Translational research priorities for bacteriophage therapeutics

The gravity of the antimicrobial resistance (AMR) threat is well recognized in terms of its impact on human morbidity and mortality and on increased health care costs [1]. This threat has intensified as antibiotic use in medicine and the livestock industry has accelerated and as travel and global interchange of goods have increasingly facilitated the rapid emergence and dissemination of AMR pathogens. Climate change and conflict-driven mass migration have created large populations of underhoused, food insecure persons who serve both as targets and amplifiers of the global AMR threat. Against this backdrop, traditional industry-based antibiotic discovery has slowed as the economic models for pharmaceuticals have increasingly incentivized investments in chronic conditions that affect large numbers of people rather than agents used over short terms for a limited number of people [2,3].

Efforts to create incentives for investment in antimicrobial discovery and development are underway and are laudatory. So far, the impact of these incentives has been limited but this movement and the attention of several influential governmental and non-governmental bodies including the WHO, the CDC, the NIH, BARDA, DARPA, the Bill and Melinda Gates Foundation and others has resulted in a broad-ranging set of initiatives to broaden our tools to combat AMR beyond traditional small molecule antibiotics. These initiatives have awakened interest in the development of novel antimicrobial approaches and have included traditional antibiotics (with modest success to date) as well as three approaches that dominated the pre-antibiotic era: vaccination, passive immunotherapy and bacteriophage therapy [4]. While immunotherapeutic approaches hold promise for some pathogens (and especially viral pathogens), vaccines will have only a limited reach for the prevention and treatment of AMR bacterial infections because of the implausibility of population-level vaccination for the wide array of potential AMR bacterial pathogens. Monoclonal antibodies face the inevitable challenge of antigenic variation and evolution as has been noted in the cases of Ebola and SARS CoV-2 although recent advances in structure-based design may provide approaches that will mitigate some of the challenges of antigenic diversity and evolution in the equally (if not more) complicated area of passive bacterial immunotherapeutics [5]. In view of the challenges with traditional antibiotic development and of immunotherapy, interest in bacteriophage therapy has accelerated over the past decade – well beyond its traditional geographic niches in Eastern Europe and the former Soviet Republics. The collective experience with phage therapy in those regions is substantial and the lessons learned from this experience have not been fully appreciated in the West [6]. Some of this is attributable to barriers imposed by a relative lack of scientific interchange historically between scientists from these regions and the West and some to primary languages of publications. Although the amount of data demonstrating efficacy from rigorously conducted prospective clinical trials conducted in these regions is limited, data from case reports and case series supports the safety of phage therapy.

Three factors that emerged just under 10 years ago led to wider interest in phage therapy in the West. These included (1) a growing number of pathogens and an enlarging patient population that are out of reach of currently available antibiotics, (2) advances in biotechnology that made phage identification, propagation and clinical grade production more accessible and (3) publication of an increasing number of case reports and case series that highlighted the use of phages in patients with limited treatment options [7–11]. Virtually all of these phage treatment experiences were undertaken under FDA individual patient (physician initiated Investigational New Drug) regulations in the United States or analogous regulatory pathways in Europe and Australia [12,13]. Since regulatory agencies do not publish the number of cases for a specific indication treated under IND regulations, the actual number of patients treated with phage therapy over the past decade under these so-called ‘compassionate plea’ regulations is unknown. The yearly number of case reports of phage therapy indexed in PubMed has increased roughly fivefold over the past decade (Figure 1). These data likely underestimate the true rate of rise in patients treated since publication of case reports might be expected to become less common as the novelty of the intervention declines. The distributions of organisms and medical indications for treatment are broad and reflect, to some extent, the referral and practice patterns of phage treatment centers and the channeling bias that results in overreporting of apparently successful outcomes [11,14]. Although Pseudomonas aeruginosa and Staphylococcus aureus are the organisms most frequently cited in most published case reports and series, a wide range of organisms including MDR enteric gram-negative bacteria, Acinetobacter baumannii and Mycobacterium abscessus also appear frequently in the literature [10,11,14]. Medical indications for treatment include lower respiratory tract infections, infected implanted medical devices, soft tissue infections, urinary tract infections and a number of others [11,14]. In virtually all reported experiences, phage therapy has been judged to be without serious complications and treatment success rates have been reported to be in the 50–70% range in reported series [9,10,14–16].

