Reassessment of Historical Clinical Trials Supports the Effectiveness of Phage Therapy

SUMMARY

Phage therapy has become a hot topic in medical research due to the increasing prevalence of antibiotic-resistant bacteria strains. In the treatment of bacterial infections, bacteriophages have several advantages over antibiotics, including strain specificity, lack of serious side effects, and low development costs. However, scientists dismissed the clinical success of early clinical trials in the 1940s, slowing the adoption of this promising antibacterial application in Western countries. The current study used statistical methods commonly used in modern meta-analysis to reevaluate early 20th-century studies and compare them with clinical trials conducted in the last 20 years. Using a random effect model, the development of disease after treatment with or without phages was measured in odds ratios (OR) with 95% confidence intervals (CI). Based on the findings of 17 clinical trials conducted between 1921 and 1940, phage therapy was effective (OR = 0.21, 95% CI = 0.10 to 0.44, P value < 0.0001). The current study includes a topic review on modern clinical trials; four could be analyzed, indicating a noneffective therapy (OR = 2.84, 95% CI = 1.53 to 5.27, P value = 0.0009). The results suggest phage therapy was surprisingly less effective than standard treatments in resolving bacterial infections. However, the results were affected by the small sample set size. This work also contextualizes the development of phage therapy in the early 20th century and highlights the expansion of phage applications in the last few years. In conclusion, the current review shows phage therapy is no longer an underestimated tool in the treatment of bacterial infections.

INTRODUCTION

The spread of multidrug-resistant (MDR) bacteria has a tremendous impact on public health and is expected to increase more in the near future. In the European Union (EU) alone, more than 30,000 lives are lost yearly to MDR infections, resulting in over one billion euro in medical costs (1, 2). The most relevant MDR species are Enterococcus faecium, Staphylococcus aureus, Klebsiella pneumoniae, Acinetobacter baumannii, Pseudomonas aeruginosa, and Enterobacter spp, collectively known as the ESKAPE group (3). The genetic variability of bacteria naturally allows some strains to resist the action of antibiotics, a feature that became evident soon after the discovery of antibiotics (4). In 1945, just a few months after the commercialization of the world’s first antibiotic penicillin, the Scottish microbiologist Alexander Fleming (who discovered penicillin in 1929) advocated limiting antibiotic use to prevent the proliferation of antibiotic-resistant microorganisms (5). Despite this early warning, in 2009 alone, over three million kilograms of antibiotics were administered to humans in the United States alone (5). It is estimated that 10% to 20% of antibiotics are administered to hospitalized patients, making it not possible to know if the regimen was applied effectively (6). Low-income countries are particularly relevant in this context because they lack reliable statistics on MDR spread and antibiotic intake is far less regulated than in high-income countries (7, 8).

The common perception of the cause of MDR spread is the misapplication of antibiotics for clinical use. It is often assumed the fault is on the patients’ side: people do not complete the assigned antibiotic regimen, facilitating the selection and spread of MDR bacteria. However, because approximately one-third of illnesses are routinely treated with antibiotics in the absence of a definite identification of the infectious agent, the process is more complicated and rests heavily on clinicians’ choices (9). Despite these problems, between 2000 and 2010, the purchase of over-the-counter antibiotics increased by 123% and 157% in China and India, respectively (10).

Despite such a large use of antibiotics for human use, about two-thirds of all antibiotics are applied in the food industry (11). In agribusiness, antibiotics are used for therapeutic, prophylactic, and growth purposes in animal farming, animal husbandry, and aquaculture. This usage contributes to an increased risk of MDR emergence (12–16). The administration of antibiotics for growth promotion is especially concerning because it is done at subtherapeutic doses and relies on an unknown mechanism of action, which most likely involves an induced dysbiosis, thus increasing the number of nutrients adsorbed by the animals (11). Antibiotics in the food industry were used in excess of 130 million kilograms in 2013 in the United States alone. It is estimated this number will rise to over 200 million kilograms by 2030 (5, 10). Antibiotics and their metabolic derivatives are present in animal meat and bodily fluids (17), implying not only that MDR species can be selected in treated animals, but also that the disposal of animal-derived wastes and sewage in the environment will cause antibiotic accumulation in soil and water sources, fostering the selection of environmental antibiotic-resistant strains that then will recirculate into human communities via zoonoses (18).

The increasing difficulty to develop antibiotics makes MDR a crucial public health peril that requires novel tools for its resolution. Among the alternatives to antibiotics, the application of bacterial viruses (bacteriophages, or phages for short) has gained increasing interest, as highlighted by the sharp increase in publications on this subject observed in the last 20 years (Fig. 1). In stark contrast to the total publications, the number of clinical trials and randomized controlled trials, recorded since the 1990s, has remained relatively stable and on a small scale over the last 2 decades.

FIG 1.

Rising interest in phage therapy. Distribution of articles retrieved with the keyword “phage therapy” from MedLine during 1946 to 2021. The total number of publications steadily increased, reaching 686 articles per year in 2021 (red line), while clinical and randomized trials remained a small subset of the total, with a maximum of six publications per year in 2018 and 2020.

There are two main strategies for applying phages (19) (Fig. 2). One strategy (known as prêt-à-porter and akin to the large-spectrum antibiotics) uses a cocktail of several types of phages to ensure the presence of at least one species capable of targeting the pathogenic bacteria. The other strategy (sur-mesure) involves isolating phages from suitable sources based on the actual species present in a microenvironment (a wound, for instance). Phages can also be administered in combination with antibiotics (20). In the presence of phages, subtherapeutic amounts of antibiotics can still be lethal for bacteria, resulting in a faster spread of the phages combined with a larger burst size (21), a phenomenon known as “phage-antibiotics synergy” (22). Recent mathematical models show that phage-antibiotic combinations require the host’s immune response to be effective (23). Because phage therapy depends on the host’s immune status, a better understanding of the process is required to implement an antibacterial treatment properly.

FIG 2.

Phage therapy approaches. In the prêt-à-porter (“ready-to-use”) approach, environmental samples are collected, typically from sources such as ponds or sewage purification plants. Phages are purified from the mixture of microbes present in the samples and stored in biological banks. Several isolates are mixed in a single preparation that can be administered to the patients, hoping that at least one of the isolates will have the pathogenic bacterium as a host. In the sur-mesure approach, the starting material is a clinical sample, which is analyzed to characterize the pathogenic bacterium. Phages against the bacterium are selected either anew from environmental specimens or from banked isolates. The phagial preparation is administered to the patient, knowing that the phages will recognize the pathogenic bacterium. This targeted method has the disadvantage of being time-consuming, making it unsuitable for emergency medical care.

Following the substantial successes obtained by phage therapy during its development in the 1920s, this type of medication has become a standard procedure in the countries of the ex-Soviet block. However, even today, this procedure encounters severe resistance from the practitioners of Western countries even in the face of an antibiotic crisis. Several points need to be addressed to make this application truly competitive, including handling phage-resistant bacterial strains, extending the shelf-life of phage preparations, minimizing the release of endotoxins upon bacterial lysis, and maximizing the phage action on biofilms (24). The current review will give a historical overview of phage therapy and compare earlier findings with more recent usage in the West to see whether this novel antimicrobial approach has lived up to its promises.

HISTORICAL CONTEXT

Both virology and bacteriology are the offspring of “germ theory,” and their gestation period spanned centuries. Marcus Terentius Varro (116 to 27 BC), an ancient Roman author, indicated the presence of “tiny animals” (“animalia minutissima”) capable of causing diseases within substances he referred to as “contagium” (25). In 1674, Antoni van Leeuwenhoek visualized such beings using the first microscope (26). Beginning with Lazzaro Spallanzani in 1795 and continuing throughout the 19th century with the work of eminent scientists such as Robert Pasteur, Jacob Henle, and Robert Koch, it was established that bacteria cause diseases (27). Dimitri Ivanofsky extracted an infectious agent from a filtrate of tobacco plants infected with mosaic, a disease later named by Adolf Mayer in 1879 (28). Ivanofsky suspected the agent could be a toxin, but Martinus Beijerinck, a former Mayer’s collaborator, observed in 1898 that highly diluted filtrates of mosaic-suffering tobacco plants could regain their “strength” after being in contact with plant tissues. Beijerinck named such agent “contagium vivum fluidum” (“contagious living fluid”) or simply “virus” (29). The term “virus” is derived from Latin for “poison” and indicates a tangible fluid such as the snake’s venom or the saliva of a rabid dog (30). The “virus” was opposed to “miasma,” still a poison but invisible and often squirting from the earth, such as the hydrogen sulfide gases released by volcanic vents.

Löffler and Frosch isolated the first animal virus in 1897 (foot-and-mouth disease) (31), and Walter Reed isolated the first human virus in 1901 (the yellow fever virus) (32). In 1939, the first image of the etiological agent of the tobacco mosaic disease (the tobacco mosaic virus [TMV]) was published, clearly demonstrating the presence of “germs” smaller than bacteria (33).

Although bacteriology and virology are frequently considered separate disciplines, microbiologists soon discovered these fields of study often overlapped. Ernest Hankin Hanbury described the antibacterial activity of water collected from the Ganges and Yamuna rivers in 1896. This phenomenon was subsequently interpreted as proof of the presence of phages in the samples. However, this conclusion is currently considered incorrect because the bacteriolytic capability was retained even after boiling the specimens (34). Two years later, Nikolay Fyodorovich Gamaleya reported how a filtrate could lyse Bacillus antracis (35). Later, in 1915, Frederick William Twort, while working on the smallpox vaccine for the British Army, noticed that some bacterial suspensions lost their turbidity and could no longer generate colonies on solid media. He called such a phenomenon “glassy transformation” (36). Samples from these cleared suspensions, even when filtered and at very high dilutions, could in turn induce the clarification of fresh bacterial suspensions (27). Due to personal issues with his commanding officers, Twort’s position within the Army was not renewed and he was forced to abandon further work on his phenomenon, His publication remained essentially a case report without a proper methodological section, generically listing a “lytic principle,” among other possibilities, as an explanation of the glassy transformation (37).

