On a population level, mailed fecal immunohistochemical tests (FITs) may catch more cases of advanced neoplasia than endoscopic methods, based on a Dutch screening study that invited more than 30,000 people to participate.
, reported lead author Esmée J. Grobbee, MD, of Erasmus University Medical Centre in Rotterdam, the Netherlands, and colleagues.
In addition to high participation, previous research has shown that successful FIT screening depends upon continued adherence to the screening program, the investigators wrote in Clinical Gastroenterology and Hepatology. They noted that, in the present study, just two rounds of FIT were needed to outperform endoscopic methods, and that these comparative findings are a first for the field.
“No literature is available on the comparison between endoscopic screening strategies and multiple rounds of FIT screening,” the investigators wrote. “It is of key importance for policy makers to know the impact of different screening programs over multiple rounds with long-term follow-up.”
To this end, the investigators invited 30,052 screening-naive people in the Netherlands to participate in the present study. Each invitation was for one of three groups: once-only colonoscopy, once-only flexible sigmoidoscopy, or four rounds of FIT. All individuals received an advanced notification by mail followed 2 weeks later by a more substantial information kit (and first FIT test when applicable). If these steps received no response, a reminder was sent 6 weeks later.
Participants in the FIT group received one test every 2 years. Patients who had a positive FIT (hemoglobin concentration of at least 10 mcg Hb/g feces) were scheduled for a colonoscopy. Similarly, colonoscopies were performed in patients who had concerning findings on flexible sigmoidoscopy (e.g., sessile serrated adenoma. This sequential system reduced the relative number of colonoscopies in these two groups; colonoscopy rates in the FIT group and flexible sigmoidoscopy group were 13% and 3%, respectively, compared with the 24% participation rate in the colonoscopy group.
At a population level, FIT screening had the highest advanced neoplasia detection rate, at 4.5%, compared with 2.3% and 2.2% for screening by sigmoidoscopy and colonoscopy, respectively.
“In the intention-to-screen analysis, FIT already detected significantly more advanced neoplasia and colorectal cancer (CRC) after only 2 rounds of FIT, and this difference increased over rounds,” the investigators noted.
Again in the intention-to-screen population, mailed FIT detected three times as many cases of CRC than either of the other two groups (0.6% vs. 0.2% for both). In contrast, colonoscopy and sigmoidoscopy had higher detection rates for nonadvanced adenomas, at 5.6% and 3.7%, respectively, compared with 3.2% for FIT, although the investigators noted that nonadvanced adenomas are “of uncertain clinical importance.” Sessile adenoma detection rates were similar across all three groups.
The as-screened analysis revealed higher detection rates of advanced neoplasia for colonoscopy (9.1%), compared with sigmoidoscopy (7.4%) and FIT (6.1%). In the same analysis, detection rates of colorectal cancer (CRC) were comparable across all three groups.
According to the investigators, the CRC-related findings require careful interpretation.
“Comparing CRC detection rates of FIT and endoscopic screening is complex … because CRCs detected in FIT screening could in theory have been prevented in a once-only colonoscopy by the removal of adenomas,” they wrote.
Still, the key takeaway of the study – that FIT screening was the most effective strategy – may have practical implications on a global scale, according to the investigators.
“Because many countries are considering implementing screening programs, the findings of this study aid in deciding on choice of screening strategies worldwide, which is based on expected participation rates and available colonoscopy resources,” they wrote.
The study was funded by the Netherlands Organization for Health Research and Development. The investigators disclosed no conflicts of interest.
SOURCE: Grobbee EJ et al. Clin Gastro Hepatol. 2019 Aug 13. doi: 10.1016/j.cgh.2019.08.015.
