Phage Therapy in the Era of Multidrug Resistance in Bacteria: A Systematic Review

Bacteriophages offer an alternative for the treatment of multidrug-resistant bacterial diseases as their mechanism of action differs from that of antibiotics. However, their application in the clinical field is limited to specific cases of patients with few or no other alternative therapies. This systematic review assesses the effectiveness and safety of phage therapy against multidrug-resistant bacteria through the evaluation of studies published over the past decade. To that end, a bibliographic search was carried out in the PubMed, Science Direct, and Google Scholar databases. Of the 1500 studies found, 27 met the inclusion criteria, with a total of 165 treated patients. Treatment effectiveness, defined as the reduction in or elimination of the bacterial load, was 85%. Except for two patients who died from causes unrelated to phage therapy, no serious adverse events were reported. This shows that phage therapy could be an alternative treatment for patients with infections associated with multidrug-resistant bacteria. However, owing to the phage specificity required for the treatment of various bacterial strains, this therapy must be personalized in terms of bacteriophage type, route of administration, and dosage.


Introduction
Bacteriophages were discovered by Frederick Twort and Félix d'Hérelle in 1915 and 1917, respectively. Since then, it has been suggested that the administration of these viruses could be used to treat bacterial infections. Initial results with phage therapy (PT) were promising; however, its application was limited owing to the discovery of penicillin in 1928 as well as the large-scale production of antibiotics in 1940 [1][2][3][4]. In recent years, the use of PT has recommenced owing to the increased bacterial resistance to antibiotics, which can arise because of mutations or be acquired by the acquisition of resistance-conferring genes via horizontal gene transfer (HGT). HGT may occur via conjugation, transformation, and transduction processes as well as via mobile genetic elements, such as transposons, and insertion sequences [5,6].
Infections caused by multidrug-resistant bacteria have become one of the leading causes of morbidity and mortality worldwide, with approximately 700,000 deaths each year [7,8]. The World Health Organization (WHO) has listed critical priority pathogens that require new antibiotics or therapeutic alternatives; these pathogens include multidrugresistant bacteria that are especially dangerous in hospitals and nursing homes and for patients who need to be treated with ventilators and intravenous catheters [9]. According to the Organization for Economic Cooperation and Development (OECD), in Europe, North America, and Australia alone, 2.4 million people could die between now and 2050 if current bacterial resistance rates continue [10], which evidences that bacterial resistance poses a global public health problem [2][3][4][5][6]11]. Likewise, the discovery and development of new antibiotics have decreased because of the large amount of time and money required, resulting in increasingly difficult clinical management of infections and, in some cases, infections that are impossible to treat [9,11,12]. For these reasons, there is a need to look for alternatives for patients for whom conventional antibiotic therapy is ineffective. Thus, PT has become an alternative as its mechanism of action differs from that of antibiotics [13].
Lytic bacteriophages have had an impact on the clinical treatment of multidrugresistant bacteria because of their capacity to naturally control bacterial populations [14,15]. These viruses provide novel advantages, such as the safe treatment of infections, as they are harmless to eukaryotic cells, do not cause harmful side effects, and demonstrate high host specificity. This is because they only replicate in the presence of bacteria causing the infection, thus reducing damage to the natural microflora. In addition, genetic exchange between phages rarely happens [16][17][18]. Therefore, as the frequency of therapy increases, the treatment doses and periods needed to achieve an optimal effect are expected to decrease [16,17]. Phage therapy has been used for the treatment of infections related to burn injuries or soft tissue and skin trauma, osteomyelitis, sepsis, bacteremia, and otitis media as well as urinary tract, pulmonary, and prosthetic device-associated infections. It is often used after the generally prolonged stay of the patients, who have been treated with different antibiotics without being able to eradicate the infection successfully [19,20]. However, effectiveness-and safety-related data for this type of therapy are scattered across individual clinical case reports owing to the few clinical studies conducted. In this review, we summarize the effectiveness-and safety-related data from studies conducted on human patients infected with various types of bacteria who underwent PT as an adjuvant or alternative therapy to antibiotics.

Search Strategy
To carry out the bibliographic search, three databases were selected: ScienceDirect, PubMed, and Google Scholar. The keywords used were phage therapy and multidrug resistant. The search was defined with the use of Boolean operators as follows: "phage therapy" AND "multidrug resistant". The search was adjusted in each database according to the descriptors and using "phage therapy in the era of multidrug resistance in bacteria" as a reference. The search protocol was conducted by two independent reviewers.