With the rapidly expanding access to phage therapy, the lack of evident toxicity and a relatively high success rate for treatment of infections with low a priori likelihood of treatment success, some have questioned whether clinical trials of phage therapeutics are still needed [17]. A stronger argument can be made that with the increasing recognition that bacteriophages are likely to be a major tool in our response to the AMR crisis, clinical and translational research focused on phage therapeutics is more urgent now than ever. Three factors strongly support the need for clinical trials of phage therapy. First, although ‘compassionate plea’ pathways have made phage therapy more broadly available over the past decade, these essentially philanthropic activities cannot be taken to the scale that will ultimately be required to meet the global clinical need. This will require commercialization which will, in turn, require approval by regulatory authorities on the basis of Phase 3 trials that demonstrate clinical evidence of safety and effectiveness for specific medical conditions. Second (and in some ways, perhaps, more importantly), rigorous clinical trials provide the most effective and direct pathway through which the underlying principles of phage therapy can be developed so that clinical impact can be maximized. Defining the fundamental principles of phage therapy will require much more than a handful of registrational trials with clinical endpoints; it will require that clinical trials be enriched with substantial investments in translational research that directly address critical elements of phage biology as they apply to their therapeutic use (Figure 2). In the case of Phase 3 trials, in particular, it is extremely important that sufficient translational research endpoints be incorporated into each trial so that mechanistic reasons for success or failure of each trial can be assessed. Maintaining a rigorous scientific framework within which clinical trials are undertaken will be critical to the design of subsequent studies and to sustain the investment that will be required to advance phage therapy from anecdotes to approved medications that are widely available through scalable commercial channels.

Finally, unambiguous data from rigorously designed and conducted clinical trials will provide support and reassurance to a currently reluctant pharmaceutical industry that phage research is a rational investment. Several challenges must be overcome to assuage this reluctance. The first relates to a general disinvestment in antimicrobial research on the part of industry for at least two decades. Antimicrobials are seen as a low margin business in which courses of therapy are limited in duration. The psychology of antimicrobial discovery and development has generally centered on the development of broad-spectrum agents capable of killing multiple bacterial species. The pharmaceutical industry has not been enthusiastic about or willing to develop antimicrobial agents for narrow indications or patient populations. Although models for reimbursement for ‘personalized therapy’ have been developed for cancer and genetic diseases, such models have not yet emerged for antimicrobial therapy. Finally, industry is intrinsically conservative and risk averse. Although major advances in our understanding of phage biology and technology have transformed opportunities for phage development, many in the pharmaceutical industry leadership remain frozen in the ‘Arrowsmith era’ and do not understand that prospects for phage development are fundamentally different from where they were even a decade ago. In the past year, Phase 1/2 clinical trials of phages based on drug development paradigms that were successfully employed in the development of small molecular antibiotics have demonstrated antimicrobial activity of the same order of magnitude that expected of antibiotics and multiple strategies for the protection of intellectual property have been developed [18,19].

There are several routes to incentivizing the pharmaceutical industry to become more heavily invested in the development of phage therapeutics. The first is that investment by government research agencies such as the NIH, BARDA, ARPA-H and analogous agencies in Europe and elsewhere will both ‘derisk’ and legitimatize investments by industry. Furthermore, government research support can facilitate the development of multidisciplinary, multicenter collaborations that include both academia and industry and from which consensus methodologies and protocols emerge. These can catalyze further engagement by industry. Clear guidance by regulatory agencies regarding clinical development pathways that would lead to commercial approval as well as further clarity around intellectual property protection would also lower barriers to industrial engagement. Finally, creativity around the economics of drug development and reimbursement will be critical. A number of approaches have been developed in the United States to incentivize the development of ‘orphan’ drugs for rare indications (https://www.fda.gov/industry/medical-products-rare-diseases-and-conditions). These benefits are generally reserved for diseases that are sufficiently rare in the population that traditional reimbursement paradigms are felt to be inadequate. Although MDR infections are, unfortunately, extremely common in the aggregate, one could make a cogent argument that infections caused by, for example, a specific strain of carbapenem-resistant Klebsiella pneumoniae are sufficiently rare that a phage product developed for that specific infection should receive orphan drug-like incentives.

