Patterns and risk factors of infection and its impact in outcomes of newly diagnosed multiple myeloma.

Abstract

e20029 Background: Infection is the leading cause of death in myeloma (MM). It is important to know its patterns and risk factors for determined the better prophylaxis, however it is unknow in Colombia. Methods: In this retrospective study, clinical and microbiological records of newly diagnosed MM from a single center were reviewed from 2014 to 2022. Infections were classified according to the International Immunocompromised Host Society. Flow cytometry (FC) were performed for all patients at diagnosis. The primary objective was to describe the impact of infections on MM outcomes and to identify variables associated with increased risk of infection. Binomial and ordinal logistic regression, Kaplan Meier and Cox proportional hazard models were used. Results: A total of 177 MM treated with bortezomib based therapy were included. 73.4% developed at least one infection. 40% 2 or 3 infections during the study. Overall incidence of infection was 0.34 per patient-year. 69.3% were defined microbiologically ( Streptococcus pneumoniae was the most frequent); 19, 3% had fever of unknown focus (FUF) and 11.4% were clinical defined (CD), being pneumonia the most important. From 2020, 23.8% had SARS-COV-2. 2,2% of infections were at diagnosis, 41% in induction. In patients who underwent stem cell transplanted (SCT), 34% had infection during 100 days after SCT, being FUF the most frequent, while 23% of SCT-ineligible patients had an infectious event during maintenance. Baseline demographic variables were not associated with risk of infection. Multivariable analyses showed that LDH (OR 1.05; 95%CI 1.02 - 1.08 p < 0.01) greater number of tumor plasma cells (TPC) (OR 1.03; 1.01 - 1.06 p 0.04), lower granulocytes (OR 0.64; 0.46 - 0.85 p 0.004) T and NK cells (OR 0.89; 0.81, 0.96 p < 0.01), evaluated with FC and albumin (OR 0.12; 0.01 - 0.084 p = 0.036) were associated with higher risk of infection. Median follow-up was 6.72 years. 11,5% of patients died due to an infection episode. In univariate analysis, infection was associated with a lower overall survival (OS) (HR 0.53 95% 95%CI 0.32-088 p = 0.014), but not in multivariate, when infection was adjusted for other variables as SCT, stage and TPC. Conclusions: This is the first study of infection in MM in Colombia, which could guide the prophylaxis in MM in our setting. A high percentage of infections occurred during treatment, especially in induction which there are a greater tumor load and density of therapy and early SCT, in which bone marrow aplasia occurs. CD and bacterial infections were the most frequent. Infections are associated with lower OS in univariate but not in multivariate analysis likely because patient status, treatment intensity and SCT are also associated with infection and but potentially lower overall mortality. LDH, albumin and immunological populations evaluated with FC were associated with increased risk of infections, thus that these parameters could guide the prophylaxis.