Inclusion and Exclusion Criteria
The review included full-text articles and clinical case reports published in Spanish and English over the last ten years, that is, from 1 January 2011 to 25 April 2021. The main objectives of these studies were to describe the use of PT in patients infected with various bacteria, provided that, at the end of the study, it would be possible to tell whether there was a positive or negative response to the treatment.
The review also excluded bibliographic material that had not been reviewed by academic peers; articles written in a language other than Spanish or English; articles published outside the selected duration; bibliographic material, such as reviews, systematic reviews, posters, conferences, book sections, and perspectives; multidrug-resistance-based studies that did not use phage therapy; studies on metagenomics in bacteriophages without subsequent application in human patients; studies on PT in animals; and studies on the use of phages in foods.
The selection of articles was carried out considering the PRISMA statement (Preferred Reporting Items for Systematic Reviews and Meta-analyses) ( Figure 1).

Data Extraction
The citations selected, along with their title, were imported into EndNote, a Bibliographic Reference Management Software, and Microsoft Office Excel, which was used to check duplicated results between the three databases. Duplicated records were eliminated.

Data Analysis
Evaluation of the records was carried out by reading the titles and the abstracts. All those that did not meet the inclusion criteria were discarded. Then, the full texts were analyzed to rule out those articles that did not meet the inclusion criteria, thus leaving us with the final records that were included in this review. Subsequently, the information of each article was collected in two Excel tables, whereas the most relevant information was captured in graphs for later analysis.

Results
The bibliographic search yielded a total of 1500 articles distributed as follows: 203 in PubMed, 306 in ScienceDirect, and 991 in Google Scholar published between 2011 and 2021. After excluding duplicated records (336 articles) and 1121 articles that did not meet the inclusion criteria, 27 articles were included in this review ( Figure 1). Subsequently, information concerning the effectiveness and safety of PT was collected along with other findings that the authors considered relevant [21].
Of a total of 165 patients who underwent phage therapy, 141 (85%) showed a reduction in or complete elimination of the bacterial load accompanied by an improvement in the signs and symptoms, whereas PT was ineffective in the remaining 24 (15%). Only 21% (n = 35) of patients received PT combined with antibiotics, with a 100% success rate. The remaining 79% (n = 130) were treated exclusively with PT, with a success rate of 81%. Bacte-rial resistance to phages was reported in 6 of the 27 articles reviewed [22][23][24][25][26][27]; in four studies, the initial therapy was modified by changing or including new bacteriophages [22][23][24][25][26][27], whereas in the remaining studies, resistance was measured at the end of the study.
The age of the patients receiving PT ranged from 2 to 88 years, with an average of 54 years. Three of these studies were conducted on minors (12%) [19,28,29], whereas 88% included patients over the age of 18. Most cases yielded positive results in terms of effectiveness (improving the medical condition), and adverse side effects were minimal. In addition, phage administration was not associated with the death of any of the patients (Table 1).

