Immune Responses to Phage Therapy in Humans: A Review

Abstract

This review explores mammalian immune responses to phages with a particular emphasis on human immune responses to therapeutic phages and their potential implications for the outcomes of phage therapy. Despite the ubiquity of phages in the human microbiome, particularly in the gut (the phageome), research on immunological mechanisms governing immune responses to both endogenous and therapeutic phages are still in their infancy. We highlight key components of the immune system that contribute to clearance of phages in vivo and examine how various factors– including patient-specific variables, treatment regimens and phage characteristics can influence immune responses and, consequently, phage pharmacokinetics during therapy. A clearer understanding of human immune responses to phages is urgently needed to inform the development of more targeted and effective personalized phage therapies—an essential step in combating the escalating threat of antimicrobial resistance.

This review explores how patient-specific, treatment-related, and phage-intrinsic factors could fine-tune the immune responses to phages and phage therapy outcomes and underscores the lack of adequate human data on antiphage immune responses.

In light of growing interest in bacteriophage therapy as 1 potential strategy to tackle antimicrobial resistance (AMR), phage researchers are keen to further explore human immune interactions with phages to understand the pharmacological and immunological properties of phages, and to inform their effective, personalized use in treating infections [1]. In most countries, phage therapy is currently administered under compassionate use protocols as an adjunct to antibiotic treatment [2]. To facilitate integration of phage therapy into mainstream modern medical practice as a viable antimicrobial strategy, carefully designed clinical trials are essential for generating robust evidence that addresses existing gaps in both basic and clinical science. Regulatory hurdles persist because conclusive evidence from well-designed randomized controlled trials (RCTs) demonstrating the efficacy of phage therapy is limited [3]. Given that phage therapeutics are living biologics and often require personalized formulations, conventional “one-size-fits-all” clinical trial designs are poorly suited to this modality. Most available evidence informing regulatory frameworks to date are derived from case reports and uncontrolled studies. Insightful data from such studies could be utilized to improve future RCT design to provide robust data on efficacy indicators (eg, evidence of in vivo phage replication, microbiological outcomes, clinical response), dosage, route of administration and duration of therapy.

While clinical trials are essential for establishing safety and efficacy in humans, animal studies remain critical for addressing mechanistic and translational questions that cannot be fully explored in limited heterogeneous patient populations. Use of suitable animal model systems are necessary to answer hypothesis-driven biological questions relevant to the clinical context, closely representing in vivo events occurring in patients during the course of clinical phage applications and enabling the study of phage interactions with host immunity, infection dynamics, and antibiotics. Human immune responses to phage therapy is one of the critical in vivo interactions that could potentially impact the pharmacokinetics (PK) of phage therapeutics. There could be multiple factors such as mammalian host factors, treatment factors as well as phage factors that might possibly modulate this immune response, and how each of these factors contributes to the induction of phage-specific immune response and their implications on clinical outcomes are poorly characterized in humans. This review summarizes insights from the literature primarily from preclinical and experimental data on mammalian immune responses to phages and how they may influence their therapeutic outcome. This review underscores the lack of adequate human data on antiphage immune responses and calls for more extensive research in this area.

IMMUNE RESPONSE INITIATION AND PROGRESSION DURING PHAGE THERAPY

In general, mammalian immune responses to phages are, to some extent, similar to those elicited by other eukaryotic viruses. Components of the nonspecific innate immune system can recognize phage elements such as nucleic acids via Toll-like receptors (eg, TLR9, TLR3), thereby triggering antiviral pathways [4, 5]. However, the extent and biological significance of these antiviral and proinflammatory pathways at cellular level remain unclear. Strong activation has been demonstrated in germ free mice [4]; these animals, however, lack normal stimuli from any microbiota, presenting immature and “silenced” immunity, which is prone to induction of inflammatory pathways when exposed to microbes or microbial DNA [6, 7]. Often, proinflammatory action of phages was observed in phage preparation containing bacterial impurities. Even when equivalent concentration of lipopolysaccharide was applied as a control, other pattern recognition receptors (PRRs), including TLRs, can be stimulated by unquantified pathogen associated molecular patterns (PAMPs) like peptidoglycan, bacterial lipoproteins, DNA, metabolites, and others [8, 9]. Notably, a recent study by Węgrzyn group showed that activation of antiviral responses to phage via TLR9 was impaired, likely due to incomplete compatibility of phages with virus-detecting intracellular PRRs [10].

