Ko, Huaibin Mabel

AimThe aim of this study was to compare the histopathologic features of MMF‐induced colitis and colonic GVHD and develop a digital tool using deep learning convolutional neural networks (CNNs) to semi‐automate the quantification of eosinophils.MethodsMMF and GVHD colitis cases were retrieved. A cohort of transplant patients who were at risk for developing GVHD and were on MMF at the time of biopsy was also identified. Cases were reviewed for histologic features and a deep learning CNN was trained to quantify lamina propria eosinophils. Clinical history was collected.ResultsNinety‐five patients were included in the study group: GVHD (n = 37), MMF (n = 25) and GVHD vs. MMF (n = 33). Within the GVHD vs. MMF group, one patient was thought to have definitive MMF colitis and one possible MMF colitis. The GVHD group had a significantly higher apoptotic count compared to the MMF group (median: 7 vs. 2, P = 0.03). Neuroendocrine aggregates were present in the GVHD (N = 12) and GVHD vs. MMF (N = 4) groups while absent in the MMF group. The MMF group had a higher semi‐quantitative eosinophil score and a higher automated eosinophil count as detected by the deep learning CNN (median eos/HPF: MMF 19.2, GVHD vs. MMF 3.6, GVHD 3.4, P < 0.001).ConclusionsLower eosinophil counts, higher apoptotic counts and neuroendocrine cell aggregates favour the diagnosis of GVHD. AI models for eosinophil quantification may serve as an ancillary tool for assessing this differential diagnosis. Given the rarity of MMF‐induced colitis in SCT patients, pathologists should have a high threshold for making this diagnosis in this population.

AimsInflammatory bowel disease (IBD)‐associated serrated lesions are categorized into three distinct subtypes: traditional serrated adenoma (TSA)‐like lesion, sessile serrated lesion (SSL)‐like lesion, and serrated lesion, not otherwise specified (NOS). Although the risk of neoplastic progression of serrated lesions without dysplasia has not been shown to exceed that of sporadic cases, the clinicopathologic features of the three serrated subtypes with dysplasia remain poorly understood in the context of IBD.Methods and ResultsWe analysed 87 serrated lesions with dysplasia (collectively referred to as serrated dysplasia) identified endoscopically in 58 IBD patients, including 51 (59%) TSA‐like dysplasia, 24 (28%) SSL‐like dysplasia, and 12 (14%) serrated dysplasia NOS. Inclusion criteria required all three serrated subtypes to show morphologic evidence of dysplasia and to be located within areas of colitis. We also compared the clinicopathologic features of serrated dysplasia with those of 239 conventional (adenomatous) dysplastic lesions from 149 IBD patients. The cohort included 39 (67%) men and 19 (33%) women, with a mean age of 54 years and a mean IBD duration of 20 years. Most patients had ulcerative colitis (n = 41; 71%) and pancolitis (n = 48; 83%). The majority of serrated lesions with dysplasia had a polypoid or visible endoscopic appearance (n = 73; 84%), with a mean size of 1.4 cm, and were found in the left colon (n = 66; 76%). Most lesions (n = 73; 84%) demonstrated low‐grade dysplasia at the time of biopsy diagnosis, whereas high‐grade dysplasia (HGD) was identified in the remaining 14 (16%) lesions. SSL‐like dysplasia was more frequently associated with ulcerative colitis (94%) compared to TSA‐like dysplasia (67%) and serrated dysplasia NOS (56%) (P = 0.042). Although only seven (12%) patients had a concurrent history of primary sclerosing cholangitis, it was exclusively identified in the TSA‐like dysplasia group (19% versus 0% in both the SSL‐like dysplasia group and the serrated dysplasia NOS group; P = 0.017). Serrated dysplasia NOS more commonly demonstrated HGD at the time of biopsy diagnosis (42%) compared to TSA‐like dysplasia (12%) and SSL‐like dysplasia (13%) (P = 0.022). Serrated dysplasia NOS was also more frequently associated with synchronous and/or metachronous nonconventional dysplasia (60%) compared to TSA‐like dysplasia (16%) and SSL‐like dysplasia (9%) (P = 0.037). Serrated dysplasia, regardless of subtype, was associated with high rates of advanced neoplasia (HGD or colorectal cancer) at the previous biopsy site or in the same colonic segment during follow‐up. Within a mean follow‐up time of 13 months, advanced neoplasia was detected in 50% of the TSA‐like dysplasia group, 67% of the SSL‐like dysplasia group, and 100% of the serrated dysplasia NOS group (P = 0.622). Moreover, at least one‐third of patients in each group (58% in the TSA‐like dysplasia group, 44% in the SSL‐like dysplasia group, and 33% in the serrated dysplasia NOS group; P = 0.332) developed synchronous/metachronous dysplasia, with at least 50% of these lesions progressing to advanced neoplasia within a mean follow‐up time of 11 months (P = 1.000). The serrated dysplasia group showed nearly six times the incidence of advanced neoplasia upon follow‐up (59%) compared to the conventional dysplasia group (10%) (P < 0.001).ConclusionTSA‐like dysplasia, SSL‐like dysplasia, and serrated dysplasia NOS show distinct clinicopathologic features. However, all three serrated subtypes were associated with high rates of advanced neoplasia (50%–100%) during follow‐up, suggesting that these lesions could potentially be combined into one diagnostic category, such as serrated dysplasia.