Diagnostic performance of endoscopic ultrasound-guided fine-needle aspiration (EUS-FNA) for the evaluation of hypermetabolic lymphadenopathy mediastinum lower, posterior, and middle, detected by PET-CT with 18F-FDG (PET) (APOGEE Study).

798 Background: In the context of a new cancer or relapse, the high sensitivity (Se) (95-100%) of PET-CT with 18F-FDG can lead to the demonstration of hypermetabolic mediastinal adenopathies. Its lower specificity (Sp) (89%) can require histological examination. We report the results of a prospective, single-center study evaluating the diagnostic performance of EUS-FNA in this indication. Methods: Prospective single-center study featuring patients in whom PET had revealed hypermetabolic mediastinal lymphadenopathy requiring diagnostic certainty. All EUS-FNA were performed with a 19-gauge needle (EchoTip, Cook Endoscopy). Main objective: To evaluate the diagnostic performance in terms of Se and Sp of EUS-FNA in the characterization of hypermetabolic mediastinal adenopathies in PET in the context of a new cancer or relapse. Secondary objectives: To evaluate the negative predictive value (NPV) of the EUS-FNA and to evaluate the percentage of surgical diagnostic procedures avoided. The standard technique was a thoraco-abdominopelvic CT scan at 6 months and at 12 months. Results: 52 patients were eligible and evaluable for the primary endpoint. The most common primary cancers were mammary (17.3%) and bronchial (13.5%). The lymph nodes were analyzed as malignant in 44.2% of cases, benign in 50% of cases and atypical or suspicious in 3.8% of cases. The malignant lymph nodes were metastatic for breast cancer in 21.7% of cases, bronchial cancer in 17.4% of cases, colorectal cancer in 17.4% of cases and prostate cancer in 13% of cases. The Se of the EUS-FNA was 92% (95% CI 0.74-0.99) and the Sp 100%. NPV was 87% (95% CI: 0.59-0.98). A diagnostic surgical procedure was necessary in 2% of the cases. PET and EUS-FNA often allowed the modification of the therapeutic strategy. Conclusions: When a confirmed diagnosis is required, the diagnostic accuracy of the minimally invasive procedure of EUS-FNA, is sufficiently robust to avoid a surgical diagnosis technique. The combination of PET and EUS-FNA may alter the therapeutic strategy that would have been considered after PET alone. Clinical trial information: NCT01892501.