Two Case Reports of Interferon-γ Therapy in Patients with Aspergillus Tracheobronchitis Who Developed an Immunocompromised State After Severe Abdominal Sepsis

Introduction

Aspergillus tracheobronchitis is a rare condition caused by an infection of the tracheal and bronchial tree [1] with a high mortality rate [2–4]. The Aspergillus species, most often Aspergillus fumigatus, produces conidia, that are normally cleared by epithelial cilia and macrophages after inhalation. Immune-compromising conditions and drugs are believed to predispose patients to these infections [5,6]. These conditions include hematological malignancies, neutropenia, AIDS, and use of immunosuppressive drugs, such as chemotherapy, steroids, and drugs indicated after organ transplantation [7]. However, not only these “classic” immunodeficient patients are prone to Aspergillus infections, but also those with sepsis. This was previously described in 4 cases by Hartemink et al [8]. Sepsis is defined as life-threatening organ dysfunction caused by a dysregulated host response to an infection [9]. The immune dysregulation in sepsis can entail both too pronounced pro-inflammatory and anti-inflammatory responses [10–12]. If pro-inflammation dominates, a hyperinflammatory state can develop, which leads to tissue damage, hemodynamic instability, shock, and ultimately organ failure [10–12]. Conversely, an overriding anti-inflammatory response can result in a severely immuno-suppressed state known as sepsis-induced immunoparalysis [10–12]. Sepsis-induced immunoparalysis is characterized by profound suppression of both the innate and adaptive immune functions, including impaired monocyte function, increased lymphocyte apoptosis, and decreased production of pro-inflammatory mediators. This renders the host unable to clear the primary infection and increases vulnerability toward secondary infections, often with opportunistic pathogens [10–12]. Furthermore, reactivation of latent viruses is commonly observed in patients with sepsis-induced immunoparalysis [10–12]. Monocytic human leukocyte antigen DR (mHLA-DR) expression is a commonly used parameter to objectify immunoparalysis, with low levels indicating an immunoparalyzed state. This report describes 2 cases of Aspergillus tracheobronchitis after severe abdominal sepsis in patients with signs of immunoparalysis. Alongside the Aspergillus tracheobronchitis, which was inadequately cleared, other signs of immunoparalysis shown by these patients were reactivation of cytomegalovirus and low mHLA-DR expression. These patients were both treated with the immunostimulant interferon (IFN)-γ, in conjunction with local and systemic antifungal treatment, resulting in microbiological clearance of Aspergillus. Thereby, this work aims to illustrate the consequences and relevance of sepsis-induced immunoparalysis and presents observations on the use of immunostimulatory therapy in this condition.

Case Reports

CASE 1:

A 50-year-old woman underwent elective laparoscopic evaluation of an ovarian cyst, which showed to be benign and was left in situ. The patient had an unremarkable medical history and, particularly, no immunocompromising conditions or medications. After a good initial recovery, the patient was discharged home, but she was re-admitted after 3 days with severe abdominal sepsis and fecal peritonitis, due to intestinal perforation. Cultures showed an infection with intestinal microbial flora, among which Escherischia coli, Streptococcus anginosus, and Enterococcus faecalis. She was empirically treated with ceftriaxone and metronidazole, as local guidelines require. Linezolid was added to the antibiotic regimen when abdominal cultures grew Enterococcus faecium. The subsequent course was characterized by the development of severe septic shock, which required high dosages of noradrenaline and terlipressin, as well as necrosis of the bladder, hands and feet, resulting in amputation of both legs. The patient underwent multiple abdominal surgeries, in which the abdomen was closed in 5 surgical sessions. During hospitalization, the patient had several infectious complications, such as an E. coli bacteremia, a cytomegalovirus pneumonitis, wound infections of the amputated limbs, and ventilator-associated pneumonias. These infections were treated according to local protocols, local antibiotic resistance patterns, and cultures. After 20 days, the patient was planned for bronchoscopic-assisted percutaneous placement of a tracheal cannula. During bronchoscopy, extensive white tracheal plaques were observed (Figure 1). Galactomannan in bronchoalveolar lavage (BAL) fluid was positive, indicating a tracheal Aspergillus fumigatus infection. Treatment with intravenous voriconazole (twice daily 4 mg/kg, resulting in adequate serum levels), anidulafungin (200 mg on the first day, followed by 100 mg daily), and amphotericine B (bi-daily 5–10 mg of nebulization) was promptly initiated. Bronchoscopic monitoring was performed weekly, which revealed no visible response to treatment up to 1 month after the start of treatment (Figure 1). The patient also developed cytomegalovirus reactivation, which, together with the Aspergillus tracheobronchitis, made her suspect for the presence of sepsis-induced immunoparalysis. Therefore, HLA-DR expression on monocytes (mHLA-DR, the most established marker of immunoparalysis [13,14]) was determined, using the Anti-HLA-DR/Anti-Monocyte Quantibrite assay (BD Biosciences, San Jose, CA, USA) on a flow-cytometer (Navios, Beckman Coulter, Hialeah, FL, USA). This is a standardized method that measures surface mHLA-DR density through fluorescent antibody binding, and measures this by flowcytometry. A decreased expression level indicates potential immunosuppression. Indeed, mHLA-DR expression was very low (Figure 1), and therefore, immunological support was started with IFN-γ (100 µg, subcutaneous injections thrice weekly).

