Early prediction of high-dose methotrexate-induced acute kidney injury in childhood acute lymphoblastic leukemia

【作者】 程道海； 陆华； 刘滔滔；

【Author】 Dao-Hai Cheng;Hua Lu;Tao-Tao Liu;Department of Pharmacy, the First Affiliated Hospital of Guangxi Medical University;

【机构】 广西医科大学第一附属医院药学部；

【摘要】 ObjectiveHigh-dose methotrexate(HDMTX)is one of the most important agents in the treatment of extramedullary infiltration of acute lymphoblastic leukemia(ALL).HDMTX is applied for this purpose to obtain an effective plasma concentration in the target organs.Thedevelopment of renal dysfunction remains a significant management challenge. The objective of this study was to identify thepotential biomarkers for early prediction of HDMTX-induced acute kidney injury(AKI)in childhood ALL. Methods Weanalyzed theclinical dataof HDMTX treatment inChinese children at the First Affiliated Hospital of Guangxi Medical University from June2014 to August 2018. These data included demographic characteristics(e.g. age,gender, nationalityand body mass index(BMI), etc.), risk group(low risk(LR),intermediate risk(IR) and high risk(HR)), MTX dose per body surface area(BSA),MTX plasma concentrations, HDMTX course number, baseline renal function(e.g.serum creatinine(Scr), cystatin C) and baseline liver function(e.g.serum total protein,total bilirubin, alanine aminotransferase, aspartate aminotransferase,etc.) before HDMTX chemotherapy. The clinical data were expressed as percentages or medians with interquartile ranges(IQR).AKI was defined as anincrease in Scr to ≥1.5 times baseline within one week after the start of MTX infusion. The courses of HDMTX were divided into two groups based on thechildren’s development of AKI or not. The clinical data were compared between two groups.All statistical analyses were performed using the SPSS 20.0 software. Categorical variables were analyzed using a chi-square test, and continuous variables withabnormal distribution were analyzed using a Mann-Whitney U-test. All variables with a p value < 0.15 in the univariate analysis were included in analysis of binary logistic regression with a enter method. The diagnostic accuracy of final model was further analyzed by calculating the area under the receiver operating characteristic(ROC) curve. We also calculated sensitivity and specificity of 48 h plasma MTX concentrationusing cut-off value generated by ROC curve.ResultsThe study totally included 336 ALL children(214 male, 122 female) with a mean age of 6.5 years(range, 1.1-15.5 years) in 1329 HDMTX infusions. Eighty-eight patients with 104 courses of HDMTX(26.2% of patients and 7.8% of HDMTX courses) developed AKI.The courses of mild(Scr 1.5～1.9 times baseline), moderate(Scr 2.0～2.9 times baseline) and severe(Scr≥3.0 times baseline) AKI were 67, 23 and 14, respectively. Univariate analysis revealed that age(6.3(3.7-12.1)vs.5.1(3.6-8.6) years, Z=-2.159), risk group(LR/IR/HR, 27/40/37(26.0%/38.5%/35.6%)vs.513/407/305(41.9%/33.2%/24.9%), χ~2=11.006), body weight(20.0(15.0-34.6) vs.17.5(14.0-27.0) kg, Z=-2.457), BMI(16.6(15.4-17.9) vs. 15.8(14.8-17.3) kg/m~2, Z=-2.953), first HDMTX course(34(32.7%)vs.280(22.9%), χ~2=5.139), BSA(0.78(0.62-1.15)vs. 0.72(0.61-0.96) m2,Z=-2.128), MTX dose per BSA(4.88(3.20-5.05)vs. 4.70(2.87-4.99) g/m2, Z=-3.739)and baseline liver function(e.g. serum total protein(65.9(63.0-69.5)vs.67.1(63.9-70.3) g/L, Z=-2.319), albumin(45.7(43.4-47.9)vs. 46.5(44.5-48.3) g/L,Z=-2.524), total bilirubin(6.0(4.6-8.6)vs. 5.6(3.9-8.1) μmol/L, Z=-2.185) and indirect bilirubin(4.3(3.2-6.4)vs. 3.9(2.4-5.6) μ mol/L, Z=-2.198)) were potentially associated with AKI(all p < 0.15). Above potential risk factors were then brought into binary logistic regression analysis and the results showed that age(odds ratio [OR] =1.349; p = 0.005), first HDMTX course(OR = 1.767; p = 0.013),MTX dose per BSA(OR = 1.944; p = 0.015)andbaseline serumtotal protein(OR = 0.929;p = 0.021) significantly correlated with AKI. The area under the ROC of combined risk factors was 0.685(95% confidence interval [CI] 0.635-0.735). The diagnostic value of 48 h plasma MTX concentration wasfurther exploredusing the ROC curve.The area under the ROC for 48 h plasma MTX concentration was 0.890(95%CI: 0.850-0.930).The sensitivity and specificity values were 78.8% and 90.4%, respectively, at a cut-off value of 48 h plasma MTX concentration > 1.20 μ mol/L.ConclusionsOur findings suggest that greaterage, higher MTX dose per BSA, lower baseline serum protein, and first HDMTX course are significant risk factors for developing HDMTX-induced AKI in childhood acute lymphoblastic leukemia. The threshold of 48 h MTX plasma concentration > 1.20 μmol/L isa valuablebiomarkerfor prediction of HDMTX-induced AKI.更多还原