Although certain physiological characteristics of phages including burst size, adsorption rate, lytic capacity, lifestyle, biofilm disruption capability and latency period have been posited to be important determinants of phage efficacy in the clinic, none of these parameters have been formally tested in the clinic [20–26]. Physiological characteristics of phages have been explored in laboratory settings but most such experiments fail to account for host conditions that may influence them in clinical settings. For example, most laboratory studies of these parameters are undertaken with laboratory strains of bacteria propagated aerobically in log phase growth. The diversity of clinical strains coupled with substantial differences in growth conditions in patients complicates interpretation of these in vitro studies but in view of the overwhelming number of phages in nature and the growing ability to genetically engineer phage biology, it would be important to understand the extent to which these characteristics might influence phage selection.

Bacterial phage susceptibility testing (PST) is usually conducted by one of two general approaches: plaque or efficiency of plating assays on solid media or in liquid media in which bacterial metabolism is quantified in real time [20,27,28]. Unlike the case with antibiotic susceptibility testing in which central organizations such as the Clinical Laboratories Standards Institute, CLSI (https://clsi.org) or the European Committee on Antimicrobial Susceptibility Testing, EUCAST, (https://www.eucast.org) have developed standardized assays, testing reagents and clinical breakpoints, PST is undertaken as ‘home grown’ (or ‘laboratory developed’) assays in which each laboratory develops its own methodology and reports the results using metrics tailored to the assay used. This lack of standardization is highly problematic in the case of phage therapeutics. Discordant result rates are unacceptably high in the same laboratory on different days and are even less reproducible when they are performed using the same technique in different laboratories [29,30]. Concordance of plaque assays and liquid phase assays is even worse. In one recently concluded study, two laboratories performing PST on panels of four phages and 46 P. aeruginosa isolates obtained from people with cystic fibrosis shedding the organism in sputum found that the two laboratories performing broth testing delivered concordant PST results only ∼80% of the time [30]. (Table 1) Concordance between the broth and plaque assay was demonstrated only 60% of the time. (Table 1). One of the critical needs for phage therapeutics is the development of a CLSI-like standards for PST. As in the case of bacterial susceptibility testing, validation of laboratory cutoffs for susceptibility and resistance will require a substantial body of clinical data in which the output from PST efforts is correlated with microbiological and clinical outcomes from clinical trials. Clinical trials offer a critical opportunity to advance the science of PST. The use of clinical and microbiological endpoint data from clinical trials to define how to configure standardized in vitro assays so that they serve as reliable predictors of clinical efficacy will provide guidance about how to standardize PST assays and can provide critical information about why phage interventions succeeded or failed in the clinic. Although the primary clinical endpoints in the Pherecydes study of the use of phages to treat patients with P. aeruginosa in burn patients were not met, a post hoc analysis of the study demonstrated that the phages did actually have a positive microbiological impact in patients with organisms that were susceptible to one or more phages in the phage treatment cocktail [31]. This trial that was initially characterized as another ‘phage failure’ is now viewed as one that demonstrated the importance of PST in selecting patients for inclusion in clinical trials and has encouraged further studies of the topical use of phages in burns and wound infections.

