Control cells were incubated with DMSO, which was used as a solvent for imatinib, where appropriate, or a single drug. After these treatments, acridine orange (10��gml?1) different was added to each well to distinguish apoptotic cells from vital cells. Apoptosis was defined as the appearance of apoptotic bodies and/or chromatin condensation, using a fluorescence microscope. Results were expressed as the percentage of apoptotic cells in a culture by counting at least 300 cells per well. All apoptosis assays were performed three times in duplicate. Western blot analysis After treatment with MegaFasL, as described above, GIST882 and GIST48 cells were lysed and the protein concentration was determined according to the Bradford method. Western blotting was performed as described previously (De Groot et al, 2005).
Primary antibodies were goat anti-Fas (polyclonal, 1:400; Santa Cruz, Heerhugowaard, The Netherlands), mouse anti-FasL and mouse anti-caspase 3 (clone 33 and clone 19, respectively, 1:1000; BD Transduction Laboratories, Alphen aan den Rijn, The Netherlands), rabbit anti-cleaved caspase 3 and rabbit anti-caspase 6 (1:1000; Cell Signalling Technology, Biok��, Leiden, The Netherlands), mouse anti-caspase 8 (clone 1C12, 1:500; Cell Signalling), and mouse anti-actin (clone C4, 1:20000; ICN Biomedicals, Zoetermeer, The Netherlands). Tissue collection The study group consisted of 45 primary GISTs derived from 44 patients. Paraffin-embedded tumour tissues of primary GISTs were retrieved from the files of the Department of Pathology of the University Medical Center Groningen.
Tumours were surgically resected between 1985 and 2005. All tumours were reviewed by a pathologist with special expertise in soft tissue tumours (AJHS). The diagnosis of GIST was based on the typical histological features of a cellular spindle cell or epithelioid mesenchymal tumour of the gastrointestinal tract. In addition, all tumours were positive for CD117 (KIT) by immunohistochemistry. Tumours were categorised into risk groups, based on tumour size and mitotic index, according to the consensus classification for prediction of metastatic potential (Fletcher et al, 2002). Tumour size was obtained from the pathology reports. Mitotic index was counted in 50 consecutive high power fields ( �� 40 objective; field diameter 0.55mm).
Clinicopathological data collected included sex and age of the patient at diagnosis, primary tumour site, and histological subtype. For the detection of KIT mutations, genomic DNA was extracted from paraffin-embedded tumour samples and KIT exon 9 or 11 was amplified by PCR. Both forward and reverse PCR products were sequenced and the results were compared with normal sequences. One patient AV-951 had two primary GISTs, which were a high risk epithelioid gastric tumour lacking a KIT exon 9 and 11 mutation and a low risk spindle-cell small intestine tumour with a KIT exon 11 mutation, respectively.