A 12-Laboratory Study Brings Phage Susceptibility Testing Closer to Routine Diagnostics
One of the practical obstacles preventing bacteriophage therapy from becoming a routine clinical option is surprisingly familiar to clinical microbiologists: before treating an infection, laboratories need a reliable way to determine whether the pathogen is actually susceptible to the proposed antimicrobial agent.
For antibiotics, this problem has been addressed through decades of standardization. Clinical laboratories routinely follow harmonized procedures developed by organizations such as EUCAST and CLSI, allowing an Escherichia coli isolate tested in one hospital to be assessed under conditions comparable with those used elsewhere. Phage therapy does not yet have an equivalent universally accepted system.
A new multicenter study published on October 12, 2026 in the Journal of Clinical Microbiology moves the field a step closer to that goal. Researchers led from University Medical Center Rostock evaluated a standardized disk-based phage susceptibility test, known as Disk-PST, across 12 routine diagnostic laboratories in Germany. Thirty-nine participants independently performed the assay under blinded conditions, and their results showed almost perfect agreement with a reference plaque assay conducted in Rostock.
The study does not establish Disk-PST as a universal clinical standard, nor does it demonstrate that a positive result predicts whether phage therapy will succeed in a patient. What it does demonstrate is something that must come first: the same relatively simple phage test can generate highly reproducible results when performed by different operators in different diagnostic laboratories.
That distinction matters because phage therapy is unusually dependent on microbiological matching. Many therapeutic bacteriophages have relatively narrow host ranges, sometimes distinguishing between bacterial strains belonging to the same species. A phage capable of efficiently infecting one E. coli clinical isolate may be ineffective against another because of differences in surface receptors, capsule composition, bacterial defence systems or other biological factors.
Selecting a phage therefore generally requires testing the patient's bacterial isolate against available viral candidates. This process is often called phage susceptibility testing, or PST, and the resulting profile is sometimes referred to as a phagogram.
Several methods are already available. The traditional spot test applies a small volume of phage suspension directly onto a bacterial lawn and looks for an area of clearing. The double agar overlay plaque assay provides much stronger evidence that a phage can productively infect its host by detecting individual plaques generated through cycles of infection and replication. Growth-kinetic assays can follow changes in bacterial density over time in the presence of different phage concentrations.
Each approach can provide useful information, but methods vary substantially between laboratories. Agar composition, bacterial inoculum, phage concentration, incubation conditions, interpretation criteria and even the volume of a phage spot can influence the result. A laboratory reporting that a bacterium is “susceptible” may therefore not necessarily be using exactly the same biological definition or experimental procedure as another laboratory making the same determination.
Disk-PST was designed to reduce that variability by translating phage testing into a format already familiar to laboratories performing conventional antimicrobial susceptibility testing.
Instead of pipetting phage directly onto the agar surface, researchers place sterile 6-mm cellulose disks onto a standardized bacterial lawn and apply the phage suspension to the disk. The concept resembles antibiotic disk testing operationally, although the biology is fundamentally different: antibiotics are diffusible chemical compounds, whereas bacteriophages are large, self-replicating biological particles whose activity depends on interactions with living bacterial hosts.
In the protocol tested in the new study, bacteria were first grown overnight and suspended in saline to a turbidity equivalent to a 0.5 McFarland standard. A defined volume of the bacterial suspension was spread onto Mueller-Hinton agar, the same broad type of standardized medium familiar to diagnostic laboratories performing antimicrobial susceptibility testing.
Cellulose disks were then placed onto the bacterial lawn, and 50 microlitres of phage suspension were applied to each disk. After a short drying period, plates were incubated for 18 to 24 hours at 35 ± 1°C.
The readout is deliberately simple. A clear, well-defined zone of bacterial lysis around a disk is interpreted as lytic activity. Absence of visible clearing is classified as non-lytic, while ambiguous patterns can be reported as inconclusive.
The simplicity is intentional. Disk-PST is not designed to replace every sophisticated method used to characterize phage–host interactions. It does not directly quantify burst size, adsorption efficiency, replication kinetics or the precise efficiency with which a phage propagates on a bacterial strain. Instead, it is intended to provide a standardized first-line answer to a narrower question: under defined conditions, does this phage reproducibly lyse this bacterial isolate?
