A deep learning system for detection of early Barrett's neoplasia: a model development and validation study

BACKGROUND

Computer-aided detection (CADe) systems could assist endoscopists in detecting early neoplasia in Barrett's oesophagus, which could be difficult to detect in endoscopic images. The aim of this study was to develop, test, and benchmark a CADe system for early neoplasia in Barrett's oesophagus.

METHODS

The CADe system was first pretrained with ImageNet followed by domain-specific pretraining with GastroNet. We trained the CADe system on a dataset of 14 046 images (2506 patients) of confirmed Barrett's oesophagus neoplasia and non-dysplastic Barrett's oesophagus from 15 centres. Neoplasia was delineated by 14 Barrett's oesophagus experts for all datasets. We tested the performance of the CADe system on two independent test sets. The all-comers test set comprised 327 (73 patients) non-dysplastic Barrett's oesophagus images, 82 (46 patients) neoplastic images, 180 (66 of the same patients) non-dysplastic Barrett's oesophagus videos, and 71 (45 of the same patients) neoplastic videos. The benchmarking test set comprised 100 (50 patients) neoplastic images, 300 (125 patients) non-dysplastic images, 47 (47 of the same patients) neoplastic videos, and 141 (82 of the same patients) non-dysplastic videos, and was enriched with subtle neoplasia cases. The benchmarking test set was evaluated by 112 endoscopists from six countries (first without CADe and, after 6 weeks, with CADe) and by 28 external international Barrett's oesophagus experts. The primary outcome was the sensitivity of Barrett's neoplasia detection by general endoscopists without CADe assistance versus with CADe assistance on the benchmarking test set. We compared sensitivity using a mixed-effects logistic regression model with conditional odds ratios (ORs; likelihood profile 95% CIs).

FINDINGS

Sensitivity for neoplasia detection among endoscopists increased from 74% to 88% with CADe assistance (OR 2·04; 95% CI 1·73-2·42; p<0·0001 for images and from 67% to 79% [2·35; 1·90-2·94; p<0·0001] for video) without compromising specificity (from 89% to 90% [1·07; 0·96-1·19; p=0·20] for images and from 96% to 94% [0·94; 0·79-1·11; ] for video; p=0·46). In the all-comers test set, CADe detected neoplastic lesions in 95% (88-98) of images and 97% (90-99) of videos. In the benchmarking test set, the CADe system was superior to endoscopists in detecting neoplasia (90% vs 74% [OR 3·75; 95% CI 1·93-8·05; p=0·0002] for images and 91% vs 67% [11·68; 3·85-47·53; p<0·0001] for video) and non-inferior to Barrett's oesophagus experts (90% vs 87% [OR 1·74; 95% CI 0·83-3·65] for images and 91% vs 86% [2·94; 0·99-11·40] for video).

INTERPRETATION

CADe outperformed endoscopists in detecting Barrett's oesophagus neoplasia and, when used as an assistive tool, it improved their detection rate. CADe detected virtually all neoplasia in a test set of consecutive cases.

FUNDING

Olympus.

Long-term oncological outcomes of endoscopic full-thickness resection after previous incomplete resection of low-risk T1 CRC (LOCAL-study): study protocol of a national prospective cohort study

BACKGROUND

T1 colorectal cancer (CRC) without histological high-risk factors for lymph node metastasis (LNM) can potentially be cured by endoscopic resection, which is associated with significantly lower morbidity, mortality and costs compared to radical surgery. An important prerequisite for endoscopic resection as definite treatment is the histological confirmation of tumour-free resection margins. Incomplete resection with involved (R1) or indeterminate (Rx) margins is considered a strong risk factor for residual disease and local recurrence. Therefore, international guidelines recommend additional surgery in case of R1/Rx resection, even in absence of high-risk factors for LNM. Endoscopic full-thickness resection (eFTR) is a relatively new technique that allows transmural resection of colorectal lesions. Local scar excision after prior R1/Rx resection of low-risk T1 CRC could offer an attractive minimal invasive strategy to achieve confirmation about radicality of the previous resection or a second attempt for radical resection of residual luminal cancer. However, oncologic safety has not been established and long-term data are lacking. Besides, surveillance varies widely and requires standardization.

