Establishment and evaluation of a prognostic prediction model for patients after complete resection of non-small cell lung cancer and multidisciplinary therapy / 中华肿瘤杂志

ABSTRACT

<p><b>OBJECTIVE</b>A bias may be produced when only TNM stage is used to predict the prognosis of non-small cell lung cancer (NSCLC) after complete resection and multidisciplinary treatment. The reason is that histological type, differentiation, and postoperative treatment which may also affect the survival are excluded in the prognosis prediction. The aim of this study is to establish and evaluate a prognostic prediction model for NSCLC patients based on pathological parameters after completely resection and postoperative treatment.</p><p><b>METHODS</b>According to the theory of Nottingham index model, a prognostic prediction model was established based on the pathological parameters and postoperative management of 899 NSCLC patients after complete resection and multidisplinary treatment in our hospital from Jan.1, 1997 to April, 2001, and its efficiency and feasibility were evaluated.</p><p><b>RESULTS</b>Univariate analysis and multivariate analysis showed that histological type (H), T stage (T), N stage (N), M stage (M), and postoperative mediastinal radiotherapy for positive lymph node (R) were independent factors affecting the survival of NSCLC after complete resection and multidisciplinary treatment. The prognostic prediction model based on these parameters is S = 0.338H + 0.178T + 0.549N + 0.647M-0.361R. The high and low risks of prognostic index (PI) were 1.6695 and 1.1160, respectively. The 5-year survival rates of the patients in the low, middle and high risk groups stratified by this model were 70.1%, 54.5%, and 22.5%, respectively, with a significant difference among the groups (chi(2) = 132.091, P = 0.000).</p><p><b>CONCLUSION</b>A model based on the pathological parameters and postoperative management has been established, which may be helpful in predicting the prognosis for NSCLC after complete resection and multidisciplinary management.</p>