Bacteriophages — viruses that infect and kill bacteria — have been floated as an answer to antibiotic resistance for well over a century. Every few years the interest cycles back, driven by the same reality that dominates modern infection medicine: the drugs we have are getting less reliable, and the pipeline of new ones is thin. So the idea of unleashing bacteria’s own natural predators, custom-tuned to each patient’s infection, is genuinely attractive.
A new paper in Biocontaminant from Junya Zhang’s group at Shenyang Agricultural University argues that the picture we’ve been carrying around — phages as clean, targeted “living antibiotics” — is too simple, and that the real biology is more interesting.
The three stages
Instead of one relationship between phage and bacterium (predator eats prey), the authors describe three relationships that any phage-host pair can move between, depending on context:
1. The arms race. In this stage phages and bacteria are locked in constant back-and-forth co-evolution. Bacteria evolve defenses (CRISPR-Cas being the famous example, but also many others), phages evolve counter-defenses. Modern applications like CRISPR-based diagnostics and gene editing all emerged from studying this stage. Clinically, this is where phage therapy behaves the way you’d hope — the phage kills the pathogen, and the pathogen’s attempt to resist actually costs it fitness.
2. Protective coexistence. Less intuitive but well-documented: phages sometimes help their bacterial hosts, even resistant ones. Some phages integrate into the bacterial genome and hitch a ride while contributing useful genes (including, uncomfortably, resistance genes). In other cases the metabolic dance between phage and host can stabilize a resistant population rather than eliminate it. Phage therapy that lands in this stage doesn’t just fail — it can make things worse.
3. Ecological feedback. The final stage is the one where the environment holds the deciding vote. Nutrient availability, temperature, the presence of other microbes, biofilm structure, host immunity — all of it feeds back into whether the phage-bacteria dynamic ends in clearance, coexistence, or the phage becoming irrelevant. This is why lab results with phages so often fail to reproduce in animals, and why animal results fail to reproduce in real patients.
Why this matters clinically
If you take the framework seriously, it changes the practical questions you’d ask about phage therapy:
- Which stage are we in for this patient’s infection? A phage cocktail that would clear an acute wound infection might be actively harmful in a chronic biofilm setting where “protective coexistence” is the more likely outcome.
- What ecological conditions do we need to force the phage into “killer” mode? This may end up mattering more than the choice of phage itself.
- Should we be modifying the host environment (dietary, immune, wound care) alongside the phage dose? Probably yes.
The paper’s framing is less about “does phage therapy work?” and more about “under what conditions do phages become therapeutic weapons versus co-conspirators?” That’s a much better question.
What this means for us
Phage therapy is not yet a routine offering in Malaysian dermatology or STD practice — the compassionate-use cases you’ll read about are still overwhelmingly in tertiary academic settings, for infections where every antibiotic has failed. That’s likely to remain the case for a while. But phages sit adjacent to a lot of what we already do: chronic wound care, biofilm-associated infections, recurrent staph colonisation, and the slower half of antimicrobial resistance conversations.
A framework that makes clinicians and researchers more careful about when phages will help — and honest about when they might do harm — is genuinely useful. Bookmarking this one for the “watch this space” list.