After two decades of relatively disappointing research, microbiome researchers are seeing bright spots of hope. By leveraging recent findings, they are expanding therapeutics beyond just the known microbiome to include a broader array of microbial strains.
Many of the challenges that thwarted previous microbiome-based projects—such as low efficacy or short duration of effects—are being overcome. Others are being approached in new ways, enabling scientists to exploit the microbiome’s many potential therapeutic pathways to create novel, and possibly more promising, therapeutics.
Scientists are seeing a glimmer of practical results, too. Positive clinical data—including MaaT Pharma’s recent Phase III results—are spurring new levels of interest, as seen at the Microbiome Movement Summit in Barcelona, Spain, this past January.
Rewiring the immune system
“We are not attempting to restore the normal function of the gut microbiome or modulate its composition,” Benjamin Hadida, CEO of Exeliom Biosciences tells GEN. Instead, we take “a different, more direct approach” that directly reprograms the immune system. Clinical trials so far show this approach “improves the efficacy of existing immune-mediated treatments in oncology and inflammatory diseases, acting as a potent adjuvant therapy.”
EXL01, a Faecalibacterium prausnitzii (F. prau)-based therapeutic, “leverages the unique peptidoglycan structure of the F. prau to activate the NOD2 pathway in macrophages and monocytes. It bypasses the need to modify the gut microbiome, and directly drives immune modulation,” Hadida says. Because EXL01 tightly targets NOD2, adverse systemic immune activation is minimized. “Additionally, since EXL01 does not alter the gut microbiome composition, it avoids the risks associated with dysbiosis or unintended microbiome changes.”
Notably, the effects of this therapeutic are mediated entirely through the immune cells, which he says enhances the consistency and predictability of outcomes. EXL01 is administered orally once a day.
Despite having only a few regulatory precedents for live biotherapeutic products, Hadida says interactions with regulatory agencies suggest that, at this point, they view EXL01 “as just another drug.” With both the safety and the manufacturing and control strategy validated by regulators, “we are now discussing the clinical path to the market.”
Mechanistic studies are underway to refine the company’s understanding of how NOD2 activation drives metabolic reprogramming in immune cells. Meanwhile, EXL01 is undergoing Phase II trials in oncology, inflammation, and infectious diseases. Hadida also says, “We are exploring its potential to synergize with checkpoint inhibitors, anti-TNF therapies, and other immune modulators.”
Programmable, living therapeutics
“Treating solid tumors, which account for 90 percent of cancers, remains a critical challenge,” Livija Deban, PhD, CSO, Prokarium, tells GEN. Hurdles include tumor heterogeneity, drug resistance, and off-target effects.
To deliver more tightly targeted living therapeutics, Prokarium combines the tumor-colonizing abilities of specific bacteria, such as Salmonella enterica, with logic-gated synthetic biologic circuits to deliver therapeutic payloads within tumors.
“Logic-gated biological circuits are engineered systems that mimic the behavior of logic gates in electronics, allowing bacteria to sense and respond to specific environmental conditions in a precise and programmed manner,” Deban explains. Specific genetic elements, such as inducible promoters, may act as molecular switches.
The combination, therefore, of bacterial capabilities with logic-gating circuits enhances precise targeting and addresses some traditional therapeutic limitations, including poor tumor penetration and systemic toxicity. By targeting universal tumor features, such as hypoxia and the neovasculature, the therapeutic becomes broadly applicable and “able to overcome common cancer escape pathways, including antigen loss,” Deban elaborates.
In preclinical studies, Prokarium’s Living Cures platform of live attenuated bacteria differentiated between healthy tissue and tumors. Now the company is working towards iterations that can deliver the payload at the right time, place, and dose. The platform is engineered to limit its own growth and to confine that growth to the tumor site, where bacterial tropism causes Living Cures to naturally accumulate.
While optimistic, Deban admits, “There is still much to learn about how genetic engineering impacts bacterial behavior.” Potential challenges, she says, include fine-tuning live attenuation, and ensuring that all known toxins—“such as the colibactin pathway in E. coli strains”—are eliminated to ensure patient safety. Additionally, strategies like pre-emptive antibiotic coverage, in which the bacterial are susceptible to antibiotics as a last resort fail-safe, must be considered to prevent the risk of widespread infection,” she says.
