Introduction

Colorectal neoplasias developing from ulcerative colitis mucosa (CRNUCs) can be divided into ulcerative colitis-associated neoplasias (UCANs) and sporadic neoplasias (SNs); however, it is often difficult to distinguish between UCAN and SN from biopsy. We previously reported the utility of cancer genomic analysis for differentiating UCAN from SN in post-resection CRNUC samples (Inflamm Bowel Dis. 2021).

Aims & Methods

This study investigates whether genomic analysis can differentiate UCANs from SNs using CRNUC biopsy tissues. In the first step, of the 36 early-stage CRNUCs (high-grade dysplasias and T1 carcinomas) resected at Hiroshima University Hospital between June 1998 and July 2018, and examined for genetic landscapes in the resection specimens, 14 CRNUC samples that underwent biopsy before endoscopic or surgical resection were analyzed. The genomic landscape of the biopsy and resection specimens was examined for concordance. In the second step, the relationship between the genomic status and pathological diagnosis was investigated using biopsy specimens from 26 CRNUC cases, not included in the first step, which provided sufficient DNA for genomic analysis. A custom panel of 27 genes, previously identified as mutated in CRNUC studies, was used for deep sequencing analysis of the cancer genome. In the evaluation of biopsy specimens, UCAN diagnosis using p53 immunostaining (IHC) was defined as cases exhibiting strongly positive p53-IHC results, and UCAN diagnosis based on TP53 mutations was defined as cases with pathogenic mutations. UCAN diagnosis, combining hematoxylin and eosin (HE) staining with p53-IHC or TP53 mutation analysis, was defined as follows: cases diagnosed with UCAN based solely on HE staining of the biopsy specimen, or cases with strongly positive p53-IHC or a pathogenic TP53 mutation when HE staining alone was insufficient for a UCAN or non-UCAN diagnosis.

Results

After the first step of analysis, 14 out of the 27 candidate genes were found to be mutated. The concordance rate for pathogenic mutations across all 14 genes between the biopsy and resected specimens was 29% (4/14), whereas that for pathogenic mutations in TP53 and KRAS was 79% (11/14). After the second step of analysis, the accuracy of diagnosing UCAN/non-UCAN using only HE staining of biopsy specimens was 38% (10/26), whereas p53-IHC achieved an accuracy of 73% (19/26, p=0.026), and TP53 mutation analysis achieved an accuracy of 77% (20/26, p=0.012). The combination of HE pathology with other methods improved UCAN diagnosis. Specifically, combining HE pathology with p53-IHC in biopsy specimens achieved a sensitivity of 73% and an accuracy of 85% (22/26), while combining HE pathology with TP53 mutation analysis resulted in a sensitivity of 87% and an accuracy of 88% (23/26).

Conclusion

Both p53 immunohistological staining and TP53 mutation analysis of biopsy tissue were found to be useful for diagnosing UCAN, with TP53 mutation analysis demonstrating comparable accuracy to p53 immunohistological staining for UCAN diagnosis.

Introduction

Patients with ulcerative colitis (UC) are at increased risk of developing colorectal cancer (CRC), and this is assumed to be due to chronic inflammation of the colorectal mucosa.

Aims & Methods

To identify risk factors for the development of dysplasia or CRC in patients with UC. Retrospective observational cohort study. Clinical data and colonoscopy findings were collected from medical records. A 1:2 Propensity Score Matching (PSM) was performed to balance covariates adjusting for disease extension, age, gender, smoking habit and disease duration. An Andersen-Gill model was used to account for multiple colonoscopies per patient and time-varying risk of dysplasia or CRC, allowing for recurrent events and intra-individual correlation.

Results

A total of 571 UC patients were identified, median age at diagnosis was 39.5 (27.3-54.2), and 304 (53.2%) were men. At diagnosis, 147 (25.7%) had proctitis, 203 (35.6%) left side colitis and 191 (33.5%) extensive colitis. During a median follow-up of 7.1 years (IQR 3.7– 13.3), 51 patients (9.8%) developed dysplasia/CRC. This was more frequent in males (72.9% vs. 27.1%, p=0.004) and in older patients [48.1 years (36.7–60.2) vs. 38.5 years (26.5–53.8), p=0.004].
After PSM adjustment, 123 patients were included, 41 of whom had dysplasia/CRC, with a total of 381 colonoscopies over a follow-up period of 14.6 years (IQR 8.0–22.4). Macroscopic inflammatory activity was present in 283 (74.3%) colonoscopies and microscopic involvement in 5 (1.3%). Older age at diagnosis (aHR 1.03, 95% CI 1.04–1.10, p<0.001) and higher degree of macroscopic inflammation (aHR 1.44, 95% CI 1.02–2.03, p=0.041) were associated with an increased risk of dysplasia/CRC development; advance therapy, family history of CRC, disease extension and microscopic inflammation were not associated with an increased risk.