Phages, unlike eukaryotic viruses, do not infect and replicate in mammalian cells. They can enter human cells, like epithelial cells in the gut via endocytosis or transcytosis, but do not generally trigger CD8+ T-cell responses [11]. Instead, CD4+ T cells mediate adaptive immunity by producing phage-specific antibodies, beginning with immunoglobulin M (IgM) and later immunoglobulin G (IgG) upon repeated exposure [12, 13]. Endogenous phages, which are abundant at sites colonized with commensal bacteria (eg, skin, gastrointestinal, urogenital, and respiratory tracts), seem to be generally well-tolerated [14, 15]. In the 1960s, non-therapeutic phages like φX174 were used to study immune deficiencies [16]. Historical reports of phage therapy in the 1980s revealed phage-specific responses [17], as well as those on modern therapeutic applications of phage [12, 18]. Therapeutic phages, especially when administered intravenously at high doses (≥10⁹ PFU/ml), are expected to trigger antibody responses within 1–2 weeks, consistent with the typical kinetics observed following exposure to other exogenous antigens. However, immune responses during therapy differ due to complex interactions involving pathogens, phages, and host factors. Variables such as antigen exposure, immunological priming, phage replication, and pathogen-derived adjuvants likely influence these responses, though the underlying mechanism remain unclear.

PHAGE CLEARANCE

Both innate (nonspecific) and adaptive (specific) immune responses play an integral role in clearing phages from the circulation. Phagocytes, particularly within the mononuclear phagocyte system (MPS) (old term: reticuloendothelial system, RES) [19] and phage-specific antibodies are the primary components of innate and humoral immunity, respectively, that contribute to the inactivation and clearance of phages from the circulation and tissues. They are supported by the complement system which can act both nonspecific or cooperating with specific antibodies [19–23] (Figure 1).

Figure 1.

Main immune components of the mammalian immune system that contribute to phage clearance from circulation, influencing phage pharmacokinetics (Created in BioRender).

Phagocytes

The MPS, comprising phagocytic cells (mainly monocytes and macrophages) in the liver, spleen, lymph nodes, blood, and many other tissues, plays a critical role in clearing foreign particles, including phages, from circulation [23–25]. Liver sinusoids are lined by specialized scavenger endothelial cells that have recently also been demonstrated as actively eliminating phages from circulation [26]. Studies in germ-free mice show that the MPS is responsible for over 90% of phage clearance, even without antibody involvement [27, 28]. Phagocytes serve as key players of the continuous immune surveillance in the mammalian system. The phagocyte surface is populated with a variety of receptors capable of recognizing diverse molecular patterns, thereby triggering the process of engulfment [29]. Phage surface protein modifications can impact immune recognition and clearance [22, 28].

Following internalization by phagocytes, phages are sequestered within phagosomes that undergo progressive maturation and fuse with lysosomes via the conventional endolysosomal trafficking pathway [29]. During this process, phagosomes fuse with lysosomes to form phagolysosomes, where the phage virion particles are processed and degraded. This process exposes antigenic epitopes for presentation to B cells and predominantly CD4⁺ T cells, thereby triggering phage-specific adaptive immune responses [12, 30]. Concurrently, both cell surface and intracellular PRRs such as TLRs within endosomes, particularly TLR9 (unmethylated DNA sensor) and TLR3 (RNA sensor), recognize phage components and initiate innate immune signaling pathways [30] (Figure 2). Phages modulate immune responses by engaging different innate signaling pathways, but either pro-inflammatory or anti-inflammatory effects, as well as antiviral cytokines have been reported [4, 5, 31].

