A prospective study to examine the accuracies and efficacies of prediction systems for response to neoadjuvant chemotherapy for muscle invasive bladder cancer

The present study established systems to predict the chemo-sensitivity of muscle invasive bladder cancer (MIBC) for neoadjuvant chemotherapy (NAC) with methotrexate, vinblastine, doxorubicin plus cisplatin (M-VAC) and carboplatin plus gemcitabine (CaG) by analyzing microarray data. The primary aim of the study was to investigate whether the clinical response would increase by combining these prediction systems. Treatment of each MIBC case was allocated into M-VAC NAC, CaG NAC, surgery, or radiation therapy groups by their prediction score (PS), which was calculated using the designed chemo-sensitivity prediction system. The therapeutic effect of the present study was compared with the results of historical controls (n=76 patients) whose treatments were not allocated using the chemo-sensitivity prediction system. In addition, the overall survival between the predicted to be responder (positive PS) group and predicted to be non-responder (negative PS) group was investigated in the present study. Of the 33 patients with MIBC, 25 cases were positive PS and 8 were negative PS. Among the 25 positive PS cases, 7 were allocated to receive M-VAC NAC and 18 were allocated to receive CaG NAC according to the results of the prediction systems. Of the 8 negative PS cases, 3 received CaG NAC, 1 received surgery without NAC and 4 received radiation therapy. The total clinical response to NAC was 88.0% (22/25), which was significantly increased compared with the historical controls [56.6% (43/76) P=0.0041]. Overall survival of the positive PS group in the study was significantly increased compared with the negative PS group (P=0.027). In conclusion, the combination of the two prediction systems may increase the treatment efficacy for patients with MIBC by proposing the optimal NAC regimen. In addition, the positive PS group would have a better prognosis compared with the negative PS group. These results suggest that the two prediction systems may lead to the achievement of ‘precision medicine’.