Study On The Prognostic Factors Of Primary Extramedullary Plasmacytoma

Objective: To investigate the factors affecting overall survival(OS)and cancerspecific mortality of patients with primary extramedullary plasmacytoma(EMP)through the SEER database.Methods: Clinical data of EMP patients enrolled in the SEER database from 1975 and 2016 were collected.Univariate and multivariate analyses were performed by Cox proportional risk model to screen the factors affecting OS.Kaplan Meier survival curves were depicted to show the difference of survival between groups.Stratified analyses were utilized to explore the role of surgery in important clinical subgroups.In order to more accurately reflect the impact of treatment modality on OS,propensity score matching was introduced to optimize the imbalance of baseline confounders across the two treatment groups for survival analysis.Furthermore,the factors affecting cancer specific mortality were evaluated via competitive risk models.Results: A total of 1185 patients were included in the study,with a male to female ratio of 1.76:1.The head and neck was the most common disease site,involving 629 cases.The median survival time of EMP patients was 54 months(quartile: 21,121).The 3-,5-,and 10-year survival rates were 76.2%(95% CI: 0.74-0.79),68.4%(95%CI: 0.65-0.71)and 52.9%(95% CI: 0.49-0.56),respectively.Univariate analyses showed that age,race,marital status,median household income,radiotherapy and surgery were the prognostic factors of patients with EMP(all P<0.05).Multivariate analyses displayed that advanced age,black race,divorced/widowed,non-head and neck tumor,and without surgery were the risk factors for prognosis of EMP patients(all P<0.05).Results of stratified analyses indicated that surgery showed a survival advantage in different subgroups of age and primary site(all P<0.05).Besides,patients who received radiotherapy combined with surgery attained the optimal survival outcomes(P<0.001).Survival analysis after propensity score matching reinforced the beneficial role of surgery in OS,while whether radiotherapy or not had no effect on OS.Competitive risk model analyses confirmed that advanced age,divorce/widowed,non-head and neck tumor,and absence of surgery were the risk factors for caner specific mortality(all P<0.05).Compared with patients without radiotherapy,the probability of death from non-EMP events is lower in patients receiving radiotherapy(P= 0.027).Conclusion: Age,marital status,primary site and surgery were the prognostic factors of OS and cancer-specific death in EMP patients.Although radiotherapy was not a protective factor for cancer-specific death,it reduced the probability of death from other events in EMP patients.