Introduction

Chronic respiratory infections are an important cause of morbidity, progressive respiratory failure, and mortality in cystic fibrosis (CF) patients. Due to continued exposure to antibiotics, CF patients are at increased risk of developing infections with multidrug-resistant (MDR) bacteria. Pseudomonas aeruginosa is a common pathogen recovered from respiratory specimens in CF patients, with colonization reported in up to 80% of adults; there are increasing rates of colonization with MDR P. aeruginosa as the patients age. Thus, the development of new antimicrobial therapies for treatment of MDR P. aeruginosa infections in this population is a high priority.

The use of lytic bacteriophages for treatment, bacteriophage therapy (BT), represents an exciting anticipated option to fight MDR bacteria. Lytic bacteriophages are viruses that infect, replicate within, and lyse bacteria, causing bacterial cell death and have been shown in vitro to have activity against P. aeruginosa biofilms [1]. There is growing evidence to support the use of BT alone and synergistically with antibiotics in patients suffering from P. aeruginosa infections [2]. BT directed against biofilms is of increasing interest in CF patients suffering from persistent infections [3]. Here, we describe the successful use of BT in a patient with CF awaiting lung transplantation.

Case presentation

A 26-year-old female with CF on the lung transplant waitlist was admitted with a pulmonary exacerbation leading to acute-on-chronic respiratory failure requiring mechanical ventilation for 12 days. Her course was complicated by a pneumothorax requiring long-term chest tube placement. On admission, she was colonized with two distinct MDR P. aeruginosa phenotypes (Supplemental Table 1; non-mucoid strain sensitive to colistin and mucoid strain sensitive to meropenem and piperacillin/tazobactam). She was treated with antibiotics for 4 weeks: colistin and azithromycin for the entire period, piperacillin–tazobactam for the first 2 weeks and a carbapenem for the latter 2 weeks (based on clinical response and antibiotic susceptibilities of recovered organisms). At the end of 4 weeks, the patient was on supplemental oxygen via nasal cannula (NC) at 6 L/min flow rate. She was deconditioned, unable to ambulate independently, and had developed acute kidney injury (AKI) with a serum creatinine (Cr) of 1.5 mg/dL and estimated creatinine clearance (CrCl) of 43 mL/min. She was transitioned to inhaled colistin as suppressive therapy. Of note, multiple P. aeruginosa strains were sensitive only to colistin.

As shown in Fig. 1, 1 week after discontinuation of intravenous (IV) antibiotics, the patient developed fever, increased sputum production, and oxygen requirements. She was restarted on IV antibiotics (vancomycin, colistin, and meropenem which were then switched to piperacillin–tazobactam due to susceptibility profiles). Despite antibiotic therapy, she worsened over the following week with progressive respiratory and renal failure; oxygen requirement was 70 L/min with fraction of inspired oxygen (FIO2) of 50% on heated high-flow blender and serum creatinine of 2.26 mg/dL with associated serum CrCl of 26 mL/min. AKI was attributed to colistin. Due to worsening clinical status and limited antimicrobial options, she was made inactive on the transplant waitlist.

Fig. 1
figure 1

Oxygen saturation and antibacterial use during the clinical course of our case patient

Full size image

At this time, we obtained approval under emergency Investigational New Drug #17710 by the US Food and Drug Administration for AB-PA01 (combination of 4 lytic phages produced in a Good Manufacturing Practices certified facility by AmpliPhi Biosciences Corporation [AMP]) for treatment of MDR P. aeruginosa infection. After informed consent, AB-PA01 was administered every 6 h (4 × 109 plaque-forming units in 5 mL, IV syringe) for 8 weeks.

