A small soy and casein trial found changes in specific blood AGE markers, but no change in total circulating or skin AGEs. Here is how to read it.
On October 2, 2026, NutraIngredients included a Singapore trial of soy and casein protein supplements in a nutrition research roundup. The study examined advanced glycation end products, or AGEs, in blood and skin. The important distinction was between changes in individual blood markers and the broader AGE measurements.
The roundup also covered probiotics and postbiotics alongside ginger research. Those were separate research items. For readers interested in protein and healthy aging, the relevant evidence comes from the protein trial rather than a combined message about all these ingredients.
According to NutraIngredients’ report on the trial, the researchers observed different responses in specific plasma AGE markers. Plasma refers here to the blood measurements. CML and pentosidine were individual markers assessed alongside total circulating AGEs and accumulated skin AGEs.
The reported results were:
The researchers described the CML increase associated with soy as statistically significant. The casein result needs a separate statistical qualification, addressed below.
Describing this simply as protein supplements raising AGEs would flatten several different findings into one conclusion. The report instead describes marker-specific responses alongside unchanged total circulating and skin AGE measurements.
That distinction is central to reading the study. An individual marker and a broader measurement are not interchangeable outcomes. A useful summary needs to preserve both the changes that were observed and the measurements that did not change.
NutraIngredients reports that the trial was published in Nutrients. It was a randomized controlled study conducted in Singapore with 65 adults aged between 60 and 80. The intervention lasted 16 weeks and used a single-blind design with three groups.
Researchers compared a control group with groups receiving casein or soy protein isolate. Participants in the supplemented groups received 20 g of powder daily with 250 mL of UHT low-fat milk. These amounts describe the research protocol rather than a recommendation for readers.
The powder weight was the same across the supplemented groups, but the protein amounts were not identical. The casein powder supplied 16.4 g of protein. The soy powder supplied 18.0 g.
All participants were asked to follow Singapore’s My Healthy Plate dietary pattern. Researchers measured dietary intake and blood pressure as well as blood and skin AGE levels before and after the intervention.
These details define the comparison being reported. The trial tested specific supplements within a requested dietary pattern. It was not a general comparison of every food or product containing soy or casein.
The most important statistical caution concerns casein. The researchers advised careful interpretation of its pentosidine result because the time-by-group interaction was not statistically significant. That comparison addresses whether the pattern of change over time differed between groups, so the reported rise should not be presented as a clear casein effect relative to the other groups.
The explanations for the marker changes were also untested. Researchers suggested that soy’s amino-acid composition and casein’s digestion characteristics might be relevant. However, they did not directly measure post-meal amino acid concentrations or AGE-formation pathways. Those ideas therefore remain proposed mechanisms rather than demonstrated explanations.
The available study summary does not provide exact effect sizes for the marker changes. Reporting that a result reached statistical significance does not establish how large or clinically important it was. It also does not supply a basis for ranking the clinical importance of the CML and pentosidine findings.
The study’s population and duration place further boundaries around interpretation. Results from older adults following this dietary pattern do not establish what would happen in younger people or with different diets. Everyday supplement use may differ from the controlled protocol, and longer-term outcomes remain outside this trial’s findings.
Finally, the reported results concern biomarkers rather than demonstrated changes in visible skin aging or clinical disease. They do not show that the marker shifts caused wrinkles, diabetes, cardiovascular disease or fractures. Equally, unchanged skin AGE measurements should not be expanded into a claim that protein supplements have no effect on any aspect of skin aging.
For someone deciding between soy and casein, this study alone does not establish a reason to avoid either protein source. It also does not identify a better option for skin aging. The practical response is to resist turning these findings into a product ranking.
A useful way to read related headlines is to keep the outcome attached to the claim. Ask whether the article is discussing a specific plasma marker, total circulating AGEs or accumulated skin AGEs. Then check whether it reports a measured health outcome or an interpretation of a laboratory result.
The question of whether a person needs a protein supplement is separate from this trial’s AGE findings. Younell’s discussion of assessing protein needs before buying a powder addresses that broader decision. When a choice depends on an individual’s health or nutritional needs, a qualified dietitian or clinician can provide personal guidance.
This is also the distinction Younell’s research-led beauty coverage is designed to make clear: an interesting biological observation is not automatically a useful product claim. Readers can remain curious about the research without treating it as an instruction to change their routine.
The useful next question is whether these marker-specific responses persist and whether they relate to meaningful health or skin outcomes. Research addressing those questions would help connect laboratory observations with decisions people actually need to make.
For now, the takeaway is narrow but informative. Different protein sources were associated with different responses in particular plasma AGE markers, while total circulating and accumulated skin AGE measurements did not change. That is a reason to follow the evidence carefully, not a verdict on which protein supports healthier-looking skin.
Stay connected for research and practical guidance on skin, hair, collagen, nutrition and beauty longevity. Clear ideas for people who want to understand how appearance changes with age and make better-informed choices over time.
Understand your skin, hair and body better without chasing every new trend, treatment or promise.
explore the Blog
