JNET Classification of Colonic Lesions
The JNET (Japanese Narrow Band Imaging Expert Team) classification is a validated NBI magnifying endoscopic system that categorizes colorectal lesions into four types (Type 1, 2A, 2B, and 3) based on vessel and surface patterns to predict histology and guide treatment decisions. 1, 2
Classification Structure
The JNET classification evaluates two key endoscopic features:
Type 1: Benign Lesions
- Vessel pattern: Invisible vessels
- Surface pattern: Regular dark or white spots
- Histology: Hyperplastic polyp or sessile serrated polyp
- Diagnostic accuracy: 87.5% sensitivity, 99.9% specificity, 99.3% overall accuracy 3
Type 2A: Low-Grade Neoplasia
- Vessel pattern: Regular caliber vessels
- Surface pattern: Regular tubular, branched, or papillary structures similar to surrounding mucosa
- Histology: Low-grade intramucosal neoplasia (conventional adenoma)
- Diagnostic accuracy: 74.3% sensitivity, 92.7% specificity, 77.1% overall accuracy 3
Type 2B: High-Grade/Superficial Invasion
- Vessel pattern: Variable caliber, irregular vessels
- Surface pattern: Irregular or obscure surface
- Histology: High-grade intramucosal neoplasia or superficial submucosal invasive cancer (T1a)
- Diagnostic accuracy: 61.9% sensitivity, 82.8% specificity, 78.1% overall accuracy 3
Type 3: Deep Submucosal Invasion
- Vessel pattern: Loose vessel areas with interruption of thick vessels
- Surface pattern: Amorphous areas
- Histology: Deep submucosal invasive cancer (T1b)
- Diagnostic accuracy: 55.4% sensitivity, 99.8% specificity, 96.6% overall accuracy 3
Clinical Application
The JNET classification guides treatment strategy algorithmically: 1
- Type 1: Can be left in place if in rectosigmoid with high confidence diagnosis
- Type 2A: Standard endoscopic resection (piecemeal acceptable for large lesions)
- Type 2B: Consider en bloc resection when feasible for accurate pathologic assessment
- Type 3: Biopsy and refer to surgery for deep submucosal invasion
Important Caveats
Type 2B has the lowest diagnostic reliability with relatively poor specificity (82.8%) and positive predictive value (50.9%) compared to other categories 3. When encountering Type 2B lesions, supplementary pit pattern diagnosis using chromoendoscopy is recommended to improve diagnostic accuracy before deciding on resection strategy 3.
The classification was developed using magnifying NBI endoscopy, though dual-focus magnification systems have shown comparable diagnostic accuracy (>90% for all types) 4. The system is primarily used in Asian countries but is gaining acceptance in Western practice 1.
Diagnostic concordance varies significantly for neoplastic lesions even among JNET core members, particularly for T1b cancers (52% agreement), highlighting the need for proper training and experience 5. The classification performs best when endoscopists use high-confidence assessments and are familiar with the specific terminology 6.
The JNET classification demonstrates equivalent diagnostic efficacy to traditional pit pattern classification (AUC 0.97 for Type 1,0.84 for Type 2A, 0.72 for Type 2B, 0.90 for Type 3) but offers simpler clinical application without requiring chromoendoscopy 7.