Independently from these scientists, Felix Hubert d’Hérelle noted, in 1917, that patinas of Shiga bacillus he isolated on solid medium sometimes presented circles of growth inhibition that he named taches vierges (“blank spots”) or “plaques” (38). Unlike Twort, d’Hérelle pursued his discovery further and proposed the cause of the spots was a filterable viable agent that he named bacteriophage (“bacterium eater”), which today is shortened in phage (39). D’Hérelle demonstrated that phages were corpuscular entities rather than soluble molecules with an elegant experiment (40). He diluted phage preparations and then infected bacterial suspensions, noting that bacteriolysis occurred only a few times at the lowest dilutions. If phages were a solute, as in the case of enzymes, the results would have been homogeneous: dilutions above a threshold would have always lysed their host, and those below would have never achieved bacteriolysis. On the contrary, the fact that a low dilution was lytic in some instances and not in others meant phages were relatively large corpuscles that, by chance, would or would not end up into a larger dilution from a more concentrated stock. Such reasoning is not trivial. For instance, it is currently used to assess the presence of trace amounts of nucleic acids upon amplification by PCR assay, where Monte Carlo models are used to determine the concentration of the sample (41). D’Hérelle used Poisson models to describe such behavior and had no less than Albert Einstein to endorse his approach (42). Following these experiments, d’Hérelle started referring to phages as “suspensions of bacteriophages” rather than filtrates. In addition, unlike Twort, d’Hérelle published his experiments and theories in a series of scientific journals and books, demonstrating he understood the meaning of his discovery in detail (43).

D’Hérelle immediately realized the immense therapeutic potential of phages and strove to apply his discovery in the medical field, essentially creating what became known as phage therapy (44). After a small pilot study on animals, d’Hérelle cooperated with the French pediatrician Victor-Henri Hutinel to find dysentery patients to treat with the phages he had already isolated. Notably, before administrating the phages to sick children, d’Hérelle and Hutinel ingested a hefty dose of phage suspensions to ascertain the lack of side effects of the preparations. In 1919, three young brothers were admitted to the Hospital des Enfants-Malades in Paris under Hutinel’s care with acute symptoms of dysentery. They were promptly treated with the phage suspension and they recovered within 1 day, whereas their (untreated) sister died from the disease the day before hospitalization. In 1925, in collaboration with Alexander Yersin, d’Hérelle treated cases of bubonic plague and cholera, again obtaining remission within days. In particular, he treated a group of 16 cholera cases, who all recovered, and compared it with a placebo group of 33 people, 40% of which died. In a subsequent cholera study, the mortality rate in the phage-treated group (n = 76) was 8%, compared with 63% in the control group.

Given the encouraging results obtained by d’Hérelle, phages started to be employed by other physicians on a more massive scale (44). Lieutenant Colonel John Morison of the British Colonial Army collaborated with d’Hérelle between 1925 and 1935 to administer phagial preparations as a preventive measure to the populations in the Indian (now Bangladesh) region of Naogaon, a zone heavily affected by cholera outbreaks (45). During this period, there was no epidemic in Naogaon, whereas in the neighboring region of Habiganj, over 1,500 people died from this disease in 1933 alone. In Brazil, José da Costa Cruz from the Oswaldo Cruz Institute applied phages he had isolated to fight dysentery, reporting recovery within a couple of days, something hitherto unheard in the Brazilian medical community (46). In 1930, U.S. physicians Crutchfield and Stout treated 57 patients for staphylococcal skin infections and obtained an over 90% success rate (47). Moreover, American bacteriologist Thurman Rice applied phages to 300 patients suffering from suppurative illness, curing virtually all of them (48). In 1932, American physicians Ward McNeal and Frances Frisbee treated 15 patients with phage preparation to fight staphylococcal septicemia, curing seven patients (47%) (49). The early results of phage therapy were so impressive that they inspired a Pulitzer Prize-winning novel (Arrowsmith by Sinclair Lewis) published in 1925 (50).

Phage therapy’s early successes were immediately translated into commercial enterprises. D’Hérelle himself opened his own laboratory for the commercialization of phage preparations (Laboratoire du Bactériophage) in the late 1920s. The establishment was directed by his son-in-law, the pharmacist Théodore Mazure and the preparations (which took the names of Bacté-pyo-phage, Bacté-rhino-phage, Bacté-coli-phage, Bacté-intesti-phage, and Bacté-staphy-phage) were marketed by the firm Robert et Carrier, then bought by Synthélabo France, a subsidiary of L’Oréal. These preparations remained on the market well into the 1970s. D’Hérelle was not so much prone to the commercial exploitation of his discoveries as to ensure the quality of the phagial preparations available to the public (40). By the 1930s, the Western pharmaceutical market was flooded with phagial preparations. The leading players were Swan-Myers (a subsidiary of Abbott Laboratories), Eli Lily, and Squibb & Sons, but health authorities never approved these preparations (44). Most of these preparations were deceptively marketed to treat even viral infection, and they frequently contained inactivated phages (due to shelf-life or the use of preservatives like mercury that disrupted the virions and were harmful to the patients) or no phage at all, as well as bacterial debris (such as endotoxins) that inactivated the phages and caused side effects in the recipients (51). Even d’Hérelle had to sue one of his associated firms because one phagial preparation was commercialized without proper testing (40). Therefore, it is possible to observe a dichotomy in phage therapy between the astonishing results obtained at the academic level and the poor results observed by the general population.

Nonetheless, the watershed in phage therapy was the introduction of chemical antimicrobials. Sulfanilamide was commercialized by the German company Bayer in 1932 opening the path to chemo-treatment of bacterial infections. In 1939, German chemist Ernst Chain and his director Howard Florey, at the University of Oxford, mass-produced penicillin, which was used intensively by the allied forces during World War II and then marketed in 1944. Both sulfanilamide and penicillin were as fast-acting as phages but had the advantage to kill several types of bacteria at the same time—a feature that today is instead considered a significant drawback. This allowed physicians to use a single compound to treat different types of infections whereas before they had to use specific phagial preparations for each type of infection. Nevertheless, it was soon recognized that bacteria could develop resistance to both compounds and the combined application of penicillin and phages could drastically reduce the spread of the resistant strains (52, 53), but funding for further research became virtually unavailable.

PHAGE THERAPY IN THE SOVIET UNION

Antibiotics were known to the Soviets: Howard Florey was sent on Winston Churchill’s mandate to Moscow, together with American delegates, to train soviet microbiologists on the production of antibiotics (44). The difference between West and East was, in the former, the commercialization of antibiotics generated very high revenues for the pharmaceutical companies producing them. In contrast, in the latter, the production of antibiotics was under the control of the government and deemed less necessary than that of steel or weaponry. Thus, penicillin became available to the former Soviet community in 1949 and streptomycin (the second antibiotic produced), marketed in 1944 in the United States, was available only a decade later. Furthermore, the quality of the soviet antibiotics was much lower than their Western counterparts and more difficult to procure. Phages were less expensive than antibiotics; thus, they remained in physicians’ daily practice throughout the countries of the former USSR. For instance, Intestiphage, derived directly from one of d’Hérelle’s preparations, is commonly used in Georgia to treat traveler’s disease and nosocomial gastrointestinal infections (54).

The introduction of phage therapy in Eastern countries is also directly attributable to d’Hérelle, with the collaboration of Georgiy Eliava. During World War I, Eliava was enlisted as a bacteriologist on the war front and then went to the Institute Pasteur for three different time periods (1919 to 1921, 1925 to 1927, and again in 1931) for additional training. There he met d’Hérelle as he was becoming one of the heralds of phage therapy. In 1921, Eliava established the production of vaccines against cholera, smallpox, and typhus in the former USSR. In 1923, he founded and directed the Institute of Bacteriology in Tbilisi (renamed Eliava Institute of Bacteriophage Microbiology and Virology, EIBMV, after the fall of the USSR), becoming the head of the Departments of Hygiene and Microbiology at the Tbilisi State University. In 1934, he founded what is now known as the Russian National Center of Disease Control (44).

At the time, Stalin had set the goal of removing the gap the Soviets had with Western countries in terms of biological knowledge. Thus, Eliava had the availability of massive resources, and his plan of establishing a national phage laboratory was encountered with enthusiasm by d’Hérelle. The latter agreed to move definitely to Tbilisi, but this intention was shattered when Eliava became a victim of Stalin’s Great Purges. Ironically, Stalin’s plan to fill the gap between the Soviets and capitalist countries dug an abyss between them. Not only did Stalin’s terror refrain d’Hérelle from moving to the former USSR (thus, the Soviets lost the input from the central figure in phage therapy), but the Iron Curtain sealed the differences between the two blocks, which took completely different trajectories in the way bacterial infections were treated. Notably, the few years d’Hérelle spent in Georgia were sufficient to inspire yet another novel, albeit in more modern times: The French Cottage by David Shrayer-Petrov (55).

Clinical trials of phage therapy continued at a steady state in the former USSR. Among the largest of them, carried out during the war against Finland (the “Winter War” from 1939 to 1940), wounded Red Army soldiers suffering from gas gangrene were treated with phages, showing a mortality rate of 18.8% to 19.2% compared with 42.2% to 54.2% of those treated with standard measures. In a group of 6,025 soldiers treated with phage therapy, only 67 (1.1%) developed gas gangrene whereas, in the control group, the incidence was 2.3% to 6.8% (39). The treatment of typhoid fevers caused by Salmonella typhi and S. paratyphi was less successful: a trial carried out in 1938 showed no reduction in the mortality rate (12%), and a study carried out in 1950 reported that the intervention group (n = 52) had a rate of postinfection complications higher than the control group (n = 40), that is 37.7% against 27.5% (39). However, the study was biased by the fact that the former group had a higher rate of severe cases than the latter. The titer of the phage preparations or whether complete remission was achieved was not reported in other studies (39).

The treatment of dysentery was more effective than that of gangrene. In 1940, an intervention group of 100 soviet patients showed remission of symptoms within 48 h of treatment compared with a mere 2% showed by the control group (n = 50). At 6 days posttreatment, the remission rates were 100% and 46%, respectively. Phages were also administered as prophylaxis and successfully applied on several occasions (56). For instance, in an outbreak of dysentery in Georgia, the disease incidence in the control group was 4-fold higher than in the intervention group. In the 1950s, scientists from the former Soviet Union conducted anti-cholera research in Afghanistan and Dhaka, but the results were disappointing (56). In 1961, a cohort of 30,769 children aged 6 months to 7 years old was enrolled in Tbilisi for a dysenteric study, showing how the dysentery incidence decreased from 6.7 cases per 1,000 in the placebo group to 1.8 cases per 1,000 in the phage treatment group (57). In another study carried out by the Red Army in the 1980s, the incidence of dysentery was 6 to 8 times higher in the control group than in the intervention group (57). Furthermore, between 1974 and 2002, 2,516 participants were enrolled in phage therapy studies in the former USSR, with a success rate ranging between 70% and 92% (58). Dysentery incidence decreased from 6.7 cases per 1,000 in the placebo group to 1.8 cases per 1,000 in the phage treatment group. Thus, Eastern countries belonging to the former Soviet block performed large-scale phage therapy for several decades, although their results were—and mostly still are—virtually unavailable in the Western countries.

