DIAGNOSE AND DISREGARD POLICY CAN BE IMPLEMENTED IN PATIENTS WITH LYNCH SYNDROME WHEN DONE BY EXPERT COLONOSCOPISTS

Aims:

To evaluate the adequacy of a “diagnose and disregard” strategy for rectosigmoid (RS) diminutive polyps in Lynch syndrome patients.

Methods:

Secondary analysis of a prospective, multicentre, randomized, controlled and parallel study (EndoLynch study; NCT02951390) comparing panchromoendoscopy with white-light endoscopy for polyp detection in Lynch syndrome that involved 26 expert endoscopists in 14 centers. Optical diagnosis was performed in real time using NBI and/or chromoendoscopy in all detected polyps. Diagnostic categories were grouped into neoplastic (sessile serrated polyps (SSP) with dysplasia; traditional serrated adenoma (TSA), adenoma; carcinoma) and non-neoplastic (hyperplastic; SSP). Histology based in modified Vienna and WHO classification were used as gold standard.

Results:

256 patients were included in the study (mean age 47 years (± SD 14), 60% were women. The frequency of mutations was MLH1 (28,9%), MSH2 (41,4%), MSH6 (22,3%), PMS2 (6,6%) and EPCAM (0,8%). A total of 277 polyps (143 adenomas, 18 SSP, 1 TSA and 115 hyperplastic) were detected which included 102 diminutive polyps (≤5 mm) in the RS (72 hyperplastic, 3 SSP, 1 TSA, 22 low-grade dysplasia tubular adenomas, 3 high grade-dysplasia tubular adenomas and 1 low-grade tubulovillous adenoma). High confidence optical diagnosis was performed in 82 cases (80,4%) (62 non neoplastic; 20 neoplastic). Overall accuracy was 96.3% [89.7 – 99.2%]; sensitivity 90% [68,3 – 98,8%]; and specificity 98,4% [91,4 – 99,7%]. Negative and positive predictive value were 96,8% [89,11 – 99,1%]; and 94,7% [71,9 – 99,2%] respectively.

Tab. 1:

Optical diagnosis with high confidence versus histology for diminutive RS polyps in Lynch Syndrome

	NO neoplastic histology	YES neoplastic histology
NO optical diagnosis of neoplastic polyp	61	2	63
YES diagnosis of neoplastic polyp	1	18	19
Total	62	20	82

Conclusions:

In this large cohort of Lynch syndrome patients, despite over 25% of RS diminutive polyps were neoplastic, optical diagnosis was highly accurate. In expert hands, a diagnose and disregard strategy is safe and provides a decrease of the RS diminutive polyps resection by 70%.