Advancing PST from the current cottage industry to a more rigorous scientific discipline will require a coordinated effort at collaboration across laboratories from academia and industry focused on developing consensus techniques for the determination of phage susceptibility. As a first phase of this effort, one might envision a series of workshops at key microbiological conferences in which a representative group of scientists with expertise in PST are invited to develop and publish consensus PST protocols for both solid and in liquid media formats. These could be convened by organizations such as the American Society for Microbiology (ASM) or the European Society for Clinical Microbiology and Infectious Diseases (ESCMID). Once this is complete, a network of 10–20 interested laboratories could be funded through a competitive process to standardize and refine these consensus protocols. This effort would involve the distribution of standardized collections of bacterial strains and phages to participating laboratories that would then conduct PST under blinded conditions. Results would be unblinded and outliers would be evaluated with the objective of minimizing discordance of results. This process could initially focus on several key bacterial species such as P. aeruginosa, Escherichia coli and S. aureus and then broadened to include additional species as consensus methodologies emerge. A quality assessment process could then be developed that would enable laboratories to be certified in PST proficiency. As in the case of bacterial susceptibility testing with antibiotics, clinical correlation will be essential to the establishment of clinically relevant breakpoints and metrics. These clinical proficiency quality assurance programs could be developed as stand-alone programs or coordinated by CLSI in the US or EUCAST in the European Union.

The development of standardized, clinically validated phenotypic assays for PST will lay the groundwork for the subsequent development of molecular assays for phage susceptibility. At this point the complexity of phage biology and counterbalancing bacterial host defense mechanisms is such that molecular testing is still in its infancy [32]. As larger collections of phenotypic and molecular data are developed and artificial intelligence-augmented analytical approaches are refined, molecular methodologies hold tremendous promise for the future of PST.

In clinical practice and in clinical trials, phage dosing has generally been chosen arbitrarily. Dose levels reported from European experiences tend to be substantially lower than those reported from the United States but at this point formal dose ranging studies are scarce [9,11,33–35]. One ongoing Phase 1/2 dose ranging study of intravenous phage is currently underway in stable volunteers with cystic fibrosis chronically shedding P. aeruginosa [36]. This stands in stark contrast with the practice with antibiotic development in which extensive pharmacokinetic data are available prior to any use in infected patients. Parenterally administered phage are rapidly cleared from the plasma into the spleen and liver by the reticuloendothelial system. In most cases in which phages are administered parenterally, the premise is that enough ‘spillover’ of the administered dose will reach and replicate at sites of infection before the clearance is complete. The kinetics of this clearance is rapid but slows with increasing doses of phage suggesting saturability of the clearance mechanism(s) [37]. The pharmacodynamic impact of phages requires both delivery to the site(s) of infection and amplification at those sites and this cannot be assessed in uninfected humans or animals. In contrast in antibiotic dosing pharmacokinetic data from uninfected patients (or animals) can strongly predict pharmacodynamic outcomes. This conundrum has complicated studies of phage pharmacology and has resulted in our current primarily empiric approach to dosing. Clinical trials, and especially those in which the sites of infection are readily accessible, provide critical opportunities to relate microbiological outcomes to phage dosing and phage concentrations at the site of infection. Such studies, augmented by mathematical modeling will provide data that will inform the design of future clinical trials and may explain why a given trial succeeded. Using the aforementioned Pherecydes burn study as an example, the decision to measure phage concentrations at the site of infection led to the realization that the phages in this 13-phage cocktail were physically incompatible and that their storage and transportation in low concentrations had further eroded their activity at the site of infection [31]. This insight has prompted more attention to formulation and stability issues in phage trials and (coupled with the PST data noted earlier) was critical in recasting this trial as one that should stimulate more rather than less research on the therapeutic use of phages in similar patient populations.

Decisions about routes of administration have not been based on rigorous comparative observations. Prior to the advent of near-GMP phage preparation, phages were generally administered topically, orally or occasionally by a nebulized route. Since phages can now be given by intravenous or other systemic routes comparative studies of different routes of administration are also indicated – especially in the case of pulmonary infection [18]. The demonstration of substantial antimicrobial activity of an anti-P. aeruginosa phage cocktail in patients with cystic fibrosis shedding P. aeruginosa in the sputum now needs to be directly compared with the microbiological impact of intravenously administered phages to the same patient population. Such a trial is currently underway [36].

