Giving Antibiotic Cycling Another Shot

Switching up the drugs used to treat bacterial infections could help clinicians battle both illness and resistance at the same time.

By | September 25, 2013

FLICKR, ROBSON#Employing a treatment framework in which clinicians administer different drugs in strategic succession could both treat bacterial infections and select against the development of resistance, Technical University of Denmark’s Lejla Imamovic and Morten Sommer argue today (September 25) in Science Translational Medicine. This new framework, which the researchers call collateral sensitivity cycling, could also help curb unnecessary antibiotic use, which is known to contribute to the emergence of drug-resistant superbugs.

The idea of alternating antibiotics to both beat bacterial infections and outsmart pathogens on the path to acquiring resistance has circulated in the minds of microbiologists for decades, but clinical data to date have been unconvincing.

“Cycling—or rotating antibiotics—has been a hope for a number of years,” said Robert Weinstein of Rush University in Chicago, who was not involved in the work. The hope is to effectively “confuse the bacteria by changing the class of antibiotics you use every month,” he explained. “So they’re turning . . . to see which direction the antibiotic is coming from, and you hit them in the back of the head when they’re not looking.”

“The thought behind traditional drug cycling is that if you alternate between drugs, you alternate between selection pressures, so if you don’t have the selection pressure for resistance, then resistance will disappear,” Sommer told The Scientist. “That is true in some cases,” he said, though in others, “the fitness cost of maintaining resistance to drugs can be very low in the absence of drug selection.” And that could be part of the reason for mixed clinical results, Sommer said. “This type of traditional drug cycling has at times been shown to be advantageous, and other times shown to have no effect at all.”

For the present study, Imamovic and Sommer analyzed both wild-type Escherichia coli and strains evolved in the laboratory to be resistant to 23 commonly used antibiotics. They performed dose-response experiments to determine the susceptibility of each isolate to a variety of compounds. The researchers then treated those strains with pairs of drugs in a cyclical fashion, such that as the bacteria began to develop resistance to drug A—as measured by the amount of drug it took to inhibit bacterial growth—the team quickly switched to drug B. Later, as the bacteria began to develop resistance to drug B, the researchers applied drug A once more. The work identified several such sets of antibiotics for which such cycling successfully killed the bacteria without allowing resistance to take hold.

Interestingly, the researchers found that drugs belonging to a specific class do not always induce the same effects among the bacteria. As such, the researchers noted that the specific drugs E. coli is exposed to may play a role in determining its sensitivity profile.

The study “is an in-depth analysis of resistance linkages and susceptibilities,” said Weinstein. “It’s an important topic because . . . development of antimicrobial drugs is expensive and time-consuming, and there aren’t very many in the pipeline, so other ways to control resistance are very important.”

Still, even if cycling were to improve treatment outcomes, some scientists wonder whether switching up antibiotics can actually help stave off drug resistance in the clinic, or if doing so might contribute to a bigger problem.

“If the cycling results in an overall reduction in antibiotic usage, resistance rates can go down,” Brad Spellberg, who investigates infectious diseases and drug-resistant pathogens at LA BioMed in Torrance, California, told The Scientist in an e-mail. “But if the cycling results in a reduction in usage of a specific antibiotic as it is replaced by another antibiotic, all that is achieved is replacement of resistance for one drug with another.” Plus, he added, “as soon as the drug that is stopped cycles back on, the resistance comes right back,” as evidenced by previous clinical data.

Spellburg is not convinced that in vitro test results are enough to support clinical use of collateral sensitivity cycling at this point. Sommer, too, noted that extensive validation studies are needed. “We want to find out how we can test this in a setting that is more clinically relevant.”

Sommer added that coming to understand the mechanisms that govern collateral sensitivity will be key. It’s important, he said, to figure out what makes certain bacteria highly sensitive to some drugs while particularly resistant to others.

“Some would say that instead of cycling, maybe we should use [fewer] drugs,” said Weinstein. When it comes to the evolution of drug resistance, he added, “the practical implication may be that the bugs are smarter than we are.”

L. Imamovic, M.O.A. Sommer, “Use of collateral sensitivity networks to design drug cycling protocols that avoid resistance development,” Science Translational Medicine, doi:10.1126/scitranslmed.3006609, 2013.

Correction: This article has been updated to reflect the correct spelling of Brad Spellberg's surname. TS regrets the error.

Add a Comment

Avatar of: You



Sign In with your LabX Media Group Passport to leave a comment

Not a member? Register Now!

