ABSTRACT
Histopathological assessment of esophageal biopsies is a key part in the management of patients with Barrett esophagus (BE) but prone to observer variability and reliable diagnostic methods are needed. Artificial intelligence (AI) is emerging as a powerful tool for aided diagnosis but often relies on abstract test and validation sets while real-world behavior is unknown. In this study, we developed a 2-stage AI system for histopathological assessment of BE-related dysplasia using deep learning to enhance the efficiency and accuracy of the pathology workflow. The AI system was developed and trained on 290 whole-slide images (WSIs) that were annotated at glandular and tissue levels. The system was designed to identify individual glands, grade dysplasia, and assign a WSI-level diagnosis. The proposed method was evaluated by comparing the performance of our AI system with that of a large international and heterogeneous group of 55 gastrointestinal pathologists assessing 55 digitized biopsies spanning the complete spectrum of BE-related dysplasia. The AI system correctly graded 76.4% of the WSIs, surpassing the performance of 53 out of the 55 participating pathologists. Furthermore, the receiver-operating characteristic analysis showed that the system's ability to predict the absence (nondysplastic BE) versus the presence of any dysplasia was with an area under the curve of 0.94 and a sensitivity of 0.92 at a specificity of 0.94. These findings demonstrate that this AI system has the potential to assist pathologists in assessment of BE-related dysplasia. The system's outputs could provide a reliable and consistent secondary diagnosis in challenging cases or be used for triaging low-risk nondysplastic biopsies, thereby reducing the workload of pathologists and increasing throughput.
ABSTRACT
BACKGROUND: This study evaluated the association of pathological tumour response (tumour regression grade, TRG) and a novel scoring system, combining both TRG and nodal status (TRG-ypN score; TRG1-ypN0, TRG>1-ypN0, TRG1-ypN+ and TRG>1-ypN+), with recurrence patterns and survival after multimodal treatment of oesophageal adenocarcinoma. METHODS: This Dutch nationwide cohort study included patients treated with neoadjuvant chemoradiotherapy followed by oesophagectomy for distal oesophageal or gastro-oesophageal junctional adenocarcinoma between 2007 and 2016. The primary endpoint was the association of Mandard score and TRG-ypN score with recurrence patterns (rate, location, and time to recurrence). The secondary endpoint was overall survival. RESULTS: Among 2746 inclusions, recurrence rates increased with higher Mandard scores (TRG1 30.6%, TRG2 44.9%, TRG3 52.9%, TRG4 61.4%, TRG5 58.2%; P < 0.001). Among patients with recurrent disease, the distribution (locoregional versus distant) was the same for the different TRG groups. Patients with TRG1 developed more brain recurrences (17.7 versus 9.8%; P = 0.001) and had a longer mean overall survival (44 versus 35 months; P < 0.001) than those with TRG>1. The TRG>1-ypN+ group had the highest recurrence rate (64.9%) and worst overall survival (mean 27 months). Compared with the TRG>1-ypN0 group, patients with TRG1-ypN+ had a higher risk of recurrence (51.9 versus 39.6%; P < 0.001) and worse mean overall survival (33 versus 41 months; P < 0.001). CONCLUSION: Improved tumour response to neoadjuvant therapy was associated with lower recurrence rates and higher overall survival rates. Among patients with recurrent disease, TRG1 was associated with a higher incidence of brain recurrence than TRG>1. Residual nodal disease influenced prognosis more negatively than residual disease at the primary tumour site.
Subject(s)
Adenocarcinoma , Esophageal Neoplasms , Humans , Prognosis , Cohort Studies , Disease-Free Survival , Combined Modality TherapyABSTRACT
BACKGROUND: This study evaluated the proportion of patients with residual neoplasia after endoscopic resection (ER) for Barrett's neoplasia with confirmed tumor-positive vertical resection margin (R1v). METHODS: This retrospective cohort study included patients undergoing ER for Barrett's neoplasia with histologically documented R1v since 2008 in the Dutch Barrett Expert Centers. We defined R1v as cancer cells touching vertical resection margins and Rx as nonassessable margins. Reassessment of R1v specimens was performed by experienced pathologists until consensus was reached regarding vertical margins. RESULTS: 101/110 included patients had macroscopically complete resections (17 T1a, 84 T1b), and 99/101 (98%) ER specimens were histologically reassessed, with R1v confirmed in 74 patients (75%), Rx in 16%, and R0 in 9%. Presence/absence of residual neoplasia could be assessed in 66/74 patients during endoscopic reassessment (52) and/or in the surgical resection specimen (14), and 33/66 (50%) had residual neoplasia. Residual neoplasia detected during endoscopy was always endoscopically visible and biopsies from a normal-appearing ER scar did not detect additional neoplasia. Of 25 patients who underwent endoscopic follow-up (median 37 months [interquartile range 12-50]), 4 developed local recurrence (16.0%), all detected as visible abnormalities. CONCLUSIONS: After ER with R1v, 50% of patients had no residual neoplasia. Histological evaluation of ER margins appears challenging, as in this study 75% of documented R1v cases were confirmed during reassessment. Endoscopic reassessment 8-12 weeks after ER seems to accurately detect residual neoplasia and can help to determine the most appropriate strategy for patients with R1v.