Many of the earlier clinical trials assessed outcomes clinically without substantial microbiological assessment. Clinical trials in which infected sites can be repetitively sampled such as the respiratory and urinary tracts or wounds offer opportunities to develop more robust approaches to quantitatively assessing bacterial load. Until recently, quantitative data on the impact of phages on bacterial ‘load’ in clinical settings has been modest. Some of the difficulty has arisen from the fact that phages are frequently given concomitantly with antibiotics and assessment of the independent microbiological impact of phages has been challenging. Clinical trials in which phages are added to an existing regimen or used in patients without other treatment options offer opportunities to better understand the quantitative and kinetic relationships among phage dosing, phage concentrations at sites of infection and impact on bacterial load. Designing such trials have been complicated by the fact it is difficult to justify ‘phage alone’ arms for patients requiring immediate clinical intervention. Clinical trial designs in less urgent medical settings such as recurrent cystitis offer an opportunity to design studies in which patients are randomized to a phage only arm or an antibiotic arm if patients on the phage arm are followed closely and antibiotic additions are made with any evidence of clinical progression. This would not be unlike clinical scenarios in which patients are treated empirically and followed clinically awaiting bacterial identification and susceptibility testing. Clinical progression in these settings results in intensifying or broadening antimicrobial coverage pending laboratory confirmation of antibiotic susceptibility tests. Studies in patients who are chronically shedding an organism without clinical immediate clinical implications such as nasal carriers of S. aureus or a subset of patients with cystic fibrosis also provide opportunities for phage alone treatment interventions.

At this point, the two approaches most frequently used based on quantitative (or semi-quantitative) cultures or molecular quantitation. Neither metric has yet been clinically validated and few studies have measured both metrics in duplicate samples. Studies of this nature provide an opportunity to assess the validity of these techniques and, as above, to develop dosing regimens that are data driven rather than empiric. In the context of Phase 3 clinical trials, whenever possible, designing trials in ways that enable assessment of microbiological outcome will be critical to understanding whether a clinical failure is the result of a microbiological failure and may guide in refinement of the phage dosing regimen and duration if longitudinal microbiological assessments are available.

Bacteria have been developing phage defense mechanisms for tens of millions of years and, as in the case of antibiotics, phage resistant bacteria readily emerge in the clinic [7,8,10,11]. Efforts to mitigate phage resistance include the use of phage ‘cocktails’ in which multiple phages with putatively orthogonal resistance pathways are combined or the concomitant use of phages with antibiotics [38]. Phage resistance may be accompanied by deleterious effects on the bacteria under treatment including loss of fitness or re-sensitization to antibiotics [7,8,11,39–43]. Clinical trials provide an opportunity to systematically collect data about the extent to which mitigation strategies are successful and on the extent to which phage-driven evolution of the bacterium under treatment may lead to beneficial clinical outcomes. Studies of phage susceptibility require that studies be designed in ways that include serial microbiological quantification and serial sampling of residual bacteria for phenotypic and molecular characterizations of the surviving pathogen.

Phages present multiple foreign antigens in clinical use and can induce cellular and humoral immune responses when administered to humans or animals [10,11,44–49]. These responses are demonstrable when phages are administered intravenously, intramuscularly, or by aerosolized routes and, to a lesser extent by oral routes. Case series suggest that neutralizing antibodies to phages may develop in roughly half of the cases in which phages are repetitively administered parenterally. When they occur, they are usually demonstrable 10–14 days into therapy. In cases in which clinical outcomes are available, very limited date suggest that antibodies to the phages administered may (or may not) be associated with a loss of clinical activity [10,11]. Clinical trials provide an opportunity to much more systematically understand how often and in which patient populations phage specific immune responses emerge and in what clinical settings such responses may compromise treatment outcomes.

