“In a study, women over 50 who applied a quinoa-derived bioester to one forearm for 30 days showed a median proteomic-age difference equivalent to approximately 16 years younger compared to a control formulation applied to the other forearm, as measured by machine-learning analysis of skin protein profiles (Camillo-Andrade et al.,…”

The study exists and the number is real. The abstract states that in this exploratory, proof-of-concept, prospective observational study, women aged 20 to 80 received 30-day topical quinoa bioester application on one forearm, with the contralateral forearm receiving vehicle. The Support Vector Regression model, trained on pre-treatment proteomes, predicted lower proteomic ages for quinoa bioester-treated skin compared to vehicle-treated skin, with observed median differences of 11 and 16 years for participants under and over 50, respectively (p < 0.01 for participants ≥50 years). The authors attach an explicit caveat in the abstract itself: while these values do not necessarily correspond to biological years, the findings demonstrate that topical bioactives can induce detectable shifts in skin proteomic profiles. The PubMed abstract repeats this: these values demonstrate differential treatment effects and do not necessarily correspond to biological years. On the biology: mass spectrometry revealed significant upregulation of barrier function proteins (desmoglein-1, filaggrin), antioxidant enzymes (SOD1, glutaredoxin-1), and protease inhibitors. ## METHODOLOGY AND CONTEXT Design and sample. Each of the 61 participants provided four skin samples, one from each forearm, before and after a 30-day treatment period, resulting in a total of 244 samples. One forearm was treated with the quinoa bioester formulation, while the contralateral forearm served as control. The cohort consisted of 61 individuals, one per age from 20 to 80, meaning the "over 50" subgroup is roughly 31 women, approximately one per year of age. Dose and regimen. The trial registry specifies a randomised, placebo-controlled, single-blind, intra-individual paired design in which the intervention arm of 61 forearms received topical application of a cosmetic formulation containing 0.1 percent Chenopodium quinoa bioester twice daily, morning and night, for 30 days, while the control arm of 61 forearms received a placebo formulation. Model accuracy. The model was evaluated using a Leave-Age-Out approach and achieved a mean absolute error of 7.17 years and an R² of 0.74, indicating a correlation between proteomic features and chronological age. A 16-year median gap should be read against a model whose typical error on held-out data is about 7 years. What the predicted numbers actually looked like. The predicted ages were high in both arms. Vehicle-treated samples showed median predicted ages of 77 years (Q1 to Q3: 61 to 90) versus 51 years (Q1 to Q3: 37 to 66) for older and younger cohorts respectively. The 16-year figure is an arm-versus-arm difference, not a measurement of skin becoming younger than its own baseline. Authors' own caveat on this exact number. These numerical differences do not necessarily reflect biological years due to inherent limitations of applying the model to post-treatment data (domain shift), but indicate a differential effect that is more pronounced in older skin. Under-50 result was not significant. Younger participants showed an observed median difference of 11 years between treatments (p = 0.14, Wilcoxon signed-rank test), while older participants showed an observed median difference of 16 years (p < 0.01, Wilcoxon signed-rank test). Primary outcome. The pre-specified primary outcome was characterising treatment-induced proteomic changes, not the proteomic-age estimate. The age model is a downstream analytic layer, not the registered endpoint. Funding and conflicts. This is industry-involved research. Nine co-authors are employees of Grupo Boticário, a company that develops and commercializes cosmetic products. Two Grupo Boticário authors contributed to formulation preparation, experimental design, project conception, and supervised the research at the company. The authors state the study does not directly promote or evaluate a specific product of the company and that the research was conducted independently, with the company-employed authors declaring their employment did not influence design, analysis, or interpretation. Grupo Boticário is also listed in the acknowledgements alongside CNPq, CAPES and Fiocruz. Minor design inconsistency. The journal article describes the work as a prospective observational study, while the Brazilian registry entry describes a randomised, placebo-controlled, single-blind design. These labels are not equivalent. The discrepancy is unexplained in the sources retrieved.