Cyclin E confers a prognostic value in premenopausal breast cancer patients with tumours exhibiting an infiltrative growth pattern.
- Pontus Berglund (pontus.berglund{at}med.lu.se)
- Center for Molecular Pathology, Dep. of Lab. Med., Lund University, Malmö University Hospital, Sweden
- Maria Stighall (maria.stighall{at}gmail.com)
- Center for Molecular Pathology, Dep. of Lab. Med., Lund University, Malmö University Hospital, Sweden
- Karin Jirström (karin.jirstrom{at}med.lu.se)
- Center for Molecular Pathology, Dep. of Lab. Med., Lund University, Malmö University Hospital, Sweden
- Lisa Rydén (lisa.ryden{at}med.lu.se)
- Mårten Fernö (marten.ferno{at}med.lu.se)
- Bo Nordenskjöld (bo.nordenskjold{at}ibk.liu.se)
- Göran Landberg (goran.landberg{at}med.lu.se)
- Center for Molecular Pathology, Dep. of Lab. Med., Lund University, Malmö University Hospital, Sweden
- Published Online First 4 May 2007
Abstract
Aim: This study investigates the prognostic value of cyclin E in relation to tumour growth pattern by analysing stage II primary breast cancers from premenopausal women not subjected to any further adjuvant treatment. In addition, the value of cyclin E as a predictor of tamoxifen response was analysed, by comparing untreated and treated patients with oestrogen receptor positive tumours.
Methods: Breast cancer samples, assembled in tissue microarrays, were immunohistochemically stained for cyclin E and evaluated regarding the presence of nuclear staining. The overall growth characteristics of each tumour were assessed using whole tissue sections.
Results: Tumours displaying a pushing margin phenotype were strongly associated with high cyclin E levels, lymph node negative disease, a high histological grade, ER negativity, and exhibited a better prognosis compared to tumours with an infiltrative growth pattern. In the total cohort of non-treated patients (N=187), cyclin E was not associated with recurrence free survival (RFS). However, when analysing the subgroup of tumours lacking a pushing growth pattern (N=141), cyclin E was significantly associated with RFS, independent of histological grade and node status. There was no significant difference in tamoxifen response with regard to different cyclin E levels.
Conclusion: The prognostic value of cyclin E in premenopausal breast cancer is limited to patients with breast carcinomas exhibiting an exclusively infiltrative growth pattern. This limitation could be explained by the presence of a small but distinct subgroup of cyclin E-high breast cancers with a pushing margin phenotype and a more favourable outcome.
