Gemelli Biotech has reported new real-world evidence linking use of its IBS-Smart blood test and Trio-Smart breath test with lower use of selected diagnostic procedures and lower captured diagnostic costs among adults being evaluated for suspected irritable bowel syndrome. The retrospective study, published in Frontiers in Gastroenterology on July 23, 2026 and highlighted by Gemelli on August 18, included 219 adults treated at two United States gastroenterology practices.
The numbers are potentially meaningful for a gastrointestinal diagnostic pathway in which repeated investigations can become expensive. Patients who received neither proprietary test underwent colonoscopy and upper endoscopy more frequently than the tested groups, while normalized monthly diagnostic and gastroenterology office-visit costs were substantially higher among controls in a subgroup of patients first seen from 2018 onward. An exploratory model estimated $526 in incremental diagnostic savings associated with IBS-Smart and $521 for Trio-Smart.
Those figures, however, require a narrower interpretation than the headline savings numbers might suggest. The study was retrospective rather than randomized, clinicians decided which patients received the tests, the combined-testing cohort contained only 11 patients, and the economic analysis captured a limited portion of healthcare spending. It therefore cannot establish that IBS-Smart or Trio-Smart caused the reduction in procedures or costs.
That distinction is commercially important for Gemelli Biotech. The study provides practice-based evidence supporting a potential economic role for its noninvasive gastrointestinal diagnostics, but prospective evidence will be needed before the results can support stronger claims that routine use meaningfully reduces healthcare expenditure or improves patient outcomes.
What did the 219-patient study find about colonoscopy, endoscopy and diagnostic spending?
Researchers divided the 219 adults into four cohorts: 48 received IBS-Smart alone, 43 received Trio-Smart alone, 11 received both tests and 117 received neither. Eligible patients had symptoms suggestive of IBS, at least three visits, at least three months of follow-up and a documented gastrointestinal diagnosis. Patients with pregnancy or alarm symptoms including fever, hematochezia or hematemesis were excluded.
Colonoscopy was recorded in 60.7% of controls compared with 50% of IBS-Smart-only patients, 53.5% of Trio-Smart-only patients and 27.3% of patients who received both tests. Upper endoscopy was performed in 48.7% of controls, compared with 31.3%, 41.9% and 0%, respectively.
The combined-testing numbers are visually striking, but that group contained just 11 patients. A zero upper-endoscopy rate in such a small cohort should therefore not be interpreted as evidence that combining IBS-Smart and Trio-Smart eliminates the need for endoscopy.
The cost comparison also deserves attention. Among patients first seen in 2018 or later, mean captured diagnostic and gastroenterology office-visit costs were $960.39 per patient per month in controls, compared with $366.85 for IBS-Smart alone, $382.87 for Trio-Smart alone and $361.83 for patients receiving both tests. The corresponding descriptive differences versus controls were approximately $594, $578 and $599 per month.
Yet these were not comprehensive healthcare-cost calculations. Costs were constructed using a payer-proxy perspective in 2022 United States dollars and covered captured diagnostic tests and gastroenterology office visits. Emergency department care, primary care, other specialist visits, productivity losses, dietary costs, travel and many downstream treatment expenses were outside the analysis.
That makes “lower captured diagnostic costs” a materially more accurate description than saying the tests reduced the overall cost of IBS care.
Does the estimated $526 saving prove that IBS-Smart can lower the cost of diagnosing IBS?
No. The modeled result is better viewed as an early health-economic signal.
Researchers used propensity matching and an exploratory generalized linear model involving 58 IBS-Smart recipients and 58 matched controls. IBS-Smart had a coefficient of minus 0.22 with a P value of 0.050, corresponding to an estimated incremental saving of $526. A separate Trio-Smart model generated an estimated $521 diagnostic-cost saving.
A P value sitting at 0.050 already argues against treating the IBS-Smart result as overwhelming evidence, and statistical adjustment cannot fully remove unmeasured differences created when clinicians themselves decide who should be tested.
The baseline data show why patient selection matters. Diarrhea was reported at the first visit in 79.2% of IBS-Smart-only patients and 76.7% of Trio-Smart-only patients, versus 64.1% of controls. Referral rates and year of first presentation also differed across cohorts. Propensity matching helps address observed differences, but it cannot turn an observational dataset into a randomized comparison.
The commercially relevant takeaway is therefore not that spending will fall by roughly $500 every time one of the tests is ordered. Rather, the study suggests that selective noninvasive testing could influence the sequence of investigations enough to justify a larger prospective economic evaluation.
That is an important distinction for payers. A diagnostic that costs money upfront becomes economically attractive only when it reliably prevents sufficiently expensive downstream activity without increasing missed diagnoses, repeat testing or later treatment costs. This study identifies that possibility but does not yet complete that equation.
Where could IBS-Smart and Trio-Smart fit within a positive diagnostic strategy for irritable bowel syndrome?
The broader clinical context strengthens the relevance of the study. American College of Gastroenterology guidance has supported a positive diagnostic strategy for IBS rather than requiring clinicians to exhaustively exclude every alternative disease before making the diagnosis. That approach still relies on history, symptom criteria and targeted investigations appropriate to the clinical presentation.
