Advancing Precision in Barrett's Esophagus and Dysplasia Diagnoses: WATS3D Study Demonstrates Exceptional Consensus Among Pathologists

SUFFERN, N.Y., March 6, 2024 – CDx Diagnostics, Inc., developer of the WATS^3D AI platform for the detection and surveillance of Barrett’s esophagus (BE) and dysplasia, today announced the publication of "WATS^3D: An Interobserver Study of Barrett's Esophagus–Associated Dysplasia Among Gastrointestinal Pathologists" in Clinical and Translational Gastroenterology. This study provides new insights into the use of CDx’s AI-powered diagnostic technology to consistently identify precancerous cells in patients with BE, thereby helping to prevent the development of esophageal cancer.

This was a collaborative study among gastrointestinal pathologists who have an interest and expertise in Barrett’s esophagus. The lead author in this study was Dr. Deepa Patil of Brigham and Women’s Hospital, Boston, MA. Additional contributors include: John R. Goldblum, MD (Cleveland Clinic), Gregory Lauwers, MD (Moffitt Cancer Center), Jason T. Lewis, MD (Mayo Clinic), Marie Robert, MD (Yale University School of Medicine), Mendel Singer, PhD, MPH, (Case Western Reserve University School of Medicine), and senior author, Robert D. Odze, MD (Tufts Medical Center).

WATS^3D Study Demonstrates Exceptional Consensus Among Pathologists

The study set out to address the challenges in grading BE and dysplasia where the histopathological diagnosis of Barrett's esophagus-associated dysplasia has poor inter-observer agreement even among experienced gastrointestinal pathologists.^1-3 Published estimates for the interobserver agreement ("kappa values") for diagnosing dysplasia on forceps biopsy in Barrett's esophagus have varied between 0.15 and 0.69.^1,4

Within this investigation, five GI pathologists were trained in-person and virtually on specimens from CDx’s proprietary diagnostic platform, WATS^3D, which leverages AI-enabled tissue analysis and 3D-imaging to reliably identify precancerous cells in the esophagus. The pathologists were then asked to evaluate digital images from 60 WATS^3D cases with BE.

“This study tackles a crucial clinical hurdle in gastroenterology,” remarked Dr. Deepa Patil, MD, lead author. “The significant variability among pathologists assessing Barrett's esophagus poses challenges for accurate diagnosis and treatment decisions, complicating patient care.”

The overall mean kappa value in this study for all diagnoses for the five observers was calculated at 0.93, demonstrating remarkable consistency in interpreting both cell block and smear specimens (kappa=0.93 and 0.97, respectively). Kappa values of above 0.80 indicate nearly perfect agreement (Landis and Koch scale).⁵ The data represents significant improvement compared to standard histopathology in the diagnosis of BE and dysplasia where the kappa value is only considered to be “fair agreement” - 0.30. In addition, agreement was perfect (100%) regarding detection of neoplasia (either LGD, HGD, or EAC).

"Our findings show a high level of accuracy and reproducibility among GI pathologists who have had no prior experience with the WATS3D platform,” stated Dr. Robert Odze.

“These outcomes are a testament to the effectiveness of CDx Diagnostics’ approach to diagnosis. The results define a path to saving more patient lives by significantly improving diagnostic practices and delivering consistent dysplasia detection rates to more clinicians. Taken together with the fact that CDx recently processed and diagnosed its 400,000th WATS^3D specimen, the study underscores the reliability and scalability of the technology.” said Duane Dorn, Chief Operating Officer at CDx Diagnostics.

References
1. Montgomery E, Bronner MP, Goldblum JR et al. Reproducibility of the diagnosis of dysplasia in Barrett esophagus: a reaffirmation. Hum Pathol 2001;32:368-78.
2. Reid BJ, Haggitt RC, Rubin CE et al. Observer variation in the diagnosis of dysplasia in Barrett's esophagus . Hum Pathol 1988 ; 19 : 166-78.
3. Kerkhof M, van Dekken H, Steyerberg EW et al. Grading of dysplasia in Barrett's oesophagus: substantial interobserver variation between general and gastrointestinal pathologists. Histopathology 2007 ; 50 : 920 - 7.
4. Pech O, Vieth M, Schmitz D et al. Conclusions from the histological diagnosis of low-grade intraepithelial neoplasia in Barrett's oesophagus. Scand J Gastroenterol 2007;42:682-8.
5. Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics 1977; 33:159-74.

About CDx Diagnostics
CDx Diagnostics’ mission is to provide clinicians with easily implemented, cost effective tools to preempt cancer through enhanced detection of precancerous change. This is accomplished by a proprietary diagnostic platform that synthesizes computer imaging, artificial intelligence, molecular biology and three-dimensional cytopathology to detect precancerous change earlier and more reliably than prior methods. CDx tests require only a few minutes of practice time, are highly cost effective, widely reimbursed, and address a recognized critical gap in the current diagnostic standard of care that results in thousands of otherwise unnecessary cancer deaths each year. Routine clinical use of CDx testing in the oral cavity and esophagus has prevented thousands of cancers, and application of the CDx diagnostic platform to prevent cancers of the throat, bile duct, stomach, and surveillance for dysplasia of the colon in patients with IBD is currently in progress.

About WATS^3D
CDx Diagnostics’ WATS^3D addresses the major inadequacies inherent in current random forceps biopsy testing of the esophagus. In just a few minutes, endoscopists can easily obtain a wide area, full-thickness transepithelial tissue sample for computer-assisted 3D laboratory analysis by expert pathologists. In large multicenter clinical trials, WATS^3D has been found to significantly increase the detection rate of both Barrett's esophagus and esophageal dysplasia. The high sensitivity and inter-observer agreement of WATS^3D is due to the larger tissue area sampled, and the proprietary 3-Dimensionial computer imaging system that is based on an algorithm developed as part of the U.S. Strategic Defense Initiative missile defense program. To learn more about WATS^3D, visit www.wats3d.com.