One person, 5 protocols, no control group

What I ran, and what I am running now

Every other page on this site reads someone else’s evidence. This one is mine. 5 protocols over fourteen months, transcribed from my notebook with the dates on them, the doses I actually drew, and the diet, fasting and training they were attached to, plus the one I am on today. I am publishing it because I ask every vendor and every study on this site to show its working, and it would be cowardly not to show my own.

It is not a plan for you to copy, and the reason is arithmetic rather than modesty. I changed what I injected, what I ate, when I ate it and how I trained in the same weeks, and I was treated for clinically low testosterone in the middle of it. Nothing here can tell you which part did what. Read it as a record of one man’s decisions, not a result.

The current one is here for a different reason. Publishing only finished protocols lets the author choose which ones to show. This one has not happened yet: I fly to Morocco on 7 October 2026 to run marathon des sables, 120 km over 5 days, sahara, self supported with a pack, and whatever happens there happens to a plan that was already public.

Aaron Cuha. Educational use only. This database summarizes published research and is not medical advice. Where a dose appears it is a record of what a trial or clinic ran, or what people describe doing, with its source attached. Nothing here is a recommendation for personal use.

Two mirror photographs of Aaron Cuha side by side. On the left, July 2025 at 255 pounds and 37 percent body fat. On the right, September 2026 at 175 pounds and 13 percent body fat. — July 2025, 255 lb at 37 percent body fat and 19 lb of visceral fat. September 2026, 175 lb at 13 percent and 0.53 lb of visceral fat. Fourteen months. Treat this as the least reliable thing on the page. The lighting is different, the pose is different, one was taken in a dim bathroom and the other by a bright window, and every before and after photograph ever published shares those problems. The numbers underneath it are what can actually be checked, and the sections below include the ones that did not improve.

The part of that picture nobody can see

In the middle of it I tore a distal biceps tendon

Torn distal biceps tendon, left arm, surgically repaired. It put my left arm in a cast for two weeks and a brace for four more, it cost me two Ironman events I had already entered, and it stopped every upper body lift I was doing. I have not loaded that arm since, and I will not until after 20 October.

It is the most useful thing in this notebook and I nearly left it out. Every protocol before it was a choice. This one was a constraint, applied by an event nobody scheduled, and it forced a plan built entirely around losing fat to reverse itself in a single afternoon. Calories went up. The GLP dose fell to about a tenth of what it had been. The written instruction became do not try to cut fat during the healing phase, which is the opposite of every page that came before it.

It also changed what I am training for. The two races I lost were triathlons, and a triathlon needs an arm that can swim. What replaced them is a race that does not.

Two Ironman events, already entered, cancelled.

Full cast for the first two weeks, then a removable brace to week six.

No upper body loading on the left arm at all, from the repair through to at least 20 October 2026.

Fat loss stopped being the goal, in writing, and did not restart until the sheet dated 9 May 2026.

One thing here does not reconcile and it is left visible rather than tidied. The surgical day by day protocol is dated 4 April 2026 and is written from week 0. His own account, given in September 2026, puts the tear about sixteen weeks earlier. Those cannot both be true, so the phases below are anchored to the notebook dates, which are at least checkable, and counted in weeks after the repair rather than in calendar months.

2 of the 24 compounds I took have never been given to a human being in a published study

Not “the evidence is early”. No human study of any kind. Every compound below carries the band this site computes for it from counted trials and participants, and that band is the same one you would see anywhere else here. I did not soften it for my own page.

AOD-9604Limited. Someone has been given it and a result was published, but fewer than 100 people. 1 completed trial, none stating how many people.5 reported1 completed trial, none stating how many people
ARA-290Limited. Someone has been given it and a result was published, but fewer than 100 people. 5 completed trials, 95 people given it.7 reported5 completed trials, 95 people given it
BPC-157Limited. Someone has been given it and a result was published, but fewer than 100 people. 3 completed trials, 31 people given it.10 reported3 completed trials, 31 people given it
CJC-1295Limited. Someone has been given it and a result was published, but fewer than 100 people. 2 completed trials, none stating how many people.9 reported2 completed trials, none stating how many people
DSIPLimited. Someone has been given it and a result was published, but fewer than 100 people. 5 completed trials, none stating how many people.8 reported5 completed trials, none stating how many people
GLOW blendLimited. Someone has been given it and a result was published, but fewer than 100 people. 2 completed trials, 22 people given it.9 reported2 completed trials, 22 people given it
MOTS-cLimited. Someone has been given it and a result was published, but fewer than 100 people. 1 completed trial, 30 people given it.6 reported1 completed trial, 30 people given it
PinealonLimited. Someone has been given it and a result was published, but fewer than 100 people. 2 completed trials, none stating how many people.6 reported2 completed trials, none stating how many people
ElamipretideModerate. At least one completed trial, and at least 100 people given it. 5 completed trials, 487 people given it.6 reported5 completed trials, 487 people given it
EpitalonModerate. At least one completed trial, and at least 100 people given it. 4 completed trials, 266 people given it.6 reported4 completed trials, 266 people given it
IpamorelinModerate. At least one completed trial, and at least 100 people given it. 1 completed trial, 117 people given it.9 reported1 completed trial, 117 people given it
SelankModerate. At least one completed trial, and at least 100 people given it. 3 completed trials, 192 people given it.8 reported3 completed trials, 192 people given it
SemaxModerate. At least one completed trial, and at least 100 people given it. 4 completed trials, 354 people given it.9 reported4 completed trials, 354 people given it
SermorelinModerate. At least one completed trial, and at least 100 people given it. 2 completed trials, 129 people given it.8 reported2 completed trials, 129 people given it
Testosterone cypionateModerate. At least one completed trial, and at least 100 people given it. 1 completed trial, 5,246 people given it.9 reported1 completed trial, 5,246 people given it
5-Amino-1MQPreclinical. Animals or cells only. No human study of any kind. 2 animal studies, no human study.6 reported2 animal studies, no human study
SLU-PP-332Preclinical. Animals or cells only. No human study of any kind. 2 animal studies, no human study.8 reported2 animal studies, no human study
GlutathionePreliminary. Studied in people, but no published study has given it to anyone. 2 human studies where nobody was given it.8 reported2 human studies where nobody was given it
IGF-1 LR3Preliminary. Studied in people, but no published study has given it to anyone. 1 human study where nobody was given it.8 reported1 human study where nobody was given it
KLOW blendPreliminary. Studied in people, but no published study has given it to anyone. 1 human study where nobody was given it, 1 registered with no result.11 reported1 human study where nobody was given it, 1 registered with no result
Methylene BluePreliminary. Studied in people, but no published study has given it to anyone. 3 human studies where nobody was given it.6 reported3 human studies where nobody was given it
NAD+Preliminary. Studied in people, but no published study has given it to anyone. 1 human study where nobody was given it.7 reported1 human study where nobody was given it
RetatrutideStrong. At least two completed trials, and at least 1,000 people given it. 5 completed trials, 1,345 people given it.8 reported5 completed trials, 1,345 people given it
TesamorelinStrong. At least two completed trials, and at least 1,000 people given it. 7 completed trials, 1,691 people given it.10 reported7 completed trials, 1,691 people given it

