BARCELONA, Spain — Extending the interval between fecal immunochemical tests (FITs) from 2 to 3 years could reduce the burden of colorectal cancer (CRC) screening while maintaining similar rates of advanced neoplasia detection in individuals at low risk, according to findings from a large, randomized trial.

Although tailoring screening intervals according to individual risk did not improve overall detection of advanced neoplasia, the findings suggest that most people with a negative FIT could potentially undergo less frequent screening.

"I think over time, we all realized maybe we should aim for similar benefits in individuals while using fewer resources," said Esther Toes-Zoutendijk, PhD, assistant professor in the Department of Public Health at Erasmus MC University Medical Center, Rotterdam, Netherlands.

However, longer-term follow-up is needed to establish whether extending screening intervals is safe, she added.

Toes-Zoutendijk presented the findings here at United European Gastroenterology (UEG) Week 2026. 

Not All Negative FITs Are Equal 

Most FIT-based CRC screening programs use a predefined fecal hemoglobin (f-Hb) concentration threshold to identify individuals requiring colonoscopy. In the Netherlands, the positivity threshold is 47 µg Hb/g feces, with individuals below this threshold routinely invited for repeat screening after 2 years.

However, previous research has demonstrated that individuals with detectable f-Hb concentrations below the positivity threshold have a higher risk of advanced neoplasia at subsequent screening rounds than those with undetectable concentrations. A meta-analysis involving approximately 4.5 million negative FITs demonstrated a dose-response relationship between f-Hb concentration and subsequent advanced neoplasia risk.

PERFECT-FIT was designed to investigate whether this information could be used to personalize screening intervals, bringing forward testing for individuals at higher risk while extending intervals for those at very low risk.

"We aimed to actually increase yield by risk-stratifying screening intervals based on prior negative f-Hb concentrations," Toes-Zoutendijk told delegates at the meeting.

The investigators initially anticipated that personalized screening would increase advanced neoplasia detection from 1.7% to 2.2%.

Nationwide Randomized Trial 

The Dutch national CRC screening program invites adults aged 55-75 years to undergo FIT screening every 2 years. The current trial recruited individuals aged 57-72 years who had previously participated in the program. Between 2022 and 2023, 71,495 individuals were invited to participate in the trial, of whom 20,088 (28.1%) accepted.

Participants were randomized to either conventional screening at 2-year intervals (n = 10,211) or screening intervals determined by their previous negative FIT result (n = 9877; the intervention group).

In the intervention group, participants with undetectable f-Hb concentrations were assigned a 3-year interval; those with concentrations greater than 0-15 µg/g retained the standard 2-year interval; and those with concentrations greater than 15 but below 47 µg/g were screened after 1 year.

The distribution was heavily weighted toward the lowest-risk group: 9175 participants (92.9%) were assigned to 3-year screening; 350 (3.5%) to 2-year screening; and 352 (3.6%) to screening after just 1 year.

This was a higher proportion of low-risk individuals than investigators had anticipated. "I think we were a little bit surprised by the fact that there were many more people that had no detectable f-Hb," Toes-Zoutendijk said.

Detection Rates and Screening Participation 

Of the 20,088 randomized participants, 19,547 returned a FIT. The FIT positivity rate was significantly higher in the personalized screening group than in the conventional screening group (4.4% vs 3.7%; P = .03). The relative risk (RR) of a positive FIT was 1.17 (95% CI, 1.02-1.34).

However, this did not translate into a statistically significant increase in advanced neoplasia detection. Advanced neoplasia was detected in 93 of 9584 participants (1.0%) in the intervention group compared with 79 of 9963 (0.8%) in the control group (RR, 1.22; 95% CI, 0.91-1.65; P = .21).

Detection rates varied considerably across the three personalized screening groups. Among participants screened after 1 year, advanced neoplasia was detected in 3.5%, compared with 5.0% among those screened after 2 years.

By contrast, the detection rate among those assigned to the extended 3-year interval was just 0.7% (95% CI, 0.6%-0.9%), comparable to the 0.8% observed in the conventional screening group.

The positive predictive value for advanced neoplasia was also virtually identical between the intervention and control groups (23.8% vs 23.9%; RR, 1.00; 95% CI, 0.74-1.35). 

Potential Implications for Screening Programs 

Commenting on the findings for Medscape Medical News, Alanna Ebigbo, MD, of Ruhr University Bochum, Germany, who chaired the session, said the absence of a significant improvement in overall detection should not overshadow the potential implications for screening efficiency.

"Even though risk stratification overall did not show any benefit, I think there was a benefit in showing that you don't need to repeat the very low-risk patients at 2 years," he said. "You may be able to push them over to 3 years."

Ebigbo suggested that extending screening intervals for the large proportion of individuals with undetectable f-Hb could offer important advantages for healthcare systems. "That's already a cost benefit," he said, although the trial did not formally evaluate cost-effectiveness.

He emphasized that the findings covered only one screening round and that longer-term follow-up would be necessary before changing screening policies.

Longer-Term Safety Remains Uncertain 

Toes-Zoutendijk acknowledged that the study had not demonstrated the anticipated improvement in advanced neoplasia detection. Nevertheless, she said, the findings raise questions about whether future screening strategies should prioritize increased detection or maintaining similar outcomes while reducing screening resources.

She suggested that future screening strategies might be simplified by using just two categories, that is detectable and undetectable f-Hb, rather than the three categories evaluated in PERFECT-FIT.

However, she cautioned that the trial assessed only a single screening round, leaving uncertainty about the consequences of repeatedly extending intervals. Interval cancer data are also awaited, with further results expected toward the end of 2027. 

"We need to wait for the interval cancers, because maybe those that had an extended interval had many more missed cancers," said Toes-Zoutendijk.

The investigators are also awaiting findings from other randomized trials evaluating risk-stratified CRC screening, which could help inform future national screening recommendations.

Toes-Zoutendijk reported research funding unrelated to the study and an in-kind contribution of FIT tests from Sentinel. Ebigbo's disclosures were not available at the time of publication.