LabX Media Group Passport Logo


Avatar of: Brian Hanley

Brian Hanley

Posts: 36

September 26, 2013

This strategy has been shown to work on the scale of nations, with long time periods of retiring of drugs. But there are several factors at work that make clinical application of this fail. 

The real-world clinical environment has multiple patient domains that do not coordinate. Ewald showed years ago that the primary vector between patients in hospitals was physicians, so unless patients are kept in reverse isolation, they will often get re-inoculated with resistant bacteria. Ewald also showed why patients who had less isolation from the outside world often did better - they were getting inoculated with non-resistant strains, and so was the patient's environment (bedding, doorknobs, sinks, toilets, walls, etc). Generally, the non-resistant strains from outside can outcompete the resistant hospital strains, so robust bacteria from outside take over the environment. That is why we rarely see resistant strains outside hospitals. Once away from the antibiotic, the bacteria get outcompeted.The hospital's sanitation perversely protects the resistant strains.

It is true that the cost of maintaining an anti-biotic resistance gene can be low, particularly for inducible genes. (e.g. resistance promoters that ) But that misses the reason why such genes get lost. Loss of the genes is basically random. Maintenance of any gene requires energy, even maintenance of a low copy-number plasmid. If one strain reproduces 0.001 times more than another one because it expends just a little less energy, how long will it take for that strain to become 99.999% of the population? About 7,000 generations. That can be 9-10 months for bacteria, assuming reproduction once an hour. E. coli under peak conditions can have generation times as low as 20 minutes.  

If it's a 1% difference, then it will take 500 - 1000 generations, which is 20-40 days. 

So staying with a fixed strategy of 30 days is unlikely to consistently work, even if there is excellent coordination by all health care practitioners in an area. E. coli has one of the fastest generation times of any bacteria, and it is the primary model organism. These days, with select agent regulations, using more realistic organisms is quite expensive and onerous. So that is unlikely to change. 

What has good chances of working are:

- At minimum, enforcing coordinated antibiotic use throughout a hospital among all practitioners. This is tough though, because physicians concerned with one patient may (rightly) perceive that for an individual, jumping the gun on an embargoed antibiotic is the best thing for that patient. Also, a patient or family with knowledge of the embargo system may sue on the grounds that use of the embargoed antibiotic was withholding of life-saving (or limb saving) medicine. It is hard to argue statistics in court without a shield law. You can find attacks on vaccination all over the internet, and accusations that the shield law for vaccines (whose argument is statistical) is nothing but evidence of big pharma stomping on little people for profit. (Which idea is completely wrong. The shield law was created so that this low margin business wouldn't die out in the USA.) 

- If coordinated antibiotic use is feasible, then long periods of embargo time is needed. Thirty days won't cut it. Instead 12-24 months of resting antibiotics should be done. There are enough antibiotics to do that, but it will take commitment. 

- Making isolation wards more porous to bacteria coming from outside, or else deliberately inoculating surfaces and environment with non-resistant strains of normal flora. Both are also tough to do. The former is done by allowing family visits into areas often off-limits, but what they bring in can be unpredictable - robust, but sometimes undesirable. So deliberate inoculation of strains that are known to be relatively benign is probably the way to go. The theory behind this is to give a push to the mathematics. Instead of just relying on random chance and long generation times, bring in lots of immigrants. 

Immigrants can sometimes be problematic, and one needs to keep an eye on that. For instance, prior to the mass aerosol distribution of Serratia marcescens, on San Francisco by the Navy, this bacteria was considered innocuous. Changes to the way a bacterium is vectored can lead to unexpected results. A few, like Bifido and Acidophilus bacteria are difficult to cause infections with and may be viable candidates. 

Do that together, and you'll have success. But doing it will probably require shield laws at least, for facilities that perform them in good faith. 

Avatar of: mightythor


Posts: 83

September 26, 2013

Brian Hanley's argument seems sound, his calculations not so much. They assume that bacteria are dividing rapidly all the time, which is not usually the case. Detailed sequencing studies have suggested that resistant clones can lie dormant in hospital environments for weeks or months and then pop up again. The amount of time required to eliminate resistant strains through overgrowth by "normal" strains may be substantially longer than his estimates.

Popular Now

  1. Thousands of Mutations Accumulate in the Human Brain Over a Lifetime
  2. Two Dozen House Republicans Do an About-Face on Tuition Tax
  3. Putative Gay Genes Identified, Questioned
    The Nutshell Putative Gay Genes Identified, Questioned

    A genomic interrogation of homosexuality turns up speculative links between genetic elements and sexual orientation, but researchers say the study is too small to be significant. 

  4. Can Young Stem Cells Make Older People Stronger?