Diagnostic Accuracy in Bronchial Carcinoid tumors is Dependent of Biopsy Size

Objective: Recently, 60% discordancy was reported for distinction between typical carcinoid and atypical carcinoid in preoperative biopsy compared to the resection specimen. This study investigated the impact of biopsy surface size, obtained with flexible and rigid bronchoscopy, on diagnostic accuracy of typical and atypical carcinoid. Methods: Biopsy-resection paired specimens of patients referred for treatment to Amsterdam University Medical Centers were retrieved. Bronchial biopsies were obtained either by flexible or rigid biopsy. The definitive diagnosis was based on the resection specimen. Diagnosis according to the 2015 WHO classification, mitoses and necrosis in biopsy and resection specimen, were independently re-evaluated by two pathologists. Results: After screening 298 patients, 64 biopsy-resection pairs with available tissue were included of which 34 (53%) were biopsied with flexible and 30 (47%) with rigid biopsy. In 35 (55%) patients, the tumor classification between the biopsy and resection specimen was concordant. The discordance in the remaining 29 cases (45%) was caused by misclassification of atypical as typical carcinoid in bronchoscopy specimens, predominantly in small flexible biopsies (59%, p=0.021). Of biopsies measuring <2 mm2, 79% were classified as discordant and 52% of the discordant biopsies measured <4 mm2. Conclusion: Histological classification in central carcinoid tumors is discordant in 45% of the biopsies, with increasing diagnostic accuracy in larger biopsies. Distinguishing carcinoid tumor into typical or atypical carcinoid on biopsies <4 mm2 should be discouraged. A cumulative biopsy surface of at least 4 mm2 tumor is preferred to increase the diagnostic accuracy which helps in optimal treatment planning.