Prognostic Value Of Risk Index Model Based On Clinic Pathological Features For Non-small Cell Lung Cancer Patients

BACKGROUND:Recently,lung cancer is the malignant tumor with the highest incidence and mortality worldwide,and is the leading cause of cancer-related deaths.Non-small cell lung cancer(NSCLC)is the most common type of lung cancer,accounting for about 80%of the whole lung cancers.As the continuous development of medicine and constant improvement in cancer diagnosis and treatment technology,the survival of NSCLC patients has been improved to a certain extent,but its five-year survival rate has still been low along with the poor prognosis.Therefore,a large number of preclinical and clinical studies focused on the prognosis estimation of lung cancer patients,and scholars at home and abroad are constantly looking for prognostic indicators and models of great predictive value.Account for clinical heterogeneity of non-small cell lung cancer in different individuals,the patients' overall survival analysis is of limited value,while to evaluate the prognosis of each individual case by using a variety of prognostic indexes for risk stratification should be much more valuable.OBJECTIVES:On the condition of reasonable experimental design and scientific analysis and based on patients' clinical data,pathological data and laboratory data,we build a risk index(RI)model with a series of high recognizable prognostic factors for postoperative NSCLC patients.This model can assess patients' condition and prognosis more reasonably for screening high-risk patients with poor prognosis,and formulating individualized treatments and follow-up plans,which can increase patient compliance,improve the treatment effect,and reduce adverse reactions and mortality.METHODS:This retrospective clinical study included a total of 1311 NSCLC patients who underwent radical surgical treatment in the department of thoracic surgery,affiliated provincial hospital of Shandong university from January 2006 to December 2011.All patients enrolled in the study met all criteria for inclusion and did not meet any exclusion criteria.By electronic system,regular telephone,and outpatient follow-up,this study collected hospitalized medical records data and postoperative follow-up data to process statistical analysis,including gender,diagnostic age,smoking situation,preoperative Hematologic examinations,tumor size,tumor position,postoperative pathological diagnosis,lymph node metastasis,number of positive lymph nodes,number of lymph nodes dissection,Tumor-Node-Metastasis(TNM)staging,surgical method and so on.We chose valuable prognostic indicators by COX multiple factors analysis to build a RI prognosis model with weighting coefficient.According to the ROC curve method,we defined RI =1.86 as the best cutoff value,and divided the patients into high RI group(RI>-1.86)or low RI group(RI<-1.86).Moreover,we analyzed the differences of general clinicopathologic features between high and low RI groups.Statistical analysis of all data in this study was performed by SPSS 22.0,and P<0.05 was considered to be statistically significant.RESULTS:The median OS of NSCLC patients enrolled in the study was 39 months,and respectively with cumulative survival rates of 89.30%,53.92%and 39.21%for one year,three years and five years.The following indexes might be prognostic factors through univariate analysis,such as gender,age,smoking index,red blood cell count(RBC),hemoglobin(Hb),white blood cell count(WBC),neutrophil count(NEUT),lymphocyte count(LYMPH),monocyte count(MONO),platelet count(PLT),prothrombin time(PT),international standardization ratio(INR),fibrinogen(FIB),albumin(A),globulin(G),white ball ratio(A/G),diameter of the tumor,number of positive lymph nodes and number of lymph nodes dissection.After further multivariate analysis,we found that age,G,A/G,PT,INR,diameter of the tumor,the number of positive lymph nodes and the number of lymph nodes should be the independent prognostic factors.We built the RI model based on those independent prognostic factors,and NSCLC patients were divided into two groups by the best cutoff value.Survival analysis showed that OS between high and low RI groups were significantly different(?2 = 157.255,P = 4.50 x 10-36),and high RI patients suggested poor outcome(HR =3.165,P<0.05).CONCLUSIONS:RI prognostic model should be an independent prognostic factor for NSCLC patients undergoing radical resection,and high RI patients are at high risk of poor prognosis.The model is more suitable for NSCLC patients with significant heterogeneity because a series of high recognizable prognostic factors were included for overall assessment of individual.Moreover,with the characteristic of simple,fast and efficient valuation,the model can play an important role in the prognosis evaluation of NSCLC patients.