THE WANING OF PHAGE THERAPY IN THE WESTERN WORLD

In the West, regardless of early enthusiasm, phage therapy waned rapidly out of fashion due to the coincidence of multiple factors (19). A series of pitfalls in the clinical studies carried out in the first half of the 20th century have been highlighted (51):

• Poor understanding of phage biology (as exemplified by the JAMA reviews mentioned above).

• Use of poorly virulent phage strains.

• Use of phages that targeted a too narrow range of hosts.

• Failure in correctly identifying the infectious agent.

• Underestimation of the insurgence of phage-resistant bacterial strains.

• Underestimation of the phage-inactivating effects of gastric juice and immune response.

• Presence of impurities in the phagial preparations.

In this section, we list the most relevant reasons that explain why phage therapy faded away in Western countries.

Scientific Reasons

The most important reason for the waning of phage therapy in the West can be ascribed to scientific reasons that included study design and execution. It is often remarked that the early clinical studies employing phages did not include proper statistical testing. Back in the day, it was deemed immoral to withdraw a promising new treatment from a group of patients. Hence, intervention studies did not have a control group (in what is now known as a “double-blind control study”), implying that the obtained results on phage application had no statistical value (9). For instance, Robert Koch, no less than one of the founding fathers of germ theory, presented only the results from the intervention group for his alleged cure for tuberculosis (the tuberculin, which is now employed for diagnostic rather than therapeutic purposes) (46). The first treatment with antibiotics, dated 1941, included a single patient, a “bobby” in Oxford, United Kingdom, who developed a systemic staphylococcal infection from a small scratch on the mouth. This patient then succumbed when the scarce penicillin available had been consumed. There was no control group in the more extensive antibiotic applications. For example, a large fire in a Boston nightclub in 1942, which resulted in 571 deaths, was the first battleground for antibiotic treatment, and World War II pushed for the administration of antibiotics to thousands of soldiers without preliminary statistical analysis (46). Indeed, the introduction of the first double-blind clinical trial is credited to Sir Austin Bradford Hill in 1946 for the application of streptomycin to tuberculosis (59, 60).

Furthermore, in other instances, phages were applied after bacteria had been washed away from a wound or treated with antibacterial drugs, introducing an enormous bias to the research. The stability of the phagial particles played a significant drawback for phage therapy. It is well known that low pH disrupts virions, and a recent review of the literature highlighted how even orange juice has the capability of reducing phagial infectivity (61). On the other hand, it is also known that phages can cross the intestinal tract maintaining their infectivity, and given the changes in pH along the gastric tract, alteration of the virion might only be transient (62). Today, neutralizers of gastric juice acidity (63, 64) and protective microshells (65) are used routinely in phage therapy to increase the stability of phages in oral administration, although it is not clear whether the efficacy of the phagial preparation is truly improved. Nevertheless, experiments on mice demonstrated phages could cross several barriers and reach the site of infection multiplying on that spot (66). Such a targeted delivery is a major advantage over antibiotics, given that the latter are absorbed by the whole body, a feature that dilutes their dosage and might generate systemic side effects. The release of endotoxins upon lysis of bacteria has also been highlighted often as a concern associated with phage therapy (67). These endotoxins, such as lipopolysaccharides, might cause inflammation and anaphylaxis in the recipients. However, recent studies indicate phages release fewer endotoxins than β-lactams antibiotics (68).

Another scientific reason that might explain the failure of early phage treatments might be the poor understanding of the phage interaction with its host. Robert Payne and Vincent Jansen at the University of Oxford, United Kingdom, highlighted how phages, unlike antibiotics, replicate inside bacteria; thus, needing to be contextualized in the growth dynamics of their prey. Hence, the pharmacokinetic of phages is fundamentally different from that of antibiotics and other drugs (69, 70). These authors suggested failure in understanding this point might explain the phage therapy’s failures.

Payne and Jansen indicated phage therapy can be subdivided into two main strategies (Fig. 3) (71). The first approach is passive therapy, where phages are applied with essentially the same pharmacokinetics as antibiotics. In this context, the antibacterial activity does not rely on the phagial replication but on the massive lysis of the hosts. The ratio of phage particles to host cells (also known as multiplicity of infection [MOI]) is so high that all bacteria are expected to become infected and lyse soon after their encounter with phages. Despite such a massive attack, even passive therapy needs to account for some parameters to be effective. First, phages will be removed from the environment (for instance, a skin wound or the gut) at a rate ω due to natural virion degradation or wash-out of biological fluids. The immune response can also reduce the total number of phages; however, given the expected short time of therapy, this factor can be ignored (at least at the first application of the treatment). Second, there must be enough phages to lyse all bacteria present in the selected environment; otherwise, the surviving bacteria will grow again after phages had been removed from the environment according to the decay rate ω. Thus, the minimum amount of phages to be administered is defined by the “clearance threshold” (V_C):

$$V_c=\frac{\mu }{\phi }+X_0{\text{e}}^{\mu t_{\varphi }}+\frac{\omega }{\phi }\text{In}\left(X_0{\text{e}}^{\mu t_{\varphi }}\right)$$

where μ is the replication rate of the host, φ is the transmission coefficient of the phage, ω is the loss rate of the free phages, X₀ is the initial concentration of the host, and t_ϕ is the time of inoculation. The quantity μ/φ is also known as the inundation threshold (V_I). If the amount of phage inoculum (v_ϕ) is above V_I, passive therapy is established, but if V_ϕ is lower than V_C, then the therapy is likely to be completely unsuccessful.

FIG 3.

Outcomes of phage therapy. In passive therapy (left panel) there is no phagial replication and the pathogenic host is expected to be eradicated in a single event due a high ratio of phage particles to host cells of the intervention. Passive therapy is established when administering phages above a critical level V_C. A dose of phages at a concentration v_φ, given at time t_φ, will result in the decay of a certain proportion of viruses at a rate λ whereas the infectious particles will infect the bacteria at a rate δ. On the other hand, bacteria multiply at a rate μ and phages will establish a sustained infectious cycle only if the bacteria reach a minimal density at time T_P. The therapy’s success is essentially dependent on administering enough virions to infect 100% of the bacteria, because if even one bacterium escapes infection, it will multiply again. There is a critical time T_C beyond which the phages will not be able to infect the bacteria at a sufficient rate to avoid such an escape. If v_φ is below the inundation threshold V_I, however, the therapy will fail regardless of the administration time because there will be not enough virions to infect all bacteria. In active therapy (right panel) there is an increase in the number of phages, which expands at a rate β after infecting a bacterium. Active therapy is established when v_φ is below V_C, providing an environment with a low ratio of phage particles to host cells, but needs to account for the decay of virions and the replication of the bacteria. Thus, if v_φ is below the critical threshold V_F, not all bacteria will be infected. In addition, the lower t_φ is from T_P, the more phages will be lost to decay, and the fewer phages present at the start of the sustainable infectious cycle, the less likely it is that phages will catch up with bacterial expansion. As in passive therapy, if t_φ is beyond T_F the bacterial infection will not be cleared. The success of active therapy is also dependent on the fact that it occurs prior to the establishment of the immune response at time T_H, which can target both bacteria and phages.

The second therapeutic approach, referred to as active phage therapy, considers phages, unlike antibiotics and other drugs, replicate in relation to their prey. In this context, a relatively small amount of phages can be administered based on the fact that viruses will later expand in the presence of bacteria. However, in this case, the replication of the bacteria must be considered as in all other infective systems. Specifically, there must be a minimum density of bacteria, known as “proliferation threshold” (X_P), to ignite a self-sustained infection cycle. Thus, just like the measles virus requires a minimum density of human hosts to sustain its life cycle (72), phages need a minimal bacteria population to propagate. X_P can be calculated from some basic parameters (life-history traits) describing both host and prey:

$$X_p=\frac{\omega \left(\eta -\mu \right)}{\phi \left(\eta \left(\beta -1\right)+\mu \right)}$$

where η is the lysis rate of the infected hosts (the reciprocal of the latency time τ). X_P is reached at a time known as “proliferation onset time” (T_P), which depends on the characteristics of both the bacterium and the phage:

$$T_P=\frac{1}{\mu }\text{In}\left(\frac{\omega \left(\eta -\mu \right)}{\phi \eta \beta X_0}\right)$$

To establish an active phage therapy, the phage inoculum must be administered after the proliferation onset time (T_P). In other words, if the administration time (t_ϕ) is greater than the proliferation onset time, then the phage will establish secondary infections that will propagate the virus. The size of the inoculum must also consider that, if the bacterial population is too large, the infective cycle might not remove all the bacteria. To account for this restriction, the critical inoculation size (V_F) has been described:

$$V_F=\epsilon \text{exp}\left[\omega \left(T_P-t_{\varphi }\right)+\frac{\omega }{\mu }{\text{e}}^{-\mu \left(T_P-t_{\varphi }\right)}-1\right]$$

where ε is the dilution factor to obtain one phage in the system. The critical inoculation size (V_F) ensures that, even if the administration time is lower than the proliferation onset time, there will be enough phages left in the environment when bacteria will eventually reach the proliferation threshold (X_P). In addition, the critical inoculation size opens the possibility of using phages as a preventive tool: if the rate of removal of phages is low enough, phages could be applied before the onset of infection.

The experimental determination of the proliferation threshold has been challenging (73). A level of 10⁴ bacterial CFU per milliliter (CFU/mL) has been obtained for Escherichia coli RR1 in the presence of phage T4 in vitro but not in vivo (74, 75). The difficulty in assessing the proliferation threshold X_P could be attributed to several factors (76). Because the decay rate (ω) is much higher in vivo than in vitro, in the latter case, the proliferation threshold may be below the experimental setting’s detection limit. Moreover, in the presence of a too large viral inoculum, the establishment of passive therapy will mask the true proliferation threshold value. In contrast, if the initial bacterial density is too low, the time to reach the proliferation threshold may exceed the experimental time frame. In addition, phage replication requires actively multiplying bacteria, whereas E. coli in the intestinal lumen is under starving conditions (77), making the assessment of the phage therapy in vivo difficult to predict. Finally, the critical inoculation size indicated by Payne and Jansen’s model may represent the upper end of the phage dose range because it does not account for the immune system’s role. Recent research shows phage therapy is ineffective in the absence of the immune system (78). Thus, phages need only reduce the number of pathogens to a level below which the immune system can eradicate the infection. As a result, phage therapy must account not only for bacterial host replication and viral particle loss, but also for the immune system’s role. The latter parameter was explicitly excluded from Payne and Jansen’s model two decades ago but is now included in phage therapy models (79–82).

