Identification Of Frailty-Related Factors For Predicting The Outcomes In Elderly Patients With Newly Diagnosed Multiple Myeloma

Background and objective:Based on International Staging System（ISS）/Revised International Staging System（R-ISS）,to identify frailty-related factors,including age,performance status,comorbidity,and biomarkers（e.g.,NT-pro-BNP,e GFR,and CRP）,for precisely predicting the outcomes and early mortality（EM）of elderly patients with newly diagnosed multiple myeloma（NDMM）.Methods:In a total of 223 elderly patients（age≥65 years）with NDMM from three centers nationwide,univariate and multivariate Cox regression analysis were conducted to identify the independent prognostic factors associated with overall survival（OS）.Chi-square test and multivariate logistic analysis were used to identify the prognostic factors associated with EM.Results:In this cohort of elderly patients with NDMM,the median OS was 22 months,with the median follow-up of 34 months.On univariate analysis,age（≥70 years）,increased NT-pro-BNP（≥300 pg/ml）,decreased e GFR（<60 ml/min·1.73m~2）,ECOG-PS≥2,elevated LDH,ISS III and R-ISS III were significantly associated with inferior OS.Among them,increased NT-pro-BNP,ECOG-PS≥2,and R-ISS III were further identified as three independent adverse prognostic factors of OS on multivariate analysis.According to the number of these prognostic factors,patients were stratified into four subgroups with 0-3 prognostic factors.Further analysis based on the R-ISS staging revealed that the median OS of patients without or with one or two prognostic factors were 49,25,and 17 months,respectively（P<0.001）.The rates of EM3,EM6,EM12,and EM24 were 12.1%,20.1%,32.2%,and 60%,respectively.The most common cause for EM6（particularly EM3）was disease-related complications resulted from ineligibility for treatment due to poor performance status or severe organ dysfunction or treatment discontinuation due to treatment intolerance,while the most common cause for EM12（particularly EM24）was disease progression or relapse mainly due to inadequate treatment.Increased NT-pro-BNP,ECOG-PS≥2,and decreased e GFR were significantly associated with EM6,EM12,and EM24.In addition,EM12 was associated with R-ISS III,ISS III,and age≥70 years,whereas EM24 was also associated with elevated LDH.On multivariate analysis,R-ISS failed to predict EM,while decreased e GFR,ECOG-PS≥2,and increased NT-pro-BNP were able to estimate the risk of EM,with increased NT-pro-BNP as a common independent factor for EM6（P=0.057）,EM12（P=0.03）,and EM24（P=0.015）,while decreased e GFR and ECOG-PS≥2 as an additional independent factor for EM6 and EM12,respectively.Conclusions:The R-ISS staging,which primarily reflects MM biology,cannot predict EM of elderly NDMM patients.However,the factors（e.g.,NT-pro-BNP,e GFR,and ECOG-PS）associated with frailty and impairment of organ functions,can be used to estimate the risk of EM,among which NT-pro-BNP may be the most important independent factor for EM.Therefore,incorporation of these frailty-related biomarkers into the R-ISS staging may more precisely estimate the prognosis and particularly early death of elderly patients with NDMM.