Anticipate More Adenomas After FIT-Positive Colonoscopies

Endoscopists’ adenoma detection rate (ADR) during colonoscopies that follow a positive fecal immunochemical test (FIT) was inversely related to the risk for developing interval colorectal cancer (CRC), Dutch researchers found, suggesting these specialists may need to expect higher ADRs in this screening scenario.

In the population-based cohort study, the adjusted CRC risk following a FIT-positive colonoscopy was 5% lower per percentage point higher ADR on multivariable cox regression modeling analysis (adjusted HR 0.95, 95% CI 0.92-0.97), reported Manon C.W. Spaander, MD, PhD, of Erasmus University Medical Center in Rotterdam, The Netherlands, and colleagues, writing in the Annals of Internal Medicine.

After 5 years, the expected number of interval CRC diagnoses for every 1,000 patients who underwent colonoscopy was far lower when performed by endoscopists with higher ADRs:

• ADRs of 70%: nearly 2 diagnoses per 1,000
• ADRs of 65%: over 2.5 diagnoses 1,000
• ADRs of 60%: nearly 3.5 diagnoses 1,000
• ADRs of 55%: over 4.5 diagnoses 1,000

“Our results further show that colonoscopy in FIT-positive persons demands a markedly higher ADR target than primary colonoscopy,” the group noted, adding that advanced ADR, mean number of adenomas per procedure, and mean number of adenomas per positive procedure were also linked to interval CRC, “but did not outperform ADR.”

The relationship between ADR and interval CRC remained strong even when looking at only advanced-stage cancer.

“These seemingly small ADR differences are deceptive — if an endoscopist increases their ADR by just 10%, their patients’ associated decrease in relative interval cancer risk is a remarkable 40% to 50%,” wrote Douglas Corley, MD, PhD, MPH, of Kaiser Permanente in Oakland, California, in an accompanying editorial.

“And this estimate is likely conservative, given the study excluded some cancers, and modeling studies suggest that, over a lifetime, higher ADRs are associated with even greater cancer prevention,” he added.

CRC incidence and mortality improves with early detection and the removal of CRC neoplasia. The quantity of colonoscopies performed that detect at least one adenoma, or ADR, serves as a valuable quality indicator for CRC screening with colonoscopy and has been linked to interval CRCs, Spaander’s group noted. High-quality colonoscopies that detect all lesions can prevent interval CRCs. However, only limited data are available on the impact of ADR on interval CRCs after patients test positive using the most common CRC screening tool — FIT.

For this study, Spaander and colleagues examined data from the Dutch CRC screening program on 116,360 individuals who underwent their first colonoscopy following a positive FIT result from 2014 to 2016. During the first 6 months of 2014, those with a minimum FIT result of 15 mg of hemoglobin per gram of feces were provided a colonoscopy, while the FIT cutoff was later raised to 47 mg of hemoglobin per gram.

The 362 endoscopists who performed those colonoscopies had a median ADR of 67% (range 40-82%). Of those, only 311 endoscopists who performed 103,900 colonoscopies were included in the analysis, which adjusted for demographics, diagnosis at colonoscopy, endoscopy setting, and surveillance interval.

Patients included were ages 55 to 75 (median 67 years), 60% were men, had biennial FIT screening, and had not been diagnosed with CRC within the following 6 months. On colonoscopy, advanced adenoma was detected in 39.7%, non-advanced adenoma in 32.6%, serrated lesions in 3.4%, while no lesions were detected in 17%, with another 7.3% listed as unknown or other.

Over a median follow-up of 52 months, 426 participants were diagnosed with CRC following a colonoscopy, including researchers detected 209 patients with interval cancers (49%). After 5 years, cumulative incidence of interval CRC showed a steady increase to 0.27%. Non-interval CRC diagnoses — those detected on subsequent screening — were not significantly linked to ADR.

“This study provides an excellent framework for evaluating nine concepts regarding effective quality metrics and how these can illustrate pathways for meaningful metrics for the care of other cancers and disorders,” Corley said. “This is crucial, as metrics can be laboriously implemented without improving outcomes.”

Those concepts included being time-stamped, gaming-resistant, simple, actionable, relevant, strategic, important, trustworthy, and owned, the editorialist said.

Compared to colonoscopies performed at academic hospitals, the risk for interval CRCs was higher for those performed both at endoscopy centers (HR 3.87, 95% CI 1.31-11.43) and non-academic hospitals (HR 3.74, 95% CI 1.31-10.66).

The authors acknowledged limitations to the data, including the short follow-up time.