In comparison with antibiotic treatment, which can be only passive, phage therapy has the advantage of ensuring a self-sustained process that naturally continues until the infection has been cleared. Therefore, active phage therapy has the advantage over antibiotic treatment of working with small amounts of bioactive particles, with obvious advantages in terms of production and distribution over conventional anti-bacterial products. Like antibiotic treatment, phage therapy can be nullified by bacterial resistance; however, because phages can clear an infection very quickly (according to the studies provided by d’Hérelle, even within a couple of days from the first administration), the probability of establishing resistance is lower than with antibiotics, which instead requires multiple rounds of applications (usually over 2 weeks). Moreover, using the sur-mesur approach, a specific strain of phages will be applied only once; thus, any resistance against such a strain will be irrelevant. However, it has been pointed out that the biological properties of phages make them more unpredictable in their results than antibiotics (74), a statement that can be contextualized within the framework of Payne and Jansen and whose corollary is that more studies are required to better understand the dynamics of phages as antimicrobial agents.

Socio-Scientific Reasons

Some authors have highlighted how the acceptance of phage therapy relies not only on objective factors but also on a blend of historical and personal reasons enclosed in the term “socio-scientific” (83). The acceptance of phage therapy by the medical community played an important role in the diffusion of this practice and the Western scientific community was deterred from phage therapy due to two reviews published by the Journal of the American Medical Association (JAMA).

The first, published in 1934, stressed the disagreement in the scientific community regarding the nature of phages and the role of biological fluids on the action of phages (84). Specifically, the authors contrasted d’Hérelle’s model of viral entities against other theories, including enzymes (heralded by Albert Krüger, University of Berkeley), a hereditary genetic element inducing autolysis (Jules Bordet, Institute Pasteur), and shrinking of the bacterial cell to dimensions that permitted filtration, akin of spores (Philip Hadley, University of Michigan). The second review, published in 1941 and written by Krüger, highlighted the lack of statistical significance in phage therapy research (85). It is of note that the authors still embraced the enzymes theory although the first microphotography of phages had been published in 1940 (86), clearly demonstrating the structural complexity of these microbial agents. Specifically, they correctly theorized that phages are assembled from proteins. However, while they remarked “the nature of bacteriophage is no longer in question,” they indicated its lytic action as a self-inducing enzyme (“inactive phage precursor + phage → phage”) without mentioning the viral essence of these entities (85), a mechanism akin to prions (87). Furthermore, the authors considered the studies independently and, because these were mainly small intervention studies with a median of 25 individuals (with the exception of a prophylactic study carried out by Morison in Naogaon, which accounted for 530,000 people), they did not provide sufficient statistical evidence for the effectiveness of phage therapy.

The JAMA review (85) negatively influenced a whole generation of Western practitioners and scientists. As a result, phage therapy was quickly phased out of medical practice in the West and phages remained within the scientific community only because they became the preferred research tool of the so-called “Phage Group,” led by German physicist Max Ludwig Henning Delbrück, Italo-American microbiologist Salvador Edward Luria, and American chemist Alfred Day Hershey. This group, active between the 1940s and 1960s at the Cold Spring Harbor Laboratory in New York, helped establish the field of molecular biology. Notably, among the first experiments that Delbrück carried out with phages was the confirmation of the corpuscular nature of phage using Poisson distribution, the statistical confirmation that each plaque is derived from a single phage, and the visualization by electron microscopy of phages; thus, sealing the diatribe regarding the nature of phages (40). On the other hand, d’Hérelle indirectly contributed to the field of molecular biology by isolating phage φX174, which was later heavily used by the Phage Group and whose genome was the first to be completely sequenced in 1977. D’Hérelle also considered phages as the basic block of life (protobe), a concept that inspired the work of the British biologist John Burdon Sanderson Haldane (one of the founders of the school of neo-Darwinism) and the Soviet biochemist Alexander Ivanovich Oparin (who investigated the origin of life) (40).

Furthermore, at the time, like many other scientists including Pastorians, d’Hérelle was a follower of Jean-Baptiste Lamarck. Even during his appointment at Yale University, where he was supposed to study phagial genetics, he maintained that phages adapted to their environment after being exposed to certain selective conditions rather than carrying the selective genes before exposure (40). In particular, d’Hérelle believed there was only one species of phage capable of adapting in millions of different ways (88). Thus, d’Hérelle’s view did not match the Darwinian mindset of the large majority of Western scientists but did fit with the framework of soviet researchers. In particular, d’Hérelle inspired the geneticist Trofim Lysenko, who rose to the top levels of the Soviet Academy of Science and founded the neo-Lamarckian school of “Lysenkoism,” which became a state-sanctioned doctrine (89). Conversely, in the West, Max Delbrück and Salvador Luria demonstrated the Darwinian process underlying the selection of phagial variants upon shifting the culturing conditions (90). The divergent trajectories of phage biology between the two geopolitical blocks could not be more explicit.

Finally, there is also a political aspect to this debate (91). In the 1940s, phage therapy was portrayed in the United States as a German invention that was a product of America’s adversary during the World Wars. Later, phage therapy was presented as Stalin’s cure as an affair from America’s worst enemy during the Cold War. Therefore, it is no surprise that the Western community shunned phage therapy.

Reanalysis of Historical Studies

Today’s meta-analyses employ a statistical approach known as a “random effects model” that allows the combination of multiple studies to boost the statistical significance of the results (92). The present review sought to reevaluate the historical studies reported in the work of Krüger and Scribner (85) using the random effects model. The studies were screened according to the Population, Intervention, Comparison, Outcomes (PICO) framework (93):

• Population: individuals with bacterial infection;

• Intervention: administration of phagial preparations;

• Comparison: standard medical procedure;

• Outcomes: death or failure in clearing the infection.

The details of the studies included in the study by Krüger and Scribner were submitted to the Deputy Head of the Library of the University of Hohenheim for acquisition. When the cited articles could not be retrieved, the values reported in the work by Krüger and Scribner (85) or other works cited therein were used. For instance, the works of d’Hérelle in 1928 and Pasricha in 1937 were reported by a review written in 1938 (89) and mentioned by Krüger and Scribner. Accepted languages for the studies were English, German, and French. Case reports and clinical trials involving drugs other than antibiotics were not considered. The retrieved articles were eligible if the article reported the total number of patients treated by both phage therapy and the comparative treatment (either antibiotic treatment or other standard procedures implemented at the times in the hospital settings). In addition, the studies were suitable for analysis when reporting the number of patients who died or did not remit from the infection during the course of the study. None of the urinary tract infections (UTI) studies included a control group and were, therefore, excluded from the analysis. Only 17 of the 79 studies (21.5%) cited by Krüger and Scribner in their review resulted eligible for meta-analysis (Table 1). The infection treated in the selected studies could be subdivided into dysentery (n = 13), which also included cases of typhoid fever and cholera; pyoderma, which included furunculosis (n = 2); and septicemia, which included appendicitis and osteomyelitis (n = 2).

TABLE 1.

Studies reported by Krüger and Scribner in the influential 1941 review regarding the effectiveness of phage therapy

First author	Year	Diseasee	Intervention group		Control group		Source
			Patients	Not effective casesa	Patients	Not effective casesa
Gay	1916	Dysentery	53	17	0	0	85
Fairley	1921	Dysentery	21	0	27	4	85
Fairley	1923	Dysentery	55	5	59	5	103
d’Hérelle	1928	Dysentery	74	6	124	78	169
Taylor	1930	Dysentery	14	2	6	0	170
Morison	1934	Dysentery	1,032	233	2411	1365	85
Morison	1934	Dysentery	191	59	307	437	171
Melmikb	1935	Dysentery	0	(1.4%)	0	(6.4%)	85
Morison	1935	Dysentery	563,000	0	563 000	2207	85
Raynald	1935	Dysentery	323	74	579	295	172
Mikeladzé	1936	Dysentery	21	1	63	10	85
Peragallo	1936	Dysentery	7	2	4	4	85
Pasricha	1936	Dysentery	398	33	369	76	85
Doorembosd	1937	Dysentery	790	122	632	405	173
Pasricha	1937	Dysentery	684	92	685	114	169
Comptonc	1938	Dysentery	0	0	0	0	174
Haler	1938	Dysentery	11	0	0	0	175
Murray	1938	Dysentery	146	0	0	0	176
Omarc	1938	Dysentery	0	0	0	0	169
Seidlmayer	1939	Dysentery	31	0	40	0	85
Bruynoghec,d	1921	Pyoderma	0	0	0	0	177
Gratiac,d	1922	Pyoderma	0	0	0	0	178
Gougerotd	1924	Pyoderma	4	2	0	0	179
Larkum	1929	Pyoderma	208	6	0	0	180
Rice	1930	Pyoderma	66	11	0	0	85
Rice	1930	Pyoderma	44	4	0	0	85
Rixford	1931	Pyoderma	7	2	0	0	181
Cipollaro	1932	Pyoderma	62	14	0	0	182
Schless	1932	Pyoderma	1	0	0	0	183
Schultz	1932	Pyoderma	17	9	0	0	184
Boyce	1933	Pyoderma	200	4	0	0	185
Ruddell	1933	Pyoderma	28	3	0	0	186
Schreuder	1933	Pyoderma	29	10	31	5	85
Stout	1933	Pyoderma	24	0	0	0	187
Lambert	1935	Pyoderma	1,000	100	0	0	188
Dechaumec,d	1936	Pyoderma	0	0	0	0	189
Frank	1936	Pyoderma	1	0	0	0	85
Freund	1936	Pyoderma	50	1	0	0	190
Halphenc,d	1936	Pyoderma	0	0	0	0	191
Hosen	1936	Pyoderma	240	48	0	0	192
Mariond	1936	Pyoderma	352	3	0	0	193
Morrison	1936	Pyoderma	1	0	0	0	85
Ruskinc	1936	Pyoderma	0	0	0	0	194
Tsouloukidzé	1936	Pyoderma	20	7	27	24	85
King	1937	Pyoderma	64	6	87	14	195
Montantd	1937	Pyoderma	14	0	0	0	196
Raigad	1937	Pyoderma	2,759	36	0	0	197
Raigad	1937	Pyoderma	31	0	20	3	198
Steinmann	1937	Pyoderma	2	2	0	0	85
Albee	1938	Pyoderma	100	22	0	0	199
Eisfelder	1938	Pyoderma	378	365	0	0	85
Moore	1938	Pyoderma	20	7	0	0	200
MacNeal	1932	Septicemia	15	8	0	0	49
Dutton	1933	Septicemia	12	1	0	0	201
Bréhant	1936	Septicemia	1	0	0	0	202
Sauvé	1936	Septicemia	0	0	0	0	203
Gilbert	1938	Septicemia	5	1	0	0	204
MacNeal	1939	Septicemia	100	75	0	0	85
Longacre	1940	Septicemia	36	17	54	44	85
Frischd	1925	UTI	7	1	0	0	205
Munterc,d	1925	UTI	0	0	0	0	206
Cowie	1926	UTI	11	5	0	0	85
Dalsace	1926	UTI	17	2	0	0	207
Larkum	1926	UTI	0	0	0	0	208
Ravinac,d	1926	UTI	0	0	0	0	209
Zdansky	1926	UTI	20	6	0	0	85
Pelouze	1927	UTI	40	7	0	0	210
Balozetc,d	1928	UTI	0	0	0	0	211
Caldwell	1928	UTI	0	0	0	0	212
Schmidtc,d	1929	UTI	0	0	0	0	213
Voss	1929	UTI	0	0	0	0	85
Christiansenc,d	1930	UTI	0	0	0	0	214
Krüger	1930	UTI	35	4	0	0	215
Cline	1931	UTI	14	2	0	0	216
Schultz	1932	UTI	151	79	0	0	217
Hinman	1933	UTI	0	0	0	0	85
Wehrbein	1935	UTI	10	3	0	0	218
Michon	1936	UTI	0	0	0	0	85
Burnet	1937	UTI	88	31	0	0	85

