Pseudomonas aeruginosa ExoU-associated virulence in HCT recipients and patients with hematologic malignancies

TO THE EDITOR:

Pseudomonas aeruginosa remains a significant pathogen in patients undergoing chemotherapy for hematologic malignancies (HMs) and hematopoietic cell transplant (HCT) recipients.^1,2 Studies on the outcome determinants of P. aeruginosa infections in these patients have largely focused on host factors, initiation of appropriate antibacterial therapy, and resistance to antibiotics.^1,3-7 In contrast, there is a relative paucity of data pertaining to the role of intrinsic P. aeruginosa virulence factors in disease severity and outcomes in these specific patient populations.

The P. aeruginosa ExoU enzyme is a phospholipase that causes cell membrane damage, leading to cell death, and is 1 of 4 enzymes delivered into the host cell by the type 3 secretion system, with the others being ExoS, ExoT, and ExoY.^8,9 Of these, ExoU plays the greatest role in P. aeruginosa pathogenesis and disease severity;^10,11 however, not all P. aeruginosa isolates contain exoU.¹² In vitro infection with exoU+ isolates results in greater cytotoxicity and increased virulence in mouse models of infection.^10,11,13-15 

In humans, the exoU+ genotype has been associated with increased early (5 day) mortality during bacteremia.^16,17 However, not all studies have recapitulated these findings.^18,19 Furthermore, although some studies have included oncology and patients with HCT as minority subsets of the overall study population,^16-19 no study, to our knowledge, has examined the role of ExoU in disease severity exclusively in patients with HMs and recipients of HCT, in whom unique host factors preclude the extrapolation of data from existing studies. Therefore, we performed this study to determine whether the exoU+ genotype is associated with early disease severity of P. aeruginosa in HCT recipients and patients with HMs.

We conducted a retrospective review of P. aeruginosa bloodstream infections (BSIs) and pneumonia that occurred between 1 October 2016 and 30 June 2020 in adult (age ≥ 18 years) patients with HMs and recipients of HCT at Oregon Health and Science University (OHSU). The study was approved by the OHSU institutional review board and conducted in accordance with the Declaration of Helsinki. The definitions and criteria for BSI episodes, pneumonia, appropriate empirical antimicrobial therapy, and multidrug resistant P. aeruginosa have been previously published.^20,21 At OHSU, levofloxacin is used for neutropenic prophylaxis, unless contraindicated. Cefepime is the first-line agent for the empirical treatment of febrile neutropenia, with piperacillin-tazobactam and meropenem being second- and third-line agents, respectively. Adjunctive empirical tobramycin is used at the discretion of medical teams, typically for patients with septic shock and/or suspicion of infection with multidrug resistant P. aeruginosa, while awaiting susceptibility data to guide definitive therapy; combination therapy for the duration of definitive therapy is not standard practice.

P. aeruginosa isolates from blood cultures and bronchoalveolar lavage fluid (BALF) from patients meeting the definition of pneumonia were obtained from the OHSU Clinical Microbiology Laboratory. Bacterial DNA preparation, whole-genome sequencing, and genome assembly of each isolate were performed as previously described.²² GenBank accession numbers for the isolated whole-genome sequencing data have been previously published.²² All genomes were inspected for the presence of exoU gene (NCBI reference sequence, WP_134607685.1).

The severity of infection was assessed based on the composite end point of death or need for intensive care unit (ICU) admission within 7 days of the initial infected blood or BALF culture. This composite end point was chosen on the basis that a simple survival outcome may not reflect infection severity given modern supportive care measures and for alignment with other studies evaluating the impact of the exoU+ genotype on early disease severity.^16-19 If an episode was associated with both a 7-day ICU admission and 7-day mortality, it was counted only once toward the composite end point.

Univariate analysis of categorical variables was performed using Fisher 2-tailed exact test. Factors with P ≤ .1 in the univariate analysis were included in a multinomial logistic regression multivariable model (SPSS version 28 [IBM Corp, Armonk, NY]).

