The Value Of ¹⁸F-FDG PET/CT Imaging In Molecular Subtype Of Breast Cancer

Objective:The purpose of our study was to analyze the relationship between the metabolic parameters of ¹⁸F-FDG PET/CT and the molecular subtypes of breast cancer,value of identification and diagnosis of different types of molecular types of breast cancer,by comparing different molecular subtypes of breast cancer hormone receptors and indicators,clinical progression free survival time and ¹⁸F-FDG PET/CT metabolic parameters?SUVmax,MTV,TLG?and other indicators.Analyzing the prognostic value of breast cancer,hope to provide a new reference for the clinical diagnosis,molecular typing and prognosis evaluation of breast cancer.Methods:We collected 324 patients who were diagnosed with breast cancer by¹⁸F-FDG PET/CT scan from February 2012 to July 2016 at the Cancer Institute and Hospital,Tianjin Medical University.Breast cancer patients confirmed by biopsy or surgical excision pathology,through histopathology,ER,PR,HER2 scores and Ki-67percent ratio were obtained.Before the ¹⁸F-FDG PET/CT scan,none of the patients had breast biopsy and any tumor related treatment.To retrieve the original image of all patients,apply the software PET VCAR on the AW4.6 workstation,use the percentage threshold method to define ROI,the SUVmax,MTV and TLG of breast tumor were measured respectively.?1?Because SUVmax does not confirm to normal distribution,Spearman rank correlation analysis was used for correlation analysis,the correlation between SUVmax,MTV,TLG and ER,PR,HER2 and Ki-67 was investigated respectively.?2?The group variables were ER,PR,HER2 and Ki-67respectively,compare SUVmax in different groups,the comparison method is Wilcoxon rank and test.?3?The molecular subtypes of breast cancer is grouped variables,comparison of the differences of SUVmax,MTV and TLG in different groups by Kruskal-Wallis rank sum test.?4?SUVmax,MTV and TLG value are used to distinguish different molecular types of breast cancer,each index are calculated the sensitivity,specificity,accuracy,the area under the ROC curve,select the maximum point of the Jordan index as the diagnostic value,the combined diagnostic model of multiple indexes was established by Logistic regression analysis.?5?Collect the diagnosis of breast cancer after 18F-FDG PET/CT examination in our hospital from February 2012 to July 2016 was confirmed by pathology as breast cancer,Follow up until September 5,2017,and a total of 239 patients with complete follow-up data,among them,there were 73 cases of disease progression,recurrence and death positive outcome,166 patients who had no disease progression,recurrence,and death?deleted?,67 of the 40%patients were randomly selected from the deletion,plus 73cases of positive outcome,140 patients were eventually included in the survival analysis.The ER,PR,HER2 scores and Ki-67 percentage of all patients were obtained by histopathology.The software PET VCAR is used on the AW4.6workstation,and the SUVmax,MTV and TLG are measured respectively.The value of each factor Kaplan-Meier test was used to evaluate the prognosis,single factor analysis using Log-rank test.The factors influencing the progression-free survival were analyzed by multivariate COX regression analysis.Results:?1?The metabolic parameters of ¹⁸F-FDG PET/CT SUVmax,TLG were negatively correlated with ER and PR?SUVmax r=-0.204,-0.232,p<0.05;TLG r=-0.156,-0.168,p<0.05?,negative correlation between MTV and PR?r=-0.110,p<0.05?,there is no correlation between MTV,ER,HER2 and Ki-67?r=-0.092,0.073,0.098,p>0.05?.SUVmax,TLG,HER2 and Ki-67 are positively correlated?SUVmax r=0.140?0.420,p<0.05;TLG r=0.125?0.292,p<0.05?.?2?ER and PR were based on positive negative,Ki-67?14%,Ki-67<14%for grouping,the difference of SUVmax and TLG in different groups was statistically significant?p<0.05?.According to HER2 positive and negative groups,there was no statistically significant difference between SUVmax and TLG in different groups?p>0.05?.According to ER,PR,HER2 based on positive negative,Ki-67?14%,Ki-67<14%for grouping there were no statistically significant difference between MTV in different groups?p>0.05?.?3?In the molecular classification of four groups of breast cancer,the difference of metabolic parameters SUVmax and TLG was statistically significant?p<0.01?,there was no significant difference between the four groups of MTV?p>0.05?.Among them,the LuminalA SUVmax is the smallest,and the triple negative is the largest.?4?SUVmax=5.94,which can be used to distinguish LuminalA breast cancer.It was 0.654,0.732,0.719,0.386 and 0.724,respectively,with the sensitivity,specificity,accuracy,the approximate index,and the area AUC under the ROC curve.TLG has a low diagnostic capacity in LuminalA breast cancer.SUVmax,MTV,and TLG have no diagnostic ability to differentiate LuminalB breast cancer.SUVmax has a low diagnostic capacity in HER2 positive non-luminal breast cancer.TLG has no diagnostic ability to differentiate HER2 positive non-luminal breast cancer.SUVmax,TLG have a low diagnostic capacity in triple-negative breast cancer.MTV has no diagnostic ability to differentiate between LuminalA,HER2 positive non-luminal and riple-negative breast cancer.?5?Analysis of no progress survival period:univariate analysis showed that age?p=0.442?,CT?p=0.188?,HER2?p=0.155?,and Ki-67?p=0.634?had no significant relationship with the prognosis.Multifactor analysis showed that SUVmax?p=0.014?,MTV?p<0.001?,and ER?p=0.019?were independent prognostic factors in the progression free survival of breast cancer.Conclusion:?1?The metabolic parameters of ¹⁸F-FDG PET/CT were negatively correlated with SUVmax,MTV,TLG and PR.SUVmax,TLG and ER are negatively correlated.There was a positive correlation between SUVmax,TLG,HER2 and Ki-67.?2?Grouped according to different ER,PR,Ki-67,the difference of SUVmax,TLG in different groups was statistically significant.According to the HER2 group,there were no statistical differences in the difference of SUVmax and TLG in the group.Grouped according to different ER,PR,Her2,Ki-67,there was no statistical difference in the difference of MTV in the group.?3?In the molecular classification of different breast cancer,the difference between SUVmax and TLG was statistically significant,but there was no significant difference between the four groups of MTV.Among them,the LuminalA SUVmax is the smallest,and the triple-negative is the largest.?4?SUVmax can be used to distinguish between LuminalA and non-LuminalA breast cancer.?5?The single factor analysis showed that the tumor diameter,SUVmax,MTV,TLG,ER,PR were related to the prognosis of breast cancer.Multifactor analysis showed that SUVmax,MTV and ER could be an independent prognostic factor in the progression free survival of breast cancer.