At the start of BT, the patient was receiving supplemental oxygen via a heated high-flow blender at 40% FIO2 and at a rate of 30 L/min. Colistin was discontinued and the patient received concomitant ciprofloxacin and piperacillin–tazobactam for 3 weeks. By Day 7 of BT, the patient was afebrile and had a dry cough with difficulty in expectorating sputum. Figure 2 provides a baseline chest X-ray image; no radiographical changes were seen in her follow-up chest X-rays as her baseline image was severely abnormal and she had a chronic pneumothorax; however, her pneumonia clinically resolved and she transitioned to NC 10 L/min. For the latter portion of BT, ciprofloxacin was discontinued and doripenem was added based on updated PA sensitivity profiles; the isolates remained sensitive to BT. At the end of the 8-week BT course, the patient was on 3 L/min NC supplemental oxygen, without sputum production and was ambulatory. AKI resolved with return to baseline renal function (peaked at 2.26 mg/dL 1 week before starting BT therapy; 0.83 at the end of therapy, EOT). Her white blood cell count (WBC) normalized to 7.7 1000/mm3 from 15 1000/mm3 prior to BT initiation; and her fever (maximum temperature prior to BT was 102.3 °F) resolved and did not recur while on BT. She did not have recurrence of P. aeruginosa pneumonia and CF exacerbation within 100 days following the end of BT; she underwent successful bilateral lung transplantation 9 months later.

Fig. 2
figure 2

Patient’s chest X-ray on admission

Full size image

No adverse events related to BT were noted clinically or on laboratory monitoring (liver function tests, complete blood counts, electrolytes). At baseline, patient’s P. aeruginosa isolates were susceptible to AB-PA01. Serial P. aeruginosa isolates remained mostly sensitive throughout treatment except for the transient appearance of one non-sensitive isolate on Day 9 (Supplemental Table 1).

Discussion

Bacteriophage therapy is an innovative antimicrobial treatment option against P. aeruginosa infections, especially MDR strains, and deserves more attention. We report the first case in the US that we are aware of in which IV BT was used as an adjunctive treatment for MDR P. aeruginosa infection in a CF patient.

Given the common occurrence of P. aeruginosa in CF patients, there has been burgeoning interest in using BT in this particular population. AB-PA01 is active against approximately 80% of > 400 global clinical P. aeruginosa isolates, including a panel of 65 genotypically different CF isolates (AMP, unpublished). French researchers tested bacteriophages from a Georgian commercial phage preparation against P. aeruginosa isolates from 47 CF patients and reported activity against 70% of the strains [4]. Another in vitro study noted that approximately 50% of P. aeruginosa isolates from CF patients were susceptible to a combination of 10 lytic phages [5]. Bacteriophages can also produce enzymes that hydrolyze biofilms, further validating their use in chronic P. aeruginosa infections in CF patients [3].

The Eliava Institute of Bacteriophages, Microbiology and Virology in Georgia successfully used nebulized BT as an adjunct to antibiotics, antimucus medications, and vitamins in 8 CF patients [6]. The treatment led to a decrease in bacterial concentration in all patients’ sputum samples, as well as improved clinical status. This correlates with our experience, as the patient had difficulty in producing sputum after 5 weeks of BT. Similar to our case patient, they noted a prolonged infection-free period between colonization episodes in all patients.

The goal of using a combination of phages rather than a single phage at a time is that multiple P. aeruginosa isolates may be targeted simultaneously, and within-product complementation can limit resistance development. This is specifically important in the complex microbiological milieu of the CF respiratory tract, which usually contains a genetically and phenotypically heterogeneous P. aeruginosa population. We observed that sensitivity to individual AB-PA01 component phages shifted over time in our patient and included the transient appearance of a resistant isolate, while still ultimately leading to clinical resolution. Clinical trials evaluating the utility of such an approach are urgently needed and microbiome assessments need to be carried out as well.

Our case report demonstrates that intravenous BT is safe and well tolerated and can be used as an adjunct to antibiotics. No adverse events related to BT were noted and the patient had a successful clinical outcome. There was no recurrent pneumonia or CF exacerbations within 100 days of follow-up after the end of BT.

Conclusion

We describe the safe and efficacious use of intravenous bacteriophage therapy in a CF patient awaiting lung transplant. Given the concern for MDR P. aeruginosa in CF patients, BT may offer a viable adjunct to antibiotic therapy in CF patients and further study in clinical trials is urgently needed.