Abstract
Background: Infectious complications are the leading cause of non-relapse mortality in adults with acute lymphoblastic leukemia (ALL), especially in low- and middle-income countries (LMICs), where over 20% of deaths during treatment are infection-related. Febrile neutropenia (FN) is a frequent and potentially fatal consequence of intensive chemotherapy. International guidelines recommend antibacterial prophylaxis in selected high-risk patients; however, in LMICs with high baseline fluoroquinolone resistance, the clinical benefit of this approach remains uncertain.
Methods: We conducted a retrospective cohort study of adult patients with ALL receiving Hyper-CVAD chemotherapy at a national referral center in Mexico City between 2014 and 2023, to assess the impact of fluoroquinolone (FQN) prophylaxis on FN incidence and infection-related mortality. The Hyper-CVAD protocol included alternating cycles of cyclophosphamide-dexamethasone-doxorubicin-vincristine (FA) and high-dose methotrexate-cytarabine (FB). The primary outcome was FN incidence. We used mixed-effects multivariable logistic regression models with random intercepts for individual patients to account for repeated cycles and adjust for relevant covariates. A propensity score–adjusted sensitivity analysis was also conducted.
Results: Fifty-six patients were included (median age 46 years, IQR 23–56), 17.9% of whom had BCR-ABL1-positive ALL and received tyrosine kinase inhibitors. A total of 245 chemotherapy episodes were analyzed (137 FA and 108 FB). FQN prophylaxis was used in 56.7% of episodes, more frequently in subsequent cycles (64.0% vs. 32.1% in first induction, p<0.001) and during FB vs. FA cycles (64.8% vs. 50.4%, p=0.027). Use of FQ prophylaxis progressively declined, from 76.9% during 2014–2016 to just 5.3% in the 2020–2023 period. Overall FN incidence was 54.3%, higher during FB cycles (63.9% vs. 46.7%, p=0.010). FQN prophylaxis was associated with a significantly lower FN incidence (43.9% vs. 67.9%; OR 0.369, 95% CI 0.218–0.626; p<0.001), both in FA (31.9% vs. 61.8%; OR 0.290, 95% CI 0.143-0-586; p=0.001) and FB cycles (55.7% vs. 78.9%; OR 0.335, 95% CI 0.135-0.835; p=0.021). This protective effect persisted in multivariable models (adjusted OR 0.53; 95% CI 0.35–0.80; p=0.002).
The overall mortality rate per chemotherapy episode was 4.9%, of which 83.3% were infection-related. FQN prophylaxis was associated with a significant reduction in mortality (1.4% vs. 9.4%; OR 0.14; 95% CI 0.03–0.65; p=0.006). Despite an increase in fluoroquinolone-resistant gram-negative isolates over time (25% to 61%; p=0.013), clinical benefits of prophylaxis remained evident. There was no increased prevalence of extensively drug-resistant organisms associated with FQ prophylaxis (66.7% vs. 66.0%; OR 1.03, 95% CI 0.43–2.46; p= 0.827). There was no significant increase in Clostridioides difficile infections in the prophylaxis group (5.0% vs. 8.5%; OR 0.572, 95% CI 0.206-1.588; p=0.278).
Conclusions: In this cohort of adults with ALL treated with Hyper-CVAD in a high-resistance, resource-limited setting, fluoroquinolone prophylaxis was associated with significantly reduced rates of febrile neutropenia and infection-related mortality. These findings support the selective use of prophylaxis within a structured antimicrobial stewardship framework, even in regions with rising resistance, as a feasible strategy to mitigate infectious morbidity and mortality in ALL.
