A Century of Clinical Use of Phages: A Literature Review

Growing antibiotic resistance and the broken antibiotic market have renewed interest in the use of phages, a century-old therapy that fell into oblivion in the West after two decades of promising results. This literature review with a particular focus on French literature aims to complement current scientific databases with medical and non-medical publications on the clinical use of phages. While several cases of successful treatment with phages have been reported, prospective randomized clinical trials are needed to confirm the efficacy of this therapy.


Introduction
Growing antibiotic resistance and the broken antibiotic market currently pose a major threat to global health. In this context, new therapeutic strategies are needed to avoid returning to the pre-antibiotic era. One potential solution is the use of phages, a century-old therapy that fell into oblivion in the West after two decades of promising results. Phages (or bacteriophages) are bacteria-specific viruses that inject their genome into bacterial cells and use the bacterial metabolism to replicate. The life cycle of phages is either lytic or lysogenic. During the lytic cycle, the injected phage genome replicates and destroys the bacterial cell, thus inhibiting bacterial growth. During the lysogenic cycle, the injected genome remains in the host's genome, where it enters a dormant state [1,2]. Lytic phages will be the focus of this literature review, as only they have the capacity to destroy infected bacteria.
Phages are well absorbed enterally and transmucosally [3]. They have short half-lives in vivo as they are rapidly destroyed by the immune system, in particular the reticuloendothelial system. When they infect their target bacteria, however, they can multiply exponentially [4]. Phages have been classified by the International Committee for Taxonomy of Viruses into 12 families, with Straboviridae, Autographiviridae, and Drexlerviridae accounting for over 95% of identified phages [5]. Phages present several advantages: they self-replicate; they have greater specificity than antibiotics; they can be used in patients with allergy to antibiotics; they have a low rate of side effects; their production cost is low; they can be administered through different routes; they can present synergistic effects when combined with antibiotics; they are effective against biofilm; and in the case of bacterial resistance to a specific phage, other phages can be used against the resistant bacterium [1,6]. However, phages also have disadvantages: phage treatment can be initiated only after the causative agent has been identified; multi-bacterial infections require cocktails containing several phages; phage preparations must be sterilized; phage preparations have to be very clean and endotoxin free; phage treatment can induce an immune response in humans; pharmacokinetic data on the action of phages in the human body are lacking; and phages have yet to be properly classified by regulatory agencies [1,6,7].
Many cases of successful treatment with phages have been reported over the last century. Yet, several of these reports are missing from current scientific databases: some fell into oblivion before the advent of databases, others were published in non-medical outlets, and others were written in languages other than English [8]. While many case reports of treatment with phages have an insufficient level of proof, a century of experimentation cannot be ignored.
This literature review with a particular focus on French literature aims to complement current scientific databases with medical and non-medical publications on the clinical use of phages in humans.

History of Phage Therapy
Phages were first described as ultra-microscopic viruses by Frederick Twort in 1915 and characterized as bacteriophages by Félix d'Hérelle two years later [9,10]. They were introduced in the former Eastern Bloc by a student of d'Hérelle, George Eliava, who founded a research institute specifically dedicated to phage therapy in Tbilissi, Georgia, in 1923.
In the 1920s and 1930s, phages were used worldwide for several indications in humans. Several large pharmaceutical companies produced commercial phage preparations at the time: Eli Lilly (Indianapolis, IN, USA) in the early 1930s, Abbott Labs (Chicago, IL, USA) in the early 1930s, and Bristol-Myers Squibb (New York, NY, USA) from the early 1930s to the 1940s [11]. However, phage therapy failed to impose itself in the West due to inconsistent treatment results, a poor understanding of phages' mechanism of action, and the advent of broad-spectrum antibiotics [1]. The therapy was discredited in two reports published in 1934 and 1941 by Eaton and Krueger [12][13][14][15][16]. While some Western companies continued to produce phages afterwards (in particular for the food processing industry), the use of phages in humans was largely abandoned.
In the former Eastern Bloc, the institute founded by Eliava in Georgia continued to carry out major research on phage therapy and supplied phage products to the USSR throughout the 20th century. Renamed the G. Eliava Institute of Bacteriophages, Microbiology, and Virology in 1988, it pursued its research activities after the fall of the Iron Curtain. In Russia, phages are now produced commercially by the company Microgen.
In 2005, a Phage Therapy Unit was created at the Hirszfeld Institute of Immunology and Experimental Therapy in Wrocław, Poland. This unit is now the second largest center for phage research in Europe after the G. Eliava Institute.
Since the mid-2000s, growing antibiotic resistance has prompted institutes and pharmaceutical laboratories all over the world to create their own phage banks on the model of the G. Eliava Institute [17]. These include: Adaptive Phage Therapeutics in the United States, Biobank, First Affiliated Hospital of Xi'an Jiaotong University, Institute for Protein Science and Phage Research in China, the Bacteriophage Bank of Korea [18], the Israeli Phage Bank at the Hebrew University of Jerusalem in Israel [19], the National Collection of Type Cultures, the Bacteriophage Collection in the United Kingdom, Fagenbank in the Netherlands, DMZ in Germany, Queen Astrid Military Hospital in Belgium, and Pherecydes Pharma in France.
Despite this renewed interest in phage therapy, the efficacy of phages has yet to be demonstrated.