Monitoring of immunological treatment was performed with subsequent mHLA-DR measurements (Figure 1). After an initial increase in mHLA-DR expression following the start of IFN-γ therapy, mHLA-DR decreased again. Shortly after, an increase in inflammatory parameters C-reactive protein, ferritin, triglycerides, and s-interleukin (IL)-2 receptor was observed (all measured by routine laboratory methods). Concurrently, there was clinical evidence for a newly developed ventilator-associated pneumonia. As hemophagocytic lymphohistiocytosis could not be ruled out as a cause of the pro-inflammatory phenotype in the setting of interferon therapy, interferon treatment was stopped. Concurrently, the pneumonia was treated. Hereafter, the patient recovered clinically, with normalization of inflammatory parameters. Bronchoscopic and microbiological studies showed good recovery of the trachea and microbiological clearance. Antifungal therapy was continued as prophylaxis. A surgical tracheal cannula was placed, and the tracheal tissue obtained during this placement was microscopically free of Aspergillus. Hereafter, weaning of the ventilator was swift, and the patient was discharged to rehabilitation.

CASE 2:

A 58-year old woman with a medical history of obesity, non-insulin-dependent diabetes mellitus and hypertension underwent a percutaneous gallbladder drainage for a cholecystitis. This patient did not have any conditions that could have rendered her immunocompromised. She developed severe septic shock the day after the intervention, due to peritonitis after biliary leakage. The cultures from this patient showed intestinal microbial flora, and the patient was treated with cefotaxime and metronidazole empirically. Several surgical interventions were performed to drain multiple abscesses. The subsequent postoperative course was complicated by acute kidney injury and Staphylococcus aureus bacteremia. Twenty-five days after surgery, Aspergillus fumigatus was detected in bronchial secretions, for which micafungine, voriconazole, and amphotericin B nebulization and intravenous treatment were initiated. On day 28 postoperatively, a bronchoscopy was performed, showing signs of a diffuse tracheobronchitis covering the mucosa with a white layer. Biopsy (from the carina) revealed a negative culture. BAL fluid galactomannan was (weakly) positive, while serum galactomannan remained negative. On day 32, the patient was transferred to an academic center and bronchoscopy was repeated, which showed severe tracheobronchitis despite continuation of broad antifungal treatment with adequate drug levels (voriconazole). Due to severe respiratory compromise, the patient was mechanically ventilated in the prone position. The presence of a low mHLA-DR expression (determined using the same methods as in case 1) prompted initiation of treatment with IFN-γ (100 µg subcutaneous injections thrice weekly). mHLA-DR expression robustly increased after start of this treatment (Figure 2). Two weeks after initiation of IFN-γ treatment, a tracheal biopsy revealed extensive hyphal elements with background necrosis. Galactomannan peaked at 5.0 at the same time. Remarkably, Aspergillus polymerase chain-reaction and culture remained negative in BAL fluid. Approximately 1 month after the start of IFN-γ, new pulmonary consolidations were observed, and ferritin levels rose. To counteract these signs of hyperinflammation, the IL-1-receptor antagonist anakinra was added to the therapeutic regimen in a dose of 300 mg once daily, followed by prednisolone 1 mg/kg. IFN-γ treatment was discontinued for a period of 2.5 weeks. While galactomannan in BAL fluid gradually declined afterward, the patient could be weaned of mechanical ventilation slowly. Anakinra, amphotericin B, and prednisolone were ceased. Biopsy showed microbiological clearance of Aspergillus, but antifungal treatment was continued as prophylaxis. Approximately 1 week after, biopsy revealed no aspergillus; however, while still on antifungal therapy, the patient developed an acute inspiratory and expiratory stridor. Emergency bronchoscopy showed an obstruction of the tracheal lumen with necrotic tissue, and evacuation of this tissue was not possible (Figure 2). As there were no treatment options, palliative sedation was started, shortly after which the patient died, 4.5 months after her initial hospitalization.

