Cardiac MRI For The Screening,risk Stratification And Treatment Efficacy In Amyloidosis Patients With Autologous Stem Cell Transplantation

Objective:The objectives of this study were to use cardiac MRI to screen population for treatment with autologous stem cell transplantation(ASCT)in patients with systematic amyloidosis,to stratify the risk with combined clinical routine staging system,and to monitor the therapy effect after ASCT treatment.Materials and methods:91 light chain systematic amyloidosis patients confirmed by pathology and 34 healthy volunteers were prospectively included.(1)Firstly,patients were divided into two groups according to treatment methods:ASCT(n=66)and non-ASCT groups(chemotherapy,n=25),and main adverse cardiac events(MACE)were recorded during follow-up in two groups.The included analyzed parameters were from clinical routine and cardiac MRI.Clinical routine parameters were demographic characteristics,laboratory examination,immune parameters and electrocardiogram,while cardiac MRI had function,perfusion at rest,late gadiolinium enhancement,native T1 value,extracellular volume(ECV)and strain imging.The predictors of MACE in ASCT group and chemotherapy group were obtained by Logistic regression model and Cox regression model,respectively,and the cut-off value of risk factors of MACE in ASCT group was obtained by using the receiver operating characteristic(ROC)curve.The Kaplan-Meier curve was used to describe the results of the cut-off values obtained from ROC.(2)Secondly,the prediction power of MACE in patients with systematic amyloidosis with risk stratification of clinical biomakers combined with cardiac MRI were compared to clinical biomakers staging system alone.Clinical biomarkers staging system were as following:NT-proBNP?1800pg/mL,cTnT?0.025ng/mL,difference between the involved and uninvolved light chain(dFLC)>180mg/L scored 1,respectively,with 0-3 scores in total.Cardiac MRI staging system were as following:late gadolinium enhancement(LGE),native T1 values,myocardium perfusion at rest and strain were also included into staging system of cardiac MRI in patients.ROC curves were conducted to obtain the difference between two staging system,and Kaplan-Meier curve was also used to describe the results among the staging.(3)Thirdly,30 consecutive patients with histopathologically proven light chain amyloidosis and prospectively underwent baseline cardiac MRI and were then treated with ASCT,after median follow-up of 6 months,follow-up cardiac MRI and clinical biomarkers(including NT-proBNP,dFLC,creatinine,uric acid,alkaline phosphatase and etc..)were preformed.20 healthy volunteers were recruited for baseline and follow-up cardiac MRI and blood routine test with 6 months' follow-up.Primary efficacy assessment endpoint was cardiac response;second efficacy assessment endpoint was hematologic response.Logistic regression model and Cox regression model were used to obtain the predictors of cardiac MRI parameters and clinical biomarkers of cardiac response and hematological response.Results:7 patients(7.7%)had major adverse cardie events(MACE)after median follow-up of 12 months,3 in ASCT group,4 in chemotheraphy group.Comparing ASCT Group with non-ASCT group(chemotherapy),low voltage,uric acid level>530 mg/dl,cardiac function(ejection fraction<3 8%),resting perfusion(maximum myocardial intensity<261.4),LGE(>77.9%),strain(global circumferential strain>-7.7,global longitudinal strain>-6.2)suggested that patients with systematic amyloidosis may not be suitable for ASCT treatment.(2)The predicting power of clinical biomakers combined with cardiac MRI was higher than that of clinical biomarkers staging system alone with P value of 0.045(AUC:0.830 vs.0.884).(3)The median follow-up time after ASCT in patients with systematic amyloidosis and healthy volunteers were both 6.5 months.As a result,cardiac response was occurred in 15 cases,and complete hematologic response in 18 cases.The decrease of N-terminal brain natriuretic peptide(P=0.014)and baseline 2D short axis diastolic circumferential strain rate is the only index(P=0.027)that could predict cardiac response with corresponding AUC of 0.96 and 0.79 using 5-fold cross-validation test,while the decrease of the left ventricular end-diastolic volume/body surface area(P=0.031)and the baseline longitudinal strain rate as well as IgG had potential to predict compelte hemalogical response with P value of 0.024 and 0.048 with corresponding AUC of 0.78 and 0.76.Conclusion:Cardiac MRI could suggest that the screening criteria of the population for the indication of ASCT in terms of imaging in patients with systematic amyloidosis,and have the ability to improve the predictive value of MACE with clinical biomarker staging system,the potential value to predict cardiac response and hematological response after ASCT treatment in amyloidosis patients was also explored preliminarily.But this required larger samples and longer follow-up to further verify the results of the study.