^aNumber of patients who died from the infection or did not show signs of recovery.

^bOnly the percentages were reported (in brackets), but not the actual number of patients.

^cNo original clinical cases identified in the text.

^dStudy not available in English-language.

^eUrinary tract infection.

The meta-analysis employed the random effects method implemented by Mantel-Haenszel (94), which computed odds ratio (OR) as a summary estimate of effect regarding the administration of phages over the use of standard methods. The OR were associated with a 95% confidence interval as a measure of dispersion. To establish the consistency between studies, the Higgins index I² was implemented as a measure of heterogeneity (95). The I² index was preferred over the Cochrane’s Q test following indications it might be more robust on small sample sets (96). Nonetheless, the Q test was included in the graphical report of the data to provide further information regarding the heterogeneity of the selected publications. The I² index was classified as low (<30%), moderate (30–60), substantial (61–75), and considerable (>75) (97). To further assess the publication bias in the analysis, funnel plots were drawn indicating the variation of the standard error over OR (98) and Egger’s regression test was implemented to assess the level of publication bias (99, 100).

The present re-analysis showed the application of phages in comparison with standard practices indeed reduces the risk of treatment failure because the OR obtained by the combination of all the studies was 0.2117 (95% CI = 0.1020 to 0.4393) with a significant effect (P value < 0.0001), considerable heterogeneity (I² = 88.4%, 95% CI = 83.1% to 92.1%), and a nonsignificant Egger’s test (P value = 0.9808) (Fig. 4). Because all the studies were therapeutic, the prophylactic study by Morison (1935) was additionally removed for consistency. The association became: OR = 0.2696 (95% CI = 0.1626 to 0.4469), P value < 0.0001, I² = 86.9 (95% CI = 80.3% to 91.3%), Egger’s test P value = 0.5948.

FIG 4.

Meta-analysis of the studies reported by Krüger and Scribner in their review published in 1941. (A) Random forest plot of the studies reporting the odds ratio (OR) of the intervention (experimental) over the control groups. An evident outlier is the Morison’s prophylactic study carried out between 1930 and 1935 in India. Removing this study increases the strength of the study, providing a random model with OR = 0.27 (95% CI = 0.16 to 0.45, P value < 0.0001) and heterogeneity characterized by I² = 86.9%. (B) Funnel plot of the studies shown in panel A, showing the trend line of the model.

The two eligible studies related to septicemia returned: OR = 0.3253 (0.1251 to 0.8457), P value = 0.0212. It was not possible to calculate the index of heterogeneity nor the Egger’s test. The 11 eligible studies related to dysentery (excluding the prophylactic study by Morison in 1935) returned: OR = 0.2358 (95% CI = 0.1446 to 0.3845), P value < 0.0001. The index of heterogeneity was considerable (I² = 88.2%, 95% CI = 83.3% to 92.6%), with nonsignificant Egger’s test (P value = 0.8456). The two eligible studies related to pyoderma returned: OR = 0.4412 (95% CI = 0.0117 to 16.6492), P value = 0.6587. It was not possible to calculate the index of heterogeneity nor the Egger’s test.

Overall, there was a high level of publication heterogeneity—that is, study variability not attributable to sampling errors (101). Such results support the use of the random effects method employed in the present meta-analysis, which is less accurate (wider confidence interval) than the fixed effect model but better accounts for small study effects (102). Nonetheless, according to the Egger’s test, the results obtained by analyzing the entire data set were not influenced by publication issues. This trend was also evident when the core of the data set, based on dysentery treatment, was examined. Conversely, because of the small sample size in the studies on septicemia and pyoderma, it was not possible to assess the level of publication bias for these groups.

Furthermore, it can be said the English language studies were less enthusiastic about phage therapy than the French studies. It is incorrect to assume these older studies lacked control groups. For example, a 1923 study clearly compared intervention (phage-treated) and control (standard treatment) groups (103). Instead, it was the statistical analysis of the data that was lacking. Such lacking can be explained by the statistical tests being still in their infancy in those years. For example, the British statistician William Gosset introduced the t test in 1908, but it remained a niche discovery until his fellow countryman Sir Roland Fisher popularized it, along with the concept of statistical significance, in the mid-1920s (104). Indeed, the need for solid statistical analysis in phage research was a well-understood requirement even in the early 20th century. In 1934, for instance, the Indian Research Fund Association that founded the work of Igor Ashoshev, who took over d’Hérelle and Morison’s work in Mumbai, remarked the “need for the presence of a statistician” in the Ashoshev’s group to truly sustain the claims on phage therapy’s effectiveness (40).

Personal Reasons

At the turn of the 20th century, phage therapy was synonymous with Felix d’Hérelle, making the fate of the former linked to that of the latter and vice-versa. At the time, the major problem d’Hérelle faced was that, despite his impressive achievements, he was not an established academic. In addition, he engaged in a dispute with the Institute Pasteur’s highest echelons that lasted decades and poisoned the establishment of the theory of phages in the West (40).

D’Hérelle was born in 1873 in Montreal, Canada and, like Pasteur himself, started his career in microbiology studying the biochemistry of fermentation, specifically exploring, with success, the conversion of excess sugars derived from the cultivation of maple and bananas into alcohol to produced spirited beverages. He then characterized diseases of coffee plants, isolating the pathogenic fungus Phtora vastatrix and, subsequently, approached the use of the bacterium Coccobacillus acridiorum to kill infesting locusts of the species Schistocerca pallens, a primeval example of biological pest controls.

In 1911, he started working at the Institute Pasteur in Paris under the patronage of the director, Émile Roux. His first position was that of laboratory assistant to Alessandro Tavelli Salimberi, but later he became director of the Laboratory of Bacteriology and was involved in the production of vaccines for French troops during World War I. Albert Calmette became acting director of the Institute Pasteur after Roux became ill. D’Herelle and Calmette went along fine at first, but, around 1920, the former raised safety concerns about the possibility of reversion in virulence of the attenuated tuberculosis strain that the latter had isolated with the contribution of Camille Guérin. Calmette did not take the criticism well and became, to use d’Herelle’s words, his “sworn enemy.” Because Calmette was directing the Institute, it is easy to imagine how d’Herelle’s life was made professionally miserable (83). For instance, after returning from Indochina, where he worked under the tutelage of Alexander Yersin in treating cases of plagues and cholera with phages, d’Herelle found himself without laboratory space and could only continue his research because his friend Edouard Pozerski gave him “the use of a stool in a corner of a table in his minuscule lab.” Notably, Pozerski also gave space to Eliava, indirectly helping to establish the friendship between the two scientists.

If having a powerful adversary was not enough, d’Herelle managed to make another with Belgian immunologist Jules Jean-Baptiste Vincent Bordet, another prominent Pastorian. Bordet characterized bacteriolysis and serum complement as well as founded the field of serodiagnosis, receiving the Nobel prize in 1919. It is unclear how the two began their feud. However, d’Herelle likely criticized Bordet’s work on immunology—whose applications he directly experienced in the development of vaccines, most of which were probably ineffective due to production problems—summarizing it as the “history of an error” (35). Offended, Bordet began to work with phages with the sole purpose of disproving d’Hérelle’s theories, putting forward an enzymatic process that was simply an extension of his previous work. Incidentally, though, Bordet expanded the field of phage biology: for instance, he was the first to describe lysogeny—although he believed it was the result of normal bacterial biochemistry. Bordet went even further than Calmette, establishing a cabal of scientists (the “Belgian group”) who systematically attacked d’Hérelle (83, 91). Among the adherents of this group were American biochemist John Howard Northrop, who received a Nobel Prize for his work on digestive enzymes, Belgian microbiologist André Gratia, and Romanian bacteriologist Mihai Ciucă. The Belgian group’s attack on d’Hérelle has been described as “cruel and vicious,” with Twort elevated as the true discoverer of phages, despite Twort’s misinterpretation of the inner principles of the “glassy transformation.” Remarkably, Twort and d’Hérelle met in 1922 at a conference organized by the British Medical Association but the former added nothing to the biology of phages (40), confirming that he only recorded a phenomenon he had no explanation for. To further understand d’Hérelle’s isolation back in the day, not only did all the textbooks at the time side with the Belgian group’s interpretation of phage biology, but d’Hérelle had to sue the editor of the Annals of the Institute Pasteur to have his work published (91).

