By Deborah Borfitz 

July 19, 2022 | In collaboration with physicians from around the world, researchers at the University of Pittsburgh and the University of California San Diego School of Medicine recently reported on a series of 20 consecutive cases where an intractable, non-tuberculous Mycobacterium infection was treated with phage therapy over an average course of six months. More than half of those patients saw a favorable clinical or microbiological outcome, suggesting the treatment has efficacy and controlled clinical trials are warranted to nail down the variables contributing to success or failure, according to Pitt biotechnology professor Graham Hatfull. 

Only seven phages—viruses that infect bacteria—were used therapeutically and they were all pulled from an on-campus collection totaling about 20,000 discovered primarily by students working in the Science Education Alliance-Phage Hunters Advancing Genomics and Evolutionary Science (SEA-PHAGES) program that Hatfull began at the University of Pittsburgh in 2008. The employed phages were isolated on a non-pathogenic strain of bacterium called M. smegmatis and their genomes were fully sequenced and analyzed. 

Pitt’s screening process involves about three dozen phages in clusters thought to be particularly effective against M. abscessus infections frequently encountered with cystic fibrosis (CF) patients or other chronic diseases that damage or destroy the lungs’ bronchi, Hatfull says. “The combination of the genomic information and what we know about the host range lets us trim that large set [of phages] down to a modest subset that we can test on a routine basis.”  

Patients in the latest case series report had phages administered twice daily (intravenously, by aerosolization, or both) by their home physician who initially sent the bacterial isolates to the University of Pittsburgh for screening, says Hatfull. The phage preparations were then sent back to the treating doctor. A description of the 20 cases and their outcomes published recently in Clinical Infectious Diseases (DOI: 10.1093/cid/ciac453). 

As with phage therapy in general, these were experimental treatments allowed through the expanded access IND pathway of the Food and Drug Administration (FDA) or the equivalent process for compassionate use outside the U.S., he continues. The administering institution submits the regulatory paperwork, obtains institutional review board approval for the various consent forms, and signs a material transfer agreement.  

The scale of commitment is different when phages are being used to fight other types of infections, such as Pseudomonas aeruginosa or Staphylococcus aureus, and treatments are given for just five to seven days, says Hatfull. Phage therapy for Mycobacterium infections can persist for many months, if not years. 

The bacterial pathogens he specializes in are also more challenging to grow and study, he adds. The University of Pittsburgh is one of very few places where physicians can turn when they have a patient with a severe, unmovable Mycobacterium infection and want to explore the possibility of using phages.  

To date, Pitt has provided phages for the treatment of roughly three dozen such patients, says Hatfull.  

No Cherry-Picking 

For the case series study, Hatfull and his colleagues enrolled 20 out of 200 screened patients whose strains of Mycobacterium were susceptible to phages in Pitt’s collection. Of these, 11 had a favorable outcome, another four showed no response, and the remaining five couldn’t be adjudicated one way or the other, Hatfull says. Every case in the series was, inescapably, an anecdote. 

But the fact that the cases were done consecutively with no cherry-picking makes the study approach “the next best thing” to a controlled clinical trial, he adds. The positive outcomes seen in so many of those cases boosts the odds that phages were responsible. 

It also provides useful insights to properly design a clinical study, or set of trials, to evaluate the variables and measure the effects—and help convince would-be sponsors they would be worthwhile doing, says Hatfull. “The four cases where we didn’t see any obvious impact are in some ways the most interesting because we would like to know why that is.” Perhaps the phages didn’t travel well, he muses, or outcomes were influenced by an underlying condition or the combined use of phages and antibiotics.  

Hatfull says the Pitt team is now talking with physician collaborators in Colorado who regularly deal with CF patients suffering from Mycobacterium infections about securing funding for some small clinical trials. They are also in active discussions with other partners in and outside the U.S. about finding financial support for the launch of larger, multi-center studies. 

“People have been talking about using phage therapeutics for a long period of time, 100 years or so, with waxing and waning interest and sometimes it has been expressed in a form of unbridled enthusiasm,” he notes. Hatfull is committed to following the data either way but believes the just-published case series provides “enough indication” of efficacy to press forward to try to figure out which patients, with what kind of infections, could benefit. 

Realistically, it could take at least another five years to come up with an FDA-approved phage therapy for infections such as M. abscessus that can cause serious lung infections in people with CF, says Hatfull. “It would behoove us to figure out whether the phages can provide some reproducible, consistent resolution for those” given the clinical predicament they can create for tens of thousands of individuals worldwide. 

In welcoming remarks at a phage therapy workshop last year, Marion Gruber, then-head of the vaccine review office of the FDA’s Center for Biologics Evaluation and Research, reports a “resurgence of interest in phage therapy” in the U.S. has led to an influx of applications from sponsors wanting to discuss the potential use of phage therapy and to conduct clinical trials. “Several such studies are now ongoing under our investigational new drug application process and, if successful, may... ultimately lead to FDA licensure of bacteriophage products.” 