Figure 2.

Overview of innate and adaptive immune responses to phages (Created in BioRender); Abbreviations: PAMPs: pathogen associated molecular patterns; PRR, pattern recognition receptors; TLR9, toll-like receptor 9; TLR3, toll-like receptor 3; TRIF, TIR-domain-containing adapter-inducing interferon-β; MyD88, myeloid differentiation primary response 88; TNFα, tumor necrosis factor; APC, antigen presenting cell; MHC-II, major histocompatibility complex class II; TCR, T-cell receptor; IL-6, interleukin-6; IFN, interferon; IL-10, interleukin-10.

Complement Proteins

Complement proteins, key components of innate immunity, can mediate in vivo phage inactivation either via phage-specific antibodies on its classical pathway or without antibodies on the alternative pathway (Supplementary Table 1). Complement can enhance antibody-dependent neutralization [32]. While complement can trigger antibody-independent lysis of enveloped viruses, therapeutic phages are predominantly nonenveloped, and neutralization may occur primarily through nonlytic processes [33, 34].

Phage-specific Antibodies

Phage-neutralizing activity can be evaluated using in vitro neutralization assays where patient sample is incubated with phages and the reduction in phage titer is assessed in terms of number of lytic plaques developed on lawns of bacteria (a standardized reference strain) in serial dilutions which are then compared with that of phage-only control. This method of quantification is widely referred to as the “efficiency of plating.” Antiphage activity of sera can be analyzed as the rate of phage inactivation (K) that allows us to standardize and compare phage inactivation rates accounting for serum dilution, phage-serum reaction time, phage titer at the beginning and at the end of the reaction [35]. The inhibitory effect of antiphage serum is largely attributed to antibodies, though complement proteins may also contribute to the effect [18, 22], so identification of neutralizing effects of antibodies on phages routinely includes complement inactivation in tested sera [32].This can be applied either in complement-inactivated or in fully functional sera. Phage-specific antibodies such as IgG, IgM, and IgA can neutralize phages through various mechanisms [21, 36] (Figure 3). Indirect enzyme linked immuno sorbant assay can be used to monitor phage-specific antibody levels during therapy, quantifying antibody classes or subclasses using whole phages as antigens [20, 37]. Monitoring neutralizing antibody responses during phage therapy can help define treatment window before adaptive immunity may potentially limit phage access to the infection site [37, 38].

Figure 3.

Possible mechanisms by which phage neutralizing antibodies could potentially block phage infectivity (Created in BioRender); A, Antibodies can bind to phage virions, allowing adsorption but hindering nucleic acid injection and infection. B, Antibodies may interact with phage proteins at locations other than receptor binding sites but block adsorption due to steric hindrance. C, Multivalent antibodies may agglutinate phage virions, reducing the number of infectious units. D, Antibodies can impede phage adsorption by binding to phage proteins crucial in bacterial adsorption. E, Antibodies deposited on the phage surface can trigger complement-mediated neutralization. F, Antibodies can also bring about neutralization by signaling other elements of the immune system toward the phage virions via antibody mediated opsonization.

VARIABLES ALTERING PHAGE CLEARANCE RATE

The rate of phage clearance mediated by both the innate and adaptive immune response may differ between patient populations depending on multiple factors (Figure 4).

Figure 4.

Key factors that may alter the rate of phage clearance in circulation (Created in BioRender).