The Rostock team initially tested the protocol internally before distributing it to other laboratories. Twenty-eight participants, including academics and laboratory technicians, performed the assay during this first stage. The results were compared with the double agar overlay plaque assay used as the reference method.
The internal validation showed strong agreement but also identified practical problems that could affect routine use. Cross-contamination between phage samples was one issue, prompting the researchers to supply separate phage aliquots and recommend filtered pipette tips. Another challenge involved interpreting small bacterial microcolonies appearing within otherwise clear zones of lysis.
After these adjustments, the researchers organized a nationwide inter-laboratory comparison involving 12 diagnostic laboratories across Germany. The participating sites included university hospitals, private medical laboratories, a tertiary-care hospital and a government testing laboratory, providing a more realistic test of whether the procedure could travel beyond the group that originally developed it.
Thirty-nine volunteers participated, consisting of 18 academics and 21 technicians. Testing was performed during May 2025, and each laboratory received a standardized kit containing bacterial strains, phages, sterile disks and other required materials together with a standard operating procedure and instructional video.
The biological panel was deliberately small. Researchers used one Escherichia coli phage, UMR_Zora, one Staphylococcus epidermidis phage, UMR_Minerva, and SM buffer as a negative control. These were tested against four bacterial strains: two E. coli and two S. epidermidis, including the propagation hosts and clinical isolates.
The experiment included both biological and technical replication. Participants performed two technical replicates on two separate test dates, allowing the investigators to examine whether results remained consistent not only between laboratories but also between repeated performances of the assay.
The results were strong.
Across the four inter-laboratory datasets, Cohen's kappa values comparing Disk-PST with the reference plaque assay ranged from 0.925 to 0.939. Values in this range indicate near-perfect categorical agreement.
Sensitivity ranged from 92.25% to 94.87%, while specificity ranged from 97.76% to 98.72%. Positive predictive values ranged from 98.67% to 100%, and negative predictive values from 99.35% to 100%.
These numbers are particularly relevant because the assay was not being performed exclusively by researchers specializing in bacteriophages. It was deliberately introduced into routine diagnostic environments and used by personnel with different professional backgrounds.
The results therefore address one of the major requirements for any clinical diagnostic method: reproducibility outside the laboratory that created it.
A technically impressive assay has limited clinical value if it requires extensive specialist expertise or if interpretation changes dramatically between operators. Diagnostic tests intended for broad deployment must survive transfer between hospitals, equipment setups and laboratory personnel without losing reliability.
Disk-PST benefits from being built around procedures microbiology laboratories already know. Preparing a standardized bacterial suspension, inoculating agar, placing disks and reading zones around those disks resemble workflows already performed routinely for antibiotic susceptibility testing.
The analogy should not be taken too far. The researchers explicitly emphasize that Disk-PST is not biologically equivalent to antibiotic disk diffusion.
An antibiotic molecule diffuses through agar according to physicochemical properties, producing a concentration gradient that can be related to susceptibility thresholds. A bacteriophage behaves differently. It must encounter the correct bacterial receptor, successfully initiate infection and potentially replicate within host cells. Its apparent activity can be affected by bacterial density, environmental conditions, particle diffusion, host physiology and viral replication.
For that reason, Disk-PST currently provides a categorical lysis readout rather than the standardized numerical clinical breakpoints available for many antibiotics.
This is one of the major unresolved questions for the entire field of phage susceptibility testing. A laboratory can observe that a phage lyses a bacterial strain in vitro, but the clinically relevant threshold separating a likely effective treatment from an ineffective one has not been established across phages, pathogens and infection types.
The Rostock study therefore validates analytical reproducibility rather than clinical predictive validity.
This is an essential distinction. High agreement with a plaque assay shows that different laboratories can identify the same phage–bacterium interactions under standardized conditions. It does not demonstrate that patients whose isolates produce a positive Disk-PST result will necessarily respond to treatment.
Phage pharmacology introduces many variables after laboratory testing. A phage may be highly active on agar but fail to reach sufficient concentrations at an infection site. The immune system may eliminate it rapidly. Biofilms, mucus, low bacterial density or unusual physiological states can alter viral replication. Antibiotics administered simultaneously may enhance or inhibit phage activity depending on the combination.
