Endoscopic ultrasound-guided fine needle tissue acquisition biopsy samples do not allow a reliable proliferation assessment of gastrointestinal stromal tumours

Background: Current prognostication of gastrointestinal stromal tumours is validated on/applies to resected tumours, mainly because surgery is recommended whenever possible. However, pre-treatment prognostication is increasingly warranted, considering the follow-up strategy recently admitted for expectedly low-risk tumours and the possible distinctive molecular features/spontaneous regression of some small cases. Aims: To investigate whether endoscopic ultrasound-guided fine-needle tissue acquisition-biopsies reflect prognosticators of resected gastrointestinal stromal tumours, for possibly developing reliable pre-treatment prognostic criteria. Methods: The applicability/reliability of mitotic index/5mm2 and MIB1 proliferative index/1000 cells were tested in 35 endoscopic ultrasound-guided fine-needle tissue acquisition-biopsies diagnosed as gastrointestinal stromal tumour, subsequently resected without intervening therapy, consecutively collected in thirty months. Size and mitotic/proliferative indexes were compared with resection specimens. The feasibility of bioptic genotyping was also tested. Results: 35 patients were studied (45.7% males; mean age 61.6 years, range 26-83 years). Mitotic/proliferative indexes were determinable in 68.6%/88.6% of biopsies, respectively; they were nevertheless underestimated, as happened with endoscopic ultrasound-determined tumour size. Bioptic genotyping revealed reliable. Conclusions: Endoscopic ultrasound-guided fine-needle tissue acquisition does not reliably reflect gastrointestinal stromal tumours' proliferation and size. Alternative parameters should be validated for a pre-surgical prognostic classification. Considering the emerging potentially prognostic genetic markers in gastrointestinal stromal tumours, the reliability of bioptic genotyping is a promising result.