Patient Factors

Patient immune status, whether immunocompetent or immunocompromised, may influence phage therapy outcomes by affecting phage clearance rates. However, there is a paucity of studies that have systematically investigated this aspect in clinical settings. Both in vivo and in silico models have been developed to explore how immune responses influence therapy success [24, 32, 39, 40]. Immunocompetent hosts may benefit from “immunophage synergy” according to a preclinical study in mouse model of acute pneumonia, where host immunity and lytic phages work together to enhance pathogen clearance [39, 41]. However, successful outcomes in immunocompromised individuals have been reported, suggesting the relationship is not straightforward. The diversity of immune-compromising conditions makes it difficult to determine how phage therapy interacts with the immune system [42]. While evaluating immunological criteria for phage therapy is valuable, more research is needed to understand these interactions and develop personalized phage therapy strategies.

Prior exposure to a therapeutic phage or related phage can influence clearance rate due to pre-existing antibodies and immunological memory [17, 43, 44]. Hence, testing for pre-existing phage-specific antibodies in serum may help with phage selection. However, pre-existing antibodies may not always neutralize phages [37]. Non-neutralizing antibodies are those that can recognize and bind to viruses but do not prevent infection of host cell. These antibodies are generally known to be elicited during viral infections (eg,; influenza, rotavirus, cytomegalovirus, HIV, and SARS-CoV-2) [45].

Different infectious syndromes or pathological conditions may also modulate the immune response against phages. Lipopolysaccharides/endotoxins on Gram-negative bacteria or lipoteichoic acids and/or exotoxins of Gram-positive bacteria appear to be potent stimulators of immune activation [46–48]. Infectious syndromes caused by such bacterial pathogens as well as effects of their PAMPs on the immune system may alter responses to other antigens (such as a therapeutic phage) [32, 49]. Furthermore, the course of the infection, whether acute or chronic, may alter host immune responses to phage therapy through modification of the general immune state of the host [50]. Lastly, there are certain anatomical sites of immune privilege, such as the eye and brain. Surprisingly, certain phages (eg,; anti-Escherichia coli and anti-Shigella phages) administered via intramuscularly/intraperitoneally in mice and chickens were found to cross the blood-brain barrier and thus enter sites of immune privilege [51, 52]. However, it remains unclear whether this is common to most phages or what specific phage characteristics facilitate their passage through this highly selective barrier, which typically restricts the entry of most foreign particles. The immune responses elicited against phages are expected to be blunted within such microenvironments [53, 54].

Treatment Factors

Route of Administration

Selecting the appropriate route for phage administration is crucial to facilitate access to bacteria within the body. Phages could be delivered orally, inhaled, topically, including via instillation, or intravenously. Oral administration shows poor penetration into the systemic circulation compared with injection methods, leading to a lower likelihood of developing a humoral response [13, 23, 55, 56]. Intravenous (IV) administration involves direct introduction into the blood, enabling rapid dissemination to organs and tissues within minutes, which makes it a preferred choice in human applications, especially for systemic bacterial infections [55, 57]. The likelihood of neutralizing antibody induction may vary depending on the route of administration [2, 12, 37, 58, 59].