Two of those rows are vendor blends, and they hide more than they show. GLOW is three peptides in one vial and KLOW is four, adding KPV, which has two animal studies and no human study of any kind. So the count of things I injected with no human evidence behind them is three, not two. And the blends themselves are worse than their parts: no study has tested either combination in any species, so the band beside them describes the ingredients rather than the thing in the syringe.

Some of what I injected is not on that list at all, because my own notebook records a product name and nothing else: NAD+ Blend, NAD Blend, GHRH 4x Blend. The NAD blend turned out to be seven actives plus an anaesthetic once I went and read the seller’s ingredient list, and one of them is why I changed my dosing. The GHRH blend is still unresolved. I injected it nightly for months on a rising dose, from 20 units in February to 40 by May, and I came off it without ever finding out what was in it. A blend I cannot name is a blend I cannot check, and that is my gap, not the database’s.

Part one

What I actually ran

Five dated pages, in order, from the morning I started to the sheet I was carrying in May. Each one gets what I injected, what I ate alongside it, what changed from the page before, what was measured, and what it cannot tell you.

Phase 1 · July 2025

The first stack

Notebook page dated 21 July 2025. Goal: Start. Fat loss, with everything running at once.

Written four days after the morning he decided to start, and it reads like it: a checklist rather than a plan, eleven items with durations attached and no phasing. Everything begins on the same day. Testosterone is the lowest dose it would ever be at 50 mg twice a week, and the GLP is written as "GLP-3 RT", which is the vendor code for retatrutide rather than a name any label uses.

What I injected and swallowed

When	What	Dose	Evidence
On waking	SLU-PP-3327 days a week, 12 weeks	1 capsuleoral	Preclinical. Animals or cells only. No human study of any kind. 2 animal studies, no human study.
On waking	NAD+ Blend7 days a week	1 ml, 100 unitsSubQ or IM	composition not recorded
On waking	Glow Blend5 days a week	25 unitsSubQ	Limited. Someone has been given it and a result was published, but fewer than 100 people. 2 completed trials, 22 people given it.
On waking	MOTS-cMonday, Wednesday, Friday, 12 weeks	3.5 mg, 70 unitsSubQ	Limited. Someone has been given it and a result was published, but fewer than 100 people. 1 completed trial, 30 people given it.
On waking	AOD-96047 days a week	0.4 mg, 16 units	Limited. Someone has been given it and a result was published, but fewer than 100 people. 1 completed trial, none stating how many people.
On waking	Testosterone cypionateMonday and Tuesday	50 mg, 25 unitsIM, glute	Moderate. At least one completed trial, and at least 100 people given it. 1 completed trial, 5,246 people given it.
11:00 AM, first meal	GLP-3 RT, his code for retatrutide5 days a week, 12 weeks	2 mg, 20 unitsSubQ	Strong. At least two completed trials, and at least 1,000 people given it. 5 completed trials, 1,345 people given it.
11:00 AM	5-Amino-1MQ7 days a week, 12 weeks	50 mg, 1 capsuleoral	Preclinical. Animals or cells only. No human study of any kind. 2 animal studies, no human study.
3:00 PM	SLU-PP-3325 days a week, 12 weeks	1 capsuleoral	Preclinical. Animals or cells only. No human study of any kind. 2 animal studies, no human study.
9:00 PM, about 3 hours after dinner	Tesamorelin7 days a week, 12 weeks	2 mg, 40 unitsSubQ	Strong. At least two completed trials, and at least 1,000 people given it. 7 completed trials, 1,691 people given it.
9:00 PM	GlutathioneMonday, Wednesday, Friday, 4 weeks	200 mg, 100 unitsSubQ	Preliminary. Studied in people, but no published study has given it to anyone. 2 human studies where nobody was given it.
9:00 PM, post workout	Glow Blendtwice a week	25 unitsSubQ	Limited. Someone has been given it and a result was published, but fewer than 100 people. 2 completed trials, 22 people given it.

What actually happened

17 July 2025: 255 lb at the start.
15 December 2025: 189 lb and 20 percent body fat, on the day he left for Italy.
66 lb over about five months, roughly 3 lb a week averaged across the whole run.

This is the only phase of the three with a start and an end weight attached to it, and it is the largest change of the three by a wide margin. It is also the phase where attribution is weakest, because it is the phase where everything changed at once: eleven compounds began on the same day, alongside a strict diet and a training programme he was following for the first time in eighteen months. A 66 lb loss over five months is what a large sustained calorie deficit produces on its own. Nothing here separates that from the compounds sitting on top of it.

