| Objective: The purpose of this study was to evaluate the usefulness ofPerfusion CT to predict the response to neoadjuvant chemotherapy (NACT) inpatients with non-small cell lung cancer (NSCLC).Materials and Methods: From October 2010 to December 2011, 41consecutive patients with pathologic confirmed NSCLC in affiliated hospitalof north Sichuan medical college were enrolled into our study, and underwentperfusion computed tomography (CT) and enhancement CT scan beforeNACT. All patient received NACT at least 1 week after CT scanning. Theresponse to treatment was evaluated according to Response EvaluationCriteria in Solid Tumors (RECIST 1.1) after two course of treatment. Patientswho showed a complete or partial response were categorized as clinicalresponders. The remaining patients with either stable or progressive diseasewere categorized as clinical non-responders. The differences in perfusionparameters between groups were analyzed, and to explore whether perfusionCT parameters can be used to predict response of NSCLC to NACT, areceiver operating characteristic curve was performed to determine the best cut-off value of perfusion parameters for differentiation of responders fromnon-responders.Result: The follow-up examinations were performed in forty-onepatients within 58 days (range, 56–65 days). On the basis of RECIST criteria,9 (21.95%) patients were classified as having partial response, 2 (4.88%) ashaving progressive disease, and the remaining 30 (73.17%) as having stabledisease. No patients showed a complete response to therapy. There weresignificant differences in pre-NACT blood flow (104.71±26.72 ml/min/100gvs. 75.30±23.95 ml/min/100g, P=0.003), pre-NACT mean transittime(5.15±1.10s vs. 6.77±3.43s, P=0.027) between clinical responders andnon-responders. Patients who had tumors with a high initial blood flow, ashort initial mean transit time showed a good response to NACT. Respondersalso had higher BV and PS, but no significant changes were found (P = 0.204and P = 0.291, respectively). There was a highly significant difference inpre-NACT blood flow between clinical responders and non-responders. Byusing 86.05ml/ 100 g/min of pre-NACT blood flow as the cut-off value, theROC curve showed a sensitivity of 77.8%, a specificity of 78.1%, a positivepredictive value of 77. 8%, a negative predictive value of 75.0% and theaccuracy of 77.6%.Conclusion: BF and MTT may be used to predict tumor response toNACT in patients with NSCLC. Tumors with a high BF and low MTT tendedto respond better to the NACT. Tumor may be sensitive to NACT when BF≥86.05 ml/min/100g. Perfusion CT may be used to predict tumor response toNACT in patients with NSCLC. This might help to make reasonabletreatment strategies. Part 2Application of the binary Logistic regression model to analyze thefactors in predict the neoadjuvant chemotherapy response for patientswith non-small cell lung cancer by computed tomographyObjective: To investigate the influential factors associated withneoadjuvant chemotherapy (NACT) response in patients with advanced nonsmall cell lung cancer (NSCLC), and to determine the correlation betweenclinical indicator or CT finding and NACT response in patients with NSCLC.The efficient information associated for predicting the response in patientswith NSCLC treated with NACT were chosen.Material and Methods: From October 2010 to December 2011, 63consecutive patients with advanced NSCLC, which treated with NACT inaffiliated hospital of north Sichuan medical college, were enrolled into ourstudy. Univariate analysis were performed by using chi square test and binarylogistic regression, the correlation between patient age, gender, histologicaltype, lesion location, pre-treatment diameter of tumor, lymph node (LN)staging, stations of positive LN and the status of distant metastases observedon CT and the response for NACT were evaluated. The influential factorsassociated with NACT response were subsequently determined.Result: By chi square test analysis, pre-treatment diameter of tumor, LNstaging (cN0/cN1 vs. cN2/cN3), LN stations, and distant metastases werefound obvious correlation with NACT response (P= 0.005, P= 0.036, P=0.041 and P=0.042, respectively). However, no significant association was found between the NACT response and age, sex, histological type, and theoriginal sites in CT. By binary logistic analysis, the major influential factorsof response for NACT in patients with NSCLC were positive LN stage(OR=5.251, P=0.012), pre-treatment diameter of tumor (OR=4.783, P=0.018)and distant metastases (OR=2.127, P=0.031).Conclusion: The binary Logistic regression can select out the valuableCT indexes, which related to the response of the NACT for the patients withNSCLC, and can describe and analyze the process. LN stage, diameter oftumor and status of distant metastases may be considered as predicting factorfor NACT response. |
PDF Full Text Request