Clinical validation will therefore require prospective studies linking PST results with patient outcomes.
The authors specifically identify this as a future priority. Further trials will need to determine how different laboratory patterns correlate with therapeutic success or failure and whether particular interpretive criteria should be used when selecting phages.
Another limitation involves the diversity represented in the current inter-laboratory study. Only two bacteriophages and four bacterial strains were included in the ring trial. Both phage–host combinations had been pre-characterized before distribution to ensure that the experiment could focus on reproducibility.
This is scientifically reasonable for a first multicenter validation, but real phage libraries contain enormous biological diversity. Phages targeting Pseudomonas, Klebsiella, Acinetobacter, Staphylococcus, Enterococcus or other pathogens can produce very different plaque morphologies and infection dynamics. Bacterial capsules, slow growth, mucoid phenotypes and unusual media requirements may further complicate interpretation.
The excellent performance demonstrated here therefore cannot automatically be generalized to every clinically relevant phage–bacterium pair.
One specific interpretation problem encountered during the study was the appearance of microcolonies within lysis zones. These small areas of bacterial growth can represent partial regrowth or heterogeneity within the population and make a simple positive–negative classification more difficult.
A second issue is “lysis from without.”
At very high phage concentrations, large numbers of viral particles binding to bacterial cells can damage the cell envelope and cause visible lysis without productive phage replication. A conventional spot test can therefore sometimes show a zone of clearing even when the phage does not establish the type of productive infection that would generally be desired for therapeutic amplification.
The study used plaque assays during the characterization stage to help rule out this phenomenon for selected phage–host combinations. In routine practice, however, ambiguous lysis patterns may still require confirmatory testing.
Cross-contamination represents another unusually important problem. Phages are self-replicating entities. A very small number of contaminating infectious particles can multiply dramatically when transferred onto a susceptible bacterial lawn, potentially generating a false-positive result. The investigators therefore emphasize strict handling procedures and quality control.
These limitations support a model in which Disk-PST becomes a first-line screening tool rather than the only laboratory method used for therapeutic decision-making.
The authors propose a decentralized–centralized diagnostic structure. Routine microbiology laboratories could perform standardized Disk-PST screening against available therapeutic phages, while bacterial isolates and promising phage combinations could then be transferred to specialized phage centres for more detailed evaluation.
Reference laboratories could perform plaque assays, growth-kinetic studies, phage–antibiotic interaction testing or other specialized analyses before a personalized product or cocktail is prepared.
Such a system would resemble established diagnostic networks in other areas of microbiology, where common tests are performed locally and more technically demanding assays are concentrated in specialist centres.
The attraction for phage therapy is scalability.
If every hospital interested in using bacteriophages needs its own highly specialized research laboratory to identify active phages, personalized treatment will remain difficult to deploy widely. A simple assay that uses equipment, media and technical skills already present in routine clinical microbiology laboratories could greatly expand the number of institutions capable of participating in the initial selection process.
The development also comes as European organizations are beginning to address phage susceptibility testing systematically.
EUCAST has established a dedicated Phage Susceptibility Testing Subcommittee composed of experts from multiple countries. Its objectives include developing reproducible reference methods, establishing interpretation criteria, defining quality-control systems and providing training for laboratories conducting phage testing.
In June 2026, members of that group published a position paper specifically addressing the need for a reference methodology for bacteriophage susceptibility testing. EUCAST currently reports that its PST subcommittee includes 23 members from 16 countries and is conducting a pilot trial involving multiple laboratories as part of the broader standardization effort.
The German Disk-PST study therefore arrives at a particularly relevant moment. Rather than solving the standardization problem independently, it provides experimental evidence that one relatively simple agar-based format can remain reproducible across a network of routine laboratories.
The method itself originated several years earlier. In 2022, Romy Skusa, Katharina Ekat and colleagues at University Medicine Rostock published the first proof-of-concept version of the single-disk approach in Applied and Environmental Microbiology. That study explored bacterial inoculum density, phage concentration and application format in an attempt to turn conventional phage spotting into a procedure closer to routine diagnostic practice.
The 2026 study represents a different level of validation. The question is no longer simply whether Disk-PST works in Rostock, but whether independent laboratories can reproduce the same answer.
The results suggest that they can, at least for the controlled set of organisms and phages tested.
