Duodenal cancer is the leading cause of cancer-related mortality in patients with familial adenomatous polyposis (FAP), yet the current Spigelman staging system provides limited predictive accuracy for advanced neoplasia. The DRACO study (Duodenal Risk Assessment in adenomatous polyposis Coli -Oncogene) is a multicenter, STROBE- and CONSORT-compliant cohort study that analyzes upper endoscopies from genetically confirmed FAP patients across independent cohorts to develop, validate, and externally test two multivariable risk models.

Duodenal cancers represent the leading cause of cancer-related mortality in individuals with familial adenomatous polyposis (FAP). While endoscopic surveillance is a critical component of duodenal cancer management, clinical decisions are primarily guided by the Spigelman staging system. Introduced in 1989, this framework stratifies patients into four stages based on duodenal polyp size, number, histology, and the grade of dysplasia.

Despite its widespread use, the Spigelman system was not originally designed to predict cancer incidence or progression and lacks formal validation for these outcomes. A recent systematic review and meta-analysis confirmed its limited sensitivity for detecting duodenal cancers, at approximately 50%, with even lower sensitivity for papillary cancer. This limitation in early cancer detection also constrains the identification of patients at elevated risk of malignant transformation during a window in which endoscopic intervention might still be feasible. As high-grade dysplasia (HGD) is a well-established precursor to adenocarcinoma and a valid therapeutic target, the inability of the current system to anticipate HGD and its progression to cancer can undermine opportunities for timely, minimally invasive intervention.

The current clinical paradigm reserves prophylactic surgery or intensified endoscopic therapy primarily for patients with Spigelman stage IV disease. However, both cancer and high-grade dysplasia frequently arise in patients with lower-stage disease. Given the already limited sensitivity of the Spigelman system for cancer detection, its sensitivity for predicting the development of HGD is likely even lower. Consequently, relying on a stage IV threshold may delay preventive intervention beyond the window of opportunity for feasible endoscopic management.

These limitations have led to a growing recognition of the need to revise the Spigelman classification system. Two broad strategies have been proposed to improve its clinical utility: reweighting the existing components to better reflect their prognostic contribution, or integrating patient-level variables, such as age and desmoid tumor status, to more accurately capture individual risk trajectories. Importantly, a staging system intended to guide endoscopic surveillance should prioritize prevention by informing timely intervention before malignant transformation occurs. Therefore, risk stratification must be grounded in the likelihood of developing pre-cancerous lesions, such as high-grade dysplasia, not solely on the risk of overt carcinoma. The DRACO study was designed to fill this gap in knowledge.

Inclusion Criteria:

• Confirmed germline diagnosis of FAP, as defined by genetic testing
• Two or more upper gastrointestinal endoscopies
• Complete documentation of all Spigelman classification variables at each endoscopic evaluation

Exclusion Criteria:

• Histological grading of duodenal polyps incomplete
• Follow-up data were unavailable
• Duodenal surgery before study baseline endoscopy

Individuals with familial adenomatous polyposis, defined by a pathogenic/likely pathogenic germline Adenonomatous Polyposis Coli (APC) gene variant, undergoing surveillance for duodenal polyposis using esophagogastroduondeoscopy (EGD)

Data collected for the study will be made available to others, including de-identified participant data, at publication, via a signed data access agreement and at the discretion of the investigators' upon approval of the proposed use of such data