Background Esophageal food impaction is relatively common and increasing over time. Treatment ranges from medications to invasive endoscopic therapies. The endoscopic push technique has been advised...
Precancerous duodenal polyps can be subtle in familial adenomatous polyposis (FAP). Artificial intelligence models can detect gastrointestinal (GI) tract pathology; thus, we aimed to develop a model to identify duodenal polyps in FAP patients. Images of duodenal polyps from upper GI endoscopic surveillance in FAP patients were obtained. Polyps were manually annotated (Label Studio; HumanSignal, San Francisco, Calif, USA), and the nnU-Net framework (Applied Computer Vision Lab of Helmholtz Imaging and the Division of Medical Image Computing at the German Cancer Research Center, Heidelberg, Germany) was used to automate polyp segmentation. Images were randomly divided into training, validation, and test sets (80%, 10%, and 10%, respectively). Primary performance metric was Dice coefficient (score 0-1). Manual counting was also used to compare the model's ability to identify polyps. Mean polyps per image by manual count was 4.2 (standard deviation [SD] = 4.8) and by prediction model 4.0 (SD = 4.7). Sixty of 87 images (69.0%) had 0 missed polyps, 74 of 87 (85.1%) missed a maximum of 1 polyp, and 83 of 87 (95.4%) missed a maximum of 4. Mean missed polyps per image was 0.9 (SD = 2.3), and the model identified 287 of 365 (78.6%) polyps. Forty of 87 (46.0%) had falsely identified polyps (mean 0.8; SD = 1.0). Missed polyps were smaller than identified polyps (mean 73.3 [SD = 45.3] versus 233.5 [SD = 274.5] pixel diameter, respectively). There was a complete match (no missed or false positive) in 33 of 87 images (37.9%). Dice coefficient was 0.73. A model to identify duodenal polyps in FAP was successfully created. Although the Dice coefficient is modest compared with that of colon polyp models, duodenal anatomy creates a challenging background for human and computer detection. Rate of polyp detection, likely a superior marker of goal achievement, was >75%, with a low false polyp rate (mean <1/image). This prototype model is the first step toward a refined algorithm to assist in identification of duodenal polyps with a need for larger prospective studies.
Topical steroids are commonly used in treatment of eosinophilic esophagitis (EoE), but currently there is lack of data to clarify most effective regimen. We aimed to study the achievement of histologic remission using the same dose of budesonide in two different delivery formulations. Patients with established EoE treated with pharmacy compounded budesonide capsule or budesonide Rincinol gel (both 3 mg twice daily) were studied retrospectively. Those with pre-treatment and post-treatment histologic assessment were included with main endpoint being histologic remission. 103 patients (62 gel, 41 capsule) were included, with higher rate of histologic remission with gel (84 vs. 59%, P=0.004). A subset of patients in both groups had lack of steroid response (<50% drop in eosinophils) (15% for gel, 32% for capsule). Formulation/delivery vehicle of steroid treatments to esophageal mucosa in EoE appears important for treatment efficacy, with budesonide gel having higher likelihood of histologic remission compared to budesonide capsules in our population. A truly steroid refractory group appears likely in our population. Larger, prospective studies may help clarify best regimen of topical steroids in EoE and may work to identify patients likely to benefit from alternative therapies.
Background and aims: Aberrations in telomere length can have important implications in cancer. Using a cohort of 1007 patients, we investigated whether leukocyte telomere length (LTL) in patients with colorectal cancer (CRC) is associated with survival. We also investigated whether some telomere maintenance genes are associated with survival in these patients. Methods: The Biobank for Gastrointestinal Health Research (BGHR), an ongoing project involving collection of biospecimens at the Mayo Clinic, was utilized to obtain data from patients diagnosed with stage II or III CRC. Blood samples were collected prior to chemotherapy/radiation and DNA was extracted for measuring median LTL. The main outcome measures were overall survival (OS) and disease-free survival (DFS) by disease stage. Results: A significant inverse relationship was observed with patient age and LTL (spearman correlation coefficient (r) = −0.48, 95%; p = 1.13 × 10−58). Females had significantly longer LTL than males (p = 3.97 × 10−5). The rs1317082 SNP in the TERC gene was significantly associated with both OS and DFS in combined stage II and stage III patients (p = 0.017 and p = 0.023, respectively). A statistically significant association of the OBFC1 SNP (rs9419958) was observed for OS for the combined stage II and stage III patients (p = 0.016). Importantly, LTL was significantly associated with both OS and DFS (p = 0.008 and 0.044 respectively) in combined stage II and stage III patients. Conclusions: Our results show that LTL is predictive of OS and DFS for stage II and III CRC patients, particularly over a longer follow-up, extending beyond five years after a diagnosis of CRC, and certain SNPs in genes involved in telomere maintenance are significantly associated with patient outcomes, independent of telomere length.