Does the delivery format affect probiotic efficacy?

Often I am asked some version of the question,: “Does the delivery format affect probiotic efficacy?” This is important to consumers trying to choose between probiotic foods and supplements. This is important to responsible probiotic product manufacturers interested in tweaking their formulations. This is important to regulators assessing whether a human efficacy trial conducted on one type of food is applicable to another type of food with the same probiotic.

The current answer is “maybe.” We don’t have a robust body of research to provide a clear answer right now. It is conceivable that different food components or structures will impact efficacy. Probiotics are live microbes and their growth and development depends greatly on their environment (Marco & Tachon 2014). The presence of substrates, exposure to stresses, and interactions with other live microbes will drive what genes are expressed by the probiotic and what metabolic, antimicrobial, or immunomodulatory products it makes. But how long will the impact of such ‘preconditions’ before consumption last after consumption? How long after the probiotics hit the human host do they start to adapt to this new environment? We don’t know. More importantly, we don’t know if differences in delivery format result in measurable differences in clinical endpoints, such as gut symptoms, incidence/duration of common infectious diseases, or reduction of LDL cholesterol.

What do we know? We know that an array of different probiotic strains, delivered in an array of different formats, have been tested in human studies assessing an array of different health endpoints. Many – but certainly not all – of these studies show or suggest that probiotics can positively impact human health. I think of this body of research as an incomplete mosaic; some areas of the image with many “tiles” of research are distinct, while other areas with very few tiles are much less clear. What we don’t have is a systematic filling in of this mosaic’s image through testing of strains delivered in different formats compared head-to-head in human studies assessing the same clinical endpoint. And that is the exact information we would need to decisively answer our original question.

A paper by Flach et al. (2017) did a comprehensive job reviewing available studies on the effect of food matrices on probiotics, specifically on probiotic survival in the product, survival through the gastrointestinal tract, and clinical efficacy. The paper makes a solid case that food matrix can change probiotic survival, which alters the dose delivered by the food. But the paper confirms that little information is available on the ultimate impact of the matrix on probiotic-mediated clinical endpoints. The authors found only two studies that directly compared the same probiotic in different matrices assessing human, ‘patient-oriented’* (not biomarker) endpoints (Isolauri et al. 1991 and Hütt et al 2015). One study on treatment of acute pediatric diarrhea showed that the matrix (fermented milk and dried format) had no effect. The other had mixed results; for systolic blood pressure, results were different for yogurt vs cheese, but for diastolic blood pressure they were not. Two other studies, although conducted in humans, looked at biomarkers, which while interesting cannot be translated into a measured health benefit.

My conclusion is that a clear matrix-dependency for clinical benefits of probiotics has NOT been established.

I fully acknowledge that matrix might be important. I also fully recognize that many pre-clinical (lab bench and animal) studies suggest that matrix does matter. But, as they say, humans are not mice. The ability of any probiotic to have a clinical benefit likely depends on many factors–diet (see Tachon et al. 2014), human genetics, colonizing microbiota, medications, stress, among others—of which matrix may be one. It’s certainly conceivable that a finely tuned probiotic paired with the right matrix/ingredients may have superior benefits. Such is the hope for the synbiotics of the future. But we don’t yet know how much of the variation among an individual’s ability to respond or not respond to a probiotic intervention is due to matrix differences. We do know that several well-studied probiotic strains (e.g., L. rhamnosus GG, B. lactis BB-12, L. reuteri DSM17938) deliver benefits when tested in several types of matrices.

Human trials are expensive and, from a business perspective, risky. When does a company need to conduct additional human studies to support matrix differences? (This topic was addressed in more detail in Sanders et al. 2014). Changes such as selling a probiotic yogurt in flavors different from the one studied are likely trivial; changes that reduce probiotic stability (either in the product or as it traverses the GI tract) are not. Until more head-to-head clinical trials are conducted, we cannot know the full weight of the delivery matrix. Available evidence suggests that probiotics may be resilient enough to endure some environmental changes and still deliver the expected benefits. The field is best served if we approach this issue with flexibility, considering the totality of mechanistic and human evidence, and develop a sound scientific rationale to justify equivalency among different delivery matrices.

*Patient-oriented outcomes refer studies that “measure things a patient would care about, such as improvement of symptoms, quality of life, cost of the intervention, morbidity or mortality, length of stay, etc.—essentially whether the intervention helps patients live longer or better lives.” (Ref)


Additional reading:

Flach J, van der Waal MB, van den Nieuwboer M, Claassen E, Larsen OFA. The underexposed role of food matrices in probiotic products: Reviewing the relationship between carrier matrices and product parameters. Crit Rev Food Sci Nutr. 2017 Jun 13:1-15. PMID: 28609116

Hütt P, Songisepp E, Rätsep M, Mahlapuu R, Kilk K, Mikelsaar M. Impact of probiotic Lactobacillus plantarum TENSIA in different dairy products on anthropometric and blood biochemical indices of healthy adults.  Benef Microbes. 2015;6(3):233-43. PMID: 25524863

Isolauri E, Juntunen M, Rautanen T, Sillanaukee P, Koivula T. A human Lactobacillus strain (Lactobacillus casei sp strain GG) promotes recovery from acute diarrhea in children. Pediatrics. 1991 Jul;88(1):90-7.  PMID: 1905394

Marco ML, Tachon S. Environmental factors influencing the efficacy of probiotic bacteria. Curr Opin Biotechnol. 2013 Apr;24(2):207-13. PMID: 23102489

Sanders ME, Marco ML. Food formats for effective delivery of probiotics. Annu Rev Food Sci Technol. 2010;1:65-85. PMID: 22129330

Sanders ME, Klaenhammer TR, Ouwehand AC, Pot B, Johansen E, Heimbach JT, Marco ML, Tennilä J, Ross RP, Franz C, Pagé N, Pridmore RD, Leyer G, Salminen S, Charbonneau D, Call E, Lenoir-Wijnkoop I. Effects of genetic, processing, or product formulation changes on efficacy and safety of probiotics. Ann N Y Acad Sci. 2014 Feb;1309:1-18. PMID: 24571253

Tachon S, Lee B, Marco ML. Diet alters probiotic Lactobacillus persistence and function in the intestine. Environ Microbiol. 2014 Sep;16(9):2915-26. PMID: 24118739