58 episodes of P. aeruginosa BSI and pneumonia were identified in 56 unique patients with HM or recipients of HCT. Two patients had 2 unique BSI episodes separated by 65 days and 120 days. Fifty-six of the 58 (96.6%) isolates obtained during the 58 episodes of infection were from the blood, and 2 (3.4%) were from the BALF. Patient characteristics at the time of infection are shown in Table 1. Cefepime or piperacillin-tazobactam was used as empirical therapy in 90% of the episodes, and most of the isolates were susceptible to these agents. Combination empirical therapy with tobramycin was infrequent (19%).

Eighteen isolates (31%) from 18 patients were exoU+, 16 (88.9%) were from blood cultures, and 2 (11.1%) were from BALF. Fourteen exoU+ isolates were sequence type (ST)-446, which is known to be 1 of the 2 dominant STs in our HM/HCT unit²²; 3 were ST-308, and 1 ST-532. No significant differences were found when comparing the characteristics presented in Table 1 between infections caused by exoU+ and exoU- isolates (data not shown).

The composite end point of 7-day mortality or need for ICU transfer was met in 22 episodes (37.9%) of infection. Seventeen episodes (29.3%) required ICU transfer without 7-day mortality, and 5 episodes (8.6%) were associated with 7-day mortality.

Host and bacterial factors were evaluated for their association with the composite end point (Table 2). The ExoU+ genotype and pneumonia were associated with the composite end point in a univariate analysis, and both were independently associated with the composite end point when analyzed in a multivariable model. Of the 5 episodes associated with 7-day mortality, 4 (80%) were caused by exoU+ isolates (P = .03). A trend toward an association of an exoU+ genotype with 14-day mortality (N = 11) was observed (6 of 11 [54.5%] vs 12 of 47 [25.5%]; P = .08); no association with 30- or 60- day mortality was observed (data not shown).

To our knowledge, the finding that an exoU+ genotype was associated with early disease severity is the first such demonstration specifically in patients with HMs and recipients of HCT and is consistent with the data from studies involving other patient populations.^16,17 This finding also represents, to the best of our knowledge, the first to associate an intrinsic P. aeruginosa pathogenicity factor with disease severity in these patients who are highly vulnerable, in whom the focus has traditionally been on unique host factors. The exoU+ genotype was a more important predictor of disease severity than host factors such as neutropenia (including profound neutropenia), steroid receipt, and age. The only other factor associated with early disease severity was pneumonia, which has been previously identified as an independent risk factor for poor outcomes in similar patients.^1,7 

The relatively small sample size and clonal relatedness of exoU+ isolates limit the generalizability of our findings, which requires confirmation in larger studies. If supported by such studies, strategies to rapidly detect exoU in P. aeruginosa isolates and to develop interventions to mitigate its impact⁸ are justified. Intriguingly, tobramycin disrupted type-III secretion system expression and reduced ExoU-mediated cytotoxicity,²³ suggesting a potential role for adjunctive tobramycin targeted to the treatment of exoU+ isolates. Because of the limitations inherent to our study and our institutional approach to tobramycin use in these patients, we were unable to evaluate the impact of tobramycin administration on the outcomes. Our findings also highlight the need to focus future efforts on defining the contributions of other P. aeruginosa virulence factors, in addition to exoU,²⁴ to the severe disease that characterizes P. aeruginosa infections in highly unique and vulnerable patients.

Contribution: M.H. designed the study, performed the research, analyzed the data, and wrote the manuscript; L.F. performed the research, analyzed the data, and reviewed the manuscript; and L.S. performed the research and reviewed the manuscript.

Conflict-of-interest disclosure: The authors declare no competing financial interests.

Correspondence: Morgan Hakki, Division of Infectious Diseases, Oregon Health and Science University, 3181 SW Sam Jackson Park Rd L457, Portland, OR 97239; e-mail: hakki@ohsu.edu.