wasted 3 months on this stack, debrief
ok so this is going to be one of those posts where the guy who ran the experiment admits the experiment was probably flawed. feel free to pile on, thats the point. the hypothesis: cerebrolysin + dihexa as a cognitive recovery stack after a mild-to-moderate tbi (around 18 months post-injury at the time). the specific endpoints i was tracking were working memory (dual n-back scores), verbal processing speed (informal but consistent test setup), and subjective mental clarity rated daily 1-10. baseline measured for 6 weeks before touching anything. protocol: cerebrolysin at 5ml IM, 10-day cycles x2 with a 4-week break between. dihexa at 1mg intranasal per session, 5 days on 2 off throughout. ran the full stack for 12 weeks total. sourced from two different vendors, only got a COA on the cerebrolysin (HPLC, looked fine). never got one for the dihexa and thats probably confession #1. actual outcome: dual n-back moved from n=2.8 avg to n=3.1 over the 12 weeks. barely outside what id call noise. verbal processing, basically flat. subjective clarity went from about a 5.2 baseline to 5.9, but i also changed my sleep schedule in week 4 so who knows what caused what. my guesses on why it flopped: dihexa dose might be off (the intranasal bioavailability thing is genuinely unclear to me), the cerebrolysin cycles might have been too short, or, and this is the uncomfortable one, the endpoints im using are too blunt for what im hoping to detect at 18 months post-concussion. could also just be that im the wrong subject at the wrong recovery stage. not blaming vendors, not saying the compounds dont work in general. this is n=1 with several confounds baked in. what would you have done differently?
2 Replies
the dihexa intranasal thing is basically an open question. nobody has nailed down human bioavailability that route and the peptide degrades fast in nasal mucosa, protease activity up there is no joke. you might have been delivering 10% of what you thought, or 60%, nobody actually knows. the COA gap is the bigger issue honestly. dihexa is one of those where the synthesis byproducts are hard to characterize without MS, HPLC alone wont catch everything. not saying your source was bad, just saying you had an uncontrolled variable in the actual compound. also 5ml cerebrolysin per cycle is on the lower end for what some researchers use as a meaningful intervention window. the 10-day cycles are pretty standard in the clinical literature but the gap between cycles might have been too long to stack the effect. your endpoints arent wrong, theyre just noisy. dual n-back has decent test-retest variance even in healthy subjects. 0.3 improvement over 12 weeks with those confounds... yeah that could be anything.
appreciate you posting this. failure logs are worth more than most success posts because people actually include the protocol when things dont work. that said, dihexa without a COA is a pretty big hole in the design. its a short peptide, synthesis variability is real, and youre trying to detect subtle cognitive effects with informal tests. thats a lot of assumptions stacked on top of each other. the sleep schedule change in week 4 would have killed my confidence in the subjective data entirely tbh. sleep and post-tbi cognition are so tightly coupled that a 0.7 point shift on your clarity scale means basically nothing in that context. not saying the stack is useless from this. just saying the trial couldnt have told you much either way. if you rerun it id nail down the sleep variable first, get proper testing on both compounds, and extend the cerebrolysin cycles before assuming dose was adequate.