One person's choices. The tag on any line below describes the cartilage literature, never my choice — and my choice never moves a tag. Most of this stack is taken for healing, methylation or healthspan, and the entries cover those literatures too. The entries are the reference; this page is a disclosure.
From the author · Appendix
What I take, and why
Not a recommendation — a disclosure. This is my current routine after years of trial and error, with each item linked to the entry that explains why it made the cut.
As an Amazon Associate I earn from qualifying purchases.
Morning
As an Amazon Associate I earn from qualifying purchases.
Collagen peptides
15 g — 40/40/20 FORTIGEL · TENDOFORTE · VERISOL
Pooled pain benefit in knee OA across many trials, tendon and nail results for the other two fractions, and one trial of this exact 30 g split dose that measured absorption but no change in synthesis. the entry →
Collagen peptides
15 g — 40/40/20 FORTIGEL · TENDOFORTE · VERISOLWhat I do, and why
Thirty grams a day, split morning and evening — a 40/40/20 blend of FORTIGEL, TENDOFORTE and VERISOL — with a gram of vitamin C. The effect I can see is my fingernails, which have gone thick and strong enough that other people comment on it. VERISOL is the fraction with the nail trial behind it, and that is the part of the blend I'd defend first.
Buy on Amazon↗Vitamin C
1 g liposomal, with each collagen dose
A randomised crossover in 133 people with OA found 2 g/day improved pain; plasma saturates completely at that dose, and the collagen trials pair their peptides with about 50 mg. the entry →
Vitamin C
1 g liposomal, with each collagen doseWhat I do, and why
A gram of liposomal vitamin C with each collagen dose, morning and evening. The enzymes that build collagen cannot run without vitamin C, so taking the two together is the obvious move.
Buy on Amazon↗NMN
1 g
NAD+ levels rise dose-dependently in human trials; functional endpoints are mixed, and the rodent cartilage work has no human counterpart. the entry →
NMN
1 gWhat I do, and why
A gram every morning. I take it for the NAD+ and healthspan literature, not for my knee — NAD+ falls with age, and raising it back is the one part of this that reproduces reliably in people. Any joint benefit would be a bonus I'm not counting on.
Buy on Amazon↗Trans-resveratrol
500 mg, with the NMN
A wide human trial record with mostly small or null results; the one publicly funded knee-OA trial found a 0.6-point difference on a 100-point pain scale. The pairing has been tested twice. In people, adding resveratrol to an NAD+ precursor gained 3.7 metres of walking distance where the precursor alone gained 17.6. In mice, it raised tissue NAD+ 1.59x in heart and 1.72x in skeletal muscle over NMN alone, in six tissues that did not include a joint. the entry →
Trans-resveratrol
500 mg, with the NMNWhat I do, and why
Taken with the NMN, on the sirtuin logic — NAD+ as the substrate, this as the activator. Sirtuins are the enzymes that do the repair-and-maintenance work the NAD+ is for, so supplying the fuel and leaving the engine alone never made sense to me. The one thing that gives me pause is a training study where it blunted the adaptation, and I'm still weighing that.
Buy on Amazon↗TMG (Betaine)
1.5–3 g
Betaine lowers homocysteine 12% at 1.5 g and 15% at 3 g in controlled trials; the cartilage work is rodent, at doses far above these. the entry →
TMG (Betaine)
1.5–3 gWhat I do, and why
Methylation support alongside the NMN. The dose-response was measured at exactly this range, and it sits below where the cholesterol signal turns up.
Buy on Amazon↗Zinc picolinate
50–100 mg
Zinc is a cofactor across matrix enzymes; the joint trials that exist are old RA studies, and the best-characterised cartilage pathway runs toward matrix breakdown. the entry →
Zinc picolinate
50–100 mgWhat I do, and why
For tissue repair and immune function. Zinc is structural to a long list of the enzymes that build and remodel connective tissue, and it is one of the minerals a hard training load actually depletes, so I would rather run it high than run it short. I cycle within the range rather than sit at the top of it, because of the copper trade-off the entry sets out.
Buy on Amazon↗Vitamin K2 (MK-7)
200 µg, with breakfast fat
Six cohorts link low vitamin K status to faster cartilage damage, and people on vitamin K antagonists show 2.5× the OA progression; the MK-7 trials measured arteries and bone. the entry →
Vitamin K2 (MK-7)
200 µg, with breakfast fatWhat I do, and why
Two hundred micrograms with the morning fat, for the matrix Gla protein pathway. MK-7 is the form that actually reaches the blood at capsule doses, and 200 µg sits inside the band the multi-year trials used.
Buy on Amazon↗Curcumin C3
1 g
Around forty knee-OA trials show pain relief matching NSAIDs with fewer adverse events; blood levels vary 57-fold between formulations without tracking outcomes. the entry →
Curcumin C3
1 gWhat I do, and why
A gram of the C3 extract daily, for the inflammatory side of the ledger. It is the most-trialled botanical there is for a knee, and it holds up: pain relief that matches ibuprofen and diclofenac across dozens of trials, with fewer stomach problems than either. That last part is the whole point — it does the job I would otherwise be reaching for daily NSAIDs to do, which is a trade I am happy with.
Buy on Amazon↗GHK-Cu
2 mg
A large wound-repair literature, plus three cartilage studies from one laboratory, including new hyaline cartilage in a minipig defect model. the entry →
GHK-Cu
2 mgWhat I do, and why
Two milligrams a day, for tissue repair. It drives matrix synthesis — collagen, chondroitin sulfate, decorin — and in the cartilage work the collagen the chondrocytes made was predominantly type II, which is the kind that matters. Worth knowing the copper alone reproduces several of those effects.
