Infections are a common complication in patients with cancer. They are a significant cause of complications and death in this population. Patients with cancer and low neutrophil counts due to chemotherapy or disease often have a fever and receive antibiotic treatment. The optimal duration of this treatment is largely unknown. Late, there have been some data suggesting the safety of early discontinuation of antibiotics, though most centers still give more prolonged antibiotic therapies in this situation. The unnecessary prolonged antibiotic use may increase infections with multi-drug-resistant bacteria, which carry a high death rate. Also, an increase in infections caused by Clostridioides difficile and an increase in fungal infections can happen. However, some are concerned that stopping antibiotics while the neutrophil count is still low will result in life-threatening infections. Our study aims to test whether shorter antibiotic treatment in these situations is as safe as more prolonged treatment, resulting in better antibiotic prescription practices in this population.

Background Febrile neutropenia is a common complication of chemotherapy-induced neutropenia and neutropenia related to the disease. It occurs in 80% of patients with hematological malignancies with a significant impact on morbidity and mortality. The issue of antibiotic treatment duration in febrile neutropenia is unresolved. The rationale for continuing antibiotics until neutrophil recovery is based on the concern that neutropenic patients may have an ongoing risk of life-threatening infection, and neutropenia may conceal classical manifestations of infection. On the other hand, prolonged broad-spectrum antibiotic treatment has been associated with the emergence of antibiotic resistance, Clostridioides difficile infection and invasive fungal infections, as well as adverse effects and allergic reactions. The evidence suggesting the beneficial effects of prolonged antibiotic treatment is derived from 2 small randomized controlled trials (RCTs) from the 1970s, which showed an increase in infections and mortality with early stoppage of antibiotics. Two recently done RCTs, the HOW LONG and the SHORT studies, suggested that early discontinuation of antibiotics is feasible and is not associated with adverse outcomes. However, the first was not powered for safety outcomes and the second included a lower-risk population and a de-escalation strategy. Despite these reassuring studies, most centres still utilize the resolution of neutropenia as one of the criteria for stopping antibiotics in patients with febrile neutropenia.

Objective This pilot RCT will assess the feasibility of conducting a full future RCT. In the full trial, the investigators will compare early antibiotic discontinuation to the continuation of antibiotics until the resolution of neutropenia in high-risk febrile neutropenic patients, aiming to prove non-inferiority.

Methods The investigators will conduct a pilot open-label, multicentre RCT involving centres in Canada and Israel. The investigators will include adult patients with acute leukemia or patients undergoing allogeneic hematopoietic stem-cell transplantation diagnosed with febrile neutropenia of unknown source. Patients who have received antibiotics for at least 72 hours and are still neutropenic will be recruited if afebrile for at least 24 hours. Patients will be randomized to either early discontinuation or prolonged treatment in a 1:1 ratio using stratified, permuted block randomization. Patients randomized to the intervention arm will have antibiotics stopped at randomization, whereas those in the control group will receive antibiotics until resolution of neutropenia. The outcomes for this pilot study will be to assess the recruitment rate and understand the barriers to obtaining physician and patient consent; assess adherence to the allocated intervention and understand the reasons for crossovers; and measure primary outcome data for sample size re-estimation. This trial will serve as an internal pilot and the outcome data generated will contribute to the full trial. The primary outcome of the full trial will be a composite of all-cause mortality, transfer to intensive care units, or any clinically or microbiologically documented infection.

Inclusion Criteria:

1. Age 18 years and older.

2. The patient either has acute leukemia (AML, ALL or mixed-phenotypic acute leukemia) and is undergoing induction, re-induction or salvage chemotherapy or undergoing allogeneic HSCT and receiving conditioning chemotherapy and/or radiation.

3. Documented febrile neutropenia as defined by the IDSA guidelines [1]:

   1. Single oral temperature of ≥38.3°C or at least two measurements of ≥38.0°C in an interval of ≥1 hour.
   2. ANC ≤ 0.5x109/L.

4. Patient without a clinically or microbiologically documented infection (CDI/MDI).

   We will require the following criteria to rule out infection:

   1. No focus of infection on a thorough history and physical examination at baseline and daily.
   2. Negative blood cultures after at least two sets of blood cultures have been taken. For example, the growth of coagulase-negative staphylococci, diphtheroids or Bacillus spp. from a single set will be considered contamination if another set of blood cultures is negative. Therefore, additional blood cultures will be taken in this case.
   3. Other cultures will be taken as indicated.
   4. A negative chest XR or CT scan (which will be performed according to the physician's discretion) for patients with symptoms of cough or chest pain.

5. The subject will comply with the following criteria:

   1. Received empirical antibiotics for at least 72 hours AND
   2. Is afebrile for at least 24 hours AND
   3. Is still neutropenic (ANC ≤0.5x109/L).

Exclusion Criteria:

1. Concurrent participation in another interventional trial.
2. The patient has received empirical antibiotics for more than seven days from the onset of the febrile neutropenic episode.
3. Septic shock at the onset of the episode or 72 hours (defined as persisting hypotension requiring vasopressors to maintain a MAP ≥ 65 mmHg and having a serum lactate level > 2 mmol/L despite adequate volume resuscitation).
4. Patients with febrile neutropenia secondary to the treatment for solid malignancies, autologous HSCT, CAR-T cell therapy, hematologic malignancies besides acute leukemia when not in the context of allogeneic HSCT, AML treated with consolidation chemotherapy, or ALL treated with intensification and maintenance phase of chemotherapy.
5. Clinically or microbiologically documented infections except for probable or proven invasive fungal disease diagnosed a-priori and treated.
6. Patients receiving their induction chemotherapy or allogeneic HSCT as outpatients.
7. We will not allow the enrollment of patients who have been previously enrolled in this study.

The study protocol and statistical analysis plan will be shared directly after publication. Individual participant data that underlie the results reported in the article will be made available, after deidentification (text, tables, figures, and appendices), for other scientific research approved by an ethical board from 6 months until 24 months after article publication. The data will only be shared to achieve the aims of the approved proposal. Proposals should be directed to the Steering Committee of the clinical trial. Data requestors need to sign a data access agreement to gain access. After 24 months, the data will be available at the UHN without investigator support besides deposited metadata.

from 6 months until 24 months after article publication

scientific research approved by an ethical board. Data requestors need to sign a data access agreement to gain access.