Shaheen, Nicholas J.

ABSTRACT Accurate identification of dysplasia in Barrett's esophagus (BE) remains a challenge. Advanced optical imaging techniques may allow for better localization of dysplasia in BE. Here, we have assessed the potential clinical utility of a previously described multimodal imaging probe combining optical coherence tomography (OCT) with angle‐resolved low coherence interferometry (a/LCI) to prospectively identify dysplasia in BE. Imaging was conducted on 37 patients undergoing endoscopic surveillance of BE, yielding co‐registered biopsies of 50 esophageal sites. The a/LCI nuclear morphology data were compared to a previous decision line to prospectively predict dysplasia, demonstrating 100% sensitivity, 93% specificity, and 94% overall accuracy. The NPV was 100%, comparable to previous a/LCI studies. The addition of OCT imaging markedly improved PPV and specificity, compared to previous studies with a/LCI alone, illustrating the clinical utility of the combined platform. These findings suggest that combining OCT and a/LCI enables better detection of dysplasia by providing better guidance.

Objective: An endoscopically placed duodenal-jejunal bypass liner (DJBL) may provide a safe adjunctive therapy for those with poorly controlled type 2 diabetes mellitus (T2DM) and obesity. Background: Although some endoscopic therapies have been shown to improve glycemic indices secondary to weight loss, small bowel interventions may have direct metabolic effects. A meta-analysis of observational studies demonstrated reduction in HbA1c by 1.3% at 1 year after DJBL in patients with T2DM and obesity. Methods: This was a multicenter, double-blind, randomized, sham-controlled trial comparing DJBL to sham procedure with medical management and lifestyle modification. Primary endpoints included mean difference in changes in HbA1c at 12 months between arms, and device-related serious adverse events (SAEs). Secondary endpoints included percent total weight loss (%TWL) and subjects achieving HbA1c≤7% and TWL≥5% at 12 months. Results: Three hundred twenty subjects were randomized to DJBL (n=212) and sham (n=108). Baseline HbA1c and BMI were 8.79±0.92% and 38.45±5.75 kg/m 2 . On modified intent-to-treat analysis, change in HbA1c at 12 months was −1.10±1.45% and −0.28±1.54% for DJBL and sham groups, respectively ( P =0.0004). Rate of device-related SAEs was 9.4% including intolerance (3.7%), hemorrhage (2.8%), and hepatic abscess (2.3% stopping study early). At 12 months, DJBL group experienced greater weight loss compared with sham (7.7±9.6% TWL and 2.1±5.4% TWL, respectively; P <0.0001), with significantly more patients achieving HbA1c ≤ 7% (28.3% vs. 9.4%; P <0.0003) and TWL ≥ 5% (60.4% vs. 21.3%; P <0.0001). Conclusions: DJBL met primary glycemic control efficacy and primary safety endpoints, while providing clinically significant weight loss, and comorbidity improvement.

INTRODUCTION: The risk of progression to esophageal adenocarcinoma in Barrett's esophagus (BE) increases with advancing degrees of dysplasia. There is a critical need to improve the diagnosis of BE dysplasia, given substantial interobserver variability and overcalls of dysplasia during manual community pathologist reads. We aimed to externally validate a previously cross-validated BE dysplasia diagnosis deep learning model (BEDDLM) that predicts dysplasia grade on whole slide images (WSIs). METHODS: We digitized nondysplastic BE (NDBE), low-grade (LGD), and high-grade dysplasia (HGD) histology slides from 3 external academic centers. A consensus read by 2 expert study pathologists was used as the criterion standard. Slide stain characteristics were normalized using cycle-generative adversarial networks. WSIs were assessed by BEDDLM using an ensemble approach, combining a “You Only Look Once” model followed by a ResNet101 classifier model. RESULTS: We included 489 WSIs. Consensus histopathology revealed 232 NDBE, 117 LGD, and 140 HGD WSIs. The mean age (SD) was 66.9 (11.4) years; 413 (84.7%) were men. Using the BEDDLM ensemble model, sensitivity and specificity for NDBE were 73.3% (95% confidence interval [CI]: 67.09%–78.85%) and 93.4% (95% CI: 89.62%–96.10%); for LGD, 84.6% (95% CI: 76.78%–90.62%) and 80.6% (95% CI: 76.26%–84.54%); and for HGD, 80.7% (95% CI: 73.19%–86.89%) and 94.8% (95% CI: 91.97%–96.91%), respectively. The F1 score was 0.81, 0 0.69, and 0.83 for NDBE, LGD, and HGD, respectively. DISCUSSION: Our externally validated deep learning model demonstrates substantial accuracy for the diagnosis of BE dysplasia grade on WSIs.

INTRODUCTION: Current guidelines recommend endoscopic surveillance of Barrett's esophagus (BE) but do not account for competing mortality unrelated to esophageal cancer (EC). We conducted a systematic review and meta-analysis to estimate EC and non-EC mortality risk in BE patients. METHODS: We searched multiple databases for studies reporting mortality in BE. We included population-based studies providing standardized mortality ratio (SMR). The primary outcome was SMR from all causes and EC and non-EC etiologies. SMR was calculated by dividing the observed mortality over the expected mortality. Logarithmic form of SMRs was pooled using random-effects model. RESULTS: Our search yielded 2,826 articles, of which 7 studies (n = 34,454) were included. All-cause mortality was elevated in BE patients compared with population controls (pooled SMR 1.24, 95% confident interval [CI] 1.01–1.53) driven in part by increased EC mortality risk (SMR 8.98, 95 CI 5.12–15.77). The mortality risk was still increased but attenuated after excluding EC mortality (SMR 1.21, 95% CI 1–1.46). There was no increased mortality risk of non-EC malignancies (SMR 1.22, 95% CI 0.82–1.82) or mortality due to noncancer etiologies (SMR 1.13, 95% CI 0.90–1.43). Death due to cardiovascular diseases was higher in BE (SMR 1.16, 95% CI 1.02–1.33). BE patients were 10 times more likely to die from noncancer etiologies than EC (risk ratio 10.71, 95% CI 5.98–19.16). Subgroup analysis of studies that excluded prevalent EC at baseline (3 studies) also showed increased all-cause (SMR 1.12, 95% CI 1.07–1.18) and EC mortality (SMR 4.7, 95% CI 3.58–6.17) among BE patients. DISCUSSION: BE patients exhibit a higher all-cause mortality, driven in part by risk of EC mortality. A personalized approach to surveillance, mitigating risk of EC while recognizing the broader mortality risks, is warranted.