METHODS/DESIGN

In this nationwide, multicenter, prospective cohort study we aim to assess feasibility and oncological safety of completion eFTR following incomplete resection of low-risk T1 CRC. The primary endpoint is to assess the 2 and 5 year luminal local tumor recurrence rate. Secondary study endpoints are to assess feasibility, percentage of curative eFTR-resections, presence of scar tissue and/or complete scar excision at histopathology, safety of eFTR compared to surgery, 2 and 5 year nodal and/or distant tumor recurrence rate and 5-year disease-specific and overall-survival rate.

DISCUSSION

Since the implementation of CRC screening programs, the diagnostic rate of T1 CRC is steadily increasing. A significant proportion is not recognized as cancer before endoscopic resection and is therefore resected through conventional techniques primarily reserved for benign polyps. As such, precise histological assessment is often hampered due to cauterization and fragmentation and frequently leads to treatment dilemmas. This first prospective trial will potentially demonstrate the effectiveness and oncological safety of completion eFTR for patients who have undergone a previous incomplete T1 CRC resection. Hereby, substantial surgical overtreatment may be avoided, leading to treatment optimization and organ preservation. Trial registration Nederlands Trial Register, NL 7879, 16 July 2019 ( https://trialregister.nl/trial/7879 ).

Colorectal anastomotic leak: transcriptomic profile analysis

BACKGROUND

Anastomotic leakage in patients undergoing colorectal surgery is associated with morbidity and mortality. Although multiple risk factors have been identified, the underlying mechanisms are mainly unknown. The aim of this study was to perform a transcriptome analysis of genes underlying the development of anastomotic leakage.

METHODS

A set of human samples from the anastomotic site collected during stapled colorectal anastomosis were used in the study. Transcriptomic profiles were generated for patients who developing anastomotic leakage and case-matched controls with normal anastomotic healing to identify genes and biological processes associated with the development of anastomotic leakage.

RESULTS

The analysis included 22 patients with and 69 without anastomotic leakage. Differential expression analysis showed that 44 genes had adjusted P < 0.050, consisting of two upregulated and 42 downregulated genes. Co-functionality analysis of the 150 most upregulated and 150 most downregulated genes using the GenetICA framework showed formation of clusters of genes with different enrichment for biological pathways. The enriched pathways for the downregulated genes are involved in immune response, angiogenesis, protein metabolism, and collagen cross-linking. The enriched pathways for upregulated genes are involved in cell division.

CONCLUSION

These data indicate that patients who develop anastomotic leakage start the healing process with an error at the level of gene regulation at the time of surgery. Despite normal macroscopic appearance during surgery, the transcriptome data identified several differences in gene expression between patients who developed anastomotic leakage and those who did not. The expressed genes and enriched processes are involved in the different stages of wound healing. These provide therapeutic and diagnostic targets for patients at risk of anastomotic leakage.

Preoperative image-guided identification of response to neoadjuvant chemoradiotherapy in esophageal cancer (PRIDE): a multicenter observational study

BACKGROUND

Nearly one third of patients undergoing neoadjuvant chemoradiotherapy (nCRT) for locally advanced esophageal cancer have a pathologic complete response (pCR) of the primary tumor upon histopathological evaluation of the resection specimen. The primary aim of this study is to develop a model that predicts the probability of pCR to nCRT in esophageal cancer, based on diffusion-weighted magnetic resonance imaging (DW-MRI), dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) and ¹⁸F-fluorodeoxyglucose positron emission tomography with computed tomography (¹⁸F-FDG PET-CT). Accurate response prediction could lead to a patient-tailored approach with omission of surgery in the future in case of predicted pCR or additional neoadjuvant treatment in case of non-pCR.

METHODS

The PRIDE study is a prospective, single arm, observational multicenter study designed to develop a multimodal prediction model for histopathological response to nCRT for esophageal cancer. A total of 200 patients with locally advanced esophageal cancer - of which at least 130 patients with adenocarcinoma and at least 61 patients with squamous cell carcinoma - scheduled to receive nCRT followed by esophagectomy will be included. The primary modalities to be incorporated in the prediction model are quantitative parameters derived from MRI and ¹⁸F-FDG PET-CT scans, which will be acquired at fixed intervals before, during and after nCRT. Secondary modalities include blood samples for analysis of the presence of circulating tumor DNA (ctDNA) at 3 time-points (before, during and after nCRT), and an endoscopy with (random) bite-on-bite biopsies of the primary tumor site and other suspected lesions in the esophagus as well as an endoscopic ultrasonography (EUS) with fine needle aspiration of suspected lymph nodes after finishing nCRT. The main study endpoint is the performance of the model for pCR prediction. Secondary endpoints include progression-free and overall survival.

