The Limits of Antimicrobial Sensitivity Testing in Cystic Fibrosis

Management of infection in cystic fibrosis (CF) is complex but plays a critical role in improving the survival of patients with CF. The pathogenesis of CF involves production of copious, hyperviscous pulmonary secretions, destructive inflammation driven primarily by neutrophils, and chronic endobronchial infection. Early and persistent colonization with multiple organisms leads to frequent infectious exacerbations, contributing to progressive impairment of pulmonary function.

The aims of infection management include prophylaxis, treatment of exacerbations, and long-term maintenance treatment with antibiotics to reduce bacterial burden. These complex management plans may benefit significantly from cooperative care from pulmonologists and infectious disease specialists, as the former may be provide detailed patient histories as well as guidance on common CR-related complications and the latter may be better positioned to order and evaluate microbial and antibiotic susceptibility testing (AST), which combined allows for personalized care.

A review in Clinical Infectious Diseases discusses the role of AST in the management of CF lung infections.¹On behalf of the Antimicrobial Resistance International Working Group in Cystic Fibrosis, Walters and colleagues argued that choice of antibiotic therapy should be based primarily on good clinical response even where discrepancies with in vitro testing suggest resistance to the current therapy. Because lung infections are common, in patients with CF, a patient’s primary pulmonologist can provide individualized information regarding clinical responses to previous treatments with varying antibiotics. This information may then allow an infectious disease specialist better leverage in the decision to advise for or against results from in vitro testing.

Further, AST is most often predictive of clinical outcomes in monomicrobial acute infection caused by free-floating or planktonic organisms such as bloodstream bacteremia.² Even in this context, the degree of correlation between in vitro testing and patient outcome is far from absolute and often requires experienced clinical judgment with regard to crafting a treatment plan and adhering with that plan in accordance with the results of such testing. Rex and Pfaller have described a “90-60 rule” where infections with susceptible isolates respond to appropriate therapy 90% of the time, whereas infections with resistant isolates (ie, inappropriate therapy) still respond in 60% of cases.³ In other words, susceptibility is a stronger predictor of a successful outcome than resistance is of failure. This is largely explained by the role of host immunity and there is evidence that AST better predicts outcomes in the immunosuppressed host.

In contrast, the lungs of patients with CF are colonized with a polymicrobial microbiota varying across a range of microenvironments. Clinical outcomes are most significantly affected by the presence of Pseudomonas aeruginosa, which has a prevalence in direct proportion with age. Other organisms typically identified include Staphylococcus aureus, Burkholderia cepacia complex, Achromobacter xylosoxidans, Stenotrophomonas maltophilia, and nontuberculous mycobacteria. Over time, and in response to repeated antibiotic exposures, adaptations of these pathogens occurs.

This history is likely best and most accurately charted by pulmonologists, and thus allows for infectious disease specialists to explore, investigate, and tailor treatments based on any genotypic changes including the gradual selection of resistance-conferring mutations and, in the case of P aeruginosa, the presence of strains with a hypermutator phenotype resulting from defective DNA repair. Phenotypically, P aeruginosa characteristically switches to a slow-growing anaerobic pattern and forms biofilms with high levels of inherent antibiotic resistance.

Standard AST, performed on sputum samples and throat swabs from patients with CF, identifies only subpopulations from the community of organisms present. This leads to an underrepresentation of the overall pathogen diversity. Moreover, the recovered isolates may or may not contribute to significant clinical disease. A further issue is the possibility of interactions between different bacterial species influencing susceptibility to antimicrobial killing. For example, an enhanced resistance of P aeruginosa to tobramycin has been documented in the presence of secreted S aureus exoproducts.⁴ Testing of single bacterial isolates is unable to capture these interactions and may misrepresent in vivo susceptibilities.

Another key reason for discrepancy is the interpretation of in vitro inhibitory concentrations. Clinical breakpoints are based on antimicrobial tissue penetration, which may differ in the CF lung. In particular, inhaled antimicrobials are likely to achieve much higher tissue concentrations and so may remain efficacious despite apparent resistance.

Various innovative approaches for improving prediction of antimicrobial efficacy have been attempted. These include biofilm susceptibility testing where clinical isolates are grown as biofilms in a microbiology laboratory to determine the antibiofilm activity of selected antimicrobials. In vitro methods have similarly been employed to test for antibiotic synergy. However, both approaches have, to date, failed to yield improvement in predictive accuracy, including a randomized controlled trial comparing combination bactericidal susceptibility testing with standard approaches, which found no difference in patient outcomes.⁵ Another future avenue of research interest is the use of whole genome sequencing to determine the presence of antimicrobial resistance mutations or more directly through RNA sequencing to determine of expression of bacterial virulence factors.

These theoretical challenges concerning the predictive utility of AST have been borne out in multiple clinical studies, which were evaluated on behalf of the Antimicrobial Resistance International Working Group in Cystic Fibrosis in a systematic review in the Journal of Cystic Fibrosis. Studies were identified which addressed 2 pertinent questions in PICO-format (participant, intervention, comparator, outcome): Is clinical response to antimicrobial treatment of bacterial airways infection predictable from AST results available at treatment initiation? Is clinical response to antimicrobial treatment of bacterial airways infection affected by the method used to guide antimicrobial selection?

The studies assessed both treatment of pulmonary exacerbations and maintenance therapy. The results showed that AST accurately predicted treatment response in only 2 of 16 studies evaluating treatment of pulmonary exacerbations and 1 of 7 studies that addressed maintenance therapy.⁶ None of the studies addressing the second question found a benefit for novel AST methods. Further, these results provide a clear demonstration of the need for the collective experience and expertise of pulmonologists and infectious disease specialists.

In the lungs of patients with CF, genotypic and phenotypic diversity in respiratory pathogens limit the utility of standard antibiotic susceptibility testing. Standard testing methods lack the sophistication to capture in vivo interactions between bacteria and poorly characterize the multiple subpopulations. AST-driven selection of antimicrobials risks unnecessarily discarding useful antibiotics and switching to less efficacious and more toxic options with greater drug-drug interactions. Treatment efficacy of pulmonary infections in patients with CF therefore is best evaluated through the lens of the clinical acumen of medical professionals from both the field of pulmonology and infectious disease.

References

1. Waters V J, Kidd TJ, Canton R, et al. Reconciling Antimicrobial Susceptibility Testing and Clinical Response in Antimicrobial Treatment of Chronic Cystic Fibrosis Lung Infections [published online May 6, 2019]. Clin Infect Dis. doi:10.1093/cid/ciz364

2. Doern G, Breacher S. The clinical predictive value (or lack thereof) of the results of in vitro antimicrobial susceptibility tests. J Clin Microbiol. 2011;49 (9supplement):S11-S14.

3. Rex JH, Pfaller MA. Has antifungal susceptibility testing come of age? Clin Infect Dis. 2002;35(8):982-989.

4. Beaudoin T, Yau YCW, Stapleton PJ, et al. Staphylococcus aureus interaction with Pseudomonas aeruginosa biofilm enhances tobramycin resistance. NPJ Biofilms Microbiomes. 2017;3:25.

5. Aaron SD, Vandemheen KL, Ferris W, et al. Combination antibiotic susceptibility testing to treat exacerbations of cystic fibrosis associated with multiresistant bacteria: a randomised, double-blind, controlled clinical trial. Lancet. 2005;366(9484):463-471.

6. Somayaji R, Parkins MD, Shah A, et al. Antimicrobial susceptibility testing (AST) and associated clinical outcomes in individuals with cystic fibrosis: a systematic review. J Cyst Fibros. 2019;18(2):236-243.