Article Text
PDF
Abstract
Chronic pulmonary aspergillosis (CPA) is often poorly responsive to antifungal treatment; secondary infections increase morbidity/mortality, particularly in progressive cases. Interferon gamma (IFNγ) has been implicated in not only Aspergillus control but also bacterial clearance. Clinical notes of patients with CPA treated with IFNγ (2011–2018) were retrospectively hand-searched. In patients treated for >12 months (n=20), the frequency of acute exacerbation reduced from 3.1 to 1.4 episodes/year (p=0.006) in the 12 months after treatment initiation compared with the 12 months before. A significant reduction in the frequency of hospital admissions/year was also observed (0.8 to 0.3, p=0.04). These findings support further prospective studies.
- aspergillus lung disease
- immunodeficiency
- respiratory infection
Statistics from Altmetric.com
Background
Chronic pulmonary aspergillosis (CPA) is characterised by persistent Aspergillus infection, usually complicating pre-existing lung diseases such as chronic obstructive pulmonary disease or bronchiectasis. Although patients are typically without immunocompromise, impaired production of interferon gamma (IFNγ) and/or interleukin 12 (IL-12) has been reported.1 IFNγ prevents conidial germination, enhances alveolar macrophage killing capacity and manipulates Th1 CD4 cell/natural killer cell influx through chemokine signalling.2–4 As such, IFNγ supplementation has been explored as a salvage therapy for severe CPA in patients with impaired IFNγ production.5 6 IFNγ immunotherapy may have additional benefits as CPA is often complicated by recurrent bacterial superinfection; in chronic granulomatous disease, therapeutic IFNγ reduces bacterial infections by up to 70%.7
This study reports the impact of subcutaneous IFNγ supplementation on the frequency of acute exacerbation and all-cause hospital admission when administered as salvage therapy to patients with severe CPA, refractory to antifungals, and proven impairment of IFNγ production.
Methods
For this descriptive study, clinic letters and electronic hospital records were retrospectively hand-searched for all patients prescribed IFNγ salvage therapy for CPA at the National Aspergillosis Centre, Manchester, UK between January 2011 and September 2018 (46 patients). Subcutaneous IFNγ was self-administered at a dose of 50 µg three times per week and prescribed according to clinical judgement when antifungal therapy appeared to be failing due to a combination of clinical, microbiological and radiological parameters in patients with impaired production of IFNγ or IL-12 after in vitro cytokine induction.
For the induction of IFNγ, whole blood samples from patients and healthy controls were diluted 1:5 (Roswell Park Memorial Institute medium) into 96-well F plates (Corning) and activated with lipopolysaccharide (List Biochemicals, 1 µg/mL) or phytohemagglutinin (Sigma, 10 µg/mL), alone or in co-stimulation with IL-12 (ImmunoTools, 5 µg/mL), and incubated at 37°C/5% CO2. For the induction of IL-12, whole blood was stimulated as above with lipopolysaccharide, alone or in combination with IFNγ (Immukin, Boehringer, 2×104 IU/mL). Supernatants were taken after 24 hours and cytokine levels measured by standard ELISA (IFNγ: PeliKine, Sanquin, the Netherlands) or multiplexed particle-based flow cytometry (IL-12: R+D Systems Fluorokinemap), according to the manufacturer’s recommendations.
Patients were excluded from analysis if they started IFNγ therapy within 12 months of the patient search (n=5), were lost to follow-up (n=4) or started IFNγ prior to attending clinic (n=1). The frequency of exacerbation, defined as an acute respiratory decline clinically requiring a course of antibiotics, and all-cause hospital admission were compared between the 12 months pre-IFNγ/post-IFNγ initiation, stratified by duration of IFNγ therapy (Wilcoxon matched-pairs signed-rank test). Incidence rates of death were compared on/off IFNγ for the 36 patients meeting inclusion criteria, starting from the day of IFNγ initiation through to September 2018 for a maximum follow-up of 60 months.
Results
Of the 36 patients meeting the inclusion criteria, 20 received IFNγ for >12 months, 8 stopped treatment due to side effects (all receiving IFNγ for <6 months, mean duration: 2.5 months) and 8 patients died within 12 months of initiating treatment (mean IFNγ duration: 4.6 months, mean survival: 7.5 months). The demographic characteristics of these 36 patients are shown in table 1.
Baseline characteristics of participants at INFγ testing
The death rate was statistically similar while receiving IFNγ therapy (0.16/year) compared with after stopping IFNγ (0.12/year) (p=0.6), with a median follow-up of 33.5 months (IQR: 13–45 months). The 28 patients alive at 12 months were included in further analysis.
Twenty-two patients (79%) were on concomitant antifungal therapy and 8 (29%) were taking azithromycin for long term while receiving IFNγ. Two patients stopped antifungals as they transitioned to IFNγ and azithromycin was started and stopped for one/two patients respectively in the 12 months preceding IFNγ initiation. All remaining patients had the same on/off antifungal and azithromycin status throughout the 24 months of comparison (12 months pre-IFNγ/post-IFNγ initiation), though most patients on antifungals changed their regimen at some point due to adverse effects, azole resistance and/or clinical failure (figure 1).
Individual and mean frequencies of acute exacerbation and hospital admission in the 12 months pre-IFNγ/post-IFNγ initiation, with an illustration of concomitant antifungal/azithromycin treatment regimen, stratified by IFNγ treatment duration. IFNγ, interferon gamma.