DISCUSSION

If the multimodal PRIDE concept provides high predictive performance for pCR, the results of this study will play an important role in accurate identification of esophageal cancer patients with a pCR to nCRT. These patients might benefit from a patient-tailored approach with omission of surgery in the future. Vice versa, patients with non-pCR might benefit from additional neoadjuvant treatment, or ineffective therapy could be stopped.

TRIAL REGISTRATION

The article reports on a health care intervention on human participants and was prospectively registered on March 22, 2018 under ClinicalTrials.gov Identifier: NCT03474341 .

Impact of Endoscopic Ultrasonography on 18F-FDG-PET/CT Upfront Towards Patient Specific Esophageal Cancer Treatment

INTRODUCTION

In patients with potentially resectable esophageal cancer (EC), the value of endoscopic ultrasonography (EUS) after fluorine-18 labeled fluorodeoxyglucose positron emission tomography with computed tomography (¹⁸F-FDG-PET/CT) is questionable. Retrospectively, we assessed the impact of EUS after PET/CT on the given treatment in EC patients.

METHODS

During the period 2009-2015, 318 EC patients were staged as T1-4aN0-3M0 with hybrid ¹⁸F-FDG-PET/CT or ¹⁸F-FDG-PET with CT and EUS if applicable in a nonspecific order. We determined the impact of EUS on the given treatment in 279 patients who also were staged with EUS. EUS had clinical consequences if it changed curability, extent of radiation fields or lymph node resection (AJCC stations 2-5), and when the performed fine-needle aspiration (FNA) provided conclusive information of suspicious lymph node.

RESULTS

EUS had an impact in 80 (28.7%) patients; it changed the radiation field in 63 (22.6%), curability in 5 (1.8%), lymphadenectomy in 48 (17.2%), and FNA was additional in 21 (7.5%). In patients treated with nCRT (n = 194), EUS influenced treatment in 53 (27.3%) patients; in 38 (19.6%) the radiation field changed, in 3 (1.5%) the curability, in 35 (18.0%) the lymphadenectomy, and in 17 (8.8%) FNA was additional. EUS influenced both the extent of radiation field and nodal resection in 31 (16.0%) nCRT patients.

CONCLUSIONS

EUS had an impact on the given treatment in approximately 29%. In most patients, the magnitude of EUS found expression in the extent of radiotherapy target volume delineation to upper/high mediastinal lymph nodes.

Abstract P4-01-01: Phase II in-human dose escalation study of the optical molecular imaging tracer bevacizumab-800cw for molecular fluorescence guided surgery in primary breast cancer patients

Introduction

Molecular Fluorescence Guided Surgery (MFGS) might be used for intraoperative detection of positive resection margins in breast conserving surgery (BCS) for the treatment of breast cancer. Currently, the presence of tumor positive resection margins can only be assessed ex vivo by histopathological analysis of the excised tissue, which takes up to 5 working days. The current study defined the optimal dose of the near-infrared (NIR) optical imaging tracer bevacizumab-800CW to enable intraoperative detection and image-guided pathology of tumor positive resection margins in breast cancer patients.

Methods

Molecular Fluorescence Guided Surgery during BCS was performed in subjects treated for primary breast cancer. The NIR optical imaging tracer bevacizumab-800CW targeting vascular endothelial growth factor A was administered intravenously three days prior to surgery in four escalating dose groups (4.5mg, 10mg, 25mg and 50mg). NIR fluorescent signals were detected real-time using an intraoperative fluorescence camera system (SurgVision BV). Standardized ex vivo analyses of tumor-to-normal ratios (TNR) were performed to define the optimal tracer dose using a BlackBox imaging system (SurgVision BV) for imaging fresh bread loaf slices, a NIR fluorescence flatbed scanner (Odyssey, Li-Cor) for imaging 10µm slices of formalin-fixed paraffin-embedded (FFPE) blocks, NIR Confocal Laser Endomicroscopy (CLE, Mauna Kea Technologies) and multi-diameter single fiber reflectance and single fiber fluorescence (MDSFR/SFF) spectroscopy quantification to enable correction for the influence of tissue optical properties on fluorescence in tumor and normal breast tissue.

Results

Currently, 12 subjects have been included and analyzed in four dosing groups. All tumors showed specific tracer uptake at macroscopic and microscopic level during ex vivo analyses, confirmed by histopathology. Quantification of NIR fluorescent signals showed higher TNRs by increasing doses up to 25mg. No further increase in TNR was seen in the 50mg dose group. CLE showed the feasibility of visualization of the tracer accumulation in tumor tissue compared to normal tissue at a microscopic level. MDSFR/SFF spectroscopy objectively confirmed the dose dependency up to 25mg.

