Nomogram Model Of Infection In Newly Diagnosed Multiple Myeloma Patients And The Application Value Of Grade≥3 Infection Predictive Models In The Real World

Objective:To explore the clinical characteristics of nosocomial infection in newly diagnosed multiple myeloma(NDMM)patients,and to establish nomogram model for predicting infection.To evaluate the application value of three existing prediction models for early grade ≥3 infection in NDMM patients(FIRST scoring system,GEM-PETHEMA scoring system,and IRMM scoring system)in the real world,with the aim of providing reference for clinical decision-making and thereby reducing the incidence of infection.Methods:The clinical data were collected from 164 patients with multiple myeloma who were treated in Shanxi Bethune Hospital from January 2017 to December 2021 and were retrospectively analyzed.The clinical characteristics of infection were analyzed.Infections were grouped as microbiologically defined infections,clinically defined infections.Univariate and multivariate regression models were used to analyze the risk factors of infection.A nomograms was established.After that,306 patients with NDMM who were enrolled in the Third Hospital of Shanxi Medical University,the First Hospital of Shanxi Medical University,and the People’s Hospital of Shanxi Province from May 1,2016 to April 30,2022 were assigned and grouped based on three different predictive models.The cumulative incidence of early grade ≥ 3 infection was estimated using the Kaplan-Meier method and log-rank test to assess the statistical significance of the difference.To compare the predictive performance in the prediction of infection,the ROC curve was used to show the area under the curve(AUC),and De Long’s test was used to analyze the difference in AUC.Results:1.164 patients with multiple myeloma were included in this study,and 122 cases(74.4%)had infection.The incidence of clinically defined infection was the highest in 89 cases(73.0%),followed by microbial infection(MDI)in 33 cases(27.0%).Among 122 cases of infection,89 cases(73.0%)had CTCAE grade 3 or above.2.The most common site of infection was lower respiratory in 52 cases(39.4%),upper respiratory tract in 45 cases(34.1%),and urinary system in 13 cases(9.8%).Bacteria(73.1%)were the main pathogens of infection.3.Univariate analysis showed that ECOG ≥ 2,ISS stage Ⅲ,C-reactive protein ≥10mg/L,serum Creatinine≥177umol/L had higher correlation with nosocomial infection in patients with multiple myeloma.Multivariate regression analysis showed that C-reactive protein ≥ 10 mg / L(P<0.001)、ECOG ≥ 2(P=0.011)and ISS stageⅢ(P=0.024)were independent risk factors for infection in multiple myeloma patients.4.The nomogram model established based on this has good accuracy and discrimination.The C-index of the nomogram was 0.779(95%CI :0.682~0.875).5.The incidence of grade ≥3 infection within the first 4 months of NDMM was 40.20%.Concerning the FIRST score,GEM-PETHEMA score,and IRMM score,the probability of early grade ≥3 infection in the different groups showed statistically significant differences(low-risk vs.high-risk: 25.81% vs.50.00%,p < 0.001;low-risk vs.moderate-risk vs.high-risk: 35.93% vs.41.28% vs.60.00%,p= 0.045;low-risk vs.moderate-risk vs.high-risk: 20.00% vs.43.75% vs.52.04%,p < 0.001).6.Statistical differences existed in the probability of early grade ≥ 3 infection among the different groups by the FIRST and IRMM scores but no statistical differences in the GEM-PETHEMA score(P < 0.001,P< 0.001,and P= 0.090,respectively).7.AUC(95% CI)of FIRST score,GEM-PETHEMA score,and IRMM score: 0.650(0.593-0.703),0.619(0.562-0.674),0.630(0.573-0.684),with no significant difference(P>0.05).However,the FIRST score has the highest C-index,and its indicators are easy to assess,with a good degree of differentiation,which is more beneficial for clinical work.8.According to the FIRST score,patients treated with bortezomib regimen had significantly higher probability of early grade ≥3 infection in the high-risk group than in the low-risk group(50.94% vs 26.36%,x2=16.252,P<0.001).The OR(95%CI)for the high-risk/low-risk group was 2.901(1.714–4.908).According to the FIRST score,the probability of early grade ≥ 3 infection in the high-risk group of frail patients was significantly higher than that in the low-risk group(53.90% vs30.00%,x2=8.453,P=0.005).The OR(95% CI)for the high-risk/low-risk group was2.728(1.369 – 5.437).Conclusion:1.Patients with multiple myeloma are prone to bacterial infection during hospitalization.2.C-reactive protein≥10mg/L 、 ECOG≥2 and ISS stage Ⅲ are the risk factors of nosocomial infection in multiple myeloma patients.3.The nomogram prediction model established based on this has great prediction value.4.Our findings indicate that the FIRST score(consisting of ECOG,β2-microglobulin,hemoglobin,and lactate dehydrogenase)is a simple and robust infection stratification tool for patients with NDMM and could be used in routine clinical work.5.The FIRST score is also applicable to frail patients and NDMM patients receiving bortezomib treatment.6.The FIRST score can early identify NDMM patients with high risk of infection,contribute to infection prevention,and improve the prognosis of patients.