Subject(s)
Barrett Esophagus , Esophageal Neoplasms , Margins of Excision , Neoplasm, Residual , Humans , Barrett Esophagus/pathology , Barrett Esophagus/surgery , Male , Female , Retrospective Studies , Aged , Middle Aged , Esophageal Neoplasms/surgery , Esophageal Neoplasms/pathology , Esophagoscopy/methods , Endoscopic Mucosal Resection , Netherlands , Adenocarcinoma/surgery , Adenocarcinoma/pathology , BiopsyABSTRACT
BACKGROUND: Long-term anti-angiogenesis leads to pruned vasculature, densely deposited extracellular matrix (ECM), and consequently reduced chemotherapy delivery in esophagogastric cancer (EGC). To address this issue, we evaluated the efficacy of adding a hyaluronidase or a NO-donor to the regimen of chemotherapy and anti-angiogenic drugs. METHODS: A patient-derived EGC xenograft model was developed. Grafted mice were randomly assigned to four experimental groups and one control group. The experimental groups received DC101, a murine angiogenesis inhibitor, and nab-paclitaxel (NPTX), with the addition of hyaluronidase (PEGPH20), or NO-donor (nitroglycerine, NTG), or their combination, respectively. We compared tumor growth during 17 days of treatment. We performed immunohistochemistry for ECM components hyaluronan (HA) and collagen, CD31 for endothelial cells, and γH2AX for DNA damage. The positively stained areas were quantified, and vessel diameters were measured using QuPath software. RESULTS: Prolonged DC101 treatment induced deposition of HA (p<0.01) and collagen (p<0.01). HA was effectively degraded by PEGPH20 (p<0.001), but not by NTG as expected. Both PEGPH20 (p<0.05) and NTG (p<0.01) dilated vessels collapsed in response to long-term DC101 treatment. However, only PEGPH20 (rather than NTG) was found to significantly inhibit tumor growth (p<0.05) in combination with NPTX and DC101. CONCLUSIONS: These findings suggest that the mechanical barrier of HA is the major reason responsible for the resistance developed during prolonged anti-angiogenesis in EGC. Incorporating PEGPH20 into the existing treatment regimen is promising to improve outcomes for patients with EGC.
ABSTRACT
BACKGROUND: HER2 targeting in esophageal adenocarcinoma (EAC) has shown potential, but often fails to show durable response. Given the contributions of the tumor immune microenvironment (TIME) to therapeutic responses, we aimed to chart the TIME characteristics of HER2 positive tumors. METHODS: 84 biopsies were taken from the TRAP cohort (neoadjuvant chemoradiotherapy (nCRT) according to CROSS with trastuzumab and pertuzumab; n = 40; HER2+n = 40) and a control cohort with nCRT only (n = 44; HER2- n = 40, HER2+n = 4) before treatment. Biopsies were analysed using targeted gene expression analysis (Nanostring immune-oncology panel, 750 genes). Differential gene expression was assessed between HER2 positive (n = 44) vs. negative biopsies (n = 40), and non-responders (n = 17) vs. responders (n = 23) to anti-HER2 treatment. Statistical significance was determined as p-value <0.05, adjusted for multiple testing correction. RESULTS: 83 biopsies were eligible for analyses following quality control (TRAP cohort n = 40; control cohort n = 43); there were no significant differences in clinical characteristics between the TRAP vs. control the cohort or HER2 positive vs. HER2 negative biopsies. HER2 expression was found to associate with epithelial markers (EPCAM p < 0.001; E-cadherin p < 0.001). Moreover, HER2 expression was associated with a lower expression of immune cell infiltration, such as NK-cells (p < 0.001) and CD8 T-cells (p < 0.001), but also lower expression of immune exhaustion markers (PDCD1LG2, CTLA4; p < 0.001). In non-responders to anti-HER2 treatment, baseline biopsies showed increased expression of immune exhaustion markers, as well as hypoxia and VEGF signalling. DISCUSSION: HER2 expression was associated with epithelial tumor characteristics. The HER2 positive TIME showed reduced immune cell infiltration but also lower expression of inhibitory signals associated with immune exhaustion, questioning the mechanism behind potential clinical benefit of co-administration of anti-HER2 agents and checkpoint inhibitors. As limited response was associated with increased VEGF signalling, studies could investigate potential synergism of targeting VEGF and HER2.