Phage Dosing and Duration of Treatment

Phage dosing and treatment duration should be optimized to ensure bacterial clearance, minimize resistance, and maintain safety and cost-effectiveness. A recent rodent study showed a saturation-like effect: lower phage doses were cleared faster and accumulated mainly in the liver, while higher doses persisted longer and appeared in the spleen, likely exceeding hepatic clearance capacity [26]. It is likely that phage dosing and treatment duration directly influence immune responses. In turn, immune responses are key parameters influencing phage PK [58, 60]. The influence of immune responses upon the number of circulating phages during phage therapy can be studied by correlating phage kinetics data with phage neutralization assay results and phage-specific antibody levels from phage therapy patients [61, 62]. Broadly speaking, there are 2 approaches to dosing in phage therapy [63]. The first involves providing a high dose of phages, known as inundative phage titers, to rapidly reduce the bacterial burden through direct lysis brought about by administered doses. The alternative strategy takes advantage of the self-replicating nature of phages. A lower titer of phage is administered, and bacterial killing is achieved by the administered phage dose but also through the subsequent production of phage virions through successful infection of the target pathogen (auto-dosing). This continuous production of phages in situ helps reduce the bacterial burden over time [64]. It is likely that induction of immune responses varies between these dosing strategies. Auto-dosing may elicit a weaker immune response, which could be beneficial for avoiding rapid phage neutralization; however, low initial titers risk being cleared by the immune system before reaching the infection site. Therefore, auto-dosing may be more suited to topical applications, whereas for systemic (eg, IV) use, higher phage doses may be preferable to ensure sufficient delivery and bactericidal activity [8]. Furthermore, delivering effective phage doses within a 1–2-week timeframe before the onset of a phage-specific antibody response would be ideal. However, in a clinical scenario, several factors, including the patient's immune status, and the nature of the infectious syndrome, may affect the duration of treatment. For example, chronic infections with slowly replicating bacteria, especially those encountered within granulomas (eg, infections caused by Mycobacterium sp., Listeria, Salmonella), are hypothesized to need prolonged phage exposure [37, 41, 65, 66]. Chronic bacterial infections can also induce a stress response, leading to phage tolerance and necessitating longer treatment or higher doses, increasing the risk of antiphage immune responses [67].

Phage Factors

The genetic makeup of phages and the resulting pattern of amino acid residues in various phage proteins can also impact how the immune system recognizes and clears them [22]. A Lambda phage mutant with a capsid E protein mutation, changing an acidic to a basic amino acid residue, significantly reduced phage clearance in mice. This effect was proposed to result from different susceptibilities of wild type phage and its mutant to elimination by MPS [28], but later a role of complement system has also been proposed [22]. Phage virion size and morphology may influence MPS capture and clearance rates, but evidence is limited [21]. The immunogenicity of phage proteins has been examined in the model system T4 E. coli phage, indicating variations in the ability of phage capsid proteins, namely Hoc (highly antigenic outer capsid protein) or gp23 (major capsid protein), and the less antigenic Soc (small outer capsid protein) or gp24 (pentameric corner protein), to induce antibody responses [43]. Immune responses could be highly phage-specific and can significantly differ between phages, for example E. coli phage JIPh_Ec70 eliciting a strong STING-mediated inflammatory response at the cellular level, while Klebsiella pneumoniae phage JIPh_Kp127 provokes minimal immune activation [68].

In the context of phage therapy, phage ability to induce specific antibodies raises concerns about therapeutic efficacy due to potential neutralization or cross-neutralization of therapeutic phages, and hypersensitivity. Limited research has examined the immunogenicity of therapeutic phages or the impact of structural or genomic differences. Berkson et al [38] investigated phage-specific immunity against phage cocktail targeting gut colonization by vancomycin-resistant Enterococcus in mice and reported strong neutralizing antibody responses against myophages in the cocktail. Similarly, variable phage-specific responses were reported in a clinical case of Mycobacterium abscessus infection where an 81-year-old patient with M. abscessus infection received a 6-month IV phage cocktail containing Muddy, BPsΔ33HTH_HRM10, and ZoeJΔ45. All 3 phages were immunogenic, producing IgM, IgG, and IgA antibodies, with the serum showing stronger neutralization of Muddy. Western blotting revealed robust IgG recognition of Muddy's capsid and major tail proteins, while BPsΔ and ZoeJΔ exhibited weaker recognition [37]. However, a key limitation of this approach is the use of denatured gels in western blotting, which may disrupt native protein structures and overlook antibodies that bind conformational epitopes. While these studies highlight variability in antiphage antibody responses, the underlying factors remain unclear. Factors such as variations in susceptibilities of pathogen strains to phages and phage amplification in vivo which might result in generation of antigenic content exceeding that of the administered doses could also potentially play a significant role in shaping the immune response. The foreignness of phage epitopes is another factor that may contribute to differences in phage immunogenicity; it is plausible to hypothesize that endogenous phages, naturally present within the human microbiome often referred as “phageome,” have co-evolved with the host and may often recognized as nonthreatening, may be subject to some degree of immunological tolerance, leading to reduced immunogenicity or minimal immune activation [4, 14, 15, 24, 31, 69, 70].

