The MIF Characterization Of Tertiary Lymphoid Structures In Lung Cancer And Its Clinical Prediction Model And Prognosis

PURPOSE:Tertiary lymphoid structures(TLS)can contribute to tumor control and good prognosis by recruiting circulating immune cells and enhancing local immunity.In this thesis,multiplex immunofluorescence(mIF)was used to observe TLS expression in non-small cell lung cancer(NSCLC),and to establish a TLS expression prediction model based on clinicopathological and hematological characteristics.This study was conducted to determine the prognostic value of TLS expression in NSCLC,with a view to providing a reference for accurate clinical treatment.METHODS:General data,pathological data,blood tests and follow-up data were collected from pathologically confirmed NSCLC patients(n=181)at Meishan Cancer Hospital between January 2017 and March 2022.Two pathologists independently assessed and recorded TLS expression on hematoxylin-eosin staining(HE)sections(TLS-positive group n=99,TLS-negative group n=82).Ten of these sections containing mature TLS were selected for mIF analysis to observe the histopathological features of TLS.Independent factors on TLS expression were analyzed by univariate and logistic regression multifactorial analysis for general data,pathological and blood parameters.A prediction TLS line graph model was constructed based on these factors.In addition,the C-index,calibration curves and decision curve analysis(DCA)were used to evaluate the discrimination,predictive power and net clinical benefit of the line graphs.Kaplan-Meier and Log-rank test were used to analyse the prognostic impact of TLS.SPSS27.0 and R4.1.3 were used for analysis.RESULTS:1.Baseline information:median age of patients was 66 years(43-82years),median follow-up time was 20.9 months(2.07-64.5 months),72 patients progressed and 109 patients died(23 patients died of advanced tumour,54 died of tumour progression,14 died of secondary infections,2 died of haemoptysis,1 died of secondary intracranial haemorrhage and 15 died of other causes)through the follow-up.2.Pathological features of TLS based on mIF staining:Early-TLS(CD3+,CD8+,CD20+,PNAd+,CD21-,BCL-6-),primary follicle-like TLS(CD3+,CD8+,CD20+,PNAd+,CD21+,BCL-6-),secondary follicle-like TLS(CD3+,CD8+,CD20+,PNAd+,CD21+,BCL-6+).3.Construction of a TLS prediction model based on clinicopathological blood parameters:five factors including N stage,M stage,total stage,TP and Cl~-were independent factors for TLS.The line graph was constructed based on the five factors with an area under the ROC curve(AUC)of 0.842[95%CI(0.783,0.900),P<0.005],with a sensitivity of 0.737 and specificity of 0.866.The Bootstrap method was internally validated with a C-index of 0.810,and the calibration curve showed that the actual and predicted probabilities were generally consistent,with the DCA curve showing a little overall net benefit.4.Prognostic role of TLS expression:The 1-year,3-year and 5-year OS of TLS-positive group and TLS-negative group were 89.58%vs.58.03%,69.84%vs.11.61,51.52%vs.2.68%.The 1-year,2-year and 3-year PFS of TLS-positive group and TLS-negative group were 80.65%vs.57.53%,68.14%vs.37.99%,66.37%vs.33.52%.The patients in the TLS-positive group had significantly longer OS and PFS compared to those in the TLS-negative group(P<0.001).CONCLUSION:The present study of mIF suggests that the composition and proportion of lymphocyte subpopulations differ between mature and immature TLS.The prediction model constructed based on the five independent factors has high value in predicting TLS expression in NSCLC.TLS expression is strongly associated with a good prognosis,which is important for patient prognosis and clinical decision making.