Conclusion and future perspective

Intravenous administration of bevacizumab-800CW in doses up to 50mg is safe and highly tumor specific, showing a plateau of TNR at 25mg and 50mg. Further expansion of the dosing cohorts of 10mg and 25mg with additional seven patients per group will be performed to establish the optimal dose for MFGS during BCS for an upcoming phase III multicenter randomized controlled trial. By enabling MFGS during BCS the surgical treatment of primary breast cancer patients might be optimized by a reduced need for re-excision surgery and thereby reducing the risk of co-morbidity, psychological burden, poor cosmesis and healthcare costs.

Citation Format: van Dam GM, Koller M, Qiu SQ, Linssen MD, de Vries J, Jansen L, Kelder W, de Jong JS, Jorritsma-Smit A, van der Vegt B, Robinson DJ, Nagengast WB. Phase II in-human dose escalation study of the optical molecular imaging tracer bevacizumab-800cw for molecular fluorescence guided surgery in primary breast cancer patients [abstract]. In: Proceedings of the 2016 San Antonio Breast Cancer Symposium; 2016 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2017;77(4 Suppl):Abstract nr P4-01-01.

Optical innovations in surgery

BACKGROUND

In the past decade, there has been a major drive towards clinical translation of optical and, in particular, fluorescence imaging in surgery. In surgical oncology, radical surgery is characterized by the absence of positive resection margins, a critical factor in improving prognosis. Fluorescence imaging provides the surgeon with reliable and real-time intraoperative feedback to identify surgical targets, including positive tumour margins. It also may enable decisions on the possibility of intraoperative adjuvant treatment, such as brachytherapy, chemotherapy or emerging targeted photodynamic therapy (photoimmunotherapy).

METHODS

This article reviews the use of optical imaging for intraoperative guidance and decision-making.

RESULTS

Image-guided cancer surgery has the potential to be a powerful tool in guiding future surgical care. Photoimmunotherapy is a theranostic concept (simultaneous diagnosis and treatment) on the verge of clinical translation, and is highlighted as an effective combination of image-guided surgery and intraoperative treatment of residual disease. Multispectral optoacoustic tomography, a technique complementary to optical image-guided surgery, is currently being tested in humans and is anticipated to have great potential for perioperative and postoperative application in surgery.

CONCLUSION

Significant advances have been achieved in real-time optical imaging strategies for intraoperative tumour detection and margin assessment. Optical imaging holds promise in achieving the highest percentage of negative surgical margins and in early detection of micrometastastic disease over the next decade.

Use of S-100B to evaluate therapy effects during bevacizumab induction treatment in AJCC stage III melanoma

AIM

To investigate the feasibility of using bevacizumab to improve the survival of American Joint Committee on Cancer (AJCC) stage III melanoma patients, we investigated how a single bevacizumab treatment affected nodal disease and a panel of biomarkers in clinically fluorodeoxyglucose positron emission tomography (FDG-PET)/computed tomography (CT)-staged, stage III melanoma patients, prior to therapeutic lymph node dissection (TLND).

METHODS

Four weeks before TLND, nine patients (median age 50, range 28.8-62.1 years; two male, seven female) with palpable lymph node metastases received 7.5 mg/kg bevacizumab. Before and after this treatment, all patients were assessed by measurements of the maximum standardized uptake value (SUVmax) by FDG-PET scan, and serum S-100B and lactate dehydrogenase (LDH). After TLND, the dissection specimen was analyzed for number of removed lymph nodes, number of metastatic lymph nodes, and tumor necrosis.

RESULTS

Median follow-up was 15.5 (2.2-32.9) months. Histopathological analysis revealed tumor necrosis in six patients, of whom five had an S-100B decline and one had an unchanged S-100B level after bevacizumab. The other three patients showed an S-100B increase and no necrosis. Tumor necrosis was correlated with S-100B decrease (P = 0.048). No association was found between necrosis and the markers SUVmax and LDH. No wound healing disturbances were encountered.

CONCLUSION

Tumor necrosis in dissection specimens was associated with declining S-100B levels, while elevated S-100B was only found in cases with no necrosis. Bevacizumab might be useful in treating AJCC stage III melanoma patients prior to TLND, and S100-B appears to be a useful marker for assessment of treatment effects.