What this phase cannot tell you

Eleven things start on the same day, so nothing that followed can be attributed to any one of them. The weight change is real and recorded, and it is the number least in need of a peptide to explain it. The retatrutide is recorded under a vendor code rather than a name, which is exactly the naming problem this site documents on the compound pages, and it means the identity rests on the seller's labelling.

Phase 2 · December 2025

The Italy cycle

Notebook page dated 8 December 2025. Goal: Hold the protocol together around travel and shift work

The first version built around a life rather than around a day. Doses move to whichever window actually exists on a given day: fasted morning on some, the pre-shift gap at 12:30 on others, pre-bed on all of them. Testosterone has doubled to 100 mg twice a week. Tesamorelin has gone from every day to three days a week, and CJC-1295 with ipamorelin appears on the days it does not. This is the page that carries the draw volume table, which is the only place in the notebook where units are tied to concentrations.

What changed from the phase before

Testosterone doubled, 50 mg twice a week to 100 mg twice a week.
Tesamorelin went from daily to three days a week, with CJC-1295 and ipamorelin filling the other nights.
Glow Blend was replaced by KLOW, which adds KPV to the same three peptides.
Epitalon and MOTS-c were added on a fixed weekly rhythm.
Retatrutide went from five days a week to two, at the same 2 mg.

What I injected and swallowed

When	What	Dose	Evidence
Monday, fasted morning	Testosterone cypionate0.5 ml, 25G 1 inch needle	100 mgIM	Moderate. At least one completed trial, and at least 100 people given it. 1 completed trial, 5,246 people given it.
Monday, fasted morning	Retatrutide	2 mg, 33 unitsSubQ	Strong. At least two completed trials, and at least 1,000 people given it. 5 completed trials, 1,345 people given it.
Monday, fasted morning	NAD Blend	1 ml, 100 unitsIM or SubQ	composition not recorded
Monday, fasted morning	5-Amino-1MQ	1 capsuleoral	Preclinical. Animals or cells only. No human study of any kind. 2 animal studies, no human study.
Monday, fasted morning	SLU-PP-332	1 capsuleoral	Preclinical. Animals or cells only. No human study of any kind. 2 animal studies, no human study.
Monday, pre-bed	Tesamorelin	2 mg, 40 unitsSubQ	Strong. At least two completed trials, and at least 1,000 people given it. 7 completed trials, 1,691 people given it.
Monday, pre-bed	KLOW Blend	1.25 mg, 25 unitsSubQ	Preliminary. Studied in people, but no published study has given it to anyone. 1 human study where nobody was given it, 1 registered with no result.
Monday, pre-bed	Epitalon	1 to 2 mg, 20 to 40 unitsSubQ	Moderate. At least one completed trial, and at least 100 people given it. 4 completed trials, 266 people given it.
Tuesday, pre-bed	CJC-1295on the days tesamorelin is not run	100 mcg, 5 unitsSubQ	Limited. Someone has been given it and a result was published, but fewer than 100 people. 2 completed trials, none stating how many people.
Tuesday, pre-bed	Ipamorelin	200 mcg, 8 unitsSubQ	Moderate. At least one completed trial, and at least 100 people given it. 1 completed trial, 117 people given it.
Wednesday, pre-shift	MOTS-c	3.5 mg, 70 unitsSubQ	Limited. Someone has been given it and a result was published, but fewer than 100 people. 1 completed trial, 30 people given it.
Most days, pre-shift	NAD+ Pure	1 ml, 100 unitsIM	Preliminary. Studied in people, but no published study has given it to anyone. 1 human study where nobody was given it.
Thursday, pre-shift	Testosterone cypionatethe second of two weekly injections	100 mgIM	Moderate. At least one completed trial, and at least 100 people given it. 1 completed trial, 5,246 people given it.
Thursday, pre-shift	Retatrutide	2 mg, 33 unitsSubQ	Strong. At least two completed trials, and at least 1,000 people given it. 5 completed trials, 1,345 people given it.

How the eating actually went

The three day types he later wrote up formally were already in use here. Highest carbohydrate days included wine, pasta and dessert; the middle days allowed some pasta; the lowest days were protein only.
Four or five of the highest days in a typical week, which is a permissive pattern rather than a restrictive one.
Training five days a week wherever the schedule allowed. Coaching ran 2 PM to 2 AM Italy time, so the gym session frequently came after finishing at 2 AM.
Around 10,000 steps a day, walked rather than programmed.

How the week was shaped

Monday, Wednesday and Friday carry tesamorelin pre-bed. Tuesday and Thursday carry CJC-1295 with ipamorelin instead.
Testosterone runs Monday and Thursday, splitting the weekly total rather than taking it in one injection. That split is a position he has published separately, and his own bloodwork does not test it.
Saturday drops to half a NAD dose and the growth hormone secretagogues become optional. Sunday is one oral capsule and a rest day.
The stated reason for the whole arrangement: shift demands, inventory, circadian rhythm and growth hormone timing.

What actually happened

Start of Italy, 15 December 2025: 189 lb, 20 percent body fat.
After about 60 days: 189 lb, 20 percent body fat.
No change either way.

Holding is the result here, and it is a more interesting one than it looks. This was a travel period on a permissive diet: wine, pasta and dessert on the highest carbohydrate days, some pasta on the middle days, protein only on the lowest, with four or five of the highest days in a typical week. He trained five days a week where the schedule allowed, which often meant going to the gym after finishing coaching at 2 AM local time, and walked around 10,000 steps daily. Sixty days of that produced no regain after a 66 lb loss, which is the phase where most of it comes back.

What this phase cannot tell you

Nothing was measured here except weight and body fat, and neither moved, so this phase has no internal signal at all. A schedule that holds a result is not evidence that the schedule held it: the training and the walking alone would account for maintenance at this intake. The draw volume table makes the doses reproducible, which is more than most published protocols manage, but reproducible is not the same as tested.