Discussion

These cases illustrate several key insights about Aspergillus tracheobronchitis in the setting of sepsis-induced immuno-paralysis. First, severe sepsis can induce a state of immuno-paralysis, as evidenced by low mHLA-DR expression, even in previously immunocompetent patients, making them susceptible to opportunistic infections, such as invasive aspergillosis. Second, conventional antifungal therapy alone can be insufficient in the context of severe immunoparalysis, as demonstrated by the persistence of infection despite adequate voriconazole treatment. Third, while immunostimulatory therapy with IFN-γ shows promise in restoring immune function (as measured by HLA-DR expression), its use requires careful monitoring, due to the potential risk of triggering excessive inflammation. Finally, these cases highlight the complex balance between providing immune stimulation while avoiding harmful hyperinflammatory states, as illustrated by the need to adjust or interrupt IFN-γ therapy when signs of excessive inflammation developed.

Isolated invasive Aspergillus tracheobronchitis is a form of invasive aspergillosis in which the infection is limited to the tracheobronchial tree. It is fairly uncommon, occurring in 8% of the Aspergillus infections, according to autopsy studies [15]. While histopathological confirmation remains the criterion standard for diagnosing invasive aspergillus tracheobronchitis, obtaining biopsies in critically ill patients can be challenging and may not always be in the patient’s best interest. In our cases, we established the diagnosis through the combination of characteristic bronchoscopic findings of white tracheal plaques, positive cultures, elevated galactomannan in BAL fluid, and compatible clinical presentation. This approach allowed for timely initiation of therapy without subjecting critically ill patients to additional invasive procedures. This is in line with the 2017 joint clinical guidelines of the European Society for Clinical Microbiology and Infectious Diseases, European Confederation of Medical Mycology, and European Respiratory Society. These mention that galactomannan testing in BAL is useful to diagnose invasive aspergillosis, and the guideline stipulates to carefully consider contraindications to invasive diagnostic procedures in patients.

Aspergillus infections can occur in cases of local susceptibility and/or systemic immune suppression. Local susceptibility of the trachea is caused by malfunctioning of mucocilliary clearance. For example, patients with congenital airway constriction are at risk for invasive aspergillosis [16]. Furthermore, ischemia of the tracheal mucosa can enhance local susceptibility, as it has been described that Aspergillus is more present in poorly perfused tracheas after lung transplantation [17,18]. Pertaining to systemic immune suppression, Aspergillus tracheobronchitis has mostly been described in patients with established chronic severe immunodeficiencies [1]. An overview of 156 cases revealed that the most important risk factors were corticosteroid therapy (71.8%), chemotherapy or neutropenia (86.5%), solid organ transplantation (44.2%), hematologic malignancy (21.2%), and chronic obstructive pulmonary disease (14.4%) [7]. Studies have shown that in patients who underwent hematopoietic stem cell transplantation or solid-organ transplantation, invasive pulmonary aspergillosis was the most common invasive fungal infection, occurring in 43% and 59% of patients, respectively [19]. The patients presented in this case report, however, had no history of these classic chronic severe immunodeficiencies, but both had severe septic shock prior to developing the Aspergillus infection. These cases illustrate how severe sepsis might reprogram the immune system toward a state that mimics classic immunocompromised conditions, potentially expanding our understanding of risk factors for opportunistic fungal infections. A case-report by Hartemink et al [8] presents 4 patients with no history of immune-compromising conditions, who developed invasive pulmonary aspergillosis after a severe sepsis. They developed sepsis, respectively, after cholecystectomy, coronary bypass surgery, pancreatico-duodenectomy, and neurosurgery. These patients were treated with antifungal regimens, but all deteriorated on this treatment and died. These 4 patients are similar to the 2 in this case report, regarding the absence of preexistent immunocompromising conditions and the presence of severe sepsis before developing invasive aspergillosis.

Sepsis-induced immunoparalysis has gained significant attention over the last decades [12]. However, diagnosing sepsis-induced immunoparalysis remains a major clinical challenge, as there are no formal diagnostic criteria. Although both patients presented in this report were initially treated with antifungal therapy according to international guidelines, there were no signs of improvement following initiation of this therapy. Together with cytomegalovirus reactivation in one of the cases, this led to the hypothesis that these patients had sepsis-induced immunoparalysis. Thus far, low mHLA-DR expression is by far the most commonly used biomarker to identify this immunologic phenotype [13,14,20]. Nevertheless, it also has limitations, including the need for specialized equipment, putative superiority of multiple measurements over time instead of a single determination [21], and inter-laboratory variation, although this is largely abolished by the Quantibrite method we used [22]. Furthermore, it lacks standardized criteria for use in patients with sepsis. Therefore, it can be considered when immunostimulatory therapy is being contemplated, rather than as a routine test for all patients with sepsis. Indeed, both of our patients showed very low mHLA-DR expression upon detection of the Aspergillus infection.