ADDRESSING PHAGE IMMUNE CLEARANCE TO IMPROVE PHAGE THERAPY OUTCOMES

Designing optimal phage therapy requires strategies to mitigate immune clearance, taking into account patient-specific factors, treatment protocols, and phage characteristics. One approach involves designing long-circulating phages with extended half-lives in the bloodstream to improve the delivery to infection sites. This may involve tailoring therapeutic phages with specific genetic modifications that enable them to evade fast elimination or conjugating phage proteins with the non-immunogenic polymer to protect phages from inactivating environment in vivo [71]. Alternatively, encapsulating phages in liposomes can protect them from rapid immune clearance and neutralization by antiphage-specific antibodies [72, 73]. Identifying immunodominant phage epitopes targeted by neutralizing antibodies could guide the selection or engineering of phages with reduced immunogenicity. Cross-reactivity between related phages is a significant clinical consideration due to prior immune exposure from environment. Moreover, cross-reactivity may occur even between distant phage strains or between phages and non-phage antigens due to structural similarities [20, 38]. In cases where adaptive immune responses against therapeutic phages impair efficacy, switching to alternative phages with the same host specificity may be viable, though cross-neutralization risks must be evaluated. Advancing research into phage immunogenicity and cross-reactivity will be key to developing personalized and resilient phage therapy strategies. Given the lack of robust and systematic data on immune responses to phage therapy in humans, there is currently no evidence-based practical guidance or conceptual framework for clinicians to decide on how frequently immunity should be assessed, which standardized assays should be used to monitor immune responses, the implications of these responses for phage therapy outcomes, and the appropriate course of action if antiphage immunity is detected (Supplementary Table 2). Centers worldwide that provide phage therapy services under compassionate use could gather limited yet critical human data from protocolized real-world patient cohorts to guide the design of future RCTs aimed at generating robust evidence [74].

CONCLUSION

While phage therapy presents a potential strategic approach to the growing threat of AMR, it is still challenging to predict the in vivo outcomes of phage therapy. Beyond the urgent need for clinical trial data to support evidence-based guidelines for dosing and treatment duration, it is equally important to monitor individual patient response to better understand the factors contributing to therapeutic success or failure. Immune responses play a critical role in shaping phage clearance rates, as demonstrated in both animal models and human studies. While immune clearance can reduce phage efficacy, it does not necessarily preclude successful treatment. Detailed immunological profiling of individual patients, combined with outcome data from controlled clinical trials, will be essential to unravel how patient-specific, treatment-related, and phage-intrinsic factors fine-tune the immune responses and influence therapeutic outcomes. This integrative approach can bridge the gap between in vitro and in vivo performance of phages, ultimately guiding the development of more personalized and effective phage therapy strategies.

Method

To identify relevant literature for this narrative review, a search strategy was implemented using combinations of keywords such as “phage therapy,” “mammalian host-phage interactions,” “phage immunogenicity,” “phage pharmacokinetics,” “therapeutic phage monitoring,” “phage-specific antibodies,” “phage dosing,” “phageome,” and “immune response” within the PubMed database. The search was limited to peer-reviewed original research articles and review papers published in English from 1931 to 2025. After the initial search, titles and abstracts of the retrieved articles were screened to determine their relevance. Subsequently, full-text articles were reviewed to evaluate their eligibility based on the following criteria: relevance to the topic, scientific rigor, and overall contribution to the field.

Supplementary Data

Supplementary materials are available at The Journal of Infectious Diseases online (http://jid.oxfordjournals.org/). Supplementary materials consist of data provided by the author that are published to benefit the reader. The posted materials are not copyedited. The contents of all supplementary data are the sole responsibility of the authors. Questions or messages regarding errors should be addressed to the author.