Phase 3 · February 2026

Hybrid recomp, cutting to sub-10 percent

Notebook page dated 16 February 2026. Goal: A 12 to 18 week cut to under 10 percent body fat while holding half marathon training

The most structured of the three, and the only one written as a single system rather than a list. It names its own goal in the first line: aggressive fat loss, preserve muscle, protect joints, maintain endurance performance. The day starts at 4:00 AM and every injection has a fixed slot inside it. Diet becomes three named day types with different carbohydrate loads, training becomes a seven day split, and a 36 hour fast is placed every ten days on a rule rather than by feel. Tesamorelin has disappeared, replaced by something written only as "GHRH 4x Blend".

What changed from the phase before

Tesamorelin and the CJC-1295 with ipamorelin pairing both disappear, replaced by a single unnamed GHRH 4x Blend at 9:30 PM with a written titration from 20 units to 25 to 30 after two to three weeks.
AOD-9604 moves to every single morning, fasted.
Retatrutide gains a titration range for the first time, 2 to 3 mg rather than a fixed 2 mg.
KLOW moves from a general pre-bed dose to a targeted post-lift injection around the knees.
Epitalon and MOTS-c are not on this page at all.
Diet and fasting stop being implicit and become written systems with numbers.

What I injected and swallowed

When	What	Dose	Evidence
4:35 AM, fasted	AOD-9604every day	12 to 16 unitsSubQ	Limited. Someone has been given it and a result was published, but fewer than 100 people. 1 completed trial, none stating how many people.
4:35 AM	NAD+ Pure or NAD Blendwhich one depends on the day	1 ml, 200 mg for the PureIM	Preliminary. Studied in people, but no published study has given it to anyone. 1 human study where nobody was given it.
4:35 AM	SLU-PP-332	250 mcgoral	Preclinical. Animals or cells only. No human study of any kind. 2 animal studies, no human study.
4:35 AM	5-Amino-1MQ	50 mgoral	Preclinical. Animals or cells only. No human study of any kind. 2 animal studies, no human study.
4:35 AM	Testosterone cypionateMonday and Thursday only	100 mgIM	Moderate. At least one completed trial, and at least 100 people given it. 1 completed trial, 5,246 people given it.
11:30 AM, before lunch	RetatrutideMonday and Thursday	20 to 30 units, 2 to 3 mg depending on titrationSubQ	Strong. At least two completed trials, and at least 1,000 people given it. 5 completed trials, 1,345 people given it.
After the afternoon lift	KLOWinjected around the knees	25 unitsSubQ	Preliminary. Studied in people, but no published study has given it to anyone. 1 human study where nobody was given it, 1 registered with no result.
9:30 PM, fasted	GHRH 4x Blendcomposition not recorded in the notebook	20 units, rising to 25 to 30 after 2 to 3 weeksSubQ	composition not recorded

The day

4:00 AM wake, water with sea salt, lemon and methylene blue.
4:05 to 4:10 AM cold plunge, 3 to 5 minutes.
4:15 to 4:35 AM sauna, 20 minutes.
4:50 AM mobility, 5:00 to 5:45 AM run.
8:30 to 9:00 AM coffee, deliberately at least 90 minutes after waking.
First meal 9:30 to 10:30 AM. Last bite 6:30 PM. Growth hormone secretagogue 7:45 to 9:30 PM. In bed 8:00 PM.

Three day types, same protein

GREEN: 225 g carbohydrate, 200 g protein, 110 g fat. Four meals at 9:30, 12:30, 15:30 and 18:00.
YELLOW: 150 g carbohydrate, 200 g protein, 110 g fat.
RED: 50 g carbohydrate, 200 g protein, 125 g fat. Meals shift later, 10:00 to 18:00.
Protein is held at 200 g on every day type. Only carbohydrate moves, and fat moves slightly to compensate.
His stated reason for four feedings rather than two: last food 90 minutes before the growth hormone pulse, carbohydrate front loaded relative to bedtime, sleep protected. He describes the result as a 14 hour fasting window and says that is enough.

The 36 hour fast, every 10 days

Finish dinner 6:30 PM, fast the whole of the next day, break it at 6:30 or 9:30 AM the morning after.
Placed after a Yellow or Green day. Never before a heavy lift or a long run.
The fast day becomes a Red day with zero carbohydrate. Light walk, mobility and Zone 2 for 30 minutes at most are allowed; heavy lifting, intervals and long runs are not.
Macros are deliberately not cut to compensate. His reasoning: the fast removes 2,200 to 2,500 calories, which across ten days is an extra 220 to 250 a day, and cutting daily calories on top of that would tank performance.
Breaking it: 40 to 50 g protein, 40 to 60 g carbohydrate, low fat, then a normal meal 3 to 4 hours later. No BCAAs during the fast. Water, electrolytes and black coffee only.

The training week

Monday: Zone 2 run 30 to 45 minutes in the morning, heavy lower body lift late afternoon.
Tuesday: intervals or tempo run, optional light kettlebell or accessory work in the evening.
Wednesday: Zone 2 run.
Thursday: hill repeats or intervals, optional short upper body lift.
Friday: run in the morning, upper body hypertrophy in the afternoon.
Saturday: long run, accessory lift or kettlebells early afternoon. No growth hormone secretagogue.
Sunday: mobility or a light walk. No growth hormone secretagogue.

No outcome was recorded for this phase. That is a gap in the notebook, not a result.

What this phase cannot tell you

This is a plan, dated at its start, and no outcome is attached to it. It sets a target of 12 to 18 weeks to sub-10 percent and states an expected rate of 1.2 to 1.5 lb of fat a week from a starting point of about 17 percent, which is a projection and not a measurement. The GHRH 4x Blend is the largest single gap: it is injected nightly, titrated upward on a schedule, and the notebook never records what is in it, so it cannot be tied to a compound page or an evidence band.

