Editorial, independent, evidence-based review, not the official DoNotAge store. To purchase, visit DoNotAge.org directly. Use DoNotAge Discount Code (BB10) for 10% off Sitewide.
DoNotAge Hydrolysed Collagen Peptides - Evidence-Based Review
DoNotAge.org sells a 10 g/day bovine hydrolysed collagen peptide powder — Type I and III, enzymatically broken into low-molecular-weight peptides for absorption. The page positions it across skin, joints, bones, recovery, hair, nails, and gut. Hydrolysed collagen is one of the better-studied nutricosmetic ingredients on the market. There are dozens of RCTs and roughly half a dozen meta-analyses on skin alone, more on knee osteoarthritis, and a small but coherent set on bone and sarcopenia. Effect sizes are modest, real, and reproducible across mechanisms that genuinely use the amino-acid building blocks the supplement delivers. The catch: when the most recent independent meta-analysis stratified RCTs by industry funding and methodological quality, the skin effect attenuated sharply in unfunded, high-quality trials — a pattern that does not invalidate the ingredient but does anchor expectations below the marketing temperature. The same caveat does not apply equally to the joint or bone literature, which is why the verdict tiers below differ by claim.
- DoNotAge recommendation: 10 g/day (one scoop)10 g/day sits squarely within the clinical evidence band — above the König 5 g/day bone dose, at the Benito-Ruiz 10 g/day knee OA dose, and below the Zdzieblik 15 g/day sarcopenia dose.
- Skin elasticity,hydration, and wrinkle depth: Multiple RCTs andmeta-analyses show small-to-moderate improvements at 2.5–10 g/day over 8–12weeks. The 2025 American Journal of Medicine meta-analysis (23 RCTs, n=1,474)found the effect was driven primarily by industry-funded and lower-qualitystudies; high-quality independent trials showed no significant effect onhydration, elasticity, or wrinkles. Real but smaller than marketing implies
- Symptomatic knee osteoarthritis: Yu 2023 meta-analysis (4 RCTs, n=507) found a standardised mean difference of −0.58 for pain (95% CI −0.98 to −0.18, p=0.004) — above the MCID. A larger 2024 trial-sequential meta-analysis (35 RCTs, n=3,165) corroborates a direction-of-effect. All four trials in Yu 2023 were rated high risk of bias, and most studies are short (<6 months). Useful adjunct, not a substitute for proven osteoarthritis care.
- Postmenopausal bone mineral density: König et al. 2018 (5 g/day, 12 months, n=131 postmenopausal women) showed statistically significant gains in lumbar spine and femoral neck BMD versus maltodextrin placebo, with corroborating shifts in P1NP and CTX-1 bone markers. The four-year follow-up replicated the trend. Evidence is essentially one well-conducted research programme — promising but not yet broadly replicated by independent groups.
- Lean-mass gains, recovery, hair and nail strength: Collagen is an incomplete protein (no tryptophan) and very low in leucine, the amino acid that drives muscle protein synthesis. Whey protein outperforms it head-to-head for muscle growth even when leucine-matched. The Zdzieblik 2015 sarcopenia trial showed striking gains at 15 g/day with resistance training, but the magnitude prompted a published critique in the same journal questioning whether the result was reproducible. Hair/nail claims rest on a small number of low-quality trials.
- 14-day money-back guarantee on unopened tubs; 50% refund on opened tubs with ≥50% remaining.
- Delivery: USA 2–3 days, UK 2 days, EU 3–4 days, Worldwide 3–7 days.
DoNotAge Hydrolysed Collagen Peptides at a glance
For readers who want the load-bearing facts before they decide. The section covers what is in the tub, who it is built for, and the regulatory and methodological context — including the items the marketing copy tends to soft-pedal (third-party testing, leucine content, oxalate metabolism).
PEPTIDE SIGNALLING · AMINO-ACID SUBSTRATE · FIBROBLAST RESPONSE
A signalling story, not a "you eat collagen, you get collagen" story
The section describes hydrolysed collagen as if the peptides slot directly into your skin and joints. The actual mechanism is more interesting and more conditional: a fraction of ingested di- and tripeptides survive intestinal digestion intact, reach circulation, and act as signals to fibroblasts and chondrocytes — not as building materials per se. The bulk of the dose is hydrolysed all the way to amino acids and enters the general pool. Both routes matter, but they have very different implications for who benefits and at what dose.
A small but real fraction of ingested collagen peptides survives the gut and appears in blood as Pro-Hyp, Hyp-Gly, and similar dipeptides — measurable for 1–4 hours post-dose.