The first-ever clinical trial involving phages was cleared by the FDA in 2018 to look at their potential in treating inflammatory bowel disease by editing the microbiome of bad bacteria, as reported in an article in a 2020 series on the topic in Clinical Research News. That investigation focuses on phages that target Escherichia coli, a strain of bacteria regularly found in the mucosa of patients with Crohn's disease, and participants were treated twice a day for 15 days.   

Safe, Well Tolerated 

Half of the 20,000 phages in the collection at the University of Pittsburgh are isolated on M. smegmatis since isolating phages on any of the pathogenic strains is for the most part “unrewarding and unyielding,” Hatfull says. Complete genome sequencing and analysis has been done on about 2,000 of that population to provide a “high-resolution view of the diversity.” 

The bottom line is that there are “lots of different types and they differ from each other in every imaginable way,” says Hatfull. Any two phages might have no genes in common or all the same genes but for one.  

Related phages have been separated into groups, and all but a handful of those clusters do not infect any of the pathogens that have been tested, continues Hatfull. “But when we look at Cluster K phages for example [PLoS One, DOI: 10.1371/journal.pone.0026750), we often find those can not only infect Mycobacterium smegmatis, but maybe also Mycobacterium tuberculosis... or some strains of Mycobacterium abscessus.” 

Phages in general have always been “very well tolerated” by patients because they’re a biological rather than a chemical entity and therefore don’t elicit similar kinds of toxicities or reactions, says Hatfull. “We have phages in us all the time and our bodies are used to seeing different types of them.” 

Compassionate use of phage therapy therefore comes with fewer safety fears and, when eventually used in clinical trials, less worry about the correct dosage, he says. With antibiotic studies, patients are started at a low level with the dose ramped up to learn what’s tolerated. Trials with phages would be “more a matter of giving as much as you have and dialing it back down... to [the minimally required] effective dose. You can essentially strike off concerns about major adverse reactions.” 

No Resistance Worries 

Importantly, there is no need to be overly concerned about resistance when administering a single phage rather than a “cocktail” of two or three of them as has traditionally been used, says Hatfull. Previously, with no data to go on, it was always one of the top worries. 

In both the lab and in the 20 patients in the case series, resistance appears to happen at “pretty low frequencies,” he says. The discovery was made serendipitously both because of the relatively small number of useful phages in the collection and the fact that only one of them was required to effectively treat most of the Mycobacterium isolates in study participants. 

As recently reported in Nature Communications (DOI: 10.1038/s41467-022-29689-4), a case of refractory cutaneous disseminated M. chelonae infection was treated with a single phage (“one of our favorites called Muddy”) in combination of antimicrobial medications and surgery with excellent clinical response—all without resistance either being observed or fundamentally being a reason for failure, Hatfull says. No bacteria were found on biopsy whereas previously they could be readily isolated.  

“If we can administer a single phage without being overly concerned about resistance, it opens up ways we may be able to use the phages for other patients as well in the future,” he says. “If we have three phages, maybe we don’t need to give all three at once but instead give one, see how that goes and, if there is an immune reaction or something else happens, we’d still have two other phages up our sleeve that we could switch to after a month or two, or maybe even rotate the phages through on a schedule over a period of years.” 

Treatment of Mycobacterium infections, particularly those caused by M. abscessus, are difficult due to growing bacterial resistance to antibiotics that has long been the standard of care, says Hatfull. The Centers for Disease Control estimates that nearly three million antibiotic-resistant infections of all sorts occur in the U.S. annually, with 35,000 resulting deaths. 

‘Low-Hanging Fruit’ Strategy 

Only three years ago, Hatfull and his colleagues published an attention-grabbing case report in Nature Medicine (DOI: 10.1093/cid/ciac453) on a 15-year-old CF patient from London with a disseminated M. abscessus infection who was successfully treated with a three-phage cocktail. The backstory on that case was subsequently covered in an article in Clinical Research News. 

Since the initial case report came out, the Pitt team has been contacted by close to 300 physicians around the world with patients in a similar circumstance where antibiotics can no longer control a longstanding infection, Hatfull says. The process of screening a particular bacterial strain against phages in the collection, even if no match turns up, adds to the knowledge base on phage biology. It can also lead to the discovery of new phages that then get added to that screen.  

Many years ago, for example, a phage known as D29 was isolated that failed to infect most of the M. abscessus strains submitted for the matching exercise. But in examining whether D29 could infect a strain called GD40 from the same patient, scientists learned that a host range mutant of that phage had acquired the ability to efficiently infect and kill a strain called GD116. 

That was the strain in one of the 20 patients in the recently published study, a Colorado man in his mid-20s who was treated with two phages for a year to resolve the infection—and put him back on the transplant list, as reported recently in Cell (DOI: 10.1016/j.cell.2022.04.024). He has new lungs, is off antibiotics, and doing better than he has for a long time, says Hatfull. 

It is currently impractical, not to mention cost-prohibitive, to “go to the ends of the earth” to find a phage for every pathogenic strain sent in for screening, he adds. “We are constantly iterating and sort of bootstrapping our way through trying to increase the number of useful phages that we have in our collection.” 

Given that only about three dozen phages are currently being tested, it often takes a broader investigation to identify phages that might infect one particular strain coming through the lab for screening, says Hatfull. It is a “low-hanging fruit” strategy, but one that bears ever-better results over time.