Recent advances in phage engineering have offered opportunities to broaden phage host range, alter phage lifestyle, enhance lytic activity and enhance biofilm disruption capabilities [14]. One of the first of these efforts to advance to clinical application was the use of BRED technology to convert lysogenic phages directed at non-tuberculous mycobacteria from a lysogenic to an exclusively lytic lifestyle [10]. Subsequently, phages engineered to broaden host range and that use CRISPR-cas mechanisms to enhance bactericidal activity have entered clinical trials [18,19,50]. As these innovations advance, it will be critical to incorporate sufficient translational endpoints to determine whether these innovations translate to enhanced killing, reduced emergence of phage resistance, improved PK/PD relationships and better clinical outcomes. Many of the metrics outlined above can be productively employed in an iterative fashion to guide additional phage engineering efforts. Although concerns have been expressed about regulatory complexity for genetically engineered phages in some jurisdictions, the US FDA has used existing regulatory ‘risk/benefit’ paradigms in the evaluation of genetically engineered phage products and has not chosen to consider them as recombinant DNA vectors [51]. Genetically engineered phages from at least three sponsors have entered clinical trials [18,19,50].

While the regulatory pathway for traditional antibiotics is relatively straightforward, that for phage therapeutics is still a work in progress. While the desired outcome for each therapeutic approach is the same, namely, to eliminate or to reduce the burden of a defined bacterial pathogen, the products are currently viewed quite differently from the regulatory perspective. In the United States, this is most clearly demonstrated by the fact that they are not even regulated by the same Centers within the FDA. Several factors contribute to this regulatory divergence including product definition, safety considerations and PK/PD relationships. Rigorous translational research can bring clarity in each of these areas.

With respect to product definition, antibiotics consist of either a single molecular entity or a fixed combination of two molecules. Well established antimicrobial testing methodologies define the (usually broad) microbial spectrum of each antibiotic and clinical trials define the clinical conditions for which a specific antibiotic is approved. Descriptions of chemical structures, molecular mass, synthetic pathways, purity and lot release criteria are straightforward. Phages require substantially more description in terms of molecular sequencing, biological quantitation, antimicrobial spectrum, purity and lot release criteria. Methods for defining these characteristics in phage therapeutics are available but not yet universally standardized.

Phage regulation is also complicated by the fact that narrow antimicrobial spectra generally require the use of fixed combinations of phages that are selected to achieve broad coverage of clinical isolates of a single bacterial species or in ‘custom’ phage combinations that are specifically tailored for an individual patient. In the case of fixed phage combinations, regulators are confronted with the fact that one (or some but usually not all) phage will be contributors to the antimicrobial effect. Regulatory decisions about combination products have generally required that each component of a combination regimen contributes to a desirable therapeutic outcome. In the case of ‘custom’ cocktails, each patient is treated with a unique ‘product’. In addition to considerations related to spectrum of activity, phage cocktails or phage-antibiotic combinations can be composed based on in vitro assays that evaluate synergy or additive activity [11,52–55]. Although a number of approaches have been developed to assess the antimicrobial activity of phage combinations in vitro, the relevance of these assays to efficacy in the clinic has not been established and will be subject to a number of variables including growth conditions and stoichiometry in clinical settings that cannot yet be fully simulated in in vitro systems [56].

Safety considerations also differ between antibiotics and phage therapeutics. In the case of antibiotics, considerations usually focus on rate limiting organ specific toxicities of each agent. Heavily influenced by the pre-existing state of organs in which toxicity is dose limiting, dose adjustment algorithms receive careful attention in the regulatory paradigm. In the case of bacteriophages, safety considerations mainly devolve to intrinsic properties the phage itself, to preparation methodology and to product purity. Molecular sequence analysis is required to determine whether phages harbor genetic elements associated with antibiotic resistance, toxic elements or transduction capability. Most regulatory agencies also require functional assays directed at defining generalized transduction capabilities of each phage. These functional assays are far from standardized. Guidance for phage production under Good Manufacturing Process (GMP) generally follows that for other biological agents. Assays for impurities and microbiological contamination also figure heavily into the regulatory process for phages. Once these aspects of phage production have been addressed, regulatory agencies are generally quite sanguine about the safety of phage administration by local or systemic routes of administration.

