To all those using, or planning to use, faecal immunochemical tests for haemoglobin (FIT) in bowel cancer screening programmes, there was some disturbing news last week. In British Columbia (BC), Canada, use of FIT ceased due to a problem with a reagent used in analysis of the samples. On the official website, it is stated that work to resolve this as quickly as possible is underway and the organisers will have a better idea of timing in the next few weeks: however, it could take a number of months for FIT analyses to resume. The details and consequences are very well documented.
In the media, it was stated the number of positive results suddenly rose in the summer from the typical 14% to over 20%, suggestive of too many false positives. Dr John Spinelli, vice-president of population oncology at the BC Cancer Agency is quoted as saying: “We have a dilemma. We have no choice but to suspend the tests. Some cancers may go undetected in the meantime and that is not good. But we need a test that is reliable…”
Was it necessary to cease the programme? The FIT used in BC seems to be a quantitative FIT and this means that numerical estimates of faecal haemoglobin concentration (f-Hb) are generated. As shown here, different f-Hb cut-offs are applied in other Provinces using the same analytical system.
Thus, if constancy of positivity is a major concern of the screening programme, then rises or falls could have been simply corrected through alteration of the f-Hb cut-off to give the desired percentage. “Would such a simple action have led to continuation of the programme while the root cause was determined?” seems a germane question, especially since confidence in the programme undoubtedly will have been considerably dented by cessation. Moreover, since f-Hb is directly related to colorectal disease severity, the higher the positivity, the greater the number of neoplastic lesions found, as well as a greater numbers of false positive results. However, higher positivity does place further burdens on often stretched colonoscopy resources. Further information on the details of the reason for the increased positivity in BC, the rationale behind cessation of the programme and the actions taken to recommence the programme are all awaited with considerable interest.
Unfortunately, suspension of a FIT-based screening programme due to a problem created by the manufacturer has happened before. In Australia, in 2009, it was noted that the positivity was lower than expected: 108,000 people who had undertaken a FIT and who had a negative result were asked to repeat the test with a new FIT.
So, does this cast doubt on the validity of all FIT used, or about to be used, in screening programmes? Fortunately, this does not seem very likely since FIT have been used all of the world for many years and there seems to have been no similar problems documented with the three FIT analytical systems in most common use. However, any analytical system used in a screening programme could be subject to failures in one or more components and ongoing vigilance is clearly mandatory.
Interestingly it was stated that the laboratories that handled the FIT detected the problem. Laboratories undertaking FIT analysis undoubtedly will have comprehensive quality control, assessment and assurance procedures in place, which would lead to early detection of analytical problems. In addition, analysis of data on outcomes, including positivity, are important markers of consistency of performance over time and are widely used as quality indicators with agreed targets for success. Use of objectively set targets and acceptable ranges for positivity, with warning limits that stimulate investigation of potential problems and action limits that lead to changes in the f-Hb cut-off or other curative actions, seem obvious necessary prerequisites to ensure the validity of a quality screening programme. It would be of considerable value to all programme organisers to see such targets and limits widely published, along with the actions taken when these were breached.