risk (chance) of missing colorectal cancer if symptoms plus a negative FIT test

Last edited 09/2023 and last reviewed 10/2023

Explaining the statistics with respect to the use of FIT for assessment of patients with symptoms suggestive of possible colorectal cancer

The Faecal Immunochemical Test (FIT) is used to detect traces of blood in faeces caused by bleeding in the lower GI tract

  • a raised level may be a marker of colorectal cancer (CRC) and is recommended in the management of patients with lower GI symptoms to identify who to refer for urgent investigations
  • for simplicity, a fHb (faecal haemoglobin) result >= 10 ug/g is considered a "positive FIT" and fHb <10 ug/g as "negative"
    • this is not strictly correct as detectable fHb <10 can be present in a "negative" result, but national and most local guidelines have recommend >= 10 ug/g as the threshold for an abnormal result
BSG/ACPGBI guidelines published in May 2022 recommend that FIT be used in all patients presenting with clinical features that may be due to CRC, except those with an abdominal or anorectal mass, or anal ulceration who need immediate referral. Importantly, the guidelines recommend that FIT be used in patients:
  • Who meet the existing NICE NG12 criteria for urgent referral;
  • With rectal bleeding (note the test detects a breakdown haemoglobin product rather than fresh blood, however advise people to take the sample from a non-bloody part of the stool);
  • With iron deficiency anaemia

How good is FIT at detecting colorectal cancer in people with symptoms?

  • on published evidence a FIT >10 ug/g identifies between 85-94% (sensitivity of 85-94%) of people with colorectal cancer (CRC) who present symptomatically (2)
      • sensitivity of a test = true positives / true positives plus false negatives (i.e. what proportion of patients with a specific condition that is being tested for will be identified via a positive test)
    • 10-15% of people with CRC who present symptomatically will have a negative FIT
    • to give this context then this can be compared the sensitivity of pre-FIT diagnostic assessment, which relied heavily on assessment against NICE NG12 clinical criteria
      • positive predictive value (PPV) measures the chance that a positive test or presence of a symptom/s indicates that the disease in question is present. PPV of FIT for CRC is substantially greater than any individual symptom specified in NICE NG12
        • in the NICE FIT study, of all patients referred urgently based on clinical presentation, CRC was detected in 3.3% of patients; in those with a FIT >10 ug/g this was 16% (3)

Use of negative predictive value in explanation of risk of CRC if negative FIT:

  • negative predictive value (NPV) = True negatives/ True negatives + False negatives
    • the quoted NPV for use of FIT in patients satisfying NG12 (2 week wait lower GI referrals) is 99% plus
      • retrospective cohort studies conducted in primary care have reported a cancer miss rate of 7 cases per 10,000 negative FIT at a threshold >10 ug/g, giving the test a negative predictive value in excess of 99% (5)
    • the significance of using NPV in this clinical scenario is that the NG12 pathway is based on a 3% prevalence of cancer in referrals from primary care
      • means that the True negatives is assumed to be 97%
      • the higher the value of True negatives in a population, the higher the NPV - no matter how effective a test is at identifying at identifying a condition (in this case a colorectal cancer). The prevalence of colorectal cancer in this referral population means the "worst" level of NPV that could be identified is, based on assumptions of NG12 guidance, 97% when it does not identify any cancers in the referral population (True negatives (97)/ True negatives (97)+False negatives (3))

How do symptoms affect the risk of identifying CRC if a negative FIT?

  • use of FIT >10 ug/g (positive FIT) plus (negative FIT and iron deficiency (IDA)) as a methodology to identify patients to investigate for CRC
    • a small study examined a cohort of referred patients in who 48 CRCs were detected of which 7/48 (14.6%) had fHb<10 ug/g (3)
      • of the 7 FIT negative CRCs, five had anaemia as well as change in bowel habits and of these, four had true IDA
      • so if a strategy to identify CRC has been to investigate mandatorily all patients with either
          • FIT >10 ug/g (positive FIT) or
          • negative FIT and iron deficiency (IDA)
        • 93.75% of CRC would have been identified (sensitivity of 93.75%)
        • 6.25% of CRC would not have been identified for mandatory investigation
    • Lazlo et al evaluated whether faecal immunochemical testing (FIT) can rule out colorectal cancer (CRC) among patients presenting with 'high-risk' symptoms requiring definitive investigation (4)
      • concluded that "1 in 6 or 1 in 8 patients with cancer respectively would have been missed. If the absence of anaemia or abdominal pain is used alongside f-Hb 10 ug/g, only 1 in 18 cancers would be missed but 56% of people without CRC could potentially avoid further investigations including colonoscopies"
        • 12.5% to 16% of CRC will be "missed" if uses a FIT of 10 but this will go down to 5.5% will be missed if you scope all who have iron deficient anaemia (so the sensitivity would increase (from 84-87.5%) to 94.5%)
        • 1 in 18 cancers would be "missed" using the criteria of mandatorily investigating all patients with either:
          • FIT >10 ug/g (positive FIT) or
          • negative FIT and iron deficiency (IDA) or
          • negative FIT and abdominal pain

