Severe fibrosis in patients with ulcerative colitis-related dysplasia: Can we predict and manage it well?

Patients with ulcerative colitis (UC) are at an increased risk of developing colorectal cancer (CRC). CRC accounts for approximately 10–15% of all deaths in patients with inflammatory bowel disease, particularly in those with UC. Current guidelines propose surveillance colonoscopy for detecting and removing dysplasia to prevent CRC.1 However, dysplasia arising from patients with UC differs from that arising from patients without UC in terms of both detection and resection. Ulcerative colitis-associated neoplasia (UCAN), which refers to CRC or dysplasia associated with UC, is distinct from sporadic neoplasia. Sporadic neoplasia, similar to adenoma in patients without UC, is relatively easy to detect. However, early detection of UCAN is difficult because of the complex background of inflamed mucosa in UC. Previously, nonpolypoid lesions were considered invisible and random biopsies were recommended. With the advancement of endoscopy instruments and diagnostic techniques, most cases of UCAN are now considered detectable using chromoendoscopy with high-definition colonoscopy in patients with UC.1 A recent study has shown that high-grade dysplasia is frequently associated with a flat/superficial elevated area and red discoloration, and the findings of chromoendoscopy and magnifying colonoscopy may also be useful in distinguishing lesions from the surrounding mucosa.2 Recently, there have been notable advancements in the development of new endoscopy systems and novel computer-aided detection/diagnosis systems, leading to the introduction of new products. With the advancement of colonoscopy and artificial intelligence, further improvements in the detection and diagnostic capabilities of UCAN are expected, and an increase in the early detection of UCAN is likely to result in an increased demand for endoscopic treatment. After the detection of neoplasia in patients with UC, treatment is the next step. Colectomy was performed more commonly when the lesion was difficult to detect on colonoscopy and was detected on random biopsy. Owing to better recognition of the lesion with high-definition colonoscopy, endoscopic resection (ER) has been commonly performed and recommended for UC-related dysplasia in current guidelines.1 However, there are several concerns in applying ER for them. First, margins of UCAN are often indistinct.1, 2 For patients with endoscopically indistinct dysplasia, referral to an endoscopist with expertise in inflammatory bowel disease surveillance is suggested, using chromoendoscopy with high-definition colonoscopy.1 However, colectomy will be considered if distinct margins of the lesion cannot be identified, even by the experienced endoscopists. Second, fibrosis is a concern during ER. Previous studies have reported a high frequency (90–100%) of submucosal fibrosis in patients with UC.3-5 Because of chronic inflammation, patients with UC often have submucosal fibrosis, which would be an obstacle to achieving complete resection using the ER. Indeed, the guidelines mention the term “endoscopically resectable,” indicating that histological examination of the resected specimen is consistent with complete removal.1 However, there is no consensus on which ER method is feasible for achieving complete removal. Endoscopic mucosal resection is technically simple and acceptable for polypoid lesions without fibrosis but unsatisfactory for nonpolypoid lesions or lesions larger than 10 mm because of its relatively low R0 resection rate (defined as en bloc resection with histologically tumor-free margins).6 Endoscopic submucosa dissection (ESD) is another option for nonpolypoid UC-related dysplasia. Recent studies have shown high en bloc and R0 resection rates of ESD for UC-related dysplasia, but the impact has been limited by the small sample size.3, 6 Moreover, ESD is technically demanding and challenging in cases with severe submucosal fibrosis because severe fibrosis is a critical risk factor for a longer procedure time, incomplete resection, and intraoperative perforation during ESD.7 In this context, the prediction of severe fibrosis will be helpful in selecting the resection method and skills of the endoscopist. Large tumor size, macroscopic type, straddling fold, and preoperative biopsy are predictors of severe fibrosis in patients without UC,8 but no study has reported these predictors in patients with UC. In this issue of Digestive Endoscopy, Nishio et al.9 reported the predictors of severe submucosal fibrosis in patients with UC and the feasibility of ESD for tumors with severe fibrosis. This was a single-center, retrospective study in Japan. ESD was performed by three experienced endoscopists who had performed ≥100 colonic ESDs. Submucosal fibrosis was assessed using a 3-point scale: F0, no fibrosis; F1, mild fibrosis; and F2, severe fibrosis.7 In addition, the authors assessed possible factors associated with severe fibrosis: the duration of UC, history of severe colitis, and endoscopic findings of background mucosa around the target lesion. The background mucosa was classified as nonatrophy/nonscarring, atrophy, and scarring.10 This study included 