Legal Framework for the Use of Phage Therapy in Humans
Phages are classified as drugs in the US and as medicinal products in the EU [20]. Like other drugs, they are subject to marketing and manufacturing authorization from the Food and Drug Administration (FDA) in the US and the European Medicines Agency (EMA) in the EU.
To obtain marketing authorization, phage therapies must first be validated in preclinical in vitro and in vivo studies. Once these are completed, phase I to IV clinical trials are required to confirm their safety and efficacy in humans. To date, no phage therapy has reached phase IV of clinical trials.
To receive manufacturing authorization, phage products must comply with good manufacturing practice (GMP) standards, which involve a high level of purification and sterilization. However, given the lack of fit between these standards and the viral nature of phages, no phage product has so far been approved for use in humans by the EMA or the FDA. Phage production is authorized only for compassionate use and clinical research under Article 37 of the Declaration of Helsinki.
Marketing regulations and GMP standards are barriers limiting pharmaceutical investment in phage therapy [20].

Safety of Phage Therapy in Humans
The human body is routinely exposed to large numbers of endogenous phages. Since phages are composed only of proteins and DNA and do not release toxins when they die, their toxicity is low in humans. Phages must nevertheless be purified to reduce their virulence and ensure their safe administration [3].
Bacteria destroyed by phages can release bacterial toxins, and excessive bacterial lysis often causes immune system reactions. These phenomena, however, occur mainly when phages are administered intravenously. Other routes of administration can be used safely.
In 2009, Merabishvili et al. proposed a quality control for the safe use of purified phage cocktails against Staphylococcus aureus and Pseudomonas aeruginosa [21]. In 2014, an international panel composed of 29 experts from ten countries developed quality, safety, and efficacy requirements for sustainable phage therapy products [22]. Such requirements could replace current GMP standards, which would facilitate the use of phage therapy in humans.

Use of Phages in Food Processing and Plant Disease Control
Several studies have shown the efficacy and safety of using phage products in food processing and plant disease control [23][24][25][26].
Intralytix (Columbia, MA, USA) markets three phage products for food processing: ListShield™ against Listeria monocytogenes, EcoShield™ against Escherichia coli, and SalmoFresh™ against Salmonella spp. Omnilytics (Sandy, UT, USA) sells several phage products, including Agriphage™, for use in agriculture. Elanco (Greenfield, IN, USA) markets Finalyse™, an anti-bacterial spray targeting E. coli O157. In Europe, Micreos (Wageningen, The Netherlands) produces Listex™ and Salmonelex™, both of which were approved by the FDA. APS Biocontrol (Dundee, SC, UK) produces Biolyse ® , a phage product sprayed on potatoes during processing. All these products have been approved by the FDA and/or the EMA [6].

Literature Review of a Century of Clinical Use of Phages
A search of the MEDLINE database was performed without language restrictions to identify articles on phage therapy published between 1922 and 2022. The following keywords were used: (phage OR phage therapy OR bacteriophage) AND (treatment OR use OR therapy) AND (clinical OR human). The titles and abstracts of identified articles were screened for inclusion. Prospective randomized clinical trials, literature reviews, prospective non-randomized clinical trials, and isolated case reports evaluating the clinical use of phages were included in the review. We excluded articles that did not deal with the use of phages in humans and those referring to articles already cited that did not bring new cases to our analysis.
The reference lists of selected articles were perused to identify other relevant publications on phage therapy. These publications consisted of books, theses, and medical and non-medical articles published between 1915 and 2022. Publications in languages other than English were considered, with a particular focus on French literature. All identified publications describing cases of phage therapy were included in the review.
The main prospective randomized clinical trials, literature reviews, and prospective non-randomized clinical trials are presented in Table 1. Isolated case reports published after 1945 are shown in Table 2, while those published prior to 1945 appear in Table 3. We took 1945 as a cut-off year as this corresponds to the period when broad-spectrum antibiotic therapy was introduced and phage therapy lost credibility in the West. Moreover, phages were mostly used in isolation before 1945 and were generally combined with antibiotics after that date.     ENT: ear-nose-throat.
Studies were considered to have demonstrated the efficacy of phage treatment if 50% of evaluated patients had a favorable outcome. Studies that did not specify the number of improvements or cures were classified as unspecified. Figure 1 shows the distribution of articles on phage therapy for the main clinical foci of infection. Studies were considered to have demonstrated the efficacy of phage treatment if 50% of evaluated patients had a favorable outcome. Studies that did not specify the number of improvements or cures were classified as unspecified. Figure 1 shows the distribution of articles on phage therapy for the main clinical foci of infection.

