The Value Of RDW In The Efficacy And Prognosis Of Newly Diagnosed Chronic Myeloid Leukemia Patients Treated With TKI

Purpose1.To evaluate the prognostic value of red blood cell distribution width(RDW)in newly diagnosed chronic phase(CP)-chronic myeloid leukemia(CML)patients treated with Tyrosine Kinase Inhibitors(TKI).2.To explore the relationship between RDW,Sokal score,Hasford score,EUTOS score,ELTS score and the clinical efficacy of CML-CP treated with TKI.MethodsWe reviewed the clinical data of all CML patients from 2009 to 2018 at the First Hospital of Lanzhou University,only there were 93 CML-CP patients newly diagnosed and treated with TKI who met the inclusion and exclusion criteria.Receiver operating curve(ROC)analysis was used to determine the optimal cut-off value of RDW for overall survival(OS)and progression free survival(PFS).We compared the demographic characteristics,laboratory indicators,prognostic scores and treatment plans of patients with different RDW,and analyzed the estimated effects of RDW,Sokal score,Hasford score,EUTOS score and ELTS score on early molecular response by 3 months(3M-EMR),complete cytogenetic response by 6 month(6M-CCyR)and major molecular response by 12 month(12M-MMR)of TKI in CML-CP patients.To evaluate the prediction effect of RDW value on the outcome of CML-CP patients,compared with the score systems.Univariate analysis(log-rank test)xand multivariate analysis(proportional risk regression model)were used to analyze the prognostic factors influencing the treatment of CML-CP with TKI.Results1.We enrolled 93 newly diagnosed CML-CP patients who met the inclusion criteria.The optimal cut-off value of RDW for OS and PFS was 18.65%,the area under the curve(AUC)was 0.761 and 0.784,the P value was 0.007 and 0.008,the 95% confidence interval(CI)was 0.646-0.877 and 0.661-0.907,the sensitivity was 80.0% and 87.5%,the specificity was 67.5% and 67.1%,respectively.According to the RDW value,35 patients were high RDW(RDW>18.65%,37.6%)and 58 patients were low RDW(RDW?18.65%,62.4%).2.The median age of all patients was 40 years,with 51 males(54.8%)and 42 females(45.2%).There were 75 patients initially treated with imatinib(50.6%),8 patients treated with dasatinib(8.6%)and 10 patients treated with nilotinib(10.8%).According to the Sokal score,there were 45 patients(48.4%)in the low-risk group,38 patients(40.9%)in the intermediate-risk group,and 10 patients(10.7%)in the high-risk group.According to the Hasford score,there were 52 patients(55.9%)in the low-risk group,38 patients(40.9%)in the intermediate-risk group,and 3 patients(3.2%)in the high-risk group.According to the EUTOS score,there were 88 patients(94.6%)in the low-risk group and 5 patients(5.4%)in the high-risk group.According to the ELTS score,there were 42 patients(45.2%)in the low risk group,36(38.7%)patients in the intermediate-risk group and 15(16.1%)patients in the high risk group.3.We analyzed the RDW value,Sokal score,Hasford score,EUTOS score,ELTS score and CML-CP clinical efficacy treated with TKI,showed that high RDW could evaluate treatment responses at 3months(P=0.034)and 6months(P=0.019),but couldn't evaluate 12 months(P=0.231),and RDW value was significantly lower in patients who responded 3M-EMR(P=0.000)and 6M-CCyR(P=0.001)than those who did not respond,while RDW values was not statistically significant between groups 12MMMR(P=0.416),Sokal score could predict 3M-EMR(P=0.036),but could not predict 6M-CCyR(P=0.763)and 12M-MMR(P=0.926),Hasford score could not predict 3MEMR(P=0.826),6M-CCyR(P=0.709)and 12M-MMR(P=0.586),EUTOS score could not predict 3M-EMR(P=0.759),6M-CCyR(P=0.810)and 12M-MMR(P=0.674),ELTS score could predict 3M-EMR(P=0.028)and 12M-MMR(P=0.034),but could not predict 6M-CCyR(P=0.815).4.The OS and PFS of CML-CP treated with TKI were 89.2% and 91.4% respectively,and survival analysis showed that high RDW had significantly lower OS(77.1% vs 96.6%;P=0.008)and PFS(80.0% vs 98.3%;P=0.002)than low RDW,in Sokal score,intermediate and high-risk had lower OS(P=0.044)and PFS(P=0.025)than low-risk,in Hasford score,intermediate and high-risk had lower PFS(P=0.040)than low-risk,but OS had no statistically difference(P=0.056),in ELTS score,high-risk had lower OS(P=0.002)and PFS(P=0.001)than intermediate and low-risk,in EUTOS score high-risk and low-risk OS(P=0.623)and PFS(P=0.477)had no statistically difference.5.We analyzed the prognostic influencing factors of OS and PFS,univariate analysis found that RDW>18.65%,age?50 years,intermediate and high-risk of Sokal score and Hasford score,high-risk of ELTS score,no 3M-EMR,6M-CCyR,12M-MMR had correlation with the prognosis of CML-CP treated with TKI(P<0.05).The proportional risk regression model(Cox regression)was used for multivariate analysis,and it found that RDW>18.65% was an independent poor prognostic factor for OS(P=0.005,HR=9.741)and PFS(P=0.009,HR=16.74)of CML-CP treated with TKI.Conclusion1.The treatment of TKI for CML-CP patients could get well clinical efficacy and prognosis,with higher OS and PFS,but high-risk of score systems had worse prognosis than low-risk.2.High RDW value could predict 3M-EMR and 6M-CCyR,Sokal score could predict 3M-EMR,ELTS score could predict 3M-EMR and 12M-MMR,but Sokal score,Hasford score,EUTOS score and ELTS score could not predict 6M-CCyR.High RDW combining with score system could better predict the clinical efficacy of CML-CP treated with TKI.3.The OS and PFS of patients with high RDW were worse than those with low RDW.High RDW(>18.65%)and age(?50 years)before TKI treatment were independent poor prognostic factors for the CML-CP treated TKI.