Collagen is the most abundant protein in the body and the only dietary protein that delivers meaningful hydroxyproline. Because hydroxyproline-containing dipeptides resist intestinal peptidases, Pro-Hyp and Hyp-Gly can be detected in plasma 1–2 hours after a 5–10 g oral dose and persist for several hours. These dipeptides are the signalling currency — they do not become collagen, they tell cells to make collagen. Total bioavailability of bioactive peptides is in the low-percent range, which is part of why doses are in grams rather than milligrams.
Pro-Hyp and Hyp-Gly stimulate dermal fibroblasts to upregulate collagen, elastin, and hyaluronic acid synthesis — the basis for the measured improvements in elasticity and hydration.
In cell-culture work cited in the Indian Journal of Dermatology meta-analysis (2025), fibroblasts exposed to Pro-Hyp at physiologically achievable concentrations increase proliferation and migrate more readily into wound-mimic gels, and they raise expression of type I procollagen and hyaluronan synthase. This is the proposed mechanism for the meta-analytic findings on skin hydration and elasticity. The mechanism is plausible, the cell-culture data are consistent, and the clinical effect sizes — small but real — match what a signalling mechanism would predict, not the dramatic effects sometimes implied in marketing.
In animal and in-vitro models, collagen peptides accumulate in cartilage and stimulate chondrocytes to increase Type II collagen and proteoglycan synthesis — even though the supplement itself contains only Type I and III.
Radiolabelled hydrolysed collagen accumulates preferentially in cartilage tissue in animal studies. Chondrocytes exposed to collagen peptides increase type II collagen synthesis and proteoglycan production, and reduce pro-inflammatory cytokine output (IL-1β, IL-6, TNF-α). This is the cleanest mechanistic explanation for why a Type I/III supplement might reduce Type II cartilage pain — peptide signalling, not substrate provision. It also explains why head-to-head trials against undenatured Type II collagen (UC-II) sometimes favour the Type II form: UC-II works via a different oral-tolerance mechanism on T-cells in Peyer's patches at much smaller doses (~40 mg).
Specific collagen peptides upregulate Runx2-mediated osteoblast differentiation and increase the type I procollagen marker P1NP while suppressing the resorption marker CTX-1.
The König 2018 trial is the clinical anchor — at 5 g/day for 12 months, bone formation markers rose and resorption markers fell, accompanied by measurable BMD gains at the spine and femoral neck. The proposed mechanism — also seen in animal and cell models — is that collagen peptides stimulate Runx2 expression, the master transcription factor for osteoblast differentiation, while simultaneously providing the proline and glycine substrate needed for bone organic matrix synthesis (which is ~90% collagen). The 5 g dose is intriguingly low for an effect on bone; it suggests the dominant mechanism is signalling, not substrate.
The bulk of the dose is hydrolysed all the way to glycine, proline, and the other amino acids, contributing to the general protein pool — useful but not the signalling story.
Glycine itself has separate, plausible roles in healthy ageing — it is rate-limiting for glutathione synthesis (the master antioxidant) and creatine synthesis, and high-dose glycine reduces inflammatory markers in some trials. Proline supports prolyl hydroxylase activity for collagen synthesis throughout the body. At 10 g of collagen peptides per day, this gives roughly 3.3 g glycine and 1.0 g proline — meaningful relative to typical dietary intake. What the amino-acid pool will not do is stimulate muscle protein synthesis the way whey does: collagen is missing tryptophan entirely and contains roughly one-third the leucine of whey. This is the load-bearing reason collagen underperforms whey in head-to-head hypertrophy trials.
What this means for who should and should not take it
The signalling-not-substrate framing predicts the trial data well: small, reliable effects on skin and joints at gram-scale doses, with no realistic prospect of a dramatic transformation. For bone, the König dose-response — a real effect at 5 g — strengthens the signalling case. For muscle, collagen will always be a poor choice on its own; the protein-quality story is not contested.
The honest version of the section would say: "Take 5–10 g for skin and joints, expect modest effects, and if you also want lean-mass benefit, take whey (or a complete plant blend) for your protein and add collagen as an adjunct."
CLAIM TIERING • EVIDENCE-GRADED
What the evidence actually supports — claim by claim
Seven benefit categories. Each is graded below by the strength of evidence specifically for hydrolysed collagen peptides at clinically relevant doses, with the relevant trials named. Two are well-supported, three are reasonable but conditional, and two are speculative.
Which two claims would justify buying it on their own
For an aging adult focused on healthspan, the two reasons to take hydrolysed collagen are:
- Modest skin and joint improvements with broad, reproducible meta-analytic support.
- If postmenopausal, the König 5 g bone-density data — which is the most interesting result in the entire literature. The other claims (hair, nails, gut, muscle, recovery) are bonuses at best and would not, on their own, justify the spend.
