Researchers at the National University of Singapore have engineered a common gut bacterium to carry a chemotherapy prodrug directly into tumors, converting it into the potent cancer-killing agent SN-38 at the tumor site in mouse models. The approach represents a growing class of experimental “living medicines” that exploit bacteria’s natural ability to colonize tumors, and it arrives as multiple independent teams race to solve the same problem, getting toxic drugs inside cancers without poisoning the rest of the body.
How a Gut Microbe Becomes a Drug Courier
The NUS team chose Lactobacillus plantarum, a bacterium already present in the human digestive tract, as its delivery vehicle. They modified the microbe to display streptavidin on its surface, a protein that acts like molecular Velcro. That surface coating lets the bacteria bind biotinylated prodrugs, inactive chemical precursors that only become toxic once they reach their target. The bacterium also latches onto cancer cells through heparan sulfate, a sugar molecule found on tumor surfaces, according to an NUS Medicine institutional release.
Once the bacteria reach the tumor, they bioconvert the prodrug into SN-38, the active metabolite of the widely used chemotherapy drug irinotecan. Standard irinotecan treatment relies on the patient’s liver to perform that conversion, which means the drug circulates systemically and causes well-documented side effects including severe diarrhea and immune suppression. By shifting the conversion step to the tumor itself, the bacterial platform concentrates the active agent where it is needed and limits exposure elsewhere. The preclinical work was conducted in mouse models, and no human trial data exist for this system, so its safety and efficacy in people remain unknown.
Parallel Efforts Using Different Bacterial Chassis
The NUS Lactobacillus work is not happening in isolation. Several independent groups have built competing bacterial platforms around the same chemical endpoint. A separate team engineered Bifidobacterium paired with chitosan-based nanomedicines that carry SN-38 to enhance tumor accumulation in mouse colorectal cancer models. That system uses a polymeric prodrug strategy rather than surface-displayed streptavidin, but the goal is identical: concentrate SN-38 inside the tumor while sparing healthy tissue.
Another group took a different enzymatic route. They transformed Bifidobacterium longum 105A with a plasmid that causes the bacterium to secrete beta-glucuronidase in tumors, an enzyme that converts SN-38 glucuronide back into active SN-38 within the tumor microenvironment. That design is mechanistically distinct because it relies on enzymatic cleavage of an already-circulating metabolite rather than carrying the prodrug on the bacterial surface. The convergence of these separate programs on SN-38 reflects the drug’s known potency and the long-standing clinical frustration with irinotecan’s systemic toxicity.
Beyond these SN-38-centered platforms, researchers are testing other microbial chassis and payloads. In one recent study, investigators described oral delivery of engineered bacteria designed to survive the gastrointestinal tract and home to tumors, underscoring the field’s interest in noninvasive administration routes. Others have reported synthetic biology circuits that let bacteria sense specific tumor metabolites and turn therapeutic genes on only when they reach malignant tissue, an added safeguard against off-target damage.
Why Bacteria Can Reach Tumors When Drugs Cannot
Conventional chemotherapy drugs face a physical barrier that bacteria can bypass. The extracellular matrix surrounding cancer cells forms a dense mesh that limits drug delivery into highly invasive tumors. Small-molecule drugs diffuse poorly through this scaffold, and even nanoparticles often stall at the tumor periphery, as highlighted in analyses of extracellular matrix barriers to intratumoral transport.
Bacteria, by contrast, are motile and chemotactic. Certain species exhibit a natural tropism for tumor tissues, drawn by the low-oxygen, nutrient-rich conditions inside solid cancers. That biological homing instinct gives engineered bacteria an advantage that synthetic nanoparticles lack: they can actively swim toward and colonize the tumor core rather than relying on passive accumulation through leaky blood vessels. Researchers have also begun engineering bacteria to be visible on ultrasound via gas vesicle structures, which would let clinicians track where the microbes go after injection and confirm they have reached the intended target.