Duplicate faecal immunochemical testing in patients at risk of colorectal cancer (6)

  • in this study:
    • all adult patients referred from primary care with suspected CRC with low-risk symptoms defined by NICE DG30 criteria were asked to complete two FIT samples on different bowel motions between August 2017 and June 2020 prior to clinical assessment in secondary care. After June 2020 all patients, including those with higher risk symptoms defined by NICE NG12 criteria, were included to facilitate decision-making and risk stratification for colonic imaging by hospital specialists during the SARS-CoV-2 pandemic. Some Trusts included in the study do test patients with FIT if they have rectal bleeding, whereas other Trusts do not. Patients with rectal bleeding were not delayed in their referral as urgent referral was made simultaneously to FIT requests
    • study results:
      • 2% of patients with colorectal cancer had two negative FIT tests (7/319)
      • 91 % of patients with colorectal cancer had two positive FIT tests (290/319)
      • 7 % of patients had one out of two FIT tests as positive (22/319)
      • i.e. only 2% of colorectal cancers in the study had two negative FIT tests at a level of 10ug/gm

Summary:

  • FIT is an excellent test for assessing a patient’s risk of having colorectal cancer. A patient with a FIT<10ug/g fHb has a less than 1% of risk of having colorectal cancer and clinicians should consider alternative diagnoses and investigatory pathways. However, this low risk partly reflects the rarity of colorectal cancer in the population and importantly 10-15% of patients with CRC have FIT<10ug/g. Therefore, if there are additional reasons for suspecting a CRC may be present it is reasonable to refer for exclusion of this diagnosis. Similarly, a patient with a FIT<10ug/g fHb and no iron deficiency has a less than 0.1% of risk of having colorectal cancer. However because of the rarity of CRC in the population, this represents that about 5% of CRC will have a FIT < 10ug/g and not have iron deficiency anaemia

  • FIT is, despite imperfections, the best method available to primary care clinicians to assess a patient's need for urgent investigation to exclude CRC. Those with a FIT >10 ug/g should be strongly considered for referral via an urgent suspected lower GI cancer pathway

  • two negative FIT tests
    • evidence suggests only 2% of CRC had two negative fit tests at a level of 10ug/gm

  • NICE have stated that (7):
    • Quantitative faecal immunochemical testing (FIT) using HM-JACKarc or OC-Sensor is recommended to guide referral for suspected colorectal cancer in adults
      • with an abdominal mass, or
      • with a change in bowel habit, or
      • with iron-deficiency anaemia, or
      • aged 40 and over with unexplained weight loss and abdominal pain, or
      • aged under 50 with rectal bleeding and either of the following unexplained symptoms:
        • abdominal pain
        • weight loss, or
      • aged 50 and over with any of the following unexplained symptoms:
        • rectal bleeding
        • abdominal pain
        • weight loss, or
      • aged 60 and over with anaemia even in the absence of iron deficiency

      • FIT should be offered even if the person has previously had a negative FIT result through the NHS bowel cancer screening programme. People with a rectal mass, an unexplained anal mass or unexplained anal ulceration do not need to be offered FIT before referral is considered

      • refer adults using a suspected cancer pathway referral (as outlined in NICE's guideline on suspected cancer) for colorectal cancer if they have a FIT result of at least 10 micrograms of haemoglobin per gram of faeces

      • for people who have not returned a faecal sample or who have a FIT result below 10 micrograms of haemoglobin per gram of faeces:
        • safety netting processes should be in place
        • referral to an appropriate secondary care pathway should not be delayed if there is strong clinical concern of cancer because of ongoing unexplained symptoms (for example, abdominal mass)
    • NICE committee note "...is a lack of evidence on using dual FIT in primary care, using FIT in people aged under 40, and using FIT in people who have conditions or medicines that increase the risk of gastrointestinal bleeding.."

Reference:

Related pages:

faecal immunochemical tests (FIT) for hemoglobin and detection of bowel cancer