Article Text

Download PDFPDF
Letter
Can the performance of a quantitative FIT-based colorectal cancer screening programme be enhanced by lowering the threshold and increasing the interval?
  1. Jayne Digby1,2,
  2. Callum G Fraser1,2,
  3. Frank A Carey3,
  4. Robert J C Steele1,2,4
  1. 1 Scottish Bowel Screening Research Unit, Ninewells Hospital and Medical School, Dundee, UK
  2. 2 Centre for Research into Cancer Prevention and Screening, University of Dundee, Dundee, UK
  3. 3 Department of Pathology, Ninewells Hospital and Medical School, Dundee, UK
  4. 4 Division of Cancer, Medical Research Institute, Ninewells Hospital and Medical School, Dundee, UK
  1. Correspondence to Jayne Digby, Scottish Bowel Screening Research Unit, Ninewells Hospital and Medical School, Dundee DD1 9SY, Scotland; jaynedigby{at}nhs.net

Statistics from Altmetric.com

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

We read with interest the work by Haug et al published in Gut.1 Longitudinal data from 4523 participants in the first round of a faecal immunochemical test for haemoglobin (FIT)-based screening programme, of whom 3427 also participated in the second round, were studied. In both first and second rounds, a threshold of 10 µg Hb/g faeces was used. The cohort was followed up for 2 years. The cumulative positivity and the number of participants diagnosed with neoplasia over the two rounds of screening were determined and compared with a hypothetical strategy involving single round screening with use of lower faecal haemoglobin concentration (f-Hb) thresholds and omission of the second round. It was suggested that lowering the f-Hb threshold and extending the screening interval could possibly enhance population-based screening programmes.

In our pilot evaluation of FIT-based screening in Scotland, a much higher f-Hb threshold (≥80 µg Hb/g faeces) was employed.2 Moreover, a much larger cohort of screened individuals (30 893) was available for study. Seven hundred and fifty-three participants with f-Hb ≥80 µg Hb/g were referred for colonoscopy. Of 30 140 with a negative screening test result, 27 890 that had participated were eligible to be invited for the next screening round (January 2011 to January 2013). Of 24 669 responders, 450 had a positive screening test result and were referred for colonoscopy. In the two rounds of screening studied, the first with quantitative FIT using a threshold of ≥80 µg Hb/g faeces and the second using the standard guaiac faecal occult blood test/FIT two-tier reflex algorithm used in Scotland,3 the positivity in the first round was 2.5%: there were 30 screen-detected cancers (SDCs) and 31 interval cancers (ICs).4 In the first round, 753 colonoscopies were performed. At the second round, there were 25 SDC, making a total of 55 SDC over two rounds.

Assuming that IC and colorectal cancer (CRC) detected at the subsequent screening round would have been present in some form during the initial screening round, the scenarios that would have eventuated by omitting the second screening round and using lower f-Hb thresholds are listed in table 1. The f-Hb threshold that would have given double the positivity rate of that found using ≥80 µg Hb/g faeces was identified as ≥28 µg Hb/g faeces.

Table 1

Alternative scenarios using lower f-Hb threshold for the first round and omitting second round

This f-Hb threshold would have generated the same number of colonoscopies in a single 4-year screening round as a threshold of ≥80 µg Hb/g faeces would have in two rounds of biennial screening. In this situation, which is colonoscopy resource neutral, three ICs and nine second round CRCs would have been detected during the first round of screening. It follows that the remaining 16 SDCs found in the second round would have become IC, or at least have been more advanced when detected at the next screening episode. This would have resulted in 42 SDCs diagnosed in a 4-year screening round, considerably fewer than the total of 55 SDCs with biennial screening, and the number of IC could have potentially increased from 31 to 44.

Therefore, the concept of lowering f-Hb cut-off and lengthening screening interval to improve test sensitivity without an increase in colonoscopy requirement is not supported by our data since a modest decrease in IC would be offset by the majority of second round SDC being missed. Other strategies such as the use of f-Hb concentration to determine the length of the subsequent screening interval for individual participants may be a better option when evolving to more intelligent use of FIT.

References

Footnotes

  • Contributors JD, CGF and RJCS designed, planned and conducted the work and produced the tables. FAC provided pathology data. All authors contributed to data interpretation and writing of the letter. All authors approved the final draft.

  • Funding Funded by the Chief Scientist Office of the Scottish Government Health Department. Project: Establishing a Bowel Screening Research Unit (Grant Ref: CZH/6/4).

  • Disclaimer The funder had no role in the study design, data collection, statistical analyses, interpretation of the result or writing of the manuscript.

  • Competing interests CGF undertook consultancy with Immunostics Inc, Ocean, New Jersey, USA, and Kyowa-Medex, Tokyo, Japan.

  • Ethics approval Ethical approval was not required for this study as it involved analysis of routinely collected screening data. The FIT pilot was approved by the Programme Board, Scottish Bowel Screening Programme as an evaluation of an established screening modality.

  • Provenance and peer review Not commissioned; internally peer reviewed.

  • Data sharing statement No further unpublished data are available to others.