Phase 4 · April 2026

The repair, when the plan stopped being about fat

Notebook page dated 4 April 2026. Goal: Heal a surgically repaired distal biceps tendon and lose as little muscle as possible doing it

This is the page where everything inverts. A torn distal biceps tendon put the left arm in a cast and then a brace, and the protocol that had been written to strip fat was rewritten in an afternoon to do the opposite. Calories go up to maintenance or a slight surplus. Retatrutide drops to 0.25 mg a week, roughly a tenth of what it had been. The written rule is blunt: do not try to cut fat during the healing phase. The training question becomes how to load a muscle you are not allowed to move, and the answer is electricity, blood flow and the other arm.

What changed from the phase before

The goal reverses. Fat loss stops being the objective and healing replaces it, in writing.
Retatrutide falls from 2 to 3 mg to 0.25 mg a week, and does not restart until week 2.
Calories go up rather than down, to 2,400 to 2,700, with carbohydrate deliberately kept high.
The unnamed GHRH 4x Blend disappears and is replaced by CJC-1295 with ipamorelin at named microgram doses, then by tesamorelin once the brace comes off.
IGF-1 LR3 appears for the first and only time in the whole notebook, tied to the moment immediately after electrical stimulation.
SS-31 and BPC-157 appear for the first time.
Upper body lifting stops completely on the left and goes heavy on the right.

What I injected and swallowed

When	What	Dose	Evidence
3:45 AM, fasted, weeks 0 to 2	CJC-1295	150 mcgSubQ	Limited. Someone has been given it and a result was published, but fewer than 100 people. 2 completed trials, none stating how many people.
3:45 AM, fasted, weeks 0 to 2	Ipamorelin	150 mcgSubQ	Moderate. At least one completed trial, and at least 100 people given it. 1 completed trial, 117 people given it.
3:45 AM, fasted	SS-31, elamipretide	2 mg	Moderate. At least one completed trial, and at least 100 people given it. 5 completed trials, 487 people given it.
3:45 AM, fasted	MOTS-c2 to 3 times a week	5 mg	Limited. Someone has been given it and a result was published, but fewer than 100 people. 1 completed trial, 30 people given it.
5:10 AM, first meal after training	KLOW	not recorded on this pageSubQ	Preliminary. Studied in people, but no published study has given it to anyone. 1 human study where nobody was given it, 1 registered with no result.
5:10 AM, first meal	BPC-157written as optional	not recorded on this page	Limited. Someone has been given it and a result was published, but fewer than 100 people. 3 completed trials, 31 people given it.
7:00 to 7:30 PM, about 90 minutes after the last meal	CJC-1295	150 mcgSubQ	Limited. Someone has been given it and a result was published, but fewer than 100 people. 2 completed trials, none stating how many people.
7:00 to 7:30 PM	Ipamorelin	150 mcgSubQ	Moderate. At least one completed trial, and at least 100 people given it. 1 completed trial, 117 people given it.
Weekly, from the start of week 2	Retatrutidedown from 2 to 3 mg in the phase before	0.25 mgSubQ	Strong. At least two completed trials, and at least 1,000 people given it. 5 completed trials, 1,345 people given it.
3:45 AM, fasted, weeks 3 to 6	Tesamorelinreplaces the morning CJC-1295 once the brace comes off	2 mgSubQ	Strong. At least two completed trials, and at least 1,000 people given it. 7 completed trials, 1,691 people given it.
3:45 AM, fasted, weeks 3 to 6	Ipamorelin	100 mcgSubQ	Moderate. At least one completed trial, and at least 100 people given it. 1 completed trial, 117 people given it.
Midday, immediately after the electrical stimulation and cuff work, weeks 3 to 6	IGF-1 LR3the written rule is that it is only ever given after stimulation, never on its own	20 to 30 mcg	Preliminary. Studied in people, but no published study has given it to anyone. 1 human study where nobody was given it.

Replacing load you are not allowed to apply

The rehab page sets it out as a substitution table. Mechanical load becomes electrical stimulation. Blood flow becomes intermittent cuff occlusion, compression and movement. Neural drive becomes training the opposite arm and contracting the injured one mentally. Metabolic stress becomes blood flow restriction, once it is allowed.
Electrical stimulation ran 45 to 60 minutes a day in the cast phase and 60 to 75 minutes a day in the brace phase, split across left chest, left front shoulder and, carefully, the left biceps.
Pad placement was the safety mechanism. Both pads sat on the upper half of the biceps, deliberately away from the repaired tendon, and the written reason is that early tendon to bone healing is fragile and an electrical contraction can pull on it.
Intensity was written on a 1 to 10 scale rather than left to feel. Biceps: nothing at all in week 0 to 1, 4 to 6 in weeks 1 to 2, 6 to 7 in weeks 3 to 4, 7 to 9 in weeks 5 to 6. Chest and shoulder ran higher throughout because they were never the injury.
Cuff occlusion five minutes on and five minutes off, three to four cycles, once or twice a day, with the cuff high on the upper arm.
Heavy training on the right side four to five days a week at a hard effort, on the stated theory that it preserves the left.
Sauna 20 to 30 minutes, four to six days a week. Compression and cooling 20 to 30 minutes, once or twice a day.
Stop rules, written down in advance: sharp pain at the tendon, a pulling sensation near the repair, or a swelling spike after stimulation.

Eating to heal rather than to shrink

2,400 to 2,700 calories, described as maintenance or a slight surplus, adjusted by output.
200 g of protein, fixed, spread across four meals. Roughly 50 to 60 g in the post training meal and 40 to 50 g in each of the other three.
Carbohydrate is explicitly not the enemy on this page. 180 to 250 g on lower body training days, 120 to 180 g on cardio days, 80 to 120 g on rest days.
The list of things not to do is the most specific part of it: do not drop calories because the appetite is gone, do not go low carbohydrate, do not skip meals, do not fast aggressively, do not try to cut fat during the healing phase.
A note on what the drug does to judgement: on retatrutide he does not feel hunger, and the page says in as many words that this is not permission to eat less. Eat on the schedule, not on the appetite. Use shakes if the food will not go down.
Collagen 10 to 20 g daily, sodium and electrolytes, and glycine listed as an optional addition.