ABSTRACT
BACKGROUND: Despite the advent of neoadjuvant chemoradiotherapy (CRT), overall survival rates of esophageal adenocarcinoma (EAC) remain low. A readily induced mesenchymal transition of EAC cells contributes to resistance to CRT. METHODS: In this study, we aimed to chart the heterogeneity in cell state transition after CRT and to identify its underpinnings. A panel of 12 esophageal cultures were treated with CRT and ranked by their relative epithelial-mesenchymal plasticity. RNA-sequencing was performed on 100 pre-treatment biopsies. After RNA-sequencing, Ridge regression analysis was applied to correlate gene expression to ranked plasticity, and models were developed to predict mesenchymal transitions in patients. Plasticity score predictions of the three highest significant predictive models were projected on the pre-treatment biopsies and related to clinical outcome data. Motif enrichment analysis of the genes associated with all three models was performed. RESULTS: This study reveals NANOG as the key associated transcription factor predicting mesenchymal plasticity in EAC. Expression of NANOG in pre-treatment biopsies is highly associated with poor response to neoadjuvant chemoradiation, the occurrence of recurrences, and median overall survival difference in EAC patients (>48 months). Perturbation of NANOG reduces plasticity and resensitizes cell lines, organoid cultures, and patient-derived in vivo grafts. CONCLUSIONS: In conclusion, NANOG is a key transcription factor in mesenchymal plasticity in EAC and a promising predictive marker for outcome.
Esophageal cancer is the sixth most common cause of cancer-related death worldwide. Although chemotherapy combined with radiotherapy (chemoradiotherapy) followed by surgery has improved survival, tumor recurrence and metastatic disease (that has spread to other parts of the body) are often observed after several months. In this study, we assessed the effect of chemoradiotherapy on esophageal cells in the lab to predict the effect in patients with esophageal cancer. To investigate this, genes were assessed from 12 different cell lines and 100 patient tissues. We revealed that levels of one of the genes, NANOG, associates with poor response in patients. NANOG could be a promising marker to predict outcome in patients with esophageal cancer. This knowledge might help clinicians to treat patients with esophageal cancer appropriately, or may lead to new or optimized treatments.
ABSTRACT
Background and Aims: Barrett's esophagus (BE) is the precursor lesion for esophageal adenocarcinoma (EAC). To detect EAC in early stage, patients with BE undergo endoscopic surveillance. Surveillance cohorts largely consist of nondysplastic BE (NDBE) patients with a low annual progression risk (<0.5%). Predictive biomarkers for malignant progression of NDBE could improve efficacy of surveillance. Biomarker research has mostly focused on aberrant protein expression on BE epithelial cells. Moreover, insight in cell signaling driving malignant transformation is unknown. This study uses a data-driven approach to analyze tumor-stroma interaction in NDBE which progressed to high-grade dysplasia or EAC. Methods: In this case-control study, we performed RNA sequencing analysis on index NDBE biopsies from 6 patients who, during long-term follow-up, progressed and 7 who did not progress to high-grade dysplasia/EAC. For control samples, squamous and duodenum tissues from BE patients were analyzed. For validation, we used quantitative PCR. Results: Significant differences in BE transcriptomic profiles between progressors and nonprogressors were found by principal component and differential expression analyses. Ingenuity pathway analysis indicated that 8 cell signaling pathways were significantly upregulated in the progressors, and 14 pathways were significantly downregulated. The most interesting finding was the upregulation of the xenobiotic metabolism pregnane X receptor signaling pathway in the progressor cohort, while of the downregulated pathways in progressors, several were related to the immune system. Conclusion: These novel transcriptomic insights are fundamental for developing (chemo-)preventive therapies. These could be therapies, which protect against toxins, including biles, responsible for pregnane X receptor activation or which enhance protective immune mechanisms. The identified RNA markers are promising biomarkers for improving risk stratification in surveillance programs.