Fecal Hemoglobin Levels From Negative FITs Signal CRC Risk

The risk of detecting colorectal cancer (CRC) increases by up to 13-fold in the presence of prior fecal hemoglobin (f-Hb) concentrations in fecal immunochemical tests (FIT), especially negative ones, according to a large international dose-response meta-analysis.

Although the association with neoplasia decreased as f-Hb levels rose, the findings support the development of risk-stratified screening strategies based on these concentrations, according to researchers led by Danica M.N. van den Berg, MSc, a PhD candidate and econometrics researcher in the Department of Public Health at Erasmus MC, University Medical Center in Rotterdam, the Netherlands.

Higher f-Hb concentrations in prior negative screening tests are strongly associated with an increased risk of detecting colorectal neoplasia in subsequent screenings, van den Berg said in an interview. “Gastroenterologists and other clinicians should consider the value of f-Hb concentrations in refining screening protocols and personalizing patient care to detect colorectal neoplasia earlier and more accurately.”

Published in Gastroenterology, the study was prompted by prior research showing individuals with f-Hb concentrations just below the positivity cutoff had an elevated CRC risk vs those with low or no f-Hb. “However, global variations in FIT positivity cutoffs and f-Hb category definitions complicated cross-study comparisons,” van den Berg said. Given the lack of an established dose-response relationship, the study aimed to clarify how f-Hb levels in previous screenings correlate with colorectal neoplasia detection. “Understanding this relationship is crucial for developing risk-stratified colorectal cancer screening strategies based on prior FIT results, which could improve the harm-benefit balance of screening,” she said.

According to van den Berg, f-Hb concentrations could help determine optimal CRC screening intervals by identifying higher-risk individuals who could benefit from more frequent testing, while those with lower concentrations could be screened less frequently.

Study Details

The systematic review and meta-analysis are the first to focus on the dose-response relationship between f-Hb levels in prior FIT screenings and colorectal neoplasia detection, van den Berg said. It included 13 ethnically diverse studies published during 2011-2023 with 4,493,223 individuals from Spain, France, the Netherlands, Taiwan, Denmark, Scotland, Ireland, Korea, Italy, and Norway. Most studies were cohort-based, and one was a randomized controlled trial.

All studies demonstrated a positive association between f-Hb in previous screenings and colorectal neoplasia detection. Almost all reported the f-Hb concentration measured in the prior screening round, while one study combined the f-Hb concentration of two previous screening rounds by using the cumulative f-Hb value. There was, however, wide variability in the stool positivity cut-offs in the included studies, ranging from 10 μg f-Hb/g to 80 μg f-Hb/g.

With an overall effect size of 0.69 (95% CI, 0.59-0.79), pooled analysis revealed that in the next screening round, individuals with f-Hb concentrations in stool of 5, 10, 20, and 40 μg/g had a threefold, fivefold, eightfold, and 13-fold higher risk for colorectal neoplasia, respectively, vs individuals showing 0 μg/g. Although there was significant study heterogeneity (I² = 97.5%, P

“This is a strategy worth pursuing and evaluating in the United States,” said gastroenterologist Theodore R. Levin, MD, a research scientist at Kaiser Permanente Division of Research in Northern California, commenting on the study but not involved in it. “However, there is no currently available FIT brand in the US that reports f-Hb concentration. All FITs in the US report as a qualitative positive-negative result.”

The Dutch investigation aligns with prior studies demonstrating a positive association between f-Hb concentrations in previous screenings and the detection of colorectal neoplasia. “Our working hypothesis was that risk increases in a decreasing manner as f-Hb concentrations rise, and the findings supported this hypothesis,” van den Berg said.

Other research has projected f-Hb level risk stratification to be effective and perhaps cost-effective in reducing delayed diagnosis of CRC.

Feasibility of Implementation

In large national screening programs in Europe, Asia, and Australia, as well as those of Kaiser Permanente and the Veterans Health Administration in the United States, information on f-Hb concentrations is already available.

“Therefore, incorporating an Hb-based approach should be relatively easy and affordable,” van den Berg said, and may help to optimize resource use while maintaining high detection rates. “However, the more critical question is whether such an approach would be acceptable to the target population.” To that end, randomized controlled trials in Italy and the Netherlands are offering tailored invitation intervals based on prior f-Hb concentrations and may provide insight into the real-world application of risk-stratified screening.

Among the many variables to be considered in the context of population-wide screening are cost-effectiveness, acceptability, and practicality, as well as invitation intervals, positivity cut-off levels, and start and stop ages for screening. “A key focus will be understanding the acceptability of risk-stratified colorectal cancer screening based on f-Hb among the target population and addressing any information needs they may have, as these are critical factors for successful implementation,” said van den Berg. Her group is currently studying the most effective and cost-effective risk-based strategy for CRC screening based on f-Hb levels.

The authors cautioned that since individuals with undetectable f-Hb levels make up the majority of those with negative FIT results, care must be taken that reducing screening frequency for this low-risk group does not lead to unfavorable outcomes at the population level.

This study was funded by the Dutch Organization for Scientific Research, which had no role in study design, data collection, analysis, interpretation, or writing. 

The authors declared no competing interests. Levin disclosed no competing interests relevant to his comments. 

Diana Swift is an independent medical journalist based in Toronto, Ontario, Canada.