Almost 100 years ago, in a Paris laboratory, Félix d’Hérelle peered at cultures packed with dysentery bacteria. He had cultured the organisms from the stool of people with the disease, and usually the bacteria blanketed the entire plate. This time, however, d’Hérelle saw something different: “plaques,” or holes, dotted the bacterial smear in cultures from people who were recovering.

D’Hérelle, a bacteriologist who had traveled from Guatemala to Turkey and Tunisia studying pathogens, finally figured out where the holes had come from: A virus that replicated only in the presence of dysentery was attacking the bacteria. He dubbed the virus a “bacteriophage” (literally, bacteria eater), and realized that these bacterial predators, which killed bacteria but spared the hosts, harbored tremendous potential. D’Hérelle began experimenting with the bacteriophages and soon used them successfully against dysentery, which at that time had no other consistently effective treatment. Phage therapy, propelled by d’Hérelle, gained an enthusiastic following throughout the 1920s and 1930s.

The enthusiasm was short-lived, however. Alexander Fleming’s discovery of penicillin, the first antibiotic, and its escalating production and use during the 1940s relegated phage therapy to a footnote in Western medical history. But now, as antibiotic resistance shelves previously effective treatments and doctors become desperate for something new, throwback phage therapy is piquing researchers’ interest.

The concept makes intuitive sense. Bacteriophages typically target and kill a single bacterial species, unlike the harmful kill-off of beneficial gut flora that happens with broad-spectrum antibiotics, and so reduces the risk of resistance. (Broad-spectrum antibiotics accelerate resistance to not just one but to many antibiotics.) In addition, only a tiny dose is required, because the viruses insert their nucleic acids into the target bacteria to replicate their genetic material and make their own new viruses until so many are produced that they kill off the target bacteria.

What’s intriguing is that phage therapy never disappeared from certain pockets of the world, particularly Eastern Europe. Poorer than the West and unable to afford the new antibiotics, doctors in the former Soviet republics of Georgia and Russia and parts of Poland clung to the treatment; the years have brought a slow evolution of phage therapy through trial and error, and to this day, phage treatment is the standard of care for bacterial infections in Georgia.

Western researchers are starting to pay attention. In 2009, researchers in the United Kingdom conducted the first controlled clinical trial of a bacteriophage treatment, in patients with chronic, painful, antibiotic-resistant ear infections caused by Pseudomonas aeruginosa. That bacterium forms tiny ecosystems—biofilms—that shield it from many antibiotics, so the researchers applied one small dose of Biophage-PA, a cocktail of six bacteriophages that target P. aeruginosa, to the ears of 12 patients, while 12 other patients received a placebo. After a few weeks, symptoms of the ear infections declined overall in the phage group, and three of the 12 patients were below the limit of detectability and essentially symptom-free.

In addition, a May 2013 study in the journal Proceedings of the National Academy of Sciencesfound that bacteriophages are not just an effective treatment for infection but also a key defense used by animals (including humans) against pathogen attack. Microbiologist Jeremy Barr and colleagues at San Diego State University showed that bacteriophages in mucus from a variety of creatures are actually symbiotic with the animals. The bacteriophages in the mucus (the first line of defense against pathogens) attack the bacteria that show up there; the phages get food, and we get protection. “This could potentially be a prophylactic treatment,” says Barr. For example, scientists could target the E. coli strain that causes food poisoning. “We might engineer a phage that attacks that strain and ingest the phage in our food and water supply, and it would stick to our mucous membranes and potentially protect against subsequent infections.”

Despite increasing interest in phage therapy, however, limitations abound. Though it is used in Eastern Europe, “they don’t really have the regulatory agencies we have in mainline Europe or in America,” says Catherine Loc-Carrillo, a phage researcher at the University of Utah. No large-scale, randomized, double-blind, placebo-controlled trials have yet examined phage therapy. In addition, no one knows exactly how the immune system deals with bacteriophages, and some phages contain the Shiga toxin gene, which causes food poisoning.

Still, scientists are optimistic that they will be able to generate phages that precisely target particular bacteria or pull from premade libraries of phages to create cocktails to treat many different types of bacteria. Interestingly, though, these cocktails would probably change over time. “Phages are very promiscuous. They tend to swap their genes quite readily, and under certain circumstances you could end up with a group of phages that is different from what you started with,” says Loc-Carrillo. That could affect phages’ effectiveness. FDA regulations would most likely require each new phage to pass intense study before allowing any changes to approved treatment formulations, making approval of phage therapy long and difficult.

Moreover, some experts are concerned that phages, meant to combat the scourge of antibiotic resistance, may themselves breed resistance. If phages do evolve to kill a broader range of bacteria, then they will give those many types of bacteria an opportunity to mount their defenses and become resistant.

Only time and much more study will tell whether phages will be deemed effective—and their fickle nature acceptable.