© 2007 European Society for Medical Oncology
gastro-intestinal cancer |
Aurora-A overexpression as an early marker of reflux-related columnar mucosa and Barrett’s oesophagus
1 Section of Medical Oncology, Department of Surgery and Oncology
2 Department of Histopathology
3 Section of Surgical Oncology, Department of Surgery and Oncology, Università di Palermo, Palermo
4 Division of Medical Oncology, National Institute of Oncology, Bari, Italy
* Correspondence to: Antonio Russo, MD Section of Medical Oncology, Department of Surgery and Oncology, Università di Palermo, Via del Vespro 127, 90127 Palermo, Italy. Tel: +39-091-6552500; Fax: +39-091-6554529; E-mail: lab-oncobiologia{at}usa.net
Background: The development of oesophageal adenocarcinoma is generally closely associated with the presence of a specialised intestinal-type epithelium such as that found in Barrett's oesophagus (BO). A particular histological condition is when the distal oesophagus showing cardiac and/or fundic mucosa without intestinal metaplasia cannot be defined as ‘Barrett's mucosa’ [condition that we call ‘columnar-lined oesophagus’ (CLO)] and up till now, there has been no agreement in literature about the management of this condition. Aurora-A overexpression leads to centrosome amplification, chromosomal instability and aneuploidy in mammalian cells.
Patients and methods: A prospective study was carried out on 28 consecutive patients who presented columnar mucosa above the gastro-oesophageal junction (GOJ) at endoscopy. As controls, two more biopsies were obtained, one on the normal-appearing squamous oesophagus above the GOJ, as far as possible from the columnar mucosa (controls A), and one taken 1 cm below the GOJ (controls B). The Aurora-A and p53 expression levels were analysed respectively by Quantitative Real Time PCR and immunohistochemistry.
Results: Twelve patients were affected by BO (43%) while the other 16 patients (57%) had a CLO. Nine of 28 (32%) cases were focally positive for p53 immunostaining. All the BO/CLO samples were positive for the Aurora-A transcript with regard to controls. Furthermore, 13 of 28 (46%) cases showed overexpression (above the median for the whole group).
Conclusion: Due to the low number of cases, we are not at present able to state that statistically significant quantitative differences in Aurora-A messenger RNA expression exist between CLO and BO cases with and without dysplasia and p53-positive immunostaining. Further studies on a larger number of cases with a follow-up period are necessary in order to establish the risk of progression and the correct management of these subjects.
Key words: Aurora-A overexpression, Barrett's oesophagus, cell cycle, columnar-lined oesophagus, p53 protein
Both authors have contributed equally to this work.