectal cancer

• COLOFIT is a risk-prediction model that incorporates faecal immunochemical test (FIT) results (faecal haemoglobin, f-Hb), age, sex, platelet count (PLT), and mean corpuscular volume (MCV) to estimate individual colorectal cancer (CRC) risk in symptomatic patients presenting under NICE NG12 criteria (1,2,3)
• Consensus Statement Guidance (2026 update, endorsed by ACPGBI, BSG, RCGP, RCPath, PCGS, IBMS, and ALM): COLOFIT 1% risk of CRC replaces the binary f-Hb 10 g/g threshold for urgent suspected cancer (USC) lower gastrointestinal (LGI) pathway referral in adults with NICE NG12 symptoms (3)

Primary Care Safety Netting and Referral Pathways (3)

• digital rectal examination (DRE) and immediate LGI USC referral: referring clinicians must perform a DRE when investigating possible CRC symptoms under NICE NG12 criteria; any patient presenting with an anal or rectal mass or anal ulceration must be referred on an LGI USC pathway immediately without waiting for FIT or blood test results
• target populations for COLOFIT: for all other symptomatic patients without a palpable anal or rectal mass, clinicians should request a FIT kit and full blood count (to obtain MCV and platelet count) to calculate the COLOFIT risk score
• electronic safety netting: robust electronic safety netting systems should be established to ensure patients return their FIT kit within an appropriate time frame
• clinical shared decision-making: once results are available, clinicians discuss the calculated COLOFIT score ( 1% triggering LGI USC referral for colonoscopy or colonic imaging) with the patient, accounting for patient context and comorbidities; discussions and risk scores are formally documented using SNOMED CT code 3000321000000102 (COLOFIT colorectal cancer risk score)

Model Performance and Diagnostic Impact

• baseline FIT limitations: standard referral thresholds (FIT 10 g/g) achieve ~90% sensitivity for CRC, but ~10 in 11 referred patients do not have cancer
• referral reduction: in derivation and internal validation cohorts, COLOFIT yielded a 4.6% to 20.2% reduction in referrals (1); external validation in the Oxford-FIT dataset (n=51,477) demonstrated an overall 8% reduction in referrals (ranging from a 23% reduction to a 2% increase depending on time period) (2)
• enhanced specificity and risk of reduced sensitivity: COLOFIT enhances specificity over FIT alone; however, external validation shows that using COLOFIT without local calibration in populations with evolving demographics or differing symptom profiles risks miscalibrating thresholds (observed/expected ratio 1.52), potentially missing a small proportion of cancers (estimated 1–3%) compared to standard FIT 10 g/g alone (2)

Implementation Criteria and Risk Mitigation

• pre-implementation prerequisites: guidance permits implementation without prior local validation only if local FIT positivity ( 10 g/g) is 17%, local CRC detection rate is 1.3–1.6%, and compatible FIT assays are used (3)
• local validation & revalidation: systems should validate locally where feasible (2); continuous service evaluation (monitoring cancer detection, endoscopy volume, and auditing patients with FIT 10 g/g but COLOFIT < 1% who develop CRC) is required to dynamically adjust risk thresholds (3)

References:

1. Crooks CJ, et al. COLOFIT: Development and Internal-External Validation of Models Using Age, Sex, Faecal Immunochemical and Blood Tests to Optimise Diagnosis of Colorectal Cancer in Symptomatic Patients. Aliment Pharmacol Ther. 2025;61(5):852-864.
2. Tamm A, et al. External validation of the COLOFIT colorectal cancer risk prediction model in the Oxford-FIT dataset: the importance of population characteristics and clinically relevant evaluation metrics. BMC Med. 2025;23(1):503.
3. COLOFIT: a consensus statement to support implementation. Endorsed by the Association of Coloproctology of Great Britain and Ireland (ACPGBI), British Society of Gastroenterology (BSG), Royal College of General Practitioners (RCGP), Royal College of Pathologists (RCPath), Primary Care Gastroenterology Society (PCGS), Institute of Biomedical Science (IBMS), and Association for Laboratory Medicine (ALM). Published October 2026.