Use of advanced endoscopic technology for optical characterization of neoplasia in patients with ulcerative colitis: Systematic review

OBJECTIVES

Advances in endoscopic technology, including magnifying and image-enhanced techniques, have been attracting increasing attention for the optical characterization of colorectal lesions. These techniques are being implemented into clinical practice as cost-effective and real-time approaches. Additionally, with the recent progress in endoscopic interventions, endoscopic resection is gaining acceptance as a treatment option in patients with ulcerative colitis (UC). Therefore, accurate preoperative characterization of lesions is now required. However, lesion characterization in patients with UC may be difficult because UC is often affected by inflammation, and it may be characterized by a distinct "bottom-up" growth pattern, and even expert endoscopists have relatively little experience with such cases. In this systematic review, we assessed the current status and limitations of the use of optical characterization of lesions in patients with UC.

METHODS

A literature search of online databases (MEDLINE via PubMed and CENTRAL via the Cochrane Library) was performed from 1 January 2000 to 30 November 2021.

RESULTS

The database search initially identified 748 unique articles. Finally, 25 studies were included in the systematic review: 23 focused on differentiation of neoplasia from non-neoplasia, one focused on differentiation of UC-associated neoplasia from sporadic neoplasia, and one focused on differentiation of low-grade dysplasia from high-grade dysplasia and cancer.

CONCLUSIONS

Optical characterization of neoplasia in patients with UC, even using advanced endoscopic technology, is still challenging and several issues remain to be addressed. We believe that the information revealed in this review will encourage researchers to commit to the improvement of optical diagnostics for UC-associated lesions.

Artificial Intelligence Program to Predict p53 Mutations in Ulcerative Colitis-Associated Cancer or Dysplasia

BACKGROUND

The diagnosis of colitis-associated cancer or dysplasia is important in the treatment of ulcerative colitis. Immunohistochemistry of p53 along with hematoxylin and eosin (H&E) staining is conventionally used to accurately diagnose the pathological conditions. However, evaluation of p53 immunohistochemistry in all biopsied specimens is expensive and time-consuming for pathologists. In this study, we aimed to develop an artificial intelligence program using a deep learning algorithm to investigate and predict p53 immunohistochemical staining from H&E-stained slides.

METHODS

We cropped 25 849 patches from whole-slide images of H&E-stained slides with the corresponding p53-stained slides. These slides were prepared from samples of 12 patients with colitis-associated neoplasia who underwent total colectomy. We annotated all glands in the whole-slide images of the H&E-stained slides and grouped them into 3 classes: p53 positive, p53 negative, and p53 null. We used 80% of the patches for training a convolutional neural network (CNN), 10% for validation, and 10% for final testing.

RESULTS

The trained CNN glands were classified into 2 or 3 classes according to p53 positivity, with a mean average precision of 0.731 to 0.754. The accuracy, sensitivity (recall), specificity, positive predictive value (precision), and F-measure of the prediction of p53 immunohistochemical staining of the glands detected by the trained CNN were 0.86 to 0.91, 0.73 to 0.83, 0.91 to 0.92, 0.82 to 0.89, and 0.77 to 0.86, respectively.

CONCLUSIONS

Our trained CNN can be used as a reasonable alternative to conventional p53 immunohistochemical staining in the pathological diagnosis of colitis-associated neoplasia, which is accurate, saves time, and is cost-effective.

Colorectal Dysplasia and Cancer Surveillance in Ulcerative Colitis

Ulcerative colitis (UC) is a risk factor for the development of inflammation-associated dysplasia or colitis-associated neoplasia (CAN). This transformation results from chronic inflammation, which induces changes in epithelial proliferation, survival, and migration via the induction of chemokines and cytokines. There are notable differences in genetic mutation profiles between CAN in UC patients and sporadic colorectal cancer in the general population. Colonoscopy is the cornerstone for surveillance and management of dysplasia in these patients. There are several modalities to augment the quality of endoscopy for the better detection of dysplastic or neoplastic lesions, including the use of high-definition white-light exam and image-enhanced colonoscopy, which are described in this review. Clinical practice guidelines regarding surveillance strategies in UC have been put forth by various GI societies, and overall, there is agreement between them except for some differences, which we highlight in this article. These guidelines recommend that endoscopically detected dysplasia, if feasible, should be resected endoscopically. Advanced newer techniques, such as endoscopic mucosal resection and endoscopic submucosal dissection, have been utilized in the treatment of CAN. Surgery has traditionally been the mainstay of treating such advanced lesions, and in cases where endoscopic resection is not feasible, a proctocolectomy, followed by ileal pouch-anal anastomosis, is generally recommended. In this review we summarize the approach to surveillance for cancer and dysplasia in UC. We also highlight management strategies if dysplasia is detected.