【Abstract】 ObjectiveHigh-dose methotrexate(HDMTX)is one of the most important agents in the treatment of extramedullary infiltration of acute lymphoblastic leukemia(ALL).HDMTX is applied for this purpose to obtain an effective plasma concentration in the target organs.Thedevelopment of renal dysfunction remains a significant management challenge. The objective of this study was to identify thepotential biomarkers for early prediction of HDMTX-induced acute kidney injury(AKI)in childhood ALL. Methods Weanalyzed theclinical dataof HDMTX treatment inChinese children at the First Affiliated Hospital of Guangxi Medical University from June2014 to August 2018. These data included demographic characteristics(e.g. age,gender, nationalityand body mass index(BMI), etc.), risk group(low risk(LR),intermediate risk(IR) and high risk(HR)), MTX dose per body surface area(BSA),MTX plasma concentrations, HDMTX course number, baseline renal function(e.g.serum creatinine(Scr), cystatin C) and baseline liver function(e.g.serum total protein,total bilirubin, alanine aminotransferase, aspartate aminotransferase,etc.) before HDMTX chemotherapy. The clinical data were expressed as percentages or medians with interquartile ranges(IQR).AKI was defined as anincrease in Scr to ≥1.5 times baseline within one week after the start of MTX infusion. The courses of HDMTX were divided into two groups based on thechildren’s development of AKI or not. The clinical data were compared between two groups.All statistical analyses were performed using the SPSS 20.0 software. Categorical variables were analyzed using a chi-square test, and continuous variables withabnormal distribution were analyzed using a Mann-Whitney U-test. All variables with a p value < 0.15 in the univariate analysis were included in analysis of binary logistic regression with a enter method. The diagnostic accuracy of final model was further analyzed by calculating the area under the receiver operating characteristic(ROC) curve. We also calculated sensitivity and specificity of 48 h plasma MTX concentrationusing cut-off value generated by ROC curve.ResultsThe study totally included 336 ALL children(214 male, 122 female) with a mean age of 6.5 years(range, 1.1-15.5 years) in 1329 HDMTX infusions. Eighty-eight patients with 104 courses of HDMTX(26.2% of patients and 7.8% of HDMTX courses) developed AKI.The courses of mild(Scr 1.5～1.9 times baseline), moderate(Scr 2.0～2.9 times baseline) and severe(Scr≥3.0 times baseline) AKI were 67, 23 and 14, respectively. Univariate analysis revealed that age(6.3(3.7-12.1)vs.5.1(3.6-8.6) years, Z=-2.159), risk group(LR/IR/HR, 27/40/37(26.0%/38.5%/35.6%)vs.513/407/305(41.9%/33.2%/24.9%), χ~2=11.006), body weight(20.0(15.0-34.6) vs.17.5(14.0-27.0) kg, Z=-2.457), BMI(16.6(15.4-17.9) vs. 15.8(14.8-17.3) kg/m~2, Z=-2.953), first HDMTX course(34(32.7%)vs.280(22.9%), χ~2=5.139), BSA(0.78(0.62-1.15)vs. 0.72(0.61-0.96) m2,Z=-2.128), MTX dose per BSA(4.88(3.20-5.05)vs. 4.70(2.87-4.99) g/m2, Z=-3.739)and baseline liver function(e.g. serum total protein(65.9(63.0-69.5)vs.67.1(63.9-70.3) g/L, Z=-2.319), albumin(45.7(43.4-47.9)vs. 46.5(44.5-48.3) g/L,Z=-2.524), total bilirubin(6.0(4.6-8.6)vs. 5.6(3.9-8.1) μmol/L, Z=-2.185) and indirect bilirubin(4.3(3.2-6.4)vs. 3.9(2.4-5.6) μ mol/L, Z=-2.198)) were potentially associated with AKI(all p < 0.15). Above potential risk factors were then brought into binary logistic regression analysis and the results showed that age(odds ratio [OR] =1.349; p = 0.005), first HDMTX course(OR = 1.767; p = 0.013),MTX dose per BSA(OR = 1.944; p = 0.015)andbaseline serumtotal protein(OR = 0.929;p = 0.021) significantly correlated with AKI. The area under the ROC of combined risk factors was 0.685(95% confidence interval [CI] 0.635-0.735). The diagnostic value of 48 h plasma MTX concentration wasfurther exploredusing the ROC curve.The area under the ROC for 48 h plasma MTX concentration was 0.890(95%CI: 0.850-0.930).The sensitivity and specificity values were 78.8% and 90.4%, respectively, at a cut-off value of 48 h plasma MTX concentration > 1.20 μ mol/L.ConclusionsOur findings suggest that greaterage, higher MTX dose per BSA, lower baseline serum protein, and first HDMTX course are significant risk factors for developing HDMTX-induced AKI in childhood acute lymphoblastic leukemia. The threshold of 48 h MTX plasma concentration > 1.20 μmol/L isa valuablebiomarkerfor prediction of HDMTX-induced AKI.更多还原