Correlation Between Body Composition Analysis And Prognosis Of Patients With Multiple Myeloma

Objective:To explore the correlation between body composition and the prognosis of patients with multiple myeloma by using deep learning model to automatically segment the body composition of patients with newly diagnosed multiple myeloma at the level of the fourth thoracic vertebra on chest CT.Methods:Patients diagnosed with multiple myeloma in China-Japan Union Hospital of Jilin University from January 2017 to December 2021 were selected for retrospective analysis.Clinical information such as age,gender,weight,height,BMI,laboratory data such as LDH,serum calcium,serum creatinine,albumin,hemoglobin,β2-MG,and serum free light chain were collected.The chest CT of 79 patients with regular efficacy evaluation was segmented into pectoralis major muscle,pectoralis minor muscle,subcutaneous fat,and mediastinal fat using the deep learning model.Image J was used to measure its area,and the correlation between its body composition and the prognosis of patients with multiple myeloma was analyzed.Ka Plan-Meier survival analysis and Log-Rank test were used to compare survival relationship.Cox proportional hazards model was used for univariate and multivariate analysis.Results:1.The overall survival of patients with low subcutaneous fat area was significantly lower than that of patients with high subcutaneous fat area(HR=2.260,95%CI 1.116-4.578,P=0.024),and the difference was statistically significant.Multivariate analysis showed that subcutaneous fat area was an independent factor affecting the prognosis of patients with multiple myeloma(HR=2.904,95%CI 1.182-7.135,P=0.020).2.Log-Rank test showed that the OS of the 79 patients with multiple myeloma in the high value group and the low value group was statistically significant(P=0.018).There was no significant difference in OS between the high value group and the low value group of subcutaneous fat area in men or women alone(P=0.161,P=0.068).In patients without hematopoietic stem cell transplantation,the OS of high value group and low value group of subcutaneous fat area was statistically significant(P=0.037).3.In multivariate analysis,albumin was an independent prognostic factor for patients with multiple myeloma(HR=2.753,95%CI 1.026-7.392,P=0.044).Conclusions:1.Low subcutaneous adipose tissue area is associated with shorter OS and is an independent risk factor for the prognosis of patients with multiple myeloma.2.The area of mediastinal adipose tissue,pectoralis major muscle,pectoralis minor muscle and BMI have no effect on the overall survival of patients with multiple myeloma,and have no predictive effect on the prognosis of patients with multiple myeloma.3.Albumin is an independent factor affecting the prognosis of patients with multiple myeloma,and low albumin is associated with poor prognosis of multiple myeloma.