Table 1. Baseline characteristics of the complete cohort and after Propensity Matching Score.

Conclusion

Older age at diagnosis and the presence of macroscopic involvement on successive colonoscopies are associated with an increased risk of developing dysplasia or CRC in patients with UC.

Disclosure

Beatriz Gros has worked as an advisor for Abbvie, Pfizer, Alphasigma and Roche and has served as an speaker for Abbvie, Pfizer, Alphasigma, Roche, Janssen, Takeda and Gilead.

Introduction

Ulcerative colitis (UC) is a chronic inflammatory bowel disease associated with an increased risk of colorectal cancer (CRC). CRC remains one of the most serious complications in UC patients, contributing to elevated mortality and underscoring the importance of preventive management. Despite the global increase in UC prevalence, large-scale, long-term epidemiological data on CRC development in UC patients are limited.

Aims & Methods

This study aimed to clarify the clinical and epidemiological characteristics of UC complicated by CRC using long-term nationwide observational data. A total of 76,172 UC patients (43,262 males and 32,910 females) registered in a national specified disease treatment research program of between 2003 and 2013 were included. The observation period ranged from 1 to 10 years after registration. CRC cases were identified based on mentions of “colorectal cancer,” “colon cancer,” or “rectal cancer” in the “complications” section or free-entry field of the individual survey forms. A machine learning model is used for prognosis prediction using B3 Analytics (Hitachi, Ltd.). Risk analyses were conducted by age, disease duration, and disease phenotype, and standardized incidence ratios were calculated using demographic data from 2022.

Results

Patients who developed CRC (n=141) were significantly older (p<0.001), and the age of UC onset was also higher (p<0.001). Endoscopic findings revealed that the cancer group had lower incidences of mucosal friability (p=0.001) and erosion (p=0.001), but a significantly higher frequency of pseudo-polyps (p<0.001). Biopsy results showed a markedly increased incidence of dysplasia in the cancer group (p<0.001). The use of 5-aminosalicylic acid (5-ASA) was significantly less frequent in these patients (p<0.001). CRC incidence increased with age and was particularly elevated among those aged 70 and above, with a notably high rate observed in males aged 70–74 years (652.88/100,000), compared to the general population.
Univariate analysis identified several significant CRC risk factors: advanced age at UC onset (HR: 1.05; 95% CI: 1.03–1.06; p<0.001), presence of pseudo-polyps (HR: 2.92; 95% CI: 1.58–5.43; p=0.001), mucosal abnormalities (HR: 3.14; 95% CI: 1.74–5.65; p<0.001), and dysplasia (HR: 11.3; 95% CI: 6.50–19.70; p<0.001). A CRC prediction model developed using machine learning achieved an AUC of 0.854. The strongest predictors of carcinogenesis were the non-use of 5-ASA and the presence of pseudo-polyps. Stratification into four groups based on 5-ASA use and pseudo-polyp status revealed that patients with both pseudo-polyps and no 5-ASA use had the highest incidence and the earliest onset of CRC.

Conclusion

The incidence of colorectal cancer (CRC) in UC patients increases with age and is significantly associated with the presence of pseudo-polyps and dysplasia. Non-adherence to 5-ASA, pseudo-polyp formation, and mucosal inflammatory findings should be recognized as important risk factors for the development of CRC in UC, necessitating close surveillance and targeted prevention strategies.

Disclosure

HS receives lecture fees from Astellas, Otsuka and Takeda Pharm. Co. and joint research funding from Tohso Co.

Introduction

Sporadic neoplasias (SNs) and ulcerative colitis-associated neoplasias (UCANs) can occur in patients with UC. Differences in the characteristics and prognoses between UCANs in patients with UC and SNs in patients without UC have been assessed in numerous previous studies. However, only a few studies have focused on the characteristics and prognoses of SNs in patients with UC. It is important to distinguish between UCANs and SNs, because they have different prognoses and therefore need different treatment strategies.

Aims & Methods

This study aimed to evaluate the clinical features and prognoses of SNs detected in patients with UC, both within and outside the affected area. A total of 141 SNs in 59 patients with UC, which were detected by surveillance colonoscopy (SCS) at Hiroshima University Hospital between January 1999 and December 2021, were included. In our hospital, we perform SCS for UC patients with left-sided or total colitis and with a disease duration of more than 7 years, if possible, in the remission stage. SNs were diagnosed based on their location, endoscopic features, and histopathological findings. The final diagnosis was made by histopathological evaluation of the excised specimens and confirmed by more than one gastrointestinal pathologist. The histopathological diagnosis of SN was made according to the pathological features described in the recent European Crohn's and Colitis Organization and the European Society of Pathology statement on histopathology. Immunohistochemical evaluation of Ki67 and p53 was performed in all cases. SNs, especially sporadic adenomas, have regular neoplastic glands with low amounts of stroma, and rarely have a mixture of benign/dysplastic crypts. Although UCANs sometimes have a surrounding flat area with overexpression of p53, SNs have no such area and have a distinct border between the neoplastic glands and non-neoplastic mucosa. SNs have Ki67 expression mainly in the superficial zone and have low levels of p53 expression.

