Nashville, Tennessee, U.S., October 12, 2026
Castle Biosciences, Inc. has announced new clinical data showing that its TissueCypher Barrett’s Esophagus test can identify patients at elevated risk of progression to high-grade dysplasia or esophageal adenocarcinoma beyond what can be determined from traditional pathology findings alone. The findings were presented at the American College of Gastroenterology (ACG) 2026 Annual Scientific Meeting, highlighting the potential role of molecular diagnostics in improving risk assessment for people with Barrett’s esophagus. Barrett’s esophagus is a condition in which the normal lining of the esophagus changes, often as a result of chronic gastroesophageal reflux disease. Although many patients do not develop cancer, a subset may progress to high-grade dysplasia or esophageal adenocarcinoma, making accurate risk stratification important for clinical management. TissueCypher analyzes tissue samples obtained during routine endoscopy and uses a proprietary algorithm to evaluate biological features associated with disease progression. The new findings add to the growing evidence base for combining molecular information with conventional pathology to help clinicians better understand individual patient risk. The test is intended to complement, not replace, established clinical assessment and surveillance.
TissueCypher Identifies Risk Beyond Routine Biopsy Findings
The announcement covers three studies, including two case reports and a multicenter cohort study, demonstrating how TissueCypher may provide additional information when conventional biopsy findings do not fully reflect a patient’s underlying risk. In one case, a patient initially diagnosed with non-dysplastic Barrett’s esophagus received a high-risk TissueCypher result indicating a 62% probability of progression to high-grade dysplasia or esophageal adenocarcinoma within five years. A repeat endoscopy six months later identified previously undetected high-grade dysplasia, leading to endoscopic eradication therapy. In another case involving conflicting biopsy classifications, the test indicated a 47% five-year progression probability and supported earlier interventional assessment. These cases illustrate how molecular risk assessment may help clinicians investigate patients whose clinical findings appear less concerning than their biological risk profile suggests. However, individual case reports cannot establish how often the test will identify otherwise missed disease across the broader patient population.
Multicenter Study Supports Personalized Patient Management
The multicenter cohort study compared TissueCypher molecular risk classification with the Progression in Barrett’s clinicopathologic risk assessment approach. The researchers reported only slight agreement between the two methods, with a Cohen’s kappa value of 0.04, suggesting that the approaches classified patients differently. TissueCypher categorized 83% of patients as low risk and 17% as intermediate or high risk, while the comparator approach classified 31% as low risk and 69% as intermediate or high risk. Following testing, management recommendations changed in alignment with TissueCypher risk classifications for 96% of patients in the study cohort. Physicians recommended five-year surveillance intervals for 38 patients and endoscopic eradication therapy for eight, including five patients with non-dysplastic Barrett’s esophagus whose molecular results indicated increased risk. These findings suggest that the test can influence clinical decision-making, although changes in management do not automatically demonstrate improved long-term patient outcomes.
Advertisement
Castle Biosciences Advances Precision Diagnostics
TissueCypher analyzes protein biomarkers and tissue architecture in routine endoscopic biopsy samples using an AI-derived spatialomics approach to estimate an individual’s five-year risk of progression. The company reports that the test has been evaluated in biopsies from more than 8,000 patients and supported by more than 17 peer-reviewed publications. The technology is designed to complement standard pathology without requiring an additional biopsy procedure. The new ACG findings reinforce the potential value of combining molecular information with histopathology to support personalized surveillance and treatment planning for Barrett’s esophagus. Further research will be important to determine whether this risk-guided approach consistently improves clinical outcomes, reduces unnecessary procedures and enables earlier intervention for patients at elevated risk of esophageal cancer.
Source: Castle Biosciences press release
