| Objective: Systemic immunoglobulin light-chain(AL)amyloidosis is the most common type of systemic amyloidosis with the poorest prognosis.Amyloid fibers can be deposited in various organs and tissues,causing damage to the structure and function of important organs and leading to lethal results.Establishing a disease risk stratification system is important in predicting prognosis,selecting treatment options,and guiding clinical trials.However,the current risk stratification systems are mainly based on the severity of cardiac involvement,and the prognosis of AL patients with other organ involvement is rarely reported.Kidney is the most frequently affected organ by AL.Currently,little is known about the prognostic factors in patients with kidney involvement,and there is no risk stratification system specialized for kidney AL amyloidosis concerning patients’ survival.Galectin-3(Gal-3),which is involved in numerous physiological and pathological processes such as inflammation and fibrosis,is expressed in the kidney and heart.The serum levels of Gal-3 proved to be related to mortality both in patients with acute and chronic heart failure and in patients with chronic kidney disease.However,little is known about the clinical utility of Gal-3 in patients with AL amyloidosis,where more than half of them have various degrees of kidney and cardiac involvement.In this study,we tested the individual and combined prognostic utilities of Gal-3 and other biomarkers to evaluate subjects with kidney AL amyloidosis.In this study,we sought to examine the relation of Gal-3 with survival of AL amyloidosis,and develop a staging system for AL amyloidosis with kidney involvement.Method: In this retrospective cohort study,we analyzed the clinical and prognosis data of 253 consecutive patients diagnosed with AL amyloidosis by renal biopsy between July 2003 and December 2014.Serum samples of these patients were obtained on the morning before biopsy and stored at-70°C.Biomarkers,including Gal-3,N-terminal B-type natriuretic peptide(NT-proBNP),high-sensitivity cardiac troponin T(hs-cTnT)and free light chain(FLC),were analyzed in the same serum sample with standard kits.Receiver operator curve(ROC)estimates for the thresholds of hs-cTnT,Gal-3,and NT-proBNP that best predict 1-year all-cause mortality were generated and used in the survival analyses.The median value was used to dichotomize the difference between involved and uninvolved free light chains(dFLC)because the area under the ROC curve(AUC)was close to 0.5.Previously reported thresholds were used to dichotomize other variables such as age.Cox proportional hazards analysis was used to identify factors that were prognostic for overall survival(OS).Multivariate analyses were performed with the use of a stepwise forward regression model with an entry probability for each variable set at 0.05.Based on the results of multivariate analyses,a risk stratification system was established.To internally validate the risk stratification system,we applied the bootstrapping method as it could provide stable estimates with low bias.Performance of the system was evaluated by assessing the discrimination and calibration.The C-index(Harrell’s concordance index)was applied to summarize the discrimination;and the calibration slope was employed to reflect the calibration.Survival curves were constructed according to the Kaplan-Meier method,and were compared using a log-rank test.Results: The median age of the 253 patients was 56 years,including 138 males and 115 females.During the follow-up period,158 patients died,7 patients were lost to follow up,and the median OS time after diagnosis was 46 months.In multivariate analysis,serum levels of Gal-3(HR: 1.46;p=0.033),high-sensitivity cardiac troponin T(hs-cTnT)(HR: 2.65;p<0.001)and dFLC(HR: 1.81;p=0.001)were independent predictors of all-cause mortality.The ROC derived best cut-point for Gal-3 and hs-cTnT were 20.24 ng/ml and 0.026 ng/ml,respectively;and the cut-point for dFLC was the median value of 75.89mg/L.Patients were assigned a score of 1 for each of Gal-3≥20.24ng/ml,hs-cTnT≥0.026ng/ml,and dFLC≥75.89mg/L,creating stages 1 to 4 with scores of 0 to 3 points,respectively.The numbers of patients with stage 1,2,3 and 4 were 43(17.0%),94(37.2%),74(29.2%)and 42(16.6%),respectively.The median OS for those patients with stages 1,2,3 and 4 were 100 months(95% CI,80 to NA),60 months(95% CI,49 to NA),29 months(95% CI,19 to 37)and 15 months(95% CI,11 to 23),respectively(p<0.01).The risk ratio per level was 2.08(95% CI,1.75 to 2.48)that of prior level(p<0.001).C-index for the original model was 0.708.Internal validation through bootstrapping yielded a C-index of 0.705 and a slope within the linear calibration of 0.976.Calibration curve at 5 years adjusted successfully to the diagonal line.Conclusion: A higher serum level of Gal-3 is closely related to the kidney and heart injury and associated with an increased risk for all-cause mortality in AL amyloidosis with kidney involvement.Meanwhile,hs-cTnT and dFLC in the traditional prognostic risk stratification system are also independent risk factors for predicting the risk of death in these patients.Based on Gal-3,hs-cTnT and dFLC,we have developed a reliable risk stratification system for evaluating prognosis,which was internally validated to be satisfactory and applicable for predicting mortality in AL amyloidosis with kidney involvement. |
PDF Full Text Request