The new study does not position Gemelli Biotech’s tests as replacements for that process. The authors specifically noted that IBS evaluation still requires assessment for alarm features and appropriate testing for conditions that can mimic an IBS phenotype. Celiac serology and inflammatory markers, for example, can remain relevant in patients presenting with diarrhea.
IBS-Smart measures antibodies against cytolethal distending toxin B and vinculin, biomarkers associated particularly with post-infectious and diarrhea-predominant IBS phenotypes. The original large validation study reported relatively high specificity for distinguishing diarrhea-predominant IBS from inflammatory bowel disease, but substantially lower sensitivity, meaning a negative biomarker result does not rule out IBS.
That performance profile is important when interpreting the new economics study. IBS-Smart may provide clinicians with additional evidence that increases confidence in a particular IBS phenotype, but it cannot function as a universal biological confirmation test for the heterogeneous syndrome.
Trio-Smart addresses a different part of the diagnostic pathway. The mail-in breath test measures hydrogen, methane and hydrogen sulfide, providing information that Gemelli positions for the evaluation of small intestinal bacterial overgrowth, intestinal methanogen overgrowth and intestinal sulfide overproduction. Hydrogen and methane breath testing is already used clinically, although test preparation, intestinal transit, substrates and interpretation can affect results.
The three-gas approach potentially broadens the information available to clinicians, but the new study was designed around healthcare resource utilization. It did not establish that Trio-Smart provides superior diagnostic accuracy or produces better clinical outcomes than alternative testing pathways.
Why is diagnostic persistence intriguing if it does not prove diagnostic accuracy?
One of the study’s largest numerical differences concerned what researchers called chart-level diagnostic persistence.
A stable charted diagnosis was observed in 79.2% of IBS-Smart-only patients and 90.7% of Trio-Smart-only patients, compared with just 25.6% of controls. Among patients receiving both tests, persistence was 72.7% after IBS-Smart and 90.9% after Trio-Smart.
For a disorder in which patients may undergo repeated investigations before diagnostic confidence develops, greater persistence could indicate that additional test information helps clinicians settle on a working diagnosis earlier and avoid reopening the diagnostic process.
But persistence is not accuracy.
The study compared diagnoses documented in medical charts rather than conducting blinded adjudication against an independent reference standard. A diagnosis that remains unchanged could reflect genuine confidence, but it could also reflect clinician preference, coding behaviour or diagnostic anchoring. The authors accordingly cautioned against interpreting the result as confirmation that the tested patients were diagnosed more correctly.
This is one area where a future prospective study could materially strengthen the commercial case. It could measure not merely whether diagnoses remain stable, but whether alternative organic diseases are subsequently identified, whether patients require reinvestigation and whether earlier diagnostic confidence leads to better treatment selection or quality of life.
What limitations will determine whether the findings can influence payer or clinical adoption?
The study’s greatest strength is also its central weakness. It examines what happened in real gastroenterology practices rather than under a tightly controlled research protocol, making the resource-use findings relevant to actual clinical workflow. But the absence of random test assignment makes it difficult to determine how much of the observed difference arose from the tests themselves and how much arose from the clinicians and patients selected for testing.
Rule-out testing was also not standardized. Celiac serology, C-reactive protein and fecal calprotectin were captured at different rates across cohorts, while other elements of differential diagnosis were not consistently recorded. Colonoscopy itself may also be clinically necessary for reasons unrelated to uncertainty over IBS, depending on age, symptoms, screening requirements and other risk factors.
The economic model consequently should not be interpreted as evidence that invasive procedures were unnecessary in the control group.
The funding relationship also warrants transparent interpretation. Gemelli Biotech funded the study and manufactures both tests. The published paper states that the company was not involved in study design, data collection, analysis, interpretation, manuscript preparation or the decision to submit the research. Three authors were employees of AHRM Inc., which received funding from the study sponsor.
Industry funding does not invalidate the findings, but independent replication would carry additional weight for payers, clinicians and guideline developers considering whether the economic signal generalizes beyond the two practices studied.
What evidence would turn this real-world signal into a stronger commercial case for Gemelli Biotech?
The next evidence threshold is relatively clear. A prospective, multicenter comparison would need predefined criteria for assigning IBS-Smart or Trio-Smart, standardized rule-out pathways and longer follow-up capable of capturing repeat consultations, subsequent procedures, treatment changes and alternative diagnoses.
Patient-centered outcomes are equally important. The current study did not measure symptom improvement or quality of life, so lower diagnostic utilization cannot yet be connected to better patient outcomes. Medication counts changed descriptively in several groups, but the archived dataset lacked enough indication-specific information to support strong conclusions about treatment effects.
For Gemelli Biotech, that makes the new publication a useful commercial evidence-building step rather than the end of the validation process. The company now has real-world data suggesting its tests may fit an economic story around more structured diagnostic sequencing, fewer selected invasive procedures and lower captured specialty-care costs.
The much larger opportunity would be demonstrating that those associations survive prospective testing across broader clinical settings.
If a future study can show that appropriately selected IBS-Smart or Trio-Smart testing reduces unnecessary investigations while maintaining low rates of missed alternative diagnoses, preserving guideline-directed evaluation and improving patient-level outcomes, the value proposition could extend beyond diagnostic convenience into payer economics and care-pathway design. Until then, the $526 and $521 figures are best treated as promising hypotheses for the next study, not guaranteed savings attached to each test.