55 tumors resected using ESD in 48 consecutive patients with UC, including 27 tumors with severe fibrosis (F2 group). Regarding the background mucosa, the rates of atrophy and scarring were 29% and 45%, respectively. The authors analyzed the clinicopathological characteristics and treatment outcomes in the F0/1 and F2 groups. Patients in the F2 group had a significantly longer UC duration than those in the other group (7 vs. 22 years, P = 0.01). The treatment time was significantly longer (35 min vs. 65 min, P < 0.01), and the intraoperative perforation rate was significantly higher (4% vs. 30%, P = 0.01) in the F2 group than in the other group. Notably, there were no significant differences between the F0/1 and F2 groups in the en bloc (100% vs. 96%, P = 0.49) and R0 resection rates (100% vs. 93%, P = 0.23). In the multivariable analysis, a longer duration of UC (≥10 years; odds ratio 6.11; 95% confidence interval 1.20–31.03; P = 0.03) and scarring of the background mucosa of the tumor (odds ratio 39.61; 95% confidence interval 3.91–400.78; P < 0.01) were independent predictors of severe submucosal fibrosis. This was the first study to elucidate the predictors of severe fibrosis in patients with UC. The study's findings have several clinical implications. First, preoperative prediction of severe fibrosis will help provide appropriate treatment for patients. The predicted probability of severe fibrosis, calculated using the coefficients from the multivariable analysis, was 84.5% in tumors arising from scarring mucosa with a UC duration of ≥10 years.9 Considering the high intraoperative perforation rate of ESD (30%) in severe fibrosis cases in this study, it is necessary to obtain sufficient informed consent when applying ESD to cases predictive of severe fibrosis, and ESD should be performed by experienced endoscopists. If endoscopists are not familiar with UC-related dysplasia treatment, a referral should be made to endoscopists with expertise in inflammatory bowel disease-related dysplasia. In contrast, tumors arising from nonatrophy/nonscarring mucosa with a UC duration of <10 years had a low probability of severe fibrosis (2.2%).9 Lesions with no risk factors for severe fibrosis will be good candidates for ESD or even endoscopic mucosal resection. However, considering the difficulty in diagnosing the tumor margin, UCAN should be treated by endoscopists who are familiar with UC-related dysplasia. Second, the en bloc, R0, and curative resection rates were satisfactory, even in cases with severe fibrosis (96%, 93%, and 93%, respectively), supporting the feasibility of ESD for UC-related dysplasia. The R0 resection rate was comparable with that in previous reports (70–80%).3-5 Although the intraoperative perforation rate was high in cases with severe fibrosis, all cases were managed conservatively. However, there was one case of delayed perforation in the severe fibrosis group. During the dissection of severe fibrosis, thermal damage to the deep muscle layer may be greater than that in cases without severe fibrosis. Because the risk factors for delayed perforation after ESD have not been elucidated and no study has shown the merits of prophylactic clipping for preventing delayed perforation, further studies are required to assess the safety of ESD for UC-related dysplasia. Despite these valuable findings, the study had some limitations that were not described in the manuscript. In the study, atrophy was defined as a whitish coloration without an ulcer scar and scarring was defined as a whitish and stretched fold in the background mucosa, as previously reported.10 However, this classification seems to be subjective and not well-validated. Although the interobserver agreement of the background mucosa was substantial (κ = 0.67), external validation is warranted to assess the generalizability of this evaluation method. In this study, a history of severe colitis, which was defined as Mayo score ≥11 or Mayo endoscopic subscore of 3, was excluded from the multivariable analysis owing to its potential correlation with mucosal scarring. The Mayo score is the best known and validated disease activity instrument for UC, which includes stool frequency, rectal bleeding, a physician's global assessment, and endoscopic evaluation. Because of its objectivity, disease activity indices such as the Mayo score may serve as potential indicators of severe fibrosis; however, further studies are warranted to evaluate the association between the Mayo score and severe fibrosis. A multicenter prospective study is warranted to assess the feasibility of ESD for UC-related dysplasia and validate the findings of the study. In summary, Nishio et al. demonstrated that longer UC duration and scarring background mucosa were predictors of severe submucosal fibrosis associated with perforation during ESD in patients with UC. These findings may guide clinicians to manage patients with UC-related lesions. Nevertheless, almost half of the patients in this study cohort had severe fibrosis. Endoscopists treating UC-related lesions that are predictive of severe fibrosis should approach ESD with sufficient preparedness and surgical support. Author declares no conflict of interest for this article. None.