Phage pharmacology and pharmacodynamic considerations also differ substantially from those encountered in antibiotic development. Regulators focus on delivery and maintenance of therapeutic concentrations of antibiotics to sites of infection based on in vitro antimicrobial and animal model data, sophisticated pharmacologic models and empiric measurements. Regulatory clarity around phage pharmacology will require substantial translational research in the context of clinical trials with patients infected with phage susceptible organisms.

The regulatory paradigm framework for phages will clearly diverge from that developed for antibiotics. These obstacles can be overcome but ongoing multidisciplinary dialogs will be required that involve phage biologists, clinical investigators and regulators and the development of novel regulatory paradigms. An underpinning of rigorous translational science will facilitate these discussions and accelerate the goal of commercialization of phage therapeutics.

The integration of translational endpoints into the fabric of each Phase 3 trial should be viewed both as an opportunity to understand the mechanistic basis for clinical success or failure and to probe phage biology in rigorously controlled clinical interventions. The former offers those conducting phage clinical investigations opportunities to improve on clinical outcomes in future interventions by defining factors that contributed to treatment failure. Inadequate exposure of the pathogen at sites of infection can be modified by changes in the dosing regimen. Rapid emergence of phage resistance might be addressed by systematic studies of phage-phage and phage-antibiotic combinations that complicate bacterial evolution. Anti-phage immunity could be addressed by designing phage cocktails that include phages of differing immunologic profiles. Better understanding of phage biology can provide insights into whether specific physiological characteristics of phages are desirable for therapeutic phages. Such knowledge can contribute to the informed selection of environmental phages for clinical trials and to prioritizing phage engineering efforts.

Development of the field phage therapeutic research as an iterative, multidisciplinary, hypothesis-driven enterprise provides an unparalleled opportunity to accelerate the successful development of phage therapeutics and should be a high priority for those engaged in therapeutic phage research from the bench to the bedside. (Figure 2) [57]. Integration of translational research into the clinical development paradigm of phage therapeutics will require that those designing and supporting the therapeutic bacteriophage research effort assimilate and apply what has been learned from nearly 100 years of antibiotic development. One of the major challenges in this regard is that clinical trials have disproportionately been designed and funded by relatively small biotechnology companies that have neither the resources nor the longer-term visions available to government funding agencies and that have (at least in the past) characterized antibiotic development by the legacy pharmaceutical industry. Governmental agencies and philanthropic entities have the responsibility to provide programmatic coordination, resources and leadership to enable the development of standardized assays, integrated platforms for phage clinical trials that transcend individual academic laboratories and biotechnology companies and repositories that enable exchanges of reagents and biological samples. Such support catalyzed the HIV therapeutic research effort in the last two decades of the twentieth century and recently accelerated the global COVID-19 response and was critical to transforming both HIV and SARS CoV-2 from highly lethal pathogens to ones associated with substantially less morbidity. Similar investments and research coordination support would transform the field of bacteriophage therapeutics. The AMR crisis deserves no less.

• A substantial body of work indicates that bacteriophage therapy poses few safety risks.

• The current primarily academic laboratory-based ‘cottage industry’ providing access to bacteriophage products under expanded access regulatory oversight mechanisms has demonstrated antimicrobial activity and clinical benefit in some, but not all, patients reported.

• Although many patients have benefitted from these expanded access efforts, the full benefits of phage therapy will not be realized until the underlying principles of phage therapy are fully delineated in rigorous clinical trials and commercialization enables access at the scale required to meet the global need.

• Clinical development of phage therapeutics can be substantially accelerated by investment in multidisciplinary research that integrates laboratory-based studies of phage biology and engineering with human studies that include substantial translational research components.

• Development of translational research tools and the quality assurance mechanisms that are essential to their refinement and use requires broad collaboration among academic, industrial and government scientists that is backed by sustained support from governmental and philanthropic funding agencies.

Dr. Schooley has provided consultive advice to SNIPRBiome, Locus Biopharmaceuticals and BiomX.

Open access for this article was enabled through a transformative open access agreement between Portland Press and the University of California.

AMR

antimicrobial resistant

ASM

American Society for Microbiology

ESCMID

European Society for Clinical Microbiology and Infectious Diseases

GMP

Good Manufacturing Process