Despite his achievements, which truly compare him to Pasteur himself, d’Hérelle never received a Nobel prize, albeit he received other awards. It is plausible that some Pastorians might have put him in a bad light with the Nobel committee but it is also possible that his position as an outsider in the academic environment might have excluded him from this prestigious recognition (40). D’Hérelle was an autodidact with no academic mentor and who never actually obtained a formal university degree, although he received honorary doctorates and was appointed professor in the United State and the former USSR. All in all, notwithstanding his contribution to microbiology and medicine, he was always deemed as a second-rate scientist by the Western academic establishment, albeit he was highly regarded in the former USSR.

For the sake of completion, this paragraph will summarize the final years of d’Hérelle’s life. At the onset of World War II, he remained in France even though his Canadian nationality made him subject to immediate arrest by the German Wehrmacht. After the occupation of Paris, he moved to Vichy, which, ironically, later became the capital of the Nazi-backed Pétain regime. Still, he remained there and continued his research under unofficial home detention, thanks to the fact that the authorities recognized his scientific value and did not persecute him. During this time, he also wrote his autobiography (Wanderings of a Microbiologist). He died in Paris in 1949 of pancreatic cancer. Recently, the phagial family Herelleviridae has been established in his honor (105).

PHAGE THERAPY RENAISSANCE

As mentioned in the preceding sections, MDR has pushed phage therapy back into the scientific spotlight. Such a phage renaissance in the West can be traced back to the work of William H. Smith and M. B. Huggins in the 1980s (51, 106). These authors compared phages and antibiotics in protecting mice from E. coli O18 infection, attaining higher protection with the former than with the latter (107). Other seminal works can be attributed to Carl Merril (National Institutes of Health) and John Morris and collaborators (University of Maryland) whose work was carried out in the mid-1990s (44). Merril tackled one of the main problems with phage administration, that of the rapid removal from the body of the exogenous phages. He isolated two phagial strains (Argo1 and Argo2) that could remain in the bloodstream for over 18 h (compared with a few minutes of the wild-type phages). Merril made a partnership with Richard Carlton, a psychiatrist turned venture capitalist, and together they funded Exponential Biotherapies to market Merril’s application.

Around the same time, physician John Glenn Morris joined forces with Alexander Sulakvelidze and Nina Chanishvili of the EIBMV, and Elizabeth Kutter from the Evergreen University in Washington, to develop a cure for the vancomycin-resistant Enterococci faecium. Even in this case, the clinical work was joined by a foundation of start-ups (PhageBiotics, Phage Therapeutics, Intralytix, and others) to commercialize the results.

Currently, the benefits of phages over antibiotics have been summarized in the following points (108, 109):

• They are specific to their target, without affecting the commensal flora. Antibiotics affect all susceptible bacteria regardless of their pathogenicity.

• They replicate at the site of infection. Antibiotics are adsorbed by the whole body.

• They have no known serious side effects. Antibiotics might have serious side effects.

• Phage-resistant bacteria are still susceptible to other phages, which are relatively easy to isolate. The isolation of new antibiotics has become increasingly more and more difficult.

• Bacteria usually develop resistance to phage infection by reducing their virulence (for instance by losing a capsule). Antibiotic-resistant bacteria have a selective advantage over the other species, and the resistance trait is spread horizontally.

Clinical Studies Performed After World War II in Western Countries

Despite the sharp increase in interest in phage therapy observed in recent years, there are very few actual applications of this approach in the medical field. The vast majority of current studies addressing the effectiveness of phage therapy as an antimicrobial implement occurs in animals, which have been covered by other excellent reviews (24, 110). Overall, only a few studies have clearly demonstrated phage-derived antimicrobial activity. Hence, phage preparations were shown to be more effective in the presence of an active and symptomatic bacterial infection than with asymptomatic pathobionts (110). Conversely, clinical studies of the application of phages to humans are quite scanty. Case reports have increased in the last few years, usually with positive medical outcomes (24, 111); however, more extensive studies are meager.

Outside of the ex-USSR, Poland is the leading Eastern country in the application of phage therapy. Between 1981 and 1986, 550 patients with persistent bacterial infections that did not respond to treatments with either antibiotics or chemotherapeutics were admitted to the Phage Therapy Unit (PTU) of the Lubwik Hirszfeld Institute of Immunology and Experimental Treatment of the Polish Academy of Sciences in Wrocław, established in 1952 (112, 113). Overall, phage therapy was successful with only 6.9% of the cases showing a transient improvement of the infection and 0.7% (four cases) not responding to the treatment. Additionally, 153 patients were treated at the PTU between 2008 and 2010 (114). Of these patients, 61 (39.9%) experienced a beneficial response to phage treatment. In particular, the treatment with phages alone failed in 65 out of 109 patients compared to 27 out of 44 patients who received phages in conjunction with antibacterials. The treatment lasted between 6 and 65 days for the patients who cleared the infection and 4 and 144 days for those who did not. The treatment consisted of 10 to 20 mL of phage preparation at a concentration of 10⁶ to 10⁹ PFU/mL three times a day. In Australia, researchers at the University of Sydney carried out a safety trial on 13 patients to treat Staphylococcus aureus infection, demonstrating the lack of side effects in the phagial preparations (115).

Conversely, in recent years, very few clinical studies have been conducted in Western countries. A topic review performed on January 6, 2022 on the MedLine platform with the term Phage Therapy[MeSH Terms] AND (randomizedcontrolledtrial[Filter]), returned only seven studies. By far, the most frequently targeted bacteria were E. coli and Pseudomonas aeruginosa.

The first clinical study conceived following the modern rules for this type of approach (randomized, placebo-controlled, double-blind trial) was a collaboration between the French Ministry of Defense and the Belgian Royal Military Academy (116). The study enrolled 25 patients with P. aeruginosa-infected burn wounds, with 12 assigned to the treatment group and 13 to the control group. The wounds in the first group were dressed with bandages soaked in a phage cocktail (PP1131) prepared by the French company Clean Cells, whereas those in the second group were injected with the broad-spectrum antibiotic sulfadiazine. PP1131 contained 12 different lytic phages against P. aeruginosa and the collective titer of solution was expected to be 10⁹ PFU/mL, resulting in a ratio of phage particles to host cells of 10 to one. However, the preparation suffered from stability issues during the study, and the actual titer was established at 10⁶ PFU/mL. Furthermore, the randomization was not effective because there were differences at the baseline between the intervention and control groups. Participants in the treatment group were older and carried bacteria more susceptible to antibiotic treatment than those in the control group. It was observed that phages resolved the infection in a median time of 144 h compared with 47 h in the control group. Furthermore, after 1 week of treatment (168 h), six of the patients in the treatment group (50%) showed clearance of the infection compared with 11 in the control group (85%). However, such a difference was not significant (P value = 0.097). Nonetheless, the study established the absence of adverse symptoms in the patients allocated in the treatment group, opening the door for other clinical trials.

Pseudomonas aeruginosa has been targeted by another clinical study carried out by the University College London in collaboration with Biocontrol Limited, United Kingdom (117). Twelve patients suffering from chronic otitis caused by antibiotic-resistant P. aeruginosa infection were treated with a single dose of a cocktail of six phages (BC-BP, 1 to 6) at a collective concentration of 10⁵ PFU and compared with a placebo group (n = 12). All the patients in the treatment group showed a significant improvement in clinical scores and reduction of P. aeruginosa load as did the patients in the placebo group. A case report described the absence of adverse side effects when using phage OMKO1 during prosthetic vascular graft infections caused by P. aeruginosa (118).

P. aeruginosa, S. aureus, and E. coli were also targeted in a phase I clinical trial that assessed the treatment of leg ulcers (119). The phage preparations (WPP-201) contained a cocktail of eight phages isolated from environmental sources kept at a concentration of 10⁹ PFU/mL. The study, carried out at the Southwest Regional Wound Care Center in Texas, USA, included 18 test subjects and 21 controls, with healing percentages of 60% and 77%, respectively.

Multiple bacteria were targeted in a study organized by the University of Zürich (120, 121). The study, preceded by an assessment of the safety of the phage cocktail itself (122), recruited patients undergoing transurethral resection of the prostate. Enrolled participants showing infection with either E. coli, Proteus mirabilis, P. aeruginosa, Enterococcus spp., Staphylococcus spp, or Streptococcus spp. were assigned to either an intervention group (n = 28) treated with EIBMV’s Pyo bacteriophage solution (a licensed version of d’Hérelle’s original), a control group (n = 37) treated with antibiotics, or a placebo group (n = 32). The Pyo bacteriophage is a commercially available cocktail containing a wide spectrum of phages against common uropathogenic bacteria, at a concentration of 10⁵ PFU/mL. For this study, the preparation was supplemented with phages against Enterococcus spp. at a concentration of 10⁴ PFU/mL. The phage cocktail was administered intravesically twice a day. After 1 week of treatment, five participants in the treatment group recovered (18%) compared with 13 in the antibiotic group (35%) and nine in the placebo group (28%). There were six participants in the treatment group (21%) who developed adverse conditions (namely, high fever) compared with 11 in the antibiotic group (30%) and 13 in the placebo group (41%). Nevertheless, the differences were not significant; thus, even this study failed to clearly determine the effectiveness of phage therapy.

Among the seven clinical trials retrieved in the present review was a safety assessment of a commercial preparation, PerforPro marketed by the American company Deerland Enzymes (123). The preparation contained a mixture of four coliphages (LH01, LL5, T4D, and LL12) for the treatment of general gastrointestinal conditions. The study, named PHAGE, enrolled 32 participants, although it was not clear how these were allocated in the treatment and placebo groups. No major side effects were reported and there was a slight increase in total carbon dioxide blood concentration in the treatment group before and after the treatment, from 23.97 to 24.94 mmol/L. Because a reduction in total carbon dioxide is observed during chronic diarrhea, the authors hypothesized that the phage preparation had resolved this symptom in the participants, although this was not recorded in the study. In addition, phage treatment resulted in lower serum alkaline phosphatase (ALT) and aspartate aminotransferase (AST) in comparison with the placebo group. The authors assumed that phage therapy reduced the density of E. coli and, consequently, the amount of serum lipopolysaccharides and associated inflammation-associated tissue damage that generates high levels of ALT and AST. Nonetheless, all of the analyzed blood samples had reported levels that were within normal clinical ranges; hence, the study did not indicate any significant differences between groups.