In the 2 cases, therapy with IFN-γ was initiated to support the immune system, and the treatment effect was monitored by frequent mHLA-DR measurements. This is in contrast to the case report of Hartemink et al [8], in which no mHLA-DR measurements are described, and patients did not receive immunostimulatory medication. Knowledge of the role of reduced mHLA-DR expression is mentioned in this report. There is some anecdotal evidence that immunological support can be used as adjunctive treatment in patients with opportunistic infections due to sepsis-induced immunoparalysis. The 2 agents that have been most extensively investigated are IFN-γ and granulocyte-macrophage colony-stimulating factor. IL-7 is another promising treatment modality, in addition to checkpoint inhibitors such as anti-programmed death ligand 1 and cytotoxic T-lymphocyte associated protein 4 [10,23]. A case series of 9 patients showed that IFN-γ increased mHLA-DR expression and production of the pro-inflammatory cytokine tissue necrosis factor by leukocytes that were ex vivo stimulated with lipopolysaccharide, indicating recovery of immune competence. Furthermore, sepsis resolved in 8 out of these 9 patients [24]. Along the same lines, IFN-γ treatment also showed improvement of leukocytic immune responses in patients with sepsis and invasive fungal infections, including Aspergillus infections [25]. It should be noted that the aforementioned reports are anecdotal or have a small sample size. Although IFN-γ shows promise in treating sepsis-induced immunoparalysis, there is no clear consensus on patient selection. Of interest in this context, the recently completed randomized placebo-controlled ImmunoSep trial investigated the efficacy of IFN-γ therapy for patients with sepsis-induced immunoparalysis (mHLA-DR <5000 and serum ferritin ≤4420 ng/mL) or anakinra therapy for patients with hyperinflammation (serum ferritin >4420 ng/mL) [26]. The eagerly awaited results of this trial will shed light on the diagnostic performance of mHLADR and to what extent it can guide immunostimulatory therapy with IFN-γ. Our experience suggests potential benefits in severe cases with opportunistic infections, but also highlights the risk of inducing hyperinflammation. This suggests that close monitoring is advisable when using these treatments, necessitating frequent measurement of inflammatory parameters.

Aspergillus tracheobronchitis has a high mortality in patients with classic severe immunodeficiencies. In recipients of hematopoietic stem cell transplantation, mortality is over 70% [3]. In a case series of patients with immunodeficiency and Aspergillus tracheobronchitis, 12 out of 17 patients died [1]. A case series of 19 patients, most of whom had radiotherapy, chemotherapy, or systemic steroid therapy as main risk factors, described a fatality rate of 26.3% [12]. Only 2 out of these 19 patients (10.5%) had no known risk factors. It is important to note that the risk of dying of an Aspergillus infection is likely to be influenced by the amendability of the underlying immunological condition. For example, Aspergillus in the context of progression of a hematologic malignancy is likely to result in a worse prognosis than in the context of immunosuppressive drugs, which can be withheld. On the other hand, in the aforementioned series of 156 cases, mortality was 39.1%, and this was related to the fungal infection in 32% of patients that died [7]. In a cohort of patients who were admitted to the intensive care unit with influenza, 19% developed invasive pulmonary aspergillosis. Within the group of patients with influenza and aspergillosis, mortality was 51%, where this was 28% in influenza patients without aspergillosis [4]. It is clear that the high mortality caused by invasive aspergillus infections calls for alternative treatment options, such as immunological support. The question nevertheless remains if this high mortality is caused by the Aspergillus infection itself, or if it is an epiphenomenon illustrating the severe immunodeficient condition the patient is in. Therefore, it is conceivable that in patients with a reversible cause of immunodeficiency (such as patients with severe sepsis), mortality related to Aspergillus infection may be lower than in patients with an irreversible immunodeficiency.

Conclusions

These cases demonstrate that Aspergillus tracheobronchitis can develop due to sepsis-induced immunoparalysis, possibly in the context of regional tracheal ischemia. When antifungal therapy alone proves insufficient, immunostimulatory therapy with IFN-γ may effectively target the underlying immune dysfunction. These findings highlight the importance of considering immunomodulatory approaches in critically ill patients with acquired immune dysfunction, although controlled trials are needed to confirm efficacy.