Association between class III β-tubulin expression and response to paclitaxel/vinorebine-based chemotherapy for non-small cell lung cancer: a meta-analysis.

Background: It has been proposed that the level of class III β-tubulin gene expression can be used to predict clinical sensitivity to paclitaxel/vinorebine-based chemotherapy in non-small cell lung cancer (NSCLC) patients. However, whereas there are published reports supporting this association, there are also reports of studies that failed to find such an association. We conducted a meta-analysis of all relevant published data to provide a combined statistical assessment of the proposed association of expression variations of class III β-tubulin with objective response and median survival in patients with NSCLC treated with paclitaxel/vinorebine-based chemotherapy.

Methods: We conducted the meta-analysis using data from ten studies, each of which evaluated the correlation between class III β-tubulin expression levels and objective response in patients treated with paclitaxel/vinorebine-based chemotherapy for NSCLC patients. All eligible studies were searched by MEDLINE, EMBASE and CNKI databases. Overall odds ratios (ORs) of the objective response were calculated using the method of Mantel-Haenszel. The differences in objective responses between Caucasian and Asian patients treated with paclitaxel/vinorebine-based chemotherapy were compared. We also compared outcomes for patients treated with paclitaxel to those treated with vinorebine.

Results: There were a total of 552 patients in the ten studies that met our criteria for evaluation. High/positive expression of class III β-tubulin was found in 279 patients (50.5%), and low/negative expression for this gene was found in 273 (49.5%) patients. The objective response rate for paclitaxel/vinorebine-based chemotherapy was significantly higher in patients with low/negative class III β-tubulin expression (OR=0.28; 95% CI, 0.20-0.41; P<0.00001). Median survival time was longer for patients with low/negative expression of class III β-tubulin compared with patients with high/positive expression (MR=1.40; CI, 0.89-0.90; P<0.00001). There was no significant difference in therapy between Caucasian and Asian patients treated with paclitaxel/vinorebine-based chemotherapy (Chi(2)=0.02, P=0.88). In our analysis, NSCLC patients treated with paclitaxel had more favorable clinical outcomes than those treated with vinorelbine (Chi(2)=3.69, P=0.05).

Conclusions: By combining data from ten different studies, we found a correlation between low TUBB3 gene expression and favorable clinical outcome to anti-tubulin therapy. The correlation for the combined data was significantly stronger than it was for any of the individual studies. This result supports the usefulness of class III β-tubulin mRNA level as a biomarker for sensitivity to paclitaxel/vinorebine-based chemotherapy in NSCLC patients.