These engineered living therapeutics will likely enter the preclinical phase in mid-to-late 2026. Prokarium’s foundational strain—live attenuated Salmonella Typh—is in clinical development for bladder cancer.
Upcoming hurdles include navigating a still-emerging regulatory landscape and “ensuring that therapeutic payload delivery does not compromise bacterial fitness, as well as matching the best bacterial strain to the best payload for a particular cancer,” Deban says.
Manufacturing, however, should be straightforward: Living Cures are made by fermentation with strains that are inactive during manufacturing.
3D mucus models shine
Mucus may not be top of mind when it comes to microbiome research, yet the properties of mucus play a huge role in microbiota microenvironments.
Bac3Gel developed models that replicate the complexity of both mucus and microbiota to help researchers gain deeper insights into host-microbiota interactions, screen therapeutic candidates, and analyze microbial behavior while reducing their reliance on animal models.
These three-dimensional mucus models feature gradients of structure, oxygen distribution, and drug and nutrient penetration. Therefore, Daniela Pacheco, PhD, co-founder and CTO, Bac3Gel, tells GEN, “They more accurately simulate the native environment.”
In contrast, “Traditional in vitro models fail to replicate the complexity and functionality of native mucus, which plays a pivotal role in microbiota-host interactions, drug/nutrient availability, and pathogen behavior.”
These ready-to-use, high-throughput, in vitro models “mimic key properties of mucus layers from various body regions, such as the gut, cervicovaginal, lung, and stomach (areas),” Pacheco says. “They bridge a critical gap in understanding the microbiota’s role in health and disease.”
These models can sustain bacterial species for up to 72 hours, as well as grow hard-to-culture microbiota (using Bac3Gel’s Growth Enhancer Beads). As Pacheco points out, the “Gut3Gel sustained 90 percent of donor bacterial species,” enabling researchers to study their interactions in an environment that closely resembles that of the human body. She says the models integrate seamlessly with standard lab equipment and protocols.
Complex microbiome analysis
Standardizing microbiome analysis is a well-acknowledged need. “Journals are asking authors to share sequencing data and bioinformatic pipelines to ensure that other scientists can reproduce published data,” Pilar Manrique, PhD, research scientist, Microviable Therapeutics, tells GEN. Beyond simply sharing data, however, “The U.K. National Institute for Biological Standards and Control developed whole cell standards for the microbiome field (that scientists) can include in their sequencing and bioinformatic pipeline to increase reproducibility.”
The ultimate goal is to reduce the significant variability that occurs, starting with how and from where on the body samples are collected and continuing through sequencing techniques, bioinformatic pipelines, and data validation.
“Even though commercially available kits are optimized, DNA extraction efficiency is very different depending on the microbes,” Manrique says, “and the results provided by different kits are still astonishingly variable.” This challenge extends beyond the gut microbiome to include very low abundance bacterial biomass, such as the skin.
Artificial intelligence (AI) and machine learning (ML) can help to some degree during the analysis process. Its greatest contribution, at this point, may be helping scientists deal with very large amounts of data, integrate different types of data, and provide meaningful information.
Additionally, AI/ML models “can be used and optimized to identify patterns and discover potential biomarkers for early disease detection,” Manrique suggests, “which is one of the main unmet needs in our healthcare systems.”
Ensuring accuracy in AI/ML analyses, however, “requires a significant amount of oversight as well as extensive work to validate the results, depending on what you’re looking at,” Manrique cautions.
Two of the most important and time-consuming areas are data preprocessing to ensure high-quality data—“As they say, ‘garbage in/garbage out,’” reminds Manrique—and model selection and validation to ensure the model addresses the specific needs of the experiment.
For drug discovery, Microviable’s Pharmabiota platform “integrates microbiology, immunology, and bioinformatic analyses with data from in vivo preclinical and clinical studies,” Manrique says. It accesses a proprietary bacterial library containing more than 2,000 bacterial isolates to inform biological development for infectious disease, cancer, and other indications.
“It is also important to mention the ethical issues that are naturally associated with these types of technologies and how we will need clear ethical guidelines and regulatory standards,” she adds.
As new tools evolve and knowledge increases, the field of microbiome and microbial therapeutics is poised to make significant contributions to a variety of therapeutic areas in the coming decade. Now, despite earlier setbacks, this new generation has a realistic potential to become important as adjuvants or as immune system–modulating therapeutics in their own right.