Sanna, M., Caocci, G., Orrù, F., Ledda, A., Vacca, A., Piras, E., . . . La Nasa, G. (2017). Safe fluoroquinolones prophylaxis in blood cancer patients with chemotherapy-induced neutropenia and Glucose-6-Phosphate-Dehydrogenase deficiency. Journal of Clinical Pharmacy and Therapeutics, 42, 733–737.

To describe how safe and effective fluoroquinolone prophylaxis is for patients with a deficiency of enzyme G6PD who are undergoing treatment for hematologic malignancies

Study participants received prophylaxis with either levofloxacin (500 mg/qd) or ciprofloxacin (500 mg/bid) beginning with initial chemotherapy administration until absolute neutrophil count recovers to greater than 1,000. Participants also received prophylactic antifungal therapy (fluconazole, posaconazole, or voriconazole).

• N = 242   
• AGE: 18-75
• MALES: 57%  
• FEMALES: 43%
• CURRENT TREATMENT: Chemotherapy, other (HSCT conditioning regimens)
• KEY DISEASE CHARACTERISTICS: Acute leukemia, Hodgkin and non-Hodgkin lymphoma, multiple myeloma
• OTHER KEY SAMPLE CHARACTERISTICS: Hematopoietic stem cell transplantation: autologous (n = 106, 44%); allogeneic (n = 49, 20%)

• SITE: Not stated/unknown   
• SETTING TYPE: Not specified    
• LOCATION: Italy

PHASE OF CARE: Active anti-tumor treatment

Retrospective cohort study

Patients were identified as having G6PD deficiency based on enzyme activity testing using the G6PD/6PGD Automatic Analyzer (KUADRO), Nurex SRL. Study definitions of “infection” included:

• “Microbiological documented infection” – When infectious agent was identified by culture or biopsy
• “Clinical documented infection” – When infection was identified, but infectious agent was not identified. 

Febrile neutropenia was assigned only when it occurred during the patient’s first episode of neutropenia (absolute neutrophil count of 1,000 mm³ or less). However, authors did not include a definition of acute hemolytic anemia.

Overall, patients with G6PD deficiency had fewer cases of febrile neutropenia (p = 0.01; hazard ratio 0.46; 95% confidence interval [0.25, 0.8]). The subset of patients treated for AML with G6PD deficiency had a higher risk of invasive fungal diseases (p < 0.0001; HR 11.4; 95% CI [3.5, 37.05]) and sepsis due to Candida (p = 0.008; HR 37; 95% CI [2.01, 680.9]). However, incidence of bacterial infection between groups was not statistically significant. Incidence of febrile neutropenia was slightly less (p = 0.01) among study participants with G6PD deficiency. Evaluation of 3,904 red blood cell units administered to study participants identified zero cases of acute hemolytic anemia for all patients regardless of G6PD status.

Fluoroquinolone prophylaxis did not appear to increase risk of acute hemolytic anemia-based on G6PD enzyme status. Patients with G6PD deficiency may be at higher risk of fungal infection, but lower risk of febrile neutropenia during intensive chemotherapy to treat hematologic malignancies.

• Risk of bias (no blinding)
• Risk of bias (no random assignment)
• Selective outcomes reporting
• Other limitations/explanation: Molecular testing results for G6PD were not available. In addition, unclear that analysis of a single potential complication (incidence of acute hemolytic anemia) is sufficient to assess safety. Unclear that single institution, retrospective study is sufficient to assess efficacy.

There is no indication that fluoroquinolone prophylaxis is unsafe or ineffective at reducing the risk of febrile neutropenia, regardless of G6PD enzyme status. If G6PD testing results are available, nurses may need to consider their patients with G6PD deficiency at higher risk for invasive fungal infections while on fluoroquinolone prophylaxis.