Molecular imaging of breast cancer

Molecular imaging of breast cancer can potentially be used for breast cancer screening, staging, restaging, response evaluation and guiding therapies. Techniques for molecular breast cancer imaging include magnetic resonance imaging (MRI), optical imaging, and radionuclide imaging with positron emission tomography (PET) or single photon emission computed tomography (SPECT). This review focuses on PET and SPECT imaging which can provide sensitive serial non invasive information of tumor characteristics. Most clinical data are gathered on the visualization of general processes such as glucose metabolism with the PET-tracer [(18)F]fluorodeoxyglucose (FDG) and DNA synthesis with [18F]fluoro-L-thymidine (FLT). Increasingly more breast cancer specific targets are imaged such as the estrogen receptor (ER), growth factors and growth factor receptors. Imaging of the ER with the PET tracer 16-alpha-[(18)F]fluoro-17-beta-estradiol (FES) has shown a good correlation between FES tumor uptake and ER density. (111)In-trastuzumab SPECT to image the human epidermal growth factor receptor 2 (HER2) showed that in most patients with metastatic HER2 overexpressing disease more lesions were detected than with conventional staging procedures. The PET tracer (89)Zr-trastuzumab showed excellent, quantifiable, and specific tumor uptake. (111)In-bevacizumab for SPECT and (89)Zr-bevacizumab for PET-imaging have been developed for vascular endothelial growth factor (VEGF) imaging as an angiogenic marker. Lastly, tracers for the receptors EGFR, IGF-1R, PDGF-betaR and the ligand TGFbeta are under development. Although molecular imaging of breast cancer is still not commonly used in daily clinical practice, its application portfolio is expanding rapidly.

Irradiation of rat brain reduces P-glycoprotein expression and function

The blood–brain barrier (BBB) hampers delivery of several drugs including chemotherapeutics to the brain. The drug efflux pump P-glycoprotein (P-gp), expressed on brain capillary endothelial cells, is part of the BBB. P-gp expression on capillary endothelium decreases 5 days after brain irradiation, which may reduce P-gp function and increase brain levels of P-gp substrates. To elucidate whether radiation therapy reduces P-gp expression and function in the brain, right hemispheres of rats were irradiated with single doses of 2–25 Gy followed by 10 mg kg⁻¹ of the P-gp substrate cyclosporine A (CsA) intravenously (i.v.), with once 15 Gy followed by CsA (10, 15 or 20 mg kg⁻¹), or with fractionated irradiation (4 × 5 Gy) followed by CsA (10 mg kg⁻¹) 5 days later. Additionally, four groups of three rats received 25 Gy once and were killed 10, 15, 20 or 25 days later. The brains were removed and P-gp detected immunohistochemically. P-gp function was assessed by [¹¹C]carvedilol uptake using quantitative autoradiography. Irradiation increased [¹¹C]carvedilol uptake dose-dependently, to a maximum of 20% above non irradiated hemisphere. CsA increased [¹¹C]carvedilol uptake dose-dependently in both hemispheres, but more (P<0.001) in the irradiated hemisphere. Fractionated irradiation resulted in a lost P-gp expression 10 days after start irradiation, which coincided with increased [¹¹C]carvedilol uptake. P-gp expression decreased between day 15 and 20 after single dose irradiation, and increased again thereafter. Rat brain irradiation results in a temporary decreased P-gp function.

Evidence for a three-nucleon-force effect in proton-deuteron elastic scattering

Developments in spin-polarized internal targets for storage rings have permitted measurements of 197 MeV polarized protons scattering from vector polarized deuterons. This work presents measurements of the polarization observables A(y), iT11, and C(y,y) in proton-deuteron elastic scattering. When compared to calculations with and without three-nucleon forces, the measurements provide further evidence that three-nucleon forces make a contribution to the observables. This work indicates that three-body forces derived from static nuclear properties appear to be crucial to the description of dynamical properties.

High-power single-mode emission from a broad-area semiconductor laser with a pseudoexternal cavity and a Fabry Perot etalon

A simple pseudoexternal cavity with a commerically available broad-area laser diode, standard optics, and a thin FabryPerot etalon provides single-mode emission with 1-W optical power and a tuning range of ~0.5nm . The linewidth is estimated at 200MHz. Such high-power semiconductor laser systems are promising light sources for atomic physics, in which narrow-band high-power light is necessary, for example, for optical pumping of alkali metal vapor. Methods for expanding the tuning range and achieving higher narrow-band optical power are discussed.