Incidence of infectious complications in Acute Myeloid Leukemia patients undergoing induction chemotherapy with or without fluoroquinolone prophylaxis in real-life: Results from INFLUENCER study

Abstract INTRODUCTION In Acute Myeloid Leukemia (AML) patients, Febrile Neutropenia (FN) is the main cause of non-relapse mortality, especially occurring during induction chemotherapy. As the most frequent complications are gram-negative bloodstream infections (BSI) caused by mucosal barrier disruption during chemotherapy, current guidelines still recommend Fluoroquinolone (FQ) prophylaxis in high-risk hematological patients with severe and prolonged neutropenia (7 days), despite a consolidated clear benefit on Infectious Related Mortality (IRM). Indeed, given the lack of robust real-life evidence in favour of FQ administration and the progressive increase of Multi Drug Resistant (MDR) bacteria rate due to prolonged antibiotic exposition, the role of FQ prophylaxis in AML neutropenic patients receiving chemotherapy should be carefully reassessed. In this light, Italian epidemiology showed a progressively increasing MDR bacteria rate thus prompting several Hematology Units to omit FQ prophylaxis as local policy.AIM: INFLUENCER study is a retrospective observational multicentric real-life study aiming to evaluate the incidence of FN, BSI, bacterial infections and IRM in AML adult patients undergoing induction chemotherapy, with or without FQ prophylaxis during a 6 years time frame.METHODS AND RESULTS: From January 2018 to December 2023, we enrolled a total of 1016 AML patients treated in 19 Italian hematological centers belonging to SEIFEM (Sorveglianza Epidemiologica Infezioni nelle Emopatie) net.The median age was 59 years, with 536/480 male/female ratio (53%, 47%). The novo and secondary AML (myelodysplastic and therapy related) were 68% and 32%, respectively. According to ELN risk, 35% scored as high-risk AML, 41% intermediate and 24% low.Regarding antibiotic prophylaxis, 418/1016 (41%) patients received FQ, while in 598/1016 (59%) patients it was omitted. Comparative data below are referred to the induction course. Median length of hospitalization was 38 days in the FQ group versus 35 days in the no-prophylaxis group, with statistically significant difference (p0.001). Median length of neutropenia (ANC500/mmc) was 27 and 25 days in the FQ group and no-prophylaxis group, respectively (p0.001). Of note, granulocyte stimulating colony factor was used in 20,2% of prophylaxed patients and in 27,6% of not prophylaxed ones (p=0.009).We documented 399/418 (95,5%) FN in FQ patients versus 521/598 (87,1%) in no-prophylaxis patients (p0.01), while BSI were similar in the two groups: 162/399 (40,6%) and 217/521 (41,7%) in FQ group and in no-prophylaxis group, respectively. Particularly, we recorded a gram-positive prevalence in both groups, with a higher rate in the prophylaxis group (65,8% versus 51,3%, p=0.022). Furthermore, organ infection rate was significantly higher in the FQ group (208/418 (49,8%)) compared to the no-prophylaxis group (175/598 (29,3%)), (p0.001).On the contrary, we recorded 3,1% septic shocks in FQ patients, an incidence significantly lower if compared to 6,5% septic shocks documented in the no-prophylaxis group (p=0.023). However, if considering overall death rate within 60 days (86/1016, 8,5%), IRM was 66,7% in FQ group and 47,6% in no-prophylaxis one, without a significant difference (p=0.187), suggesting that overall survival is not related to FQ prophylaxis use.CONCLUSIONS: To our knowledge, INFLUENCER study is the first real-life study analysing a wide homogeneous population of AML patients, reporting infectious bacterial complications during induction chemotherapy according to FQ prophylaxis. These results do not confirm a negative effect of omitting FQ prophylaxis considering incidence of infections, indeed they question FQ prophylaxis real role, highlighting the need to revise its use, considering local epidemiology and rising MDR.