Predictive value of survivin alternative transcript expression in locally advanced breast cancer patients treated with neoadjuvant chemotherapy
- Romain Boidot
- Frédérique Vegran
- Sarab Lizard-Nacol
Published online on: February 1, 2009
HTML 0 views
| PDF 0 views
Survivin, a member of the apoptosis inhibitor protein family, is expressed in numerous human tumours, and its expression is described as a negative prognostic marker. Four alternative splice variants (survivin-ΔEx3, survivin-3B, survivin-2B and survivin-2α) have been described. To date, little is known about the prognostic or predictive role of all five survivin transcripts in breast cancer. In this study, we analysed, by means of real-time quantitative PCR, the five survivin transcripts in a population of 60 breast carcinoma patients treated with 5-fluorouracil + epirubicin + cyclophosphamide (FEC, n=32) or with docetaxel + epirubicin (Tax-Epi, n=28). For each patient, samples were obtained before and after one course of chemotherapy. Before treatment, the ratio of survivin-2α was significantly higher in resistant than in sensitive tumours treated by the FEC regimen (p=0.0161), while the ratio of survivin-ΔEx3 was higher in sensitive than in resistant samples treated with Tax-Epi (p=0.0234). After one course of chemotherapy, expression of survivin-3B was significantly associated with resistance (p=0.0448) in the FEC treatment group, and the ratios of survivin-ΔEx3 (p=0.0071) and survivin-2B (p=0.0380) were significantly higher in sensitive than in resistant tumours in the Tax-Epi treatment group. Notably, increased expression and ratio of survivin-3B after one course of Tax-Epi was associated with reduced disease-free survival (p=0.0299 and 0.0277, respectively) and with reduced overall survival (p=0.0145 and <0.0001, respectively) of the patients. These results indicate that an imbalance in the alternative transcript ratios may make the cells resistant or sensitive to apoptosis. They also demonstrate for the first time that alternative survivin transcript expression levels may be predictive markers in FEC and Tax-Epi treatment in breast carcinoma.