PRIMARY EVIDENCE • RCTs AND META-ANALYSES
Ten studies that carry the weight in the collagen literature
The list is curated, not exhaustive. It includes the meta-analyses that summarise the field, the two most-cited individual RCTs that drive verdicts on skin and bone, the contested sarcopenia trial, and one key independent re-analysis. Limitations — sample size, blinding, funding — are noted where they matter.
Significant favourable effects of hydrolysed collagen on skin hydration, elasticity, and wrinkles versus placebo. Effect sizes modest. Heterogeneity across collagen sources (bovine, marine, porcine) was non-significant — type of source does not appear to drive outcome.
The most important methodological finding in the collagen literature: the pooled effect on hydration, elasticity, and wrinkles was significant overall — but in subgroup analysis, studies not receiving pharmaceutical-company funding showed no effect on hydration, elasticity, or wrinkles, while industry-funded studies did. Similarly, high-quality studies showed no significant effect; low-quality studies showed significant improvement. The headline meta-analytic estimate likely overstates the true effect.
Hydrolysed collagen significantly improved skin hydration (Z=4.94, p<0.00001) and elasticity (Z=4.49, p<0.00001) compared to placebo. Source of collagen and measurement method did not significantly modify outcomes — bovine peptides (as in this product) performed comparably to marine.
Significant improvements in skin elasticity (p=0.009), hydration (p ranged from 0.003 to <0.001), and roughness (p ranged from 0.002 to <0.001). No safety signals. This is the modal trial design in the collagen-skin literature — a typical 8-week, 10 g/day bovine peptide RCT producing positive results with industry sponsorship.
Significant pain relief versus placebo (standardised mean difference −0.58, 95% CI −0.98 to −0.18, p=0.004; I²=68%). The SMD exceeds the minimal clinically important difference of −0.37. Adverse events: not significantly different from placebo. All four included trials rated high risk of bias.
Confirmed direction of effect for collagen derivatives on osteoarthritis symptoms across a much larger evidence base. Sub-analysis: UC-II (undenatured Type II at ~40 mg) and hydrolysed collagen (gram-scale Type I/III) appear to act via different mechanisms but both show benefit. Hydrolysed collagen — i.e. this product class — has the more consistent pain-relief signal.
Statistically significant increases in BMD at lumbar spine and femoral neck versus placebo. Bone formation marker P1NP increased; resorption marker CTX-1 decreased. The clinical anchor for the bone claim. A 4-year observational follow-up reported progressive BMD gains. Industry sponsorship: Gelita AG funded the work and supplied product.
Fat-free mass: +4.2 kg (treatment) vs +2.9 kg (placebo). Fat mass: −5.4 kg vs −3.5 kg. Isokinetic quadriceps strength: +16.5 Nm vs +7.3 Nm. Magnitudes are unusually large — a published BJN response argued the FFM gain exceeded what protein supplementation typically produces (about 2.7–5.6× the meta-analytic estimate) and approached the gain seen in testosterone-supplemented men with COPD. Possible explanations include training novelty, fluid shifts, or methodology; the result has not been clearly replicated.
Greater muscle thickness gains in biceps and quadriceps in the whey group versus the collagen group, even when leucine was matched. Performance variables (training load, average power, peak power) increased similarly. The likely explanation: collagen delivered ~7.7 g of EAAs per serving versus ~13.9 g for whey — below the threshold for maximal muscle protein synthesis stimulation. This is why the muscle-claim verdict above is WEAK.
Consistent increases in skin hydration measured by the Corneometer CM825 across four trials (Yoon, Kim, Bolke, Tak). Proposed mechanism elaborated: Gly-Pro-Hyp peptide levels peak 1–2 hours post-ingestion, recognised by dermal fibroblasts, triggering collagen, elastin, and hyaluronic acid synthesis (Asserin et al.). Pro-Hyp has been detected in urine after collagen intake, confirming systemic exposure and tissue distribution.
What the evidence pattern tells us
Three things stand out across this literature. First, the effect is real but modest: small-to-moderate SMDs are the norm, not large. Second, the funding-source signal in skin trials matters — the 2025 AmJMed analysis is the most methodologically careful look at this question, and it suggests the headline pooled estimates are inflated by industry-funded work. Third, the strongest single result in the literature is bone: König 2018 at 5 g/day delivered measurable BMD change over 12 months in a placebo-controlled design, replicated at 4-year follow-up.
A reader who only wanted one practical takeaway: take 5 g/day if your goal is bone (cheaper, evidence-aligned); take 10 g/day if your goal is skin or joints (matches the trial doses where effects were demonstrated). Either way, expect modest improvement and pair it with sleep, sun protection, and resistance training, which are far larger levers.
Frequently asked about Collagen Peptides
It depends on the goal. For bone (postmenopausal women), König 2018 used 5 g/day and showed BMD change — there is no good reason to take more for that indication. For skin, trials run from 2.5 g to 10 g and many positive trials use 5 g. For knee osteoarthritis pain, the Benito-Ruiz trial used 10 g. For sarcopenia, Zdzieblik used 15 g.