Why the meals sit where they do

The stated rule is that the growth hormone peptides need a low insulin window, so meals are placed between the pulses rather than around them.
No food for 60 to 90 minutes before the evening peptides. No carbohydrate immediately before any of them.
Four meals a day at 5:10 AM, 9:00 AM, 1:00 PM and 5:30 PM, with the last bite at 5:30 PM and the night dose at 7:00 to 7:30 PM.
The coaching day version of the same plan is deliberately lazier, because it had to survive a working day: a yogurt bowl with protein powder and granola every morning without variation, a fallback rule that if the 9 AM meat is missed it becomes two scoops of protein, then a real lunch and a real dinner, and the kitchen closed after that.

What actually happened

Two Ironman events, already entered, cancelled.
No upper body loading on the left arm from the repair through to at least October 2026.
No body composition figure was recorded at any point during this phase.

The outcome that matters here is not a number. He healed, kept training, and is still racing, and the only honest thing to say is that a repaired tendon heals on a surgical timeline whether or not anything is injected alongside it. What this phase can show is the decision making: the plan was rewritten within days to stop cutting, and the thing most likely to have preserved muscle is the one with the least glamour attached, which is eating 200 g of protein a day and stimulating the muscle every single day.

What this phase cannot tell you

Nothing was measured. No weight, no body fat, no bloodwork, no imaging, and no strength testing on either arm, which is the measurement that would actually have told him whether any of the substitution worked. The most aggressive single item in the whole notebook is here: IGF-1 LR3, whose entire human record on this site is one study in which nobody was actually given it, injected into the window immediately after stimulating a recently repaired tendon. It is the one thing on this page I would expect a surgeon to object to, and no surgeon was asked.

Phase 5 · May 2026

Back to the cut with one arm out

Notebook page dated 9 May 2026. Goal: Restart fat loss while the left arm is still off limits

A printed sheet for the binder, headed "shots only, by day and time". It is the least literary page in the notebook and the most useful, because it is the one he actually carried. Fat loss is back on: retatrutide returns at 30 units with a written step to 40 in week two, and the nightly unnamed 4x blend is back with it. The healing stack is stripped back to KLOW, and the days are colour coded green, red and yellow by how hard the day is going to be, which is the same system the diet uses.

What changed from the phase before

Retatrutide restarts and climbs, 0.25 mg a week back up to 30 then 40 units.
The unnamed nightly 4x blend returns at 40 units, twice the February starting figure and above the top of the February titration.
CJC-1295, ipamorelin, tesamorelin, IGF-1 LR3, SS-31 and BPC-157 all come off.
MOTS-c and epitalon come back.
Testosterone is not written on this sheet at all, which is a gap in the sheet rather than evidence it stopped.

What I injected and swallowed

When	What	Dose	Evidence
AM fasted, every day	AOD-9604	16 unitsSubQ	Limited. Someone has been given it and a result was published, but fewer than 100 people. 1 completed trial, none stating how many people.
AM fasted, every day	KLOWplus 15 to 25 units after a hard workout	25 unitsSubQ	Preliminary. Studied in people, but no published study has given it to anyone. 1 human study where nobody was given it, 1 registered with no result.
AM fasted, Monday and Thursday	MOTS-c	50 unitsSubQ	Limited. Someone has been given it and a result was published, but fewer than 100 people. 1 completed trial, 30 people given it.
Mid morning, after food, seated	NAD+the seated instruction is his own, and it is there because IM NAD drops blood pressure	0.5 ml, 100 mg on easy days; 0.75 ml, 150 mg on coaching daysIM	Preliminary. Studied in people, but no published study has given it to anyone. 1 human study where nobody was given it.
30 to 60 minutes before lunch, Monday and Thursday	Retatrutide	30 units, rising to 40 in week 2SubQ	Strong. At least two completed trials, and at least 1,000 people given it. 5 completed trials, 1,345 people given it.
Pre-bed, every day	Epitalon	50 unitsSubQ	Moderate. At least one completed trial, and at least 100 people given it. 4 completed trials, 266 people given it.
Bedtime, fasted, Monday to Friday	GHRH 4x Blendstill no composition recorded, and now at double the units of February	40 unitsSubQ	composition not recorded

The one instruction on the page that is not a dose

"4x blend nights need a real fasted window. Since you eat last meal 5 to 6 pm and sleep 8:30 to 9, keep dinner lower fat on 4x nights so gastric emptying does not lag from retatrutide."
That is the only sentence on the sheet, and it is a good one: it is the only place in the notebook where he writes down that one drug is interfering with the timing of another.
Saturday and Sunday drop the 4x blend entirely. Tuesday, Wednesday and Thursday are marked as coaching days and carry the higher NAD dose.

No outcome was recorded for this phase. That is a gap in the notebook, not a result.

What this phase cannot tell you

It is a dosing sheet, so it records no diet, no training, no bodyweight and no outcome. Two things on it are worth flagging rather than admiring. The nightly blend is still unnamed and is now at 40 units, which is more of a thing he cannot identify, not less. And testosterone is absent from a sheet that is meant to be a complete shots list, which means either it stopped or the sheet is incomplete, and the notebook does not say which.

The three things I did instead of lifting, and what has actually been measured for them

None of these is a compound, so none of them has a profile on this site and none of them carries an evidence band. That is a gap in the database rather than a verdict, and two of the three have a better human record than most of what I was injecting at the time.