In patients treated for >12 months (n=20), there was a significantly lower number of acute exacerbations and hospital admissions in the year following IFNγ initiation compared with the year prior (table 2). No significant changes were seen in patients treated with IFNγ for <6 months for both outcomes (n=8). Patients receiving IFNγ for >12 months (versus <6 months) were observed to have a greater difference between pre-IFNγ/post-IFNγ initiation frequencies of acute exacerbation and hospital admission, although not significantly so for exacerbation. Figure 1 demonstrates the number of acute exacerbations and hospital admissions in the 12 months pre-IFNγ/post-IFNγ therapy initiation on an individual basis.
Number of acute exacerbations and hospital admissions pre-IFNγ/post-IFNγ initiation
Of the 10 patients who remained on the same antifungal treatment throughout the 24-month observation period, 8 received IFNγ for >12 months (negative outcomes lower/equal/higher in 6/1/1 patients) and 2 received IFNγ for <6 months (negative outcomes lower/equal/higher in 0/1/1 patients).
Discussion
In this retrospective, descriptive study, patients with CPA refractory to antifungals alone appeared to have lower frequencies of acute exacerbation and all-cause hospital admission after the introduction of IFNγ therapy. These effects were observed when IFNγ was given for >12 months, but not for a duration of <6 months. Our observation is promising as acute exacerbations and hospital admissions are a substantial cause of morbidity and mortality in this patient population.
Almost all patients in our study had impaired IFNγ production and those who did not have reduced levels of IL-12, a cytokine that stimulates IFNγ production. It is not clear if this represents a primary immune defect or a consequence of chronic pulmonary disease or infection. The apparent beneficial effect of IFNγ suggests that this is clinically relevant and could be mediated by its antibacterial, antiviral or antifungal action. Due to the chronicity of CPA and the number of patients treated, it was not possible to document a meaningful effect of IFNγ on CPA’s clinical course; there was no significant effect on Aspergillus serology, extent of CPA on imaging, microbiological eradication of Aspergillus or significant change in the profile of possible pathogenic organisms from sputum culture (data not shown). The frequency of adverse events and treatment discontinuation was high in this population of severely ill patients.
Long-term antifungals are the treatment of choice for CPA; most of our patients were on azoles. In addition, long-term macrolide treatment is often used for its anti-inflammatory properties and to prevent bacterial exacerbations in patients with chronic lung disease; several of our patients were taking azithromycin during the observation period.8 9 Admittedly, the concomitant use of antifungal and antibacterial treatment, with changes in regimen occurring in the majority of patients during the 24-month observation, may confound the results of this study. Only a minority of patients, however, had changes in their overall on/off treatment status during this time.
The patients included in this study were highly heterogeneous and treatment was individualised. Given this heterogeneity, using patients as their own comparator was considered to be the fairest way to present this retrospective data but is not without its flaws. Having failed to respond to antifungal therapy, they had poor prognosis from the offset. There is potential for introduction for bias if patients with a poorer physiological reserve responded less favourably to IFNγ therapy, self-selecting into the <6-month treatment group. Patients lost to follow-up may also introduce further bias.
Patients who died within 12 months of therapy initiation were not included in the final analysis, arguably introducing bias by excluding the most physiologically vulnerable or patients with the most severe CPA. It is important to note, however, that all patients had declining prognostic indicators prior to starting IFNγ. Inclusion of patients who died during the study period would itself have introduced its own bias; the methodology of a self-controlled case series was considered but deemed inappropriate due to death being a common outcome and the number of patients available.10 There was no evidence of IFNγ therapy being detrimental to health; side effects leading to withdrawal were all mild and there was no difference in death rates between patients on/off IFNγ.
In summary, the frequencies of acute exacerbation and hospital admission were lower after the introduction of IFNγ replacement therapy in patients with severe, refractory CPA. Prospective data are needed to further evaluate the role of IFNγ as adjunctive therapy. In addition to control of fungal burden, IFNγ may improve CPA morbidity/mortality by reducing the frequency and severity of bacterial or viral superinfection. These findings could apply to patients with other chronic lung diseases.
Acknowledgments
The authors would like to thank patients at the National Aspergillus Centre, Manchester, UK for their continued engagement and support.
Footnotes
Contributors CK: conceived the project. RD: coordinated and interpreted immunodeficiency testing. CH and CK: performed the patient search. EJMM and CK: performed data collection; analysed and interpreted the data; drafted the manuscript. CH, RD, GH and DWD: critically revised the manuscript for intellectual content. All authors read and approved the final manuscript.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests DWD and family hold Founder shares in F2G Ltd, a University of Manchester spin-out antifungal discovery company. He acts or has recently acted as a consultant to Scynexis, Cidara, Pulmatrix, Zambon, iCo Therapeutics, Roivant and Fujifilm. In the last 3 years, he has been paid for talks on behalf of Dynamiker, Hikma, Gilead, Merck, Mylan and Pfizer. He is a longstanding member of the Infectious Disease Society of America Aspergillosis Guidelines group, the European Society for Clinical Microbiology and Infectious Diseases Aspergillosis Guidelines group and the British Society for Medical Mycology Standards of Care committee.
Patient consent for publication Not required.
Provenance and peer review Not commissioned; externally peer reviewed.