The assay may also have uses beyond identifying an initial therapeutic candidate. The researchers suggest that a standardized disk format could eventually help verify the biological activity of therapeutic phage preparations after transport between manufacturing centres and hospitals.
Future ready-to-use disks preloaded with defined phages could simplify the system further, although such products were not evaluated in the present study. Combining standardized phage testing with conventional antibiotic testing could also provide a route toward assessing phage–antibiotic combinations in a format accessible to diagnostic laboratories.
These applications remain prospective.
For now, the strongest conclusion from the study is more specific: a phage susceptibility test derived from familiar disk-based microbiology methods produced highly concordant results when performed by 39 people in 12 independent German laboratories.
That may sound like a technical achievement rather than a major therapeutic breakthrough, but diagnostic standardization is one of the foundations on which routine antimicrobial medicine is built.
Antibiotic therapy became scalable not simply because active compounds were discovered, but because laboratories developed reproducible systems for identifying pathogens, measuring susceptibility and communicating results using common standards.
Phage therapy faces a similar requirement, complicated by the biological diversity and strain specificity of its active agents.
A clinician considering a personalized bacteriophage needs confidence that a laboratory report identifying a candidate phage has a reproducible biological meaning. Researchers running clinical trials need comparable susceptibility data across centres. Regulators need defined methods capable of supporting consistent interpretation.
The Disk-PST study does not yet provide all of those elements. It lacks validated therapeutic breakpoints, has been tested against a limited biological panel and has not been prospectively linked to patient outcomes.
But it answers an important question that precedes them: can a simple phage test be transferred from a specialist centre into ordinary diagnostic laboratories without its result becoming unpredictable?
Across 12 German laboratories, the answer was largely yes.
The next challenge is considerably harder. The field must establish what those reproducible laboratory results mean for the patient.
Sources :
Ekat K., Scharnagl L., Hahn A. et al. — “Toward routine implementation of phage testing: an inter-laboratory comparison of a disk-based susceptibility testing (Disk-PST).” Journal of Clinical Microbiology, published online October 12, 2026. The study evaluated Disk-PST with 39 participants across 12 German diagnostic laboratories and reported κ values of 0.925–0.939, sensitivity of 92.25–94.87% and specificity of 97.76–98.72%. The article is published under CC BY 4.0.
https://doi.org/10.1128/jcm.00029-26
Journal of Clinical Microbiology — Full article and study protocol.
https://journals.asm.org/doi/10.1128/jcm.00029-26
Skusa R., Groß J., Kohlen J. et al. — “Proof-of-Concept Standardized Approach Using a Single-Disk Method Analogous to Antibiotic Disk Diffusion Assays for Routine Phage Susceptibility Testing in Diagnostic Laboratories.” Applied and Environmental Microbiology 88, e00309-22 (2022). This earlier study established the original standardized disk-based PST approach subsequently evaluated in the multicenter trial.
https://doi.org/10.1128/aem.00309-22
EUCAST — Phage Susceptibility Testing Subcommittee. EUCAST describes its programme to develop standardized reference methods, interpretation criteria, quality assurance and training for phage susceptibility testing.
https://www.eucast.org/about-eucast/subcommittees/subcommittee-pst/
Classen A.Y. et al. — “Towards a reference methodology for bacteriophage susceptibility testing — position paper of the EUCAST Phage Susceptibility Testing Subcommittee.” Clinical Microbiology and Infection (2026).
https://doi.org/10.1016/j.cmi.2026.06.005
Daubie V., Chalhoub H., Blasdel B. et al. — “Determination of phage susceptibility as a clinical diagnostic tool: A routine perspective.” Frontiers in Cellular and Infection Microbiology 12, 1000721 (2022). The review discusses spot testing, double agar overlay assays, Appelmans testing and growth-kinetic approaches, and the requirements for routine clinical PST.
https://doi.org/10.3389/fcimb.2022.1000721
Schooley R.T. et al. / broader review context — “Phage susceptibility testing methods or ‘phagograms’: where do we stand and where should we go?” Journal of Antimicrobial Chemotherapy (2024). The review highlights the need for a reference PST methodology and harmonized interpretation.
https://pubmed.ncbi.nlm.nih.gov/39271114/
.png){kind=logo}
Comments
Post a Comment