Prospective Randomized Clinical Trials
Our search of the MEDLINE database identified nine prospective randomized trials evaluating phage therapy. In

Prospective Randomized Clinical Trials
Our search of the MEDLINE database identified nine prospective randomized trials evaluating phage therapy.
In . Three cohorts of three patients each received successive intranasal doses of AB-SA01 twice daily at a concentration of 3 × 10 8 PFU for seven days (cohort 1), 3 × 10 8 PFU for 14 days (cohort 2), and 3 × 10 9 PFU for 14 days (cohort 3). Treatment was well tolerated overall, and no serious adverse events or deaths were reported. Preliminary efficacy results showed favorable outcomes, with clinical and microbiological evidence of infection eradication in two of the nine evaluated patients.
The results of the "PhagoBurn" phase I/II randomized clinical trial were published in 2018 by Jault et al. [52]. The aim of this trial was to compare the efficacy and tolerability of a cocktail of anti-P. aeruginosa phages to standard of care (silver sulfadiazine) in the treatment of burn wound infections. Twenty-seven patients were randomly allocated to receive a topical application of either treatment daily for 7 days, with 14 days of follow-up. Safety was evaluated in all patients who received at least one phage dressing (treatment group) or one silver sulfadiazine dressing (control group). However, in this study, the applied phage titer was extremely low, which contributed to clinical failure. The trial was stopped before termination due to insufficient efficacy of their phage cocktail.
The 2016 trial by Sarker et al., supported by a grant from Nestlé Nutrition and Nestlé Health Science, assessed the safety and efficacy of a T4-like phage cocktail compared to the Microgen ColiProteus phage cocktail and a placebo in Bangladeshi children hospitalized with acute bacterial infection or diarrhea [52]. No adverse event attributable to the oral application of phages was reported. Treated children had higher fecal phage prevalence and titers than those who received a placebo. However, the oral phages failed to achieve intestinal replication and to improve diarrhea outcome, possibly because phage coverage was insufficient and E. coli pathogen titers were too low.
In 2009, Rhoads et al. conducted a phase I trial to evaluate the safety of a phage cocktail for the treatment of venous leg ulcers in humans [59]. Forty-two patients with chronic venous leg ulcers were included, 39 of whom completed the trial. The ulcers were treated with either a saline control or a phage preparation targeting P. aeruginosa, S. aureus, and E. coli for a period of 12 weeks. Follow-up continued through week 24. No adverse events attributable to the phage cocktail were observed. No significant difference was found between the test and control groups in terms of frequency of adverse events, rate of cure or frequency of cure.
In 2009, Wright et al. published the results of a phase I/II trial assessing the safety and efficacy of the phage preparation Biophage-PA in patients with chronic otitis caused by an antibiotic-resistant strain of P. aeruginosa [60]. A total of 24 patients randomized into two groups of 12 were treated with a single dose of Biophage-PA or a placebo and were followed up 7, 21, and 42 days after local application by the same otologist. The main outcomes were the symptoms observed by physicians (erythema, inflammation, ulceration, granulation, polyps, amount of discharge, type of discharge, and odor) and those reported by patients (discomfort, itchiness, wetness, and smell). No adverse events were observed. In the phagetreated group, pooled clinical indicators significantly improved, and P. aeruginosa counts significantly decreased from baseline, demonstrating the efficacy of the phage preparation.
In 2005, Bruttin et al. evaluated the safety of E. coli phage T4 in 15 healthy adult volunteers [62]. All patients included in this crossover study received a low dose of phage T4 (103 plaque-forming units (PFU)/mL), a high dose of phage (10 5 PFU/mL), and a placebo in their drinking water. No adverse event related to the application of phage T4 was reported. Phage T4 was detected in a dose-dependent manner in the feces of patients orally exposed to phages. However, oral phage application did not result in a decrease in total fecal E. coli counts. Moreover, no substantial replication of phage T4 was observed in the commensal population of E. coli.
Three of the four trials evaluating the clinical efficacy of phages showed negative results. As the other trials were designed to evaluate the safety, tolerability and/or prelimi-nary efficacy of phage therapy, one cannot conclude on the clinical efficacy of phages based on their results.

