by Kevin Blake
Reports on antibiotic resistance almost invariably begin with a roll call of several very big numbers. 4.95 million deaths per year globally, calculates the World Health Organization. 2.8 million infections per year in the United States resulting in 35,000 deaths, per the US Centers for Disease Control. 10 million deaths per year by 2050 globally costing $100 trillion to the global economy, estimates the Wellcome Trust.
Millions of lives. Trillions of dollars. The scale is overwhelming. And for scientists working in this haze of zeros, the human cost can quickly become abstracted. Though we shouldn't miss the forest of antibiotic resistance for the trees of individual patients, the inverse can be just as blinding. Therefore, it can be helpful now and again to set aside global statistics and focus on specific cases.
One such story was recently reported in Clinical Infectious Diseases, by Dr. Fiona Senchyna and co-authors from Stanford University Hospital. Here, they recounted the fatal case of a severely immunocompromised 12-year-old boy with multidrug-resistant Escherichia coli bacteremia.
The child's case was complex. He had T-cell lymphoblastic lymphoma, further complicated by a genetic disorder that increases the risk of developing childhood cancers and the autoimmune disease lupus erythematosus. In the months preceding this hospitalization, he'd taken ampicillin and cefepime to treat bloodstream infections with Enterococcus faecalis and Streptococcus salivarius, as well as ceftriaxone, meropenem, and vancomycin.
On hospital day 1 the child was admitted for sepsis. He immediately received floxacin and metronidazole while the lab prepared blood cultures to identify the source of his infection. The answer came the next day: carbapenem-resistant E. coli. Aztreonam and ceftazidime-avibactam antibiotics were started, and for nearly a month it appeared that they worked. His fever dissipated, and his blood cultures turned negative.
But three weeks later, on hospital day 26, the sepsis returned. E. coli was again the culprit. Testing this new isolate revealed its minimum inhibitory concentration (MIC) to aztreonam-avibactam had increased from 16 to 64 μg/ml. The E. coli had evolved resistance to the drugs used to treat it. The now ineffective antibiotics were replaced with cefiderocol on hospital day 28, and for a time this too seemed to work. But just one week later, on day 36, the E. coli returned for a third time. Only now its MIC to cefiderocol had increased from 0.5 to >256 μg/ml.
The cefiderocol was stopped, and tobramycin and polymyxin B were started. But the child did not respond well. With his body becoming progressively acidic and his blood pressure plummeting his family transitioned him to comfort care. Less than a week later the boy died of septic shock.
Fig. 1. E. coli isolates from this case study. Image credit: Kevin Blake. Frontispiece: E. coli. Source
To understand the genetic underpinnings of this E. coli's sequential resistance, the authors performed whole-genome sequencing on three isolates. This revealed what an accompanying Commentary described as, "an elegant, detailed, and truly frightening genetic description" of the evolution of antibiotic resistance.
- Isolate 1 (day 1): Isolated pre-treatment. In addition to its New Delhi metallo-β-lactamase (NDM) this isolate also had a pre-existing 4 amino acid insertion in PBP3, which has been shown to confer resistance to aztreonam and reduced susceptibility to cefiderocol. Thus, the subsequent therapy with aztreonam, as well as the future therapy with cefiderocol, were both undermined from the start.
- Isolate 2 (day 26): Isolated after failure with aztreonam and ceftazidime-avibactam. Resistant to ceftazidime, ceftazidime-avibactam, and aztreonam. This coli was a descendent of Isolate 1 and had acquired 5 nonsynonymous mutations in 5 genes. One was in the efflux pump AcrD which has been implicated in resistance to avibactam, and another in the efflux pump EmrA which is known to export multiple classes of antibiotics. The others occurred in genes involved with efflux, biofilm formation, and virulence, but whose contributions to resistance are less clear.
- Isolate 3 (day 36): Isolated after treatment failure with cefiderocol. Resistant to cefiderocol and ceftazidime (ceftazidime-avibactam and aztreonam not tested). Also a descendent of Isolate 1 (but not Isolate 2), this coli had acquired 1 nonsense mutation, 6 nonsynonymous mutations, and 1 promotor mutation in 7 genes. The most significant of these was a premature stop codon in the siderophore receptor CirA, which confers cefiderocol resistance by impairing its transport into the periplasmic space. The remaining mutations likely compensate for the CirA truncation.
This case highlights just how quickly bacteria can evolve resistance. The boy's death wasn't the result of medical ignorance or malpractice; the antibiotics used were appropriate based on careful microbiological testing and currently available evidence, and the authors should be recognized for their expert care. Evolution simply outpaced that care. With just a handful of mutations, E. coli was able to circumvent every antibiotic prescribed until none were left.
Simply put we've fallen behind in the antibiotic arms race. Both in terms of the armaments available and the speed with which we can respond. Fewer new drugs are a problem, and more could help in the short term. But bacteria will evolve ways to render those ineffective as they have for all the drugs that came before. And, as this case demonstrates, they can do so very quickly.
Though we know resistance will inevitably evolve, we currently have a limited ability to anticipate exactly how it will do so. Here, laboratory testing determined which antibiotics the E. coli was resistant to, but not how easily it could evolve resistance to others. Had the authors known that it was only a couple of mutations away from becoming resistant to ceftazidime or cefiderocol, perhaps others could have been prescribed. Currently, this approach is limited by a lack of the knowledge and rapid sequencing resources necessary. But it can save lives and should be prioritized.
Stories like this one aren't new, but they are becoming more frequent. To address rising levels of resistance current treatment options need to be reconsidered and our response times must be accelerated. Failure to do so will only see those very big numbers grow ever bigger.
Kevin Blake is a Scientific Editor at Washington University in St. Louis, in the Department of Pathology & Immunology. He earned his PhD from WashU in 2023 where, in the lab of Dr. Gautam Dantas, he studied antibiotic resistance, microbiomes, and microbial ecology.
Do you want to comment on this post? We would be happy about it! Please comment on Mastodon, Bluesky, or on 𝕏 (formerly Twitter).
Comments