Electrical stimulation, 45 to 75 minutes a day

This is the best supported thing I did in that phase. A review of interventions against disuse atrophy concludes that reintroducing contraction through neuromuscular electrical stimulation raises muscle protein synthesis both fasted and after meals, and can prevent or blunt muscle loss during short term disuse. PubMed 28970205.

The same review contradicts the other half of my plan. Eating protein above habitual intake does not prevent muscle loss during disuse in otherwise healthy people. I held 200 g a day through the whole thing on the assumption that more protein protected the arm. On the evidence, the stimulation was doing that work and the extra protein was along for the ride.

Training the other arm heavy, four to five days a week

My notes call this CNS crossover and treat it as obvious. It is real and it has a number on it. A meta-analysis of 31 randomised trials pooling 785 people found that training one limb produced an 11.9 percent strength increase in the untrained opposite limb, 95 percent confidence interval 9.1 to 14.8. For upper limbs specifically it was 9.4 percent, and eccentric training produced the largest effect at 17.7 percent. PubMed 28936703.

Two honest caveats. The authors report a high risk of bias across the included studies. And those trials measured strength gained in a healthy untrained limb, not strength held in an immobilised and surgically repaired one, which is what I was asking it to do.

Cuff occlusion, compression and cooling

Blood flow restriction after surgery is an active research area with a real rehabilitation literature, mostly in knees rather than elbows. I am not going to summarise a field on the strength of my own use of it, and this site has no profile for it yet, so the honest statement is that I used it, I cannot tell you what it did, and the elbow evidence is thinner than the knee evidence I was reasoning from.

The thing I would change: I never tested either arm. No grip strength, no range of motion figures, nothing. Six weeks of daily electrical stimulation and heavy contralateral training, and the one measurement that would have told me whether any of it worked is the one I did not take.

What was actually measured

Numbers from my own scans and lab reports, transcribed. The reports themselves carry a medical record number, a date of birth and a home address, and none of that is on this site: a record number next to a date of birth is how somebody else gets into your medical records, and it cannot be unpublished. The values are mine to show. The identifiers are not.

Body composition, including the part going the wrong way

Date	Weight	Body fat	How it was measured
7 October 2024	228.9 lb	32.1 percent	InBody 770 bioimpedance scanNine months before he started. Skeletal muscle mass 88.2 lb, body fat mass 73.5 lb, BMI 33.8, visceral fat level 15, visceral fat area 158.4 cm2.
17 July 2025	255 lb	37 percent	His own record of the day he started26 lb heavier than the scan nine months earlier. This is the start of the first stack.
15 December 2025	189 lb	20 percent	His own record, the day he left for Italy66 lb below the start, after about five months on the first stack.
Mid February 2026	189 lb	20 percent	His own record, after about 60 days in ItalyUnchanged over the Italy cycle. Held rather than lost, on a diet that included wine, pasta and dessert on the highest carbohydrate days.
September 2026	175 lb	13 percent	DEXA80 lb below the July 2025 start and 24 points of body fat below it. Visceral fat mass 0.53 lb against 19 lb at the start. This point sits about six months after a surgically repaired biceps tendon and a phase where the written instruction was to stop cutting, and he describes himself as still working the figure down.

Visceral fat, on two different scales

Date	Value	Instrument	Note
7 October 2024	Level 15, area 158.4 cm2	InBody 770 bioimpedance	A level and an area, on the InBody scale. Not comparable to a mass in pounds.
July 2025	19 lb	DEXA	Visceral fat mass at the start of the first stack.
September 2026	0.53 lb	DEXA	The largest proportional change in the whole record. Visceral fat is the fat around the organs and the compartment most closely tied to metabolic risk, which is why it is worth showing separately from total body fat.

The 2024 row cannot be compared with the two below it. A bioimpedance level and a DEXA mass in pounds are different measurements on different scales, and putting them in one column would show a change that neither machine measured.

The October 2024 scan is the one I would rather leave out. It is nine months before I started and it puts me 26 lb lighter than the day I began. I was not holding steady and then deciding to fix things. I was getting worse, and it was measured at the time.

Bloodwork: February 2025 before anything, August 2025 four weeks in

Fasting insulin26.2 uIU/mLto8.8 uIU/mLreference at or below 18.4

The largest change in the file, and the one with the clearest cause. It fell 66 percent in four weeks, from above the range to comfortably inside it. Lowering fasting insulin is the measured effect of a GLP-1 receptor agonist, and retatrutide was in the stack from the first day. Of everything he took, it is the compound with the most human evidence behind it. See what has actually been measured for it.

Fasting glucose110 mg/dLto96 mg/dLreference 65 to 99

Above range to inside it, in the same four weeks and for the same reason. See what has actually been measured for it.

HbA1c5.8 percentto5.9 percentreference below 5.7

Went up slightly, which is the honest result and not a contradiction. HbA1c reflects roughly three months of blood sugar, so at four weeks it had not caught up with the insulin and glucose underneath it. Both draws sit in the prediabetic band.

Apolipoprotein B155 mg/dLnot retestedreference high

The number that actually predicts cardiovascular events, and the worst result in the baseline file. Not retested in the August draw, so there is no after.

LDL particle number2,273 nmol/Lnot retestedreference below 1,138

Double the reference ceiling. Small LDL particles were 561 against a ceiling of 142, four times over. Not retested.

Lipoprotein(a)below 10 nmol/Lnot retestedreference below 75

The one cardiovascular number he cannot change and does not need to. Lp(a) is almost entirely genetic and his is very low. Worth stating because it is a piece of good luck rather than anything he did.

Total testosterone244 ng/dLto815 ng/dL on 6 August 2026reference 250 to 1,100

Below the bottom of the range at baseline, which is what clinically low means. The later figure is on 100 mg of testosterone cypionate a week. Starting testosterone replacement from a level that low changes body composition on its own, and that is the single largest confound on this whole page. See what has actually been measured for it.

