Persistent Pulmonary Nodules: CT Volume Growth And The Related Factors

Part 1 The CT Volume and Mass Growth of Pulmonary Nodules Detected in low-dose CT Lung Cancer Screening and its Influence FactorsPurpose:To evaluate the changes of volume and mass of pulmonary nodules that detected in low-dose computed tomography(LDCT)lung cancer screening,and to analyze the influencing factors.Materials and Methods:This retrospective study analyzed the CT images of the participants who underwent at least two chest LDCT scanning from March 2009 to December 2015 in National Cancer Center/Cancer Hospital,Chinese Academy of Medical Sciences and Peking Union Medical College.The inclusion criterion were the nodule mean diameter ≥6mm;the volume growth was defined as ≥ 20%.Fifty-one pulmonary nodules(PNs)were selected among 51 enrolled participants(26 men and 25 women).According characteristic of nodules,the nodules were classified into different groups[(10 pure ground-glass nodules(pGGNs),17 mixed ground-glass nodules(mGGNs)and 24 solid nodules(SNs)].According to NCCN lung cancer screening guidelines(2018.V2),the PNs were classified into three different risk stratification groups(14 in high-risk group,12 in moderate-risk group and 25 in low-risk group).The Lung-VCAR software was used to measure the three-dimentional diameters,volumes and volume doubling time(VDT)of the PNs,and then PNs were calculated mass doubling time(MDT).Results:There were 33 PNs increased≥ 1.5 mm in mean diameter of 51 PNs increased in volume ≥ 20%,and the rest 18 PNs were increased ＜ 1.5mm in diameter.The median VDT of mGGNs was 364 days,which was lower than that of pGGNs and SNs(761 and 819 days,respectively).The difference was statistically significant(P ＜ 0.05).The median MDT of mGGNs was 351 days,which was lower than that of pGGNs and SNs(772 days and 840 days,respectively).The difference was statistically significant(P＜0.05).The median VDT and MDT of the PNs in the high-risk group were 181 days and 256 days,respectively,which were lower than those in the low risk group(1037 days and 1035 days,respectively).VDT had good correlation with MDT(r = 0.909,P<0.05).Conclusion:Both the characteristic of PNs and the individual risk status of the participants could affect the growth of nodules in LDCT lung cancer screening.The mGGNs and high-risk group nodules grew faster relatively,which should be closely focused on.Compared with two-dimensional diameter,the three-dimensional volume and mass were more sensitively quantitative indicators for growth.The mass changes of mGGNs were earlier than that of volume.Part 2 Applying CT Texture Analysis to Determine the Prognostic Value of Growth of the Subsolid Nodules be Detected in LDCT Lung Cancer ScreeningPurpose:We analyzed the subsolid nodules(SSNs)that were detected in low-dose computed tomography(LDCT)lung cancer screening and investigated whether CT texture parameters can predict the growth trends of SSNs.Materials and Methods:Screening participants underwent LDCT from March 2009 to December 2016 in National Cancer Center/Cancer Hospital,Chinese Academy of Medical Sciences and Peking Union Medical College.In this ethics committee-approved retrospective study,89 SSNs were detected in 86 LDCT lung cancer screening participants(47 men,39 women;median age,55 years;range,41-75 years),including 42 pGGNs and 47 mGGNs.In these participants,28 SSNs were diagnosed as lung cancer by pathology,and 61 SSNs from the participants in which were undergone at least two LDCT to ensure that they had two volume measurements for comparison purposes.According to the pathologic result and volume growth ≥ 20%,all nodules were divided into three groups:cancer group,growth group and stable group.The volumes of the SSNs were assessed with Picture Archiving and Communication System(PACS)and Lung-VCAR software(General Electric Co.，USA),and texture parameters(mean value,uniformity,entropy and energy)were simultaneously assessed with CT-Kinetics texture analysis software(General Electric Co.,USA).Results:For both pGGNs and mGGNs the entropy of the cancer group was higher than those of the growth group and stable group,and the energy of the cancer group was lower than those of the growth group and stable group,which had significantly statistical differences(P ＜ 0.05).The entropy and energy of the growth group and stable group had no statistical differences(P ＞ 0.05).The mean valve in three group had no statistical differences(P ＞ 0.05).For pGGNs,the uniformity was different between the stable group and the growth group and had statistical differences(P ＜ 0.05).The uniformity had diagnostic accurancy for the growth of pGGNs.Receiver operating characteristic(ROC)area under the curve(AUC)were 0.781;when the cutoff value of uniformity was 1.12,the sensitivity was 81.30%,specificity was 72.70%.Conclusion:CT texture parameters had been showed the predictive value to the growth of the SSNs in LDCT lung cancer screening.The lung cancer presenting with SSNs had higher entropy and lower energy.We could use uniformity to predict the growth trends of the pGGNs,the growth pGGN might have lower uniformity.According to texture the follow-up and treatment plan of the SSNs could be more targeted.Part 3 A preliminary reacerch on EGFR mutation of pulmonary adenocarcinoma presenting with subsolid nodules:the differences between nodules with and without growthPurpose:To analyze the related factors of CT volume growth of pulmonary adenocarcinoma presenting with subsolid nodules(SSNs)and the correlation between the growth of SSNs and the mutation status of epidermal growth factor receptor(EGFR).Materials and methods:This retrospective study analyzed the 33 SSNs in 32 patients(14 men,19 women;average age 55.39 years,range 35-73 years),whom were confirmed pulmonary adenocarcinoma by surgical resection from March 2009 to March 2018 in National Cancer Center/Cancer Hospital,Chinese Academy of Medical Sciences and Peking Union Medical College.All SSNs were detected by EGFR mutation.All patients could obtain the whole CT imaging data from PACS,including baseline and preoperative final follow-up CT images.The Lung-VCAR software was used to measure the volumes of SSNs(General Electric Co.,USA).Then the volume growth rates of SSNs were calculated.According to the CT volume growth rate ≥20%,SSNs were divided into growth group and stable group.According to the response of epidermal growth factor receptor-tyrosine kinase inhibitor(EGFR-TKI),SSNs were divided into effective mutation group and non-effective mutation group.We analyzed the correlation between EGFR mutation status and growth of SSNs.Results:A total of 33 SSNs,22 were in growth group and the other 11 were in stable group.Sixteen SSNs had EGFR effective mutation,and 17 had non-effective mutation.In growth group,the numbers of SSNs with EGFR effective mutation were more than that in non-effective group(14 vs 8)and there was significant difference(P<0.05).In growth group,12 were high differentiatied adenocarcinoma,7 were moderate differentiatied adenocarcinoma and 3 were poor differentiatied adenocarcinoma.In stable group,3 were adenocarcinoma in situ(AIS),6 were high differentiatied adenocarcinoma and 2 were moderate differentiatied adenocarcinoma.The difference between the two groups was statistically significant(P<0.05).In growth group,1 was minimally invasive adenocarcinoma(MIA),7 were lepidic predominant adenocarcinoma(LPA),10 were acinar predominant adenocarcinoma(APA),4 were papillary predominant adenocarcinoma(PPA);in stable group,3 were AIS,3 were LPA,4 were APA,and 1 was PPA.There was no significant difference between the two groups(P>0.05).In the growth group,pure ground-glass nodules(pGGNs)were 17,mixed ground-glass nodules(mGGNs)were 5;in the stable group,pGGNs were 10,mGGN was 1.There was no significant difference between the two groups(P>0.05).Conclusion:The mutation status of EGFR and the growth of pulmonary adenocarcinoma presenting with SSNs had the correlation:The SSNs with EGFR effective mutation had great probability to grow.