Results

The mean age at first SN detection was 58.8 years, and the mean duration of UC was 9.9 years. Male patients represented 64.4% of the enrolled patients. The most common type of UC was the total colitis type (47.5%). With regard to their clinical course, approximately half of the patients (49.2%) presented with the relapse-remitting type. Among the 141 SNs, 129 lesions (91.5%) were diagnosed as adenomas, five (3.5%) as Tis carcinomas, two (1.5%) as T1a carcinomas (superficially submucosal invasive carcinoma: <1,000 µm), and five (3.5%) as T1b carcinomas (deeply submucosal invasive carcinoma: ≥1,000 µm). There were 86 (61.0 %) SNs in the right colon, 44 (31.2%) in the left colon, and 11 (7.8%) in the rectum. When classified as outside or within the affected area of UC, 100 (70.9%) SNs occurred outside and 41 (29.1%) within. Most SNs (135 lesions, 95.7%) underwent endoscopic resection, and only six lesions (4.3%) underwent surgical resection. Among the 59 patients included in this study, synchronous SNs occurred in 14 (23.7%), and metachronous multiple SNs occurred in 24 (40.7%) patients. The 5-year cumulative incidence of metachronous multiple SNs was higher in patients with synchronous multiple SNs (54.2%) than in those without (46.4%). The histology, size, and background mucosa of the initial SNs were not associated with the incidence of metachronous SNs in our study.

Conclusion

UC patients with synchronous multiple SNs were at higher risk of developing metachronous multiple SNs and may require close follow-up compared to patients without synchronous SNs.

Introduction

The prevalence of primary sclerosing cholangitis (PSC) among ulcerative colitis (UC) is known to be lower in Eastern Asia, including Japan, than in Western countries. There have been few studies describing the details of ulcerative colitis-associated colorectal cancer (CAC) in UC with PSC (PSC-UC) in Japan so far, owing to its rarity.

Aims & Methods

This retrospective study aimed to evaluate the clinicopathologic features and prognosis of CAC in PSC-UC in the largest nationwide cohort of its kind in Japan. A total of 797 eligible patients diagnosed with CAC from 1983 to 2020 were categorized into those with PSC (PSC-UC group, n = 26) and those without PSC (UC-alone group, n = 771). Clinicopathologic features, 10-year overall survival (OS) and 10-year cancer-specific survival (CSS) were compared between the two groups.

Results

In PSC-UC group, duration of UC was shorter (12.8 vs 17.0 year, p=0.042), and age at CAC diagnosis was younger (47.8 vs 53.4 years old, p=0.042) compared to UC-alone patients. Right-sided colon cancer was significantly more common in PSC-UC group than in UC-alone group (34.6% vs 15.5%, p=0.032). Left-sided colon cancer (38.5%) and rectal cancer (26.9%) accounted for a similar proportion with right-sided colon cancer in PSC-UC group. There were no differences between the two groups in terms of sex, age at UC diagnosis, disease extent of UC, medical therapies for UC, diagnostic opportunity for CAC, histological types, and pathological stages of CAC. PSC-UC group showed a comparable 10-year OS (70.6% vs 83.1%, p=0.072) and 10-year CSS (84.1% vs 87.6%, p=0.50) as UC-alone group.

Conclusion

Our study highlighted the differences and the similarities of CAC between UC with PSC and without PSC patients. In patients with PSC-UC, it may be beneficial to start surveillance colonoscopy earlier because of the shorter duration of UC before the CAC diagnosis.

Introduction

Lateral lymph node dissection (LLND) is considered for patients with advanced lower rectal cancer because of possibility of metastasis. The standard surgical procedure for ulcerative colitis (UC)-associated cancer is total colectomy; however, the significance of LLND in UC-associated rectal cancer remains unclear.

Aims & Methods

The aim of this study is to clarify the outcomes of LLND for UC-associated rectal cancer. Of 1,249 patients treated for UC-associated colorectal neoplasms between 1983 and 2020 at 43 Japanese institutions, we retrospectively analyzed 415 patients with rectal cancer.