The PerforPro preparation was additionally tested in a subsequent study termed PHAGE2 (124). The study included 68 participants subdivided into a placebo (n = 21) group, an intervention group assuming Bifidobacterium animalis subspecies lactis as probiotic alone (n = 24), and a comparative intervention group assuming the probiotic together with PerforPro (n = 23). After 4 weeks, the study found a slight improvement in gastric function with the probiotic alone, as well as a reduction in gastric inflammation and colon pain in the group that received the probiotic along with the phagial preparation. There were no significant differences in the microbiomes of the three groups. However, the genera Lachnobacterium and Lactobacillus increased in the group receiving the probiotic alone, whereas Atopombium, Gardnerella, Bifidobacterium, and Clostridium increased in the group receiving probiotics along with PerforPro. In addition, in the latter group, there was a slight decrease in the prevalence of Citrobacter and Desulfovibrio, which are associated with gastrointestinal dysfunction.

One of the identified studies from the topic review was a collaboration between the International Center for Diarrheal Diseases Research in Bangladesh and the Nestlé Research Center, Switzerland (125). A safety study was prepared for the actual clinical trial (126, 127). A cocktail of 11 T4-like phages at a concentration of 3.6 × 10⁸ PFU was orally administered to a group of 39 children and compared with a group of 40 children receiving 1.4 × 10⁹ PFU of a phage preparation (Microgen ColiProteus) developed at the EIBMV, and a placebo group of 41 children. All of the children had diarrhea, and there was no difference in symptomatology or fecal microbial composition between the groups. After 4 days, none of the phage preparations showed any benefit over the placebo group. A deeper characterization of the fecal microbial composition was performed using MPS on a subset of 56 children, without significant differences between groups found. The description of this subgroup, however, referred to two other studies that added no information about the microbial composition of these subjects (128, 129). Furthermore, the authors reported that the symptomatology was linked to the abundance of Streptococcus spp. rather than E. coli, even when enteropathogenic strains of this species were included. As a result, the failure of this study can be attributed to targeting the wrong bacterial species. Furthermore, the authors stated that the purpose of the study was to evaluate the safety of the phage preparations rather than actual medical application.

Another of the seven identified studies was a metagenome comparison of the commercially available preparation Pyophage produced by EIBMV and the Russian Microgen (130). The analysis showed how the two preparations, despite having the same name, had different mixtures of phages, albeit targeting the same bacterial species. The metagenomic analysis reported the presence of several prophage that did not encode genes that could pose a safety risk to the patients. The last of the seven identified studies focused on poultry rather than humans (131). Nevertheless, the study reported how the administration of a single dose of a preparation of phage YSP2 (8 × 10⁸ PFU) 2 h after exposure to 5 × 10⁷ CFU of Salmonella pullorum protected the chickens from diarrhea.

Meta-Analysis of Contemporary Studies

The present review sought to compare the effectiveness of modern phage therapies to those carried out almost a century ago. Contemporary clinical trials on phage therapy were retrieved using the topic review described above (see section “Clinical studies performed after World War II in Western countries”). The meta-analysis was implemented as reported above (see section “Re-analysis of historical studies”) and limited to studies performed after the year 2000. In particular, case reports and studies related to assessing the safety of phagial preparations were excluded. To avoid bias in the results, studies involving testing of drugs were also not suitable for this analysis.

One major limitation of contemporary clinical trials was the lack of an explicit enumeration of healed patients because most of the studies focused on addressing the presence of side effects and describing the microbiota in the treated patients. Thus, most modern clinical trials could be classified as safety tests rather than treatments for bacterial infections, resulting in a small pool of articles. Only five studies of the seven retrieved by the topic review, and one retrieved from a narrative review (132), were suitable for meta-analysis. The diseases treated by the clinical trials performed in the last 20 years included pyoderma (116, 117, 119), urinary tract infections (121), diarrhea (125), and multiple infection types (114).

The results of the meta-analysis showed a nonsignificant association with substantial heterogeneity: OR = 1.6538 (0.7637 to 3.5809), P value = 0.2019, I² = 71.7% (34.4% to 87.8%), but it was not possible to calculate the Egger’s test (Fig. 5). Removing the study by Międzybrodzki et al. (114), due to the fact that the control group received phages simultaneously to antibiotics, the association remained nonsignificant: OR = 1.1400 (95% CI = 0.1394 to 9.3237), P value = 0.9027, I² = 71.1% (95% CI = 26.6% to 88.6%). Again, it was not possible to calculate the Egger’s test. The visualization of the results by forest plot suggested that the work of Wright and coworkers (117) could be an outlier. This study was removed from the analysis, providing a significant effect: OR = 2.8400 (95% CI = 1.5320 to 5.2646), P value = 0.0009. The heterogeneity was overall low but with wide confidence interval: I² = 0.0% (95% CI = 0.0% to 84.7%). However, it was not possible to calculate the Egger’s test.

FIG 5.

Meta-analysis of the contemporary clinical studies employing phage therapy. (A) Random forest plot of the studies reporting the odds ratio (OR) of the intervention (experimental) over the control groups. The model showed an odds ratio above unity but not statistically significant (OR = 1.65, 95% CI = 0.76 to 3.58, P value 0.219) with heterogeneity characterized by I² = 71.7%. The works by Międzybrodzki et al. (2012) and Wright et al. (2009) (114, 117) might be considered outliers. Their removal increased the strength of the study: OR = 2.84 (95% CI = 1.53 to 5.26, P value = 0.0009) but raised heterogeneity (I² = 81.3%, 95% CI = 0% to 84.7%). (B) Funnel plot of the studies shown in panel A, showing the trend line of the model.

The fact that the OR was consistently greater than one suggested that, contrary to the historical studies reported in the “Re-analysis of historical studies” section, contemporary clinical trials indicated that the use of phages was detrimental to the healing of bacterial infections. This finding could imply that, despite the absence of side effects in the phagial preparations, the effective antibacterial value is still unclear. However, the paucity of clinical studies carried out in the last 2 decades makes the meta-analysis not exhaustive. Compared with historical studies, modern-day clinical trials display an asymmetric distribution that highlights publication bias. The results can be interpreted, therefore, as a call to increase the number of studies on this topic, including those reporting negative results.

Modern Landscape of Phage Therapy

After a crisis following the fall of the Soviets, the EIBMV has recovered and now stands out as the reference center for phage therapy. Spin-off companies such as Eliava BioPreparations, Eliava Diagnostics, Eliava Phage Therapy Center, and JSC Biochimpharm supply phage preparations, recently also abroad, and collaborate with international endeavors. Among the most promising products marketed by EIBMV is a special bandage (PhageBioDerm) that slowly releases phages and antibiotics into wounds, speeding up healing (133). The second most renowned phage center is the PTU in Wrocław, Poland. Other phage reference institutes and repositories in Western countries include the Félix d’Hérelle Reference Center at the University of Laval in Canada and the Therapeutic Phage Bank at the German Cancer Research Center (DKFZ), the Center for Innovative Phage Applications and Therapeutics (IPATH) of the University of San Diego, the Tailored Antibacterials and Innovative Laboratories for phage Research (TAILOR) of the Baylor College of Medicine in Texas, and the Center for Phage Biology and Therapy at Yale University, just to cite a few. Furthermore, several worldwide start-ups are involved in phagial applications, and their influence in the field may grow once the effectiveness of this procedure is demonstrated with greater clarity (19).

Conversely, the Institute Pasteur discontinued the production of phage preparations in the 1990s (39), although phages are still employed on a small scale in France. Belgium, Australia, and the United States are among the countries where pharmaceutical companies are now marketing phage preparations for the food industry (134). The purpose of this strategy is to circumvent the stark rules required for clinical applications present in Western countries by providing a body of evidence on the effectiveness of phage preparations in animals (44). Moreover, the rightful ban on the use of antibiotics in livestock in Europe (135) opens the need for alternative antimicrobials in agribusiness. Several studies, yielding promising results, have been carried out to determine the effectiveness of phages for the treatment and prophylaxis of infections in animals and to disinfect surfaces (51).

Phage production is not straightforward for private companies (19). The sur-mesure approach requires a very large collection of phages, mostly developed on-demand, that cannot be produced following the strict regulations and quality levels associated with the current classification of phages in the West (drugs in the United States and medicinal products in the EU). Alternatives are actively pursued. In Belgium, for instance, phages have been classified in 2017 as “active product ingredients” to ease their application. However, this is a small step in terms of production streamlining. It has been proposed that phage preparations could be legally classified under the European Medicinal Products Directive of 2001, which governs the use of a variety of pharmaceutical products, or that a new directive be created entirely (136). Under the proposed normative, phage preparations would be tested in the same way as food preparations for microbiological, physical, and chemical hazards at each stage of production. Preparations also face problems in terms of the structural stability of the virions. The selected phagial strains must be kept genetically stable to remain effective against the chosen bacterial strain, and phages are difficult to patent for legal reasons. All of these limits increase the price of phage development while reducing the avenue for the pharmaceutical companies, explaining why funding is diverted to more conventional avenues such as vaccines and drugs, leaving the start-ups to carry over basic research in phage therapy.

Despite these drawbacks, phage therapy today is expanding in ways that were unthinkable when it was first developed as an antimicrobial treatment a century ago. Through genetic engineering, it is possible to avoid lysing the target bacteria, thereby reducing the release of endotoxins, and on the other hand, to modify the phagial receptors, thereby purposefully expanding the host range of the phage preparation (137). Additionally, phages are used in gene delivery and the development of nanomaterials (138), including biopanning for selecting antibiotics and antimicrobial molecules (139), delivery of molecules inside target cells (140), biosensors (141–143), development of vaccines (144), and even battery production (145, 146).

The field of phage therapy application is also expanding. While dysentery, urinary tract infection, and septicemia were the primary targets of this antibacterial treatment at the turn of the century, phages are now used to treat bacterial sexually transmitted infections caused by Clamidia trachomatis and Neisseria gonorrhoeae (147). In addition, it has been reported since the 1970s that phages can access human cells, a feature that opens the possibility of targeting intracellular bacteria (148). For instance, phages have been applied to eradicate intracellular Staphyloccocus aureus (149) and E. coli (150) from epithelial cells. Phage therapy can also be applied in aquaculture and agriculture, reducing the employment of antibiotics in these sectors (151). Furthermore, several preparations are already commercially available for the removal of bacteria from foodstuff, reducing the risk of zoonoses (152).