This active migration could be especially important for tumors that are poorly vascularized or heavily fibrotic, where conventional drugs struggle to penetrate. By establishing bacterial colonies deep within such lesions, clinicians might be able to turn the tumor itself into a factory for chemotherapy or immunotherapy, rather than trying to push ever-higher systemic doses through the bloodstream.
Beyond Prodrug Delivery: Immune Activation
Drug delivery is only part of the story. Bacteria inside a tumor also provoke an immune response, and several groups are engineering that reaction on purpose. An immunologist at Columbia University led studies in which engineered E. coli shrank tumors in mice by combining direct bacterial colonization with immune-stimulating payloads, according to reporting from cancer research agencies. That dual mechanism, killing cancer cells chemically while simultaneously alerting the immune system, is what makes bacterial therapy conceptually different from standard chemotherapy or even antibody-drug conjugates.
Tumor-targeting bacteria have also been engineered to express prodrug-converting enzymes such as cytosine deaminase variants, which convert the antifungal drug 5-fluorocytosine into the chemotherapy agent 5-fluorouracil directly at the tumor. The variety of enzyme–prodrug combinations under investigation suggests that if the platform concept works in humans, it could be adapted to many existing drugs whose systemic toxicity currently limits their use.
At the same time, the immunological consequences of seeding tumors with bacteria remain only partially understood. Some studies suggest that bacterial colonization can turn an immunologically “cold” tumor into a “hotter” one, more visible to T cells and potentially more responsive to checkpoint inhibitors. Others warn that excessive inflammation could damage surrounding tissues or trigger sepsis-like syndromes if bacteria escape into the bloodstream. Future clinical protocols will likely need finely tuned dosing schedules and built-in “kill switches,” such as antibiotic sensitivity genes, to halt bacterial activity if adverse reactions emerge.
Risks, Unknowns, and the Road to the Clinic
Despite the enthusiasm, all of the bacterial SN-38 systems described so far remain in preclinical stages. Mouse tumors are far smaller and more homogeneous than human cancers, and rodent immune systems differ substantially from those of patients undergoing chemotherapy. Scaling up from controlled laboratory conditions to the complexity of human disease will require careful phase 1 trials focused first on safety and biodistribution.
Key questions include how long engineered bacteria persist in the body, whether they can be fully cleared after treatment, and how they interact with a patient’s existing microbiome. Regulators will also scrutinize the risk of horizontal gene transfer, in which therapeutic genes might jump from engineered strains into wild microbes. Strategies such as auxotrophy (making bacteria dependent on nutrients absent from normal tissues) and multi-layered genetic safeguards are being explored to minimize those dangers.
Manufacturing and quality control pose additional hurdles. Live bacterial products must be produced under stringent conditions to ensure genetic stability, consistent dosing, and absence of contaminating organisms. Unlike traditional small-molecule drugs, which are chemically identical from batch to batch, living medicines can evolve. Developers will need robust assays to confirm that each lot retains its intended traits, from prodrug-binding capacity to enzyme expression levels.
Even if early safety data are encouraging, it remains to be seen how bacterial therapies will fit into existing cancer treatment regimens. They might be used to resensitize tumors that have stopped responding to standard chemotherapy, or combined with immunotherapies to amplify anti-tumor responses. Given their complexity, these treatments are unlikely to replace conventional drugs outright in the near term. Instead, they may find niche roles in patients whose tumors are particularly resistant or difficult to reach.
For now, the engineered Lactobacillus platform from Singapore and its Bifidobacterium-based counterparts illustrate how rapidly the field is moving from conceptual designs to sophisticated, tumor-homing drug factories. If ongoing research can demonstrate that these microbes safely concentrate toxic agents like SN-38 inside cancers while sparing healthy tissue, they could mark a shift in how oncologists think about chemotherapy, not just as molecules delivered to the body, but as therapies built into living, programmable carriers that work from the inside out.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