A reasonable default is 5 g/day for bone or skin, and 10 g/day if joints are the main reason. The DoNotAge page recommends 10 g/day across the board, which is reasonable.
Most of the dose is hydrolysed to free amino acids. But a small bioactive fraction — di- and tripeptides like Pro-Hyp and Hyp-Gly — survives intestinal digestion intact, appears in plasma for several hours after ingestion, and reaches dermal fibroblasts and chondrocytes where it acts as a signal (not a building block). This signalling fraction is what differentiates collagen from "just eating chicken." Whether the marginal benefit justifies the price over a high-quality whey or pea protein depends on your goal: for skin and joints, yes; for muscle, no.
The DoNotAge page describes "rigorous in-house testing" and references GMP and ISO 9001 manufacturing standards. At the time of this review, the product page does not display an independent third-party CoA. This is a routine industry practice rather than a red flag — most consumer collagen products do not publish independent assays — but it is a fair question to ask and a CoA can usually be obtained on request.
Caution warranted. Collagen is exceptionally high in hydroxyproline, which is metabolised to glyoxylate and then oxalate. Older work showed gelatin ingestion increased urinary oxalate excretion. For healthy kidneys this is inconsequential. For anyone with a history of calcium-oxalate stones, this is one of the few supplements where a real metabolic concern exists. Talk to your nephrologist; staying well-hydrated and taking calcium with meals (which binds dietary oxalate in the gut) helps.
No, and not close. Collagen is an incomplete protein — it has no tryptophan — and contains roughly one-third the leucine of whey, which makes it a poor stimulator of muscle protein synthesis. Even when leucine is artificially matched in head-to-head trials (Oikawa et al. 2020), whey produces more muscle thickness over 10 weeks of resistance training. Collagen is a useful adjunct for connective-tissue support, not a primary muscle-building protein.
There is no specific clinical data on collagen peptide supplementation in pregnancy. Collagen is a food-grade animal protein consumed regularly in cooking, so the absolute risk is presumably low, but the responsible answer is to discuss with your obstetrician before adding any non-essential supplement during pregnancy or while breastfeeding.
Skin and joint trials typically show measurable change at 8–12 weeks. Bone (König 2018) requires 12 months of consistent supplementation. Subjective changes in skin texture and joint comfort are reported earlier by some users but should be treated as anecdotal until the timeframe matches the trial data.
Less than you would think. The 2023 Nutrients meta-analysis (Pu et al.) found no significant difference in skin outcomes across collagen sources. The DoNotAge hydrolysed collagen peptides supplement is bovine, which is the most common source in trials and well-established. Marine collagen is sometimes preferred for ethical or allergen reasons. Chicken-derived collagen is more often used for Type II / cartilage formulations.
Reasonable. Vitamin C is a cofactor for prolyl hydroxylase, the enzyme that produces hydroxyproline in endogenous collagen synthesis. Some trials co-supplement; others do not. If you already have adequate vitamin C intake from diet (citrus, peppers, berries), there is no clear benefit to adding more. If you take a multivitamin or eat a varied diet, you are likely covered.
Some industry-funded trials report modest improvements in self-assessed skin "firmness." The independent literature does not support strong claims here. If skin elasticity is the goal, expect small, gradual changes that may or may not be visible to the reader. Topical retinoids and adequate sun protection produce larger measurable effects.
Type I and III are the dominant collagen types in skin, tendon, ligament, bone matrix, and vascular tissue. Type II is the dominant type in articular cartilage. The supplement is positioned for general structural support; its joint benefit, where it exists, comes from peptide signalling to chondrocytes rather than direct cartilage incorporation. For undenatured Type II collagen (the cartilage-specific approach), look for UC-II products at the much smaller ~40 mg dose — a different mechanism entirely.
Read deeper
Read deeper on Collagen Peptides
What the evidence actually shows across skin, joints, bones, and muscle — and where the honest gaps are.
Read full article
The mechanism, the source, and what the powder in the scoop actually does inside you — in plain English.
Read full article
A doctor's direct answer to one of the most common questions in the category — and what actually matters if the terms don't.
Read full article
How the manufacturing process determines what actually reaches your bloodstream — and one form that breaks the rules.
Read full article
A skeptic-charitable read of the collagen trial literature — including the funding-source finding that changes the skin picture.
Read full article
The best-studied indication for hydrolysed collagen — and the one where the marketing runs strongest against the honest evidence.
Read full articleYour protocol starts here
Ready to rebuild your body's structural foundation?
Purchase directly from DoNotAge using our clinical code to guarantee authenticity and the lowest available price.
We earn a commission when you purchase through this link. This relationship funds our independent medical reviews.