Accuracy of Tokuhashi score system in predicting survival of lung cancer patients with vertebral metastasis.

Controversy exists in the accuracy of Tokuhashi score (TS) system in predicting survival of patients with vertebral metastasis, including vertebral metastases from primary lung cancer. To calculate the accuracy of the TS in predicting survival of lung cancer patients with vertebral metastasis and investigate which subgroup's survival is suitable for TS to predict, we conduct this retrospective study. From January 2008 to December 2013, 151 patients with vertebral metastases from primary lung cancer in our institute were included. The data were censored at December 2014. Patients were classified according to TS. The actual and predicted survivals were compared to calculate the accuracy rate. Spearman's correlation analysis was conducted to explore the relation between overall survival and TS. Cox regression was carried out to investigate potential factors affecting prognosis. Only 8.6 % (13 patients) actually followed the survivorship pattern as predicted by the TS. Spearman's correlation analysis demonstrated that actual overall survival correlated significantly with the TS. Cox model indicated that status of visceral metastases (i.e., none, involved with lung, involved with extrapulmonary organ) and response to first-time systemic therapy (effective or not) are associated significantly with overall survival. For patients with visceral metastases and progression to first-time systemic therapy simultaneously, predicted survival by TS and actual survival matched well. TS could only predict survival accurately in those with visceral metastases and disease progression to first-time systemic therapy simultaneously. Thus, it may not be a reliable tool to predict survival in lung cancer patients with vertebral metastasis. Status of visceral metastases and response to first-time systemic therapy should be added into TS system to increase validity and accuracy.