DISCUSSION

The spread of antibiotic-resistant bacteria has raised concerns about the indiscriminate use of antibiotics in clinical and agribusiness settings. The management of MDR bacteria is multifaceted. On the one hand, there is antimicrobial stewardship, which is the judicious application of antibiotics (153, 154). In 2006, the European Community banned the employment of antibiotics for growth promotion, planning to ban their use for disease prevention; in 2017, the WHO followed suit (11). Conversely, antibiotics stewardship is less stringent in the United States, where these antibacterial are used more extensively than in Europe (11). On the other hand, alternatives to antibiotics span a wide range of current research (155). Nanoparticles (like those attached to gold and silver), quorum sensing inhibitors, porins to selectively lyse bacterial cells, and antimicrobial peptides are all recent developments in antibiotic substitutes. Additionally, fecal transplants, plant-derived compounds, and synbiotics (the simultaneous administration of prebiotics and probiotics) show promise as antibacterial strategies. Nonetheless, phage therapy will play a significant role in the fight against MDR bacteria.

Despite its origins in the West, phage therapy has fallen out of favor in Western countries, whereas it remains a popular medical treatment in Eastern countries, mainly in Georgia and Poland. The data gathered in the present review suggest the dismissal of phage therapy in Western countries might be due not so much to scientific flaws but rather to early improper commercial exploitation and socio-economic reasons. It is conceivable that the widespread commercialization of low-quality phage preparations in the 1940s created skepticism about this therapy among the general public and medical practitioners. The publication of the JAMA reviews in the 1940s (85) further increased the alienation of phages from the scientific community. The introduction of antibiotics in the same time period supplied companies with features that could not be found in phages: a relatively simple product that could simultaneously target several bacteria with a predictable outcome (pharmacokinetic). Consequently, the dismissal of phage therapy in Western countries can be interpreted in a socio-economical context rather than merely scientific. As shown in the present work, there are only a few clinical studies performed in the last 20 years; thus, the statistical significance of phage therapy remains elusive.

Just as oncology is using precision medicine with therapies designed ad hoc for each patient (156), the application of phage therapy can also provide antibacterial therapy that might evolve into more personalized approaches. Nowadays, precise medicine is carried out by running a battery of sequencing assays to establish the most effective line of oncological treatment based on the genetic profiles of each patient. It is plausible that in the near future the laboratories of microbiology in hospitals or other large health institutions might determine the exact bacteriome of a patient and develop sur-mesure phage therapy that will maintain the healthy microbiome avoiding the spread of antibiotic-resistant strains. Nonetheless, reaching this goal will require the development of fast diagnostic pipelines coupled with large pools of phages that would be available only by international collaborative efforts. Steps in this direction have already been taken, for instance with the nonprofit initiative Phage Directory (https://phage.directory), which facilitates the sharing of phagial strains around the world, and the Phage for Global Health, part of the Bill and Melinda Gates Foundation, which provides laboratory training and development of phage preparations that are provided to developing countries free of charge (157, 158). Training on phage isolation and characterization is provided by the Science Education Alliance-Phage Hunters Advancing Genomics and Evolutionary Science (SEA_PHAGES), established by the University of Pittsburgh and the Howard Hughes Medical Institute’s Science Education division (159).

Another step toward sur-mesure phage therapy is the development of detection methods as an alternative to the classic double layer agar assay (160), which is widespread but labor intensive. A popular alternative is PCR (161–163), but this approach is hampered by both the need to design primers that can account for the wide genetic variability of phages and the incapability of specifically pinpoint infecting virions. Phages can be quantified directly by enumerating the virions using flow cytometry (164) or indirectly by measuring the amount of light given up by disrupted bacterial cells in microvesicles by normal fluorescence readers (165). Capture-ELISA is also an inexpensive and fast method (166), although this method requires the development of phage-specific antibodies. Another promising method, which is fast and inexpensive, is the comparison of optical density between a pure bacterial culture and one with lytic phages (167). Just as precision medicine has been built on the improvements in molecular biology and bioinformatics, it is plausible that in the near future phage therapy will require further advancements in fast microbial characterization, phage production at low-scale and low-cost, and the fast delivery of the required phages on a global network of operators.

In conclusion, the present review showed phage therapy demonstrated a medical advantage in treating bacterial infections in the clinical studies performed at the beginning of the 20th century. The dismissal of this approach in Western countries might be due to issues not related to the actual performance of the therapy, including personal issues between the formidable figure of the therapy’s inventor, Felix d’Hérelle, and the academic establishment. In modern times, despite the enormous progress in the understanding of phage biochemistry, there is a substantial paucity of clinical studies that obfuscates the understanding of the true value of phage therapy in fighting MDR infections.

The current study illustrated how the dismissal of phage therapy in Western countries after World War II may have been due to personal issues surrounding the first proponent of this therapy, Felix d’Hérelle, rather than problems related to the effectiveness of this antibacterial treatment. Herein, a reevaluation of historical studies by modern meta-analytical methods showed the effectiveness of clinical trials carried out about a century ago. Thus, the dismissal of phage therapy in the West in the mid-1900s may have slowed the growth of this promising field of medical microbiology. Nevertheless, the resurgence of phage applications currently taking place in Western nations shows the significance of these viruses has been appropriately assimilated by the medical and academic communities without the blind enthusiasm demonstrated a century ago, in particular by commercial enterprises.

For instance, the meta-analysis of contemporary clinical trials described in the present work showed that phage therapy might not be as effective as expected, albeit proving the absence of severe side effects in the recipients. Such a result might be due to the small number of clinical trials available for the analysis, but also to the fact that phage therapy is based on an extremely complex kinetics that involves not only bacterial growth and phage infectivity but also the immune system and the conditions of the microenvironment where the infection is taking place. Once the processes underlying phage therapy are more properly understood, it will be easier to predict a further boost of antibacterial treatments based on phages. Moreover, to minimize the degradation of phage particles, a new method to improve the stability of phagial preparations is actively being investigated. For instance, microencapsulation of phages or coadministration with probiotics have shown promising results in improving the effectiveness of phage treatment (168).

The expanding field of phage application (which now includes agriculture, diagnostics, and materials science) suggests that we are entering a new era of phage applications once unthinkable when d’Hérelle and Twort first describe these viruses. The number of medical applications for phages may rapidly increase once ongoing clinical trials show phages are effective antibacterial agents. Phage agents will almost certainly become commonplace. As a result, it is easy to imagine the current study’s small number of clinical trials is just a warm-up for the eventual use of phage therapy.

Biographies

Luigi Marongiu graduated with a bachelor’s degree in molecular biology and obtained a Ph.D. in virology from University College London, United Kingdom. Dr. Luigi Marongiu investigated the presence of bacteriophages and other viruses in colon cancer while at the University Clinic of Mannheim, University of Heidelberg, Germany. He is currently working at the Global Health Institute of the University of Heidelberg on a study regarding arbovirus epidemiology. He is also holding a research position at the University of Tuebingen, Germany to investigate the role of nutrition in viral physiology and cancer development. Dr. Luigi Marongiu’s interest in phage therapy originates from the treatment’s significant potential to cure malignant disorders combined with the intriguing history of the development of this therapy. He has over 20 years of experience in the field of virology.

Markus Burkard studied pharmacy at the Eberhard Karls University of Tuebingen, Germany and did his doctorate there in the Department of Pharmacology, Toxicology, and Clinical Pharmacy and the University Eye Hospital. From 2014, Dr. Markus Burkard worked as postdoctoral fellow with a focus on oncoimmunological issues, epigenetics, and nutrition in the Department of Internal Medicine I at the University Hospital of Tuebingen and then in the Department of Vegetative and Clinical Physiology, where he has started working closely with the Virotherapy Centre Tuebingen (VCT) until today, studying the effects of nutrition on viruses and phages. Since 2020, Markus Burkard has been head of the laboratories in the Department of Nutritional Biochemistry at the Institute of Nutritional Sciences at the University of Hohenheim, Germany.

Ulrich M. Lauer graduated from medical school at the University of Erlangen-Nuremberg, Germany and completed a residency in internal medicine and a fellowship in oncology at the University of Tuebingen, Germany. Prof. Dr. Ulrich M. Lauer received a postdoctoral training at the Department of Virology at Max Planck Institute for Biochemistry, Martinsried, Germany, and specialized in virotherapy of solid tumors and gene therapy of monogenetic diseases, such as Morbus Wilson. He is a clinical professor of Medicine at the University of Tuebingen, where he is Deputy Director of the Department of Medical Oncology and Pneumology. He is a medical oncologist at the German National Center for Tumor Diseases SouthWest (NCT-SW). He is head of the Virotherapy Center Tuebingen (VCT), Head of the Early Clinical Trials Unit Tuebingen (ECTU), and Head of the European Neuroendocrine Tumor Society Center of Excellence Tuebingen (ENETS).

Ludwig E. Hoelzle is a Full Professor and head of the Department of Livestock Infectiology and Environmental Hygiene, Institute of Animal Sciences, University of Hohenheim, Germany. Prof. Dr. Ludwig E. Hoelzle studied veterinary medicine and is specialized in veterinary microbiology and animal hygiene. Ludwig Hoelzle has longstanding experience in classical and molecular microbial diagnostics. During his Ph.D. thesis in veterinary virology, he worked with bacteriophage libraries. One current research focus is on infection biology of bacterial pig diseases (e.g., Mycoplasma suis) including molecular pathobiology, epidemiology, immunity, and vaccine development. Another focus lies on the hygiene of animal husbandry, including the analyses of microbial diversity (including bacteriophages), resistance development, and transmission as well as disinfection.

Sascha Venturelli graduated in biology as well as in medicine and obtained his Ph.D. at the Faculty of Natural Sciences, University Hohenheim, Germany and his M.D. at the Faculty of Medicine Tuebingen, Germany. Prof. Dr. Sascha Venturelli underwent postdoctoral training at the renowned Peter MacCallum Cancer Centre Melbourne Australia. After his return, he established his own research group focusing on immunomodulation, epigenetics, and nutrition in the context of oncology at the University Hospital Tuebingen. In 2016 Sascha Venturelli was appointed Deputy Director of the Department of Vegetative and Clinical Physiology and academic director in 2017. Since 2019, Sascha Venturelli has been a Full Professor at the University Hohenheim and head of the Department Biochemistry of Nutrition. He has been working closely with the Virotherapy Centre Tuebingen (VCT) for a long time, studying the effects of nutrition on viruses and phages.