Over the last two decades, it has been well clarified that the serrated/methylated pathway of colorectal carcinogenesis is responsible for about 20 %–30 % of all colorectal cancers (CRCs) and is distinct from the traditional adenoma–carcinoma sequence [1]. Three main subgroups of serrated polyps have been recognized, with different malignant potential [2]. Hyperplastic polyps (HPPs) are believed to have a benign course, unless larger than 10 mm, whereas sessile serrated lesions (SSLs), especially those with dysplasia, and traditional serrated adenomas (TSAs), regardless of their size, have the potential for malignant transformation. Therefore, there is no doubt that these precancerous lesions are clinically relevant and deserve all our attention, above all they are often flat in appearance with indistinct borders and difficult to detect.

Despite the above considerations, there is however a reluctancy to implement the assessment of serrated polyp detection in our clinical practice and quality metrics. Indeed, fecal immunochemical test (FIT)-based screening programs for CRC do not consider serrated polyps in the definition of the diagnostic yield of the test. In addition, despite a significant inverse association between the serrated lesion detection rate (SDR) and the risk of post-colonoscopy CRCs having been observed [3], the SDR is not routinely assessed and a benchmark has not been set. One of the reasons is that the interobserver agreement in the histological diagnosis of serrated colonic lesions is only moderate, even among experienced pathologists [4]. Furthermore, which kind of serrated lesions (i. e. histology, size, and location within the colon) should be considered clinically relevant is still a matter of discussion and not widely uniform. Finally, a wide variation in the sensitivity of FIT for the diagnosis of serrated lesions has been reported, as these neoplasms are unlikely to bleed, even when large in size [5].

The study by van Toledo et al. [6] in the current issue of Endoscopy is an important step forward in the understanding of the detection rate and yield of advanced, clinically relevant serrated polyps, defined as any serrated polyp ≥ 10 mm, SSL with dysplasia, or TSA, in a FIT-based screening program.

The study is noteworthy for several reasons. First, it is a large population-based study including more than 322 000 subjects who underwent a high quality colonoscopy (i. e. adequate bowel preparation and the cecum reached in all the cases) within the national Dutch screening program.

“… the added value of the routine implementation of the advanced serrated polyp detection rate within quality metrics, if preceded by a serious program of quality assessment and regular auditing, would improve the quality of colonoscopy itself.”

Second, it showed that advanced serrated polyps are detected in a not negligible proportion of patients (i. e. 5.9 %), more frequently in women and the elderly. This is slightly higher than the prevalence reported in other European countries and closer to that reported in the USA [1]; however, differences in study design, definition of advanced lesions, and population may easily explain the discrepancy. Serrated polyps ≥ 10 mm were more frequently detected in the proximal colon, along with SSLs with dysplasia, while TSAs were more common in the distal colon. Therefore, the suggestion of considering the proximal SDR as a quality metric, in addition to the adenoma detection rate (ADR), is an excessive simplification that does not take into account the true distribution of these lesions across the whole colon.

Third, with regard to the diagnostic yield, the inclusion of advanced serrated polyps as relevant lesions, together with colorectal cancer and advanced adenomas, increased the yield of screening by 2.7 percentage points, from 41.1 % to 43.8 %. Although modest, this allows a proper classification of patients that were incorrectly classified as false-positives and enables future comparison with other screening tests, such as the upcoming multitarget stool tests.

Finally, one of the main strengths of this study was the strict quality control of both endoscopy and pathology. Indeed, all endoscopists and pathologists required accreditation and regular monitoring; the reporting pathologists were also required to complete a validated e-learning program on the histopathologic diagnosis of the subclassification of serrated polyps. This is not routinely performed worldwide and explains the wide variation in SDRs and low interobserver agreements among pathologists. Therefore, the added value of the routine implementation of the advanced SDR within quality metrics, if preceded by a serious program of quality assessment and regular auditing, would improve the quality of colonoscopy itself.

Barriers to the implementation of the SDR among colonoscopy performance measures mostly relate to issues in pathology, as its clinical value has been established. This paper not only provides useful insights to our comprehension of the importance of advanced serrated lesions in the correct assessment of diagnostic yield within FIT-based screening programs, but also proposes a model for collegial quality improvement among gastroenterologists and pathologists that will help to break down these barriers.

Article published online:
03 March 2023

© 2023. Thieme. All rights reserved.

Georg Thieme Verlag KG
Rüdigerstraße 14, 70469 Stuttgart, Germany