Haematocrit46.9 percentto48.6 percentreference 38.5 to 50.0

Rising, inside the range, and the expected direction on testosterone. It is the number to keep watching rather than the number to celebrate. See what has actually been measured for it.

Ferritin435 ng/mLnot retestedreference 38 to 380

Above range at baseline. Not retested.

hs-CRP1.8 mg/Lto1.8 mg/Lreference below 1.0 is optimal

Did not move at all. In a file where the metabolic markers changed sharply, the inflammation marker sat still, and it is still above the optimal threshold.

ALT23 U/Lto37 U/Lreference 9 to 46

Rose within the range. Worth a note rather than an alarm, and worth retesting.

Four of the worst numbers in the baseline, apolipoprotein B, LDL particle number, small LDL and ferritin, were never retested. They are the ones I would most want an after for and the ones I do not have. A page that showed only the markers that improved would be a different and worse page.

Screening

3 March 2025Galleri multi-cancer early detection blood test

No cancer signal detected

The report states that in its clinical study, fewer than 2 of every 100 people with this result went on to receive a cancer diagnosis. A negative result does not rule cancer out and the test does not detect all cancers.

27 June 2025Colonoscopy

Adenomatous polyps found and fully removed

The pathologist reported the specimens as adenomatous, a precancerous polyp type that can raise colon cancer risk if left in place. All were removed during the procedure.

Repeat colonoscopy in 3 years, so due June 2028.

14 July 2025CT coronary calcium score

Agatston score 16

Left main 0, left anterior descending 1, left circumflex 0, right coronary 15. Between the 25th and 50th percentile for his age and sex. The radiologist recorded coronary atherosclerosis present with a mild plaque burden and that significant flow-limiting disease was unlikely. Taken three days before he started.

The same scan carried a non-cardiac finding: probable asymmetric gynecomastia, with a small right subareolar mass recorded as felt less likely but not excluded, and a note to correlate with history and exam. That correlation had not been done at the time of writing and is being arranged.

The blend I stopped taking daily, and why

My notebook called this “NAD Blend” and left it there. It ran on most days across all three phases. It is not NAD. It is seven actives and a local anaesthetic, and it is mostly carnitine by mass. I only worked out what was in it when I went looking to write this page.

NAD+ / Carnitine Based Amino Blend, 20 ml glass vial. Per 1 ml, drawn as 100 units on a U-100 syringe:

Ingredient	Per ml	What it is
NAD+	20 mg	The compound the blend is named after, and the smallest active in it by mass after the B vitamins.
L-Carnitine	250 mg	By far the largest ingredient. The blend is mostly carnitine by mass, not NAD.
Methylcobalamin, vitamin B12	500 mcg	Excess B12 is generally excreted and a high serum level from supplementation is common and usually benign.
Pyridoxine, vitamin B6	25 mg	The one ingredient here with a dose-dependent toxicity. Excess B6 causes peripheral neuropathy, and this is many times the daily dietary requirement, given by injection so none of it is lost to the gut.
L-Inositol	50 mg	Lipotropic cofactor. Not on this site.
L-Choline bitartrate	50 mg	Methyl donor. Not on this site.
L-Methionine	10 mg	Methyl donor. Not on this site.
Lidocaine 1.5 percent	vehicle	A local anaesthetic, about 15 mg in a 1 ml dose. It is there so the injection does not sting, and it is an active drug rather than an inert carrier.
Benzyl alcohol 0.9 percent	vehicle	Preservative, the standard bacteriostatic agent.

What the labs changed

My B vitamins came back high on my most recent test. So I cut the blend from most days to once a week, and I now take 100 mg of pure NAD on four days instead. That is the only change on this page that a blood test caused rather than a plan, and it is the one I am most glad about.

Here is what I could not see while I was taking it. The blend carries 25 mg of vitamin B6 per millilitre. On the days I ran it, that was 25 mg injected, which skips the gut entirely, against a dietary requirement measured in single milligrams. B6 is the one ingredient in there with a dose-dependent toxicity: a 2025 expert consensus on its therapeutic use states that both deficiency and excess can cause complications, particularly peripheral neuropathy, and calculates a washout of 20 to 40 days for it to clear completely. PubMed 40395441.

B12 and B6 are not the same problem. Excess B12 is mostly excreted and a high reading is usually unremarkable. Excess B6 is the one with a nerve injury attached to it. A test that says “B vitamins high” without saying which one does not tell you which of those you are in, and that is the next thing I am finding out.

The general point is bigger than this vial. Two of the things I injected are still recorded in my own notes as a product name and nothing else, including one I took nightly on a rising dose. I could not have found this without reading the seller’s ingredient list, and there was no ingredient list in my notebook at all.

Part two

What I am running now

Dated 18 September 2026, with 19 days at home, then a 10 October to 20 October travel and race window. The race is Marathon des Sables, 120 km over 5 days, Sahara, self supported with a pack. The goal, written at the top of the page, is to arrive able to walk and run on consecutive days in heat with a pack, rather than arrive lighter.

This one is different from the five above it in a way worth naming. It has no outcome because it has not happened. Everything in part one is a page I chose to publish after I knew how it went. This is the page I am publishing before I know, which is the only version of this exercise that costs me anything.

What came off, which is the actual change

Every protocol on this page until now got longer than the one before it. This one is shorter. Nineteen days out from a race, the list of things I stopped matters more than the list of things I kept.

Tesamorelin and ipamorelinoff, and they had been the spine of two earlier phases
Injectable GHK-Cuoff as an injection; a topical serum stays
AOD-9604, 5-Amino-1MQ and SLU-PP-332held back unless fat loss fully stalls and energy is still high

Retatrutide stays, but as maintenance at 3 mg a week split across two injections, for appetite control rather than cutting, with an instruction not to raise it before the race. That instruction is the whole shift on this page: at 175