Results

LLND was performed in 45 (10.8%) patients with UC-associated rectal cancer: 33 of 139 (23.7%) patients with pT3–4 cancer and 12 of 276 (4.3%) patients with pTis–T2. Of the 45 patients with LLND, 11 (24.4%) had lateral lymph node metastasis (LLNM). LLNM-positive group (n = 11) had a significantly higher proportion of patients with pT3-4 cancer (100% vs. 64.7%, P < 0.0046) and poorly differentiated adenocarcinoma/ signet-ring cell carcinoma/ mucinous adenocarcinoma (por/sig/muc) (70% vs. 17.7%, P = 0.002) than LLNM-negative group (n = 34). There was no significant difference in R0 resection rate between the LLND and non-LLND groups (81.8% vs. 91.2%, P = 0.41). The LLNM-positive group showed significantly worse 5-year overall (30.8% vs. 82.8%, P = 0.0005) and 5-year disease-free (0% vs. 72.4%, P < 0.001) survival rates than the LLNM-negative group. There was no significant difference in the administration of adjuvant chemotherapy between the two groups (54.6% vs. 40.6%, P = 0.42).

Conclusion

The LLND rate for UC-related rectal cancer was 23% in patients with T3-4 cancer in Japan. The LLNM-positive group had more advanced disease, more por/sig/muc histology, and showed poor prognostic outcomes even after LLND.

Introduction

Long-standing Inflammatory bowel disease (IBD) increases the risk of colonic neoplasia, necessitating effective screening strategies. Various endoscopic modalities are used for surveillance colonoscopy in IBD patients, yet their comparative effectiveness and safety remain unclear. Previous guidelines have supported using both dye-based and virtual chromoendoscopy as equivalent¹; however, their additional benefit in the era of high-definition (HD) white light remains controversial.

Aims & Methods

This network meta-analysis (NMA), part of the British Society of Gastroenterology's initiative to update IBD surveillance guidelines, aims to compare the efficacy and safety of different endoscopic techniques to inform clinical practice and guidelines. We searched CENTRAL, ClinicalTrials.gov, Embase via Ovid, MEDLINE via Ovid, WHO ICTPR for randomised controlled trials (RCTs), and reference lists of relevant studies up to February 2024 for RCTs comparing endoscopic modalities for screening colonoscopy in patients with IBD. The primary outcome for efficacy was the detection of any dysplastic lesion (Vienna classification 2-5) per patient and, for safety, serious adverse events. Review authors independently screened and extracted data and assessed the risk of bias using the Cochrane tool. We conducted an NMA under a frequentist framework to estimate the relative effectiveness and safety of the interventions. The GRADE approach was employed to assess the certainty of the evidence for each outcome.

Results

A total of 27 RCTs involving 4,159 participants were included. The network comprised 6 endoscopic modalities: HD white light endoscopy (HD-WLE), HD virtual chromoendoscopy (HD-VCE), HD dye-based chromoendoscopy (HD-DCE), HD-WLE with segmental re-inspection (SR), auto-fluorescence imaging (AFI), and full-spectrum endoscopy (FUSE). There were no RCTs comparing HD to standard definition (SD) WLE/DCE; thus, the SD modalities could not be connected to the main network. HD-DCE may be more effective for detecting dysplasia, with a risk ratio (RR) of 1.42 (95% CI: 1.02-1.98) and low-GRADE certainty compared to HD-WLE. All other modalities had very low-GRADE certainty to draw conclusions about them, with RRs of 0.99 (95% CI: 0.69-1.43) for HD-VCE; 1.35 (95% CI: 0.66-2.77) for HD-WLE with SR; 3.24 (95% CI: 0.66-15.87) for FUSE; and 1.18 (95% CI: 0.55-2.57) for AFI. A similar trend was noted in sensitivity analyses, excluding studies reporting on serrated lesions (RR for HD-DCE: 1.91, 95% CI: 1.36-2.69) and excluding single endoscopist studies (RR for HD-DCE: 1.57, 95% CI: 1.10-2.26) compared to HD-WLE. Similar performances were noted for HD-VCE subtypes and varying concentrations of dye used in HD-DCE. Data on serious adverse events were limited, precluding any meta-analysis. In studies that reported this outcome, 2 serious adverse events were reported out of 2164 patients; however, this outcome had very low-GRADE certainty.

Conclusion

HD DCE is suggested over other modalities for surveillance in IBD, considering its potential to detect a small number of additional patients with dysplastic lesions compared to HD WLE. HD-VCE was not supported as an equivalent alternative to HD-DCE. However, the certainty of this recommendation is low due to the limited evidence. When choosing the surveillance modality, clinicians should also consider patient preferences, resource availability, and individualised risk assessments.

References

1 Murthy et al. Gastroenterology 2021