Evening
As an Amazon Associate I earn from qualifying purchases.
Daily
time variesDry sauna
20–30 min, traditional
Trials in rheumatic disease show short-term pain and stiffness relief; the cartilage question has not been measured in people. the entry →
Dry sauna
20–30 min, traditionalWhat I do, and why
Twenty to thirty minutes most days, and the best part of the day. The honest reading is relief and recovery rather than cartilage repair — this is the one habit I'd keep even if the joint story collapsed entirely.
Cycles
not dailyBPC-157
250 µg twice daily; 30–45 days on, 14–30 days off
Around 230 published studies, mostly rodent healing models, with a first randomised human trial in hamstring injury under way. the entry →
BPC-157
250 µg twice daily; 30–45 days on, 14–30 days offWhat I do, and why
Two doses a day, 30 to 45 days on and 14 to 30 off. I take it for healing — the tendon, ligament, muscle and gut work is a large body of research, and the mechanism papers that matter most for tendon come from groups with no connection to the original lab. A phase 2 hamstring trial is running now, and that readout is the one I'm waiting on.
TB-500
250 µg twice daily, in the same shot as the BPC-157; two cycles after a surgery or PRP, otherwise not
Repair results across cardiac, corneal and dermal models; human trials in eye and skin disease missed their primary endpoints. the entry →
TB-500
250 µg twice daily, in the same shot as the BPC-157; two cycles after a surgery or PRP, otherwise notWhat I do, and why
Two cycles after a surgery or a PRP injection, and otherwise not. When I run it, it goes in the same shot as the BPC-157, 250 µg of each twice a day, on the same 30-to-45-day cycle. The repair literature is why I reach for it when there is something to heal. Thymosin beta-4 is overexpressed across tumours and oncology's direction of travel is to block it rather than supply it — nobody has shown that giving it to a person causes anything, but it is a risk I take on for a recovery, not as a routine.
Cartalax
2 ml daily × 20 days, every 6 months
Two 2023 cell studies with cartilage endpoints and one animal study; no human trial, and nearly all of it from the originating institute. the entry →
Cartalax
2 ml daily × 20 days, every 6 monthsWhat I do, and why
Twenty days, twice a year, paired with Sigumir. The tripeptide itself has chondrocyte data behind it now — it raised SOX9, aggrecan and type II collagen in aging human stem cells — which is more than I could have said for it when I started.
Sigumir
30-day course, alongside the Cartalax
The cartilage polypeptide complex has cell-culture data; no published work establishes that the studied complex is the marketed preparation. the entry →
Sigumir
30-day course, alongside the CartalaxWhat I do, and why — against my own tag
A 30-day course alongside the Cartalax, twice a year. The idea behind the bioregulators is that short peptides pulled from a tissue signal to that same tissue, and for a cartilage extract that is a mechanism I find worth a course. What the published work has not done is establish that the complex it studies is the one in the capsule.
Epithalon
500 µg daily, 20 days, twice a year
A telomerase result from the originating institute reproduced by an outside laboratory in normal cells, lifespan tails in mice, a melatonin rise in one small human study, and no joint study of the peptide. the entry →
Epithalon
500 µg daily, 20 days, twice a yearWhat I do, and why
Twenty days, twice a year, in the same block as the Cartalax and Sigumir. I take it for the ageing biology — the telomerase work, which a laboratory with no tie to the institute has now reproduced in normal cells, and the lifespan tails in mice — and for sleep, because the pineal makes less melatonin as we age and this is the one peptide with a human study showing it brings that output back. It belongs to the same bioregulator series as the other two courses, which is why the three run together. It is amazing for resetting my circadian rhythm and getting back to a natural sleep cycle.
SS-31
4 mg daily × 25 days, twice a year
Chondrocyte rescue after impact injury in explants, within a six-hour window; the human trials are in mitochondrial, heart and eye disease, and the blinded ones mostly missed their primary endpoints. the entry →
SS-31
4 mg daily × 25 days, twice a yearWhat I do, and why
Twenty-five days, twice a year, for mitochondrial function. The cartilage-explant rescues are the most striking preclinical result in anything I take, and the molecule has an FDA approval behind it in another disease, so the safety picture is unusually well characterised for something this experimental.
Pentosan polysulfate
one subcutaneous course a year, off-label prescription
Two placebo-controlled knee trials, one on symptoms and one on synovial biomarkers, a third in viral arthralgia, decades of veterinary use, and a pivotal phase 3 with results expected in 2027. the entry →
Pentosan polysulfate
one subcutaneous course a year, off-label prescriptionWhat I do, and why
A subcutaneous course once a year, by off-label prescription. It has the strongest human trial evidence of anything I take, and the phase 3 will settle whether it stays. I have my eyes checked before each course. This drug can damage the retina after years of use, and while the reported cases were people swallowing it daily for a decade, barely any of a tablet gets absorbed — an injection delivers the lot.
What I don't take
Hyaluronic acid injections
Tens of thousands of trial patients; the pooled effect sits below clinical significance. the entry →
What I do, and why
I had a course once and felt nothing from it, so I didn't go back — and the pooled trial data reads the same way my knee did. It stays on this list because people keep being offered it.
“The tags describe the literature. The routine describes my risk tolerance.”