Literature Reviews
Twenty-one literature reviews evaluating the efficacy of phage therapy were identified. The following nine reviews covered the largest number of patients.
The 2021 literature review by Genevière et al. included 20 case reports of 51 patients treated with phages for bone and joint infections [28]. The overall success rate was 71%.
In  [56]. They found phages to be well tolerated and to have overall efficacy.
In 2009, Chanishvili et al. reported several cases of patients treated with phages at the G. Eliava Institute in Georgia [58]. One cannot not conclude on the efficacy of phages due to the disparity in methods used and the range of treated infections.
In their 2000 literature review, Weber-Dabrowska et al. from the Hirszfeld Institute of Immunology and Experimental Therapy in Poland reported 1307 cases of treatment with phages in patients with suppurative bacterial infections caused by different species of multidrug-resistant bacteria [64]. Their conclusions indicated a very high efficacy of phage therapy: complete cure was observed in 85.9% of cases, partial cure in 10.9% of cases, and failure in 3.8% of cases.
Two theses were identified that reviewed cases of patients treated with phages. The thesis by Domrault, published in 1998, reported 557 cases of treatment with phages in patients infected with resistant P. aeruginosa. Treatment was successful in the majority of cases [65]. In his 1931 thesis, the oldest to evaluate phage therapy, Pesce reviewed 622 cases of phage treatment and described his own use of phages in 14 patients with cutaneous infections. Here again, results were mainly positive [67].
The majority of reviews showed positive results. Seven reviews were not referenced in the MEDLINE database. Of these, two were found in articles (one in French and one in English), three in theses (all in French), and two in books (both in English). Thus, our literature review adds more than 1000 patients to the existing data.

Prospective Non-Randomized Clinical Trials
Three prospective non-randomized clinical trials were identified. The 2021 prospective study by Patel et al. evaluated the efficacy of phage therapy in 48 patients presenting with a wound that had failed to heal after 6 weeks of conventional therapy [32]. The patients received either a phage for single bacterial infection or a cocktail of phages specific to two or more infecting bacteria. Phage treatment was applied on the wound surface five to seven times over a period of 9 months, and patients were followed for 3 months. A cure rate of 81.2% was obtained.
In 2019, Gupta et al. conducted a prospective study in 20 Indian patients treated with phages for chronic nonhealing wounds infected with E. coli, S. aureus, and P. aeruginosa [41]. A cocktail of customized phages was applied over the wounds in three to five doses. A significant improvement in wound healing was observed, with seven patients cured at Day 21.
In their 1987 prospective study, Cislo et al. evaluated the effectiveness of phage therapy in 31 patients with chronic suppurative skin infections caused by Pseudomonas spp., Staphylococcus spp., Klebsiella spp., Proteus spp., and Escherichia spp. [66]. The outcome was favorable in 23 patients.

Isolated Case Reports
Our search of the literature identified 95 case reports published since 1945. These reports concerned more than 2500 patients, and the majority showed positive results (87/95). Phage therapy was associated with antibiotic therapy in 71 of the reports. A total of 12 case reports were not referenced in the MEDLINE database. Of these, eight were found in articles (four in French and four in English), and four in theses (all in French). These case reports add 200 patients to the existing data ( Table 2). The main indications and the success rate associated were osteoarticular (94%, 16/17), pulmonary (86%, 12/14), skin (91%, 10/11), and digestive (77%, 7/9) infections.
A total of 119 case reports published before 1945 were identified. The majority showed positive results (101/118). These reports concerned nearly 4000 patients. A total of 102 case reports were not referenced in the MEDLINE database. Of the identified case reports, 78 were found in articles (45 in French and 33 in English), and 23 in theses (all in French). More than 2600 patients are thus added to the existing data ( Table 3). The main indications and the success rate associated were bacteremia (79%, 15/19), skin (92%, 24/26), and digestive (86%, 12/14) infections.

Research Prospects
In 2017, Leitner et al. proposed a methodology for randomized controlled trials evaluating the use of phages in the treatment of urinary tract infections [282].
A 2020 literature review by Melo et al. reported the results of preclinical studies on phage treatments conducted in Western countries over the previous ten years [283]. Some of these results are encouraging and suggest the need to conduct randomized clinical trials to confirm the efficacy of phage treatments.

Conclusions
This literature review identified multiple cases of successful treatment with phages in patients infected with different types of microorganisms. A total of 120 publications dealing with approximately 4000 patients were added to the existing data. However, the significant publication bias and the non-standardization of methods for the assessment of phage therapy do not allow us to conclude on the efficacy of phages. Relevant pharmacological data, including treatment dosage and duration, should be collected to help standardize evaluation methods. As we have shown previously, the literature also seems to show good results from the combination of phages and antibiotics, by a synergistic effect [284]. Prospective randomized clinical trials could then be conducted to confirm or disprove the clinical efficacy of different phage treatments. Lastly, literature reviews should be performed in different countries to identify non-referenced and/or non-translated publications on phage therapy.