Pilot feasibility study of multispectral imaging in Barrett's

• Research type

  Research Study

• Full title

  Prospective pilot cohort study to assess feasibility of multispectral endoscopic imaging for detection of early neoplasia in Barrett's oesophagus

• IRAS ID

  233522

• Contact name

  Massimiliano di Pietro

• Contact email

  md460@cam.ac.uk

• Sponsor organisation

  Cambridge University Hospitals NHS Foundation Trust and the University of Cambridge

• Clinicaltrials.gov Identifier

  NCT03388047

• Duration of Study in the UK

  0 years, 11 months, 31 days

• Research summary

  Summary of Research
  Barrett’s oesophagus (BO) is a metaplastic change in the lining of the oesophagus that predisposes to oesophageal adenocarcinoma. Patients with BO are monitored with a periodic endoscopy and multiple random biopsies (endoscopic surveillance) to detect early neoplastic changes also known as dysplasia. However this practice is highly controversial due to the sampling error inherent in untargeted biopsies. Therefore it is crucial to improve endoscopic visualisation of dysplasia with advanced imaging techniques.

  Multispectral imaging represents an exciting new field of investigation in endoscopic research. Multispectral imaging uses a specialised camera to detect multiple colours, allowing us to build a rough spectrum from each point in our image. It is widely reported that these spectra are different for different tissue types, but this is difficult to study ex vivo since de-oxygenation of the blood and decay of the tissue changes these spectra considerably.

  We have therefore designed this study to investigate the different multispectral imaging spectra of Barrett's Oesophagus and dysplasia, which we believe might be utilised as a method to detect dysplasia in the future.

  Summary of Results
  Early cancer of the oesophagus can be invisible on standard endoscopy (camera) test. Multi-spectral imaging uses a specialised camera to detect multiple colours, allowing to measure the intensity of each colour in an image.. It is widely reported that these colours (spectra) are different for different tissue types based on the blood content and oxygenation levels, but this is difficult to measure with standard cameras. In this study we analysed patients with pre-cancerous and early cancerous condition of the gullet with a multispectral camera during a standard endoscopic examination. We found that early cancer has different spectra com-pared to normal tissue and we were able to predict the presence of early cancerous changes with 85% accuracy. The new colour camera increased 12-folds the contrast between normal lining and cancer compared to standard white-light endoscopy.

  Detailed Description:
  We have developed a custom multispectral endoscope based around a CE marked device, the PolyScope disposable endoscope (PolyDiagnost). This is a combined sterile catheter and fibre optic device designed to optimize light delivery to the anatomy and to acquire and transmit endoscopic images back to a camera. This custom spectral endoscope is deployed in the lumen of GI organ via the accessory channel of the endoscope and captures detailed attenuation spectra of GI mucosa. We undertook a pilot clinical study to acquire in vivo esophageal tissue spectra matched with gold standard histopathological diagnosis of disease state. The acquired spectra were then analyzed using both model-based and learning-based methods We recruited 20 patients (15 completed the image acquisition) and captured 715 in vivo tissue spectra matched with gold standard diagnosis from histopathology. Spectral endoscopy was sensitive to changes in neovascularization during the progression of disease; both non-dysplastic and neoplastic Barrett’s oesophagus showed higher blood volume relative to healthy squamous tissue (P= 0.001 and 0.02, respectively), and vessel radius appeared larger in neoplasia relative to non-dysplastic Barrett’s esophagus (P= 0.06). We further devel-oped a deep learning algorithm capable of classifying spectra of neoplasia versus non-dysplastic Barrett’s oesophagus with high accuracy (84.8% accuracy, 83.7% sensitivity, 85.5% specificity, 78.3% positive predictive value, and 89.4% negative predictive value). Exploiting the newly acquired library of labeled spectra to model custom color filter sets identified a po-tential 12-fold enhancement in contrast between neoplasia and non-dysplastic Barrett’s oesophagus using application-specific color filters compared with standard-of-care white-light imaging (perceptible color difference = 32.4 and 2.7, respectively). This work demonstrates the potential of endoscopic spectral imaging to extract vascular properties in Barrett’s oesophagus, to classify disease stages using deep learning, and to enable high-contrast endoscopy.
  Significance: The results of this pilot first-in-human clinical trial demonstrate the potential of spectral endoscopy to reveal disease associated vascular changes and to provide high-contrast delineation of neoplasia in the oesophagus

• REC name

  North West - Liverpool Central Research Ethics Committee

• REC reference

  18/NW/0134

• Date of REC Opinion

  26 Feb 2018

• REC opinion

  Favourable Opinion