Discussion
This systematic review displays effectiveness-and safety-related data as well as some relevant findings that have been reported in PT studies in humans from studies published in English over the last ten years. Based on the findings from a total of 165 patients treated in 27 selected studies, it can be concluded that PT produced encouraging results for the treatment of infections caused by various bacterial species, especially those that are difficult to manage, such as infections caused by bacteria resistant to multiple antibiotics [22,25,28,34,41]. In 85% of the cases, the therapy was successful through the administration of either a single phage or a phage cocktail. One of the primary factors responsible for this success is the use of specific bacteriophages for each bacterial strain. Although 100% of the studies reported having conducted in vitro studies of bacteriophage activity at the beginning of the therapy, in 15% of the cases, the infection had not been resolved at the end of the therapy. Some authors attributed these failures to defective modes of administration [38], a low phage concentration [23], or co-infection with other species of bacterial strains [46], rather than to a low effectiveness of the therapy.
One of the main concerns with PT is the development of phage-resistant strains [19,25]. An in vitro study by Oechslin et al. found that the frequency of spontaneous phage resistance mutations in a susceptible strain of P. aeruginosa was around 10 -7 . Additionally, the same authors concluded that, although some mutations confer resistance to phages, these mutations can decrease bacterial fitness when compared to normal growth under less strict in vitro conditions in animals [47]. This could explain the success of PT in some patients despite the presence of phage-resistant bacteria [23,26], which would prove that a patient's immune system plays an important role in the success of PT in the short, medium, and long terms.
Although PT is presented as an emerging strategy to fight multidrug-resistant bacterial infections, evidence suggests that coadministration of phages and antibiotics could be a more effective strategy for the elimination of infection [19]. In cases where combined therapy was administered, patients recovered from the infection with 100% success, whereas 88% of those patients who received PT alone recovered. This result can be explained as follows: combination therapy may decrease the selection of phage-resistant bacteria [47], in addition to increasing the sensitivity to antibiotics [25,47,48]. Additionally, antibiotics could act as adjuvants that minimize the appearance of other pathogens in immunocompromised patients.
Owing to serious public health threats posed by health care-associated infections (HAIs), such as those caused by E. coli, S. aureus, K. pneumoniae, coagulase-negative Staphylococcus, and Enterococcus faecalis [49], and given the call made by the WHO for a search for new alternatives in the treatment of multidrug-resistant bacteria, especially those belonging to the ESKAPE group [7], it is not surprising that infections caused by these pathogenic bacteria were more frequently treated with PT (50.8%). A single phage or a cocktail of phages proved effective in the treatment of these infections regardless of their sensitivity or resistance to antibiotics. However, phage cocktails were more widely used (84%) to increase bacterial sensitivity to therapy.
In all studies, the administration of PT was considered safe for patients. The age of the treated patients ranged between 2 and 88 years, with an average of 55 years. Duplessis et al. (2017) reported the case of a 2-year-old patient with DiGeorge syndrome and a series of comorbidities that were difficult to manage, in addition to recalcitrant bacteremia/sepsis owing to multidrug-resistant P. aeruginosa. Phage therapy was initiated in this patient to treat the infection. Although sterile blood cultures were obtained for P. aeruginosa in the short term, the treatment had to be interrupted owing to decompensation of the infant caused by anaphylaxis. A refractory infection led to the reinitiation of PT, which once again proved effective, resulting in sterile blood cultures 24 h after therapy onset. Unfortunately, the patient's decompensation and worsening prognosis owing to severe arrhythmias as well as cardiogenic and septic shocks (caused by the same bacteria), possibly owing to a history of influenza infection and end-stage heart failure, among others (although none of these effects were related to the administration of PT), led relatives to withdraw care, and the child died soon afterwards [36]. Jault et al. (2019) reported the death of an adult man over 70 years of age who was recruited to participate in the controlled clinical trial PhagoBurn, once the inclusion and exclusion criteria were met. The patient received PT topically for the treatment of P. aeruginosa, and his death occurred sometime after completing the treatment regimen, although no association with PT was found [23].
The potential development of adverse events associated with the concentration of endotoxin residues, i.e., the amount of bacterial endotoxin lipopolysaccharide released during phage-mediated lysis [19], has apparently been discarded based on reports that the maximum dose is below 5 units of endotoxin per kg of body weight per hour (required by international regulators), and this is reflected by the absence of serious adverse events during the administration of PT in all the studies. This confirms the safety of PT for the treatment of bacterial infections.
Additionally, bacteriophages offer new advantages, namely, their high specificity toward the host cell, which would reduce the damage to the normal microbiome of the patient and decrease colonization by other pathogens in the absence of in vivo drug interactions, bactericidal activity, minimal variability in pharmacokinetics and pharmacodynamics, agnostic bacterial targeting irrespective of the antibacterial susceptibility profile, minimal environmental footprints, and potential inducement of susceptible bacterial profiles [36]. Moreover, their ability to alter the formation of biofilms enhances the action of antibiotics in combination therapies [21]. In addition, free genetic exchange of phages rarely occurs [3,17,18]. Thus, as the frequency of therapy increases, the treatment doses and periods needed to achieve an optimal effect would be expected to decrease [17,18]. However, one of the main challenges that must be faced for the definitive implementation of phage therapy is the unification of criteria and standardized procedures that allow dealing with infections caused by multidrug-resistant bacteria. This is because the few studies that have been reported present a great variability of results because they not only depend on the multidrug-resistant bacteria, but also on the type and place of the infection that they are causing. This in turn affects the dosage and application of combination therapies, which makes comparison between studies difficult. All this highlights the need to generate guidelines with unified criteria to determine the validity of phage therapy.

Conclusions
Phage therapy has proven to be effective and safe for the treatment of infectious diseases caused by various bacterial species, including multidrug-resistant strains. This effectiveness has been demonstrated both through local and systemic administration, with no associated serious adverse events. However, much of the success of PT depends on the selection of specific bacteriophages for each bacterial species and strain, owing to their specificity toward the host cell, which is why extensive libraries of phages are required to personalize treatments. On the other hand, coadministration of antibiotics as adjuvants is recommended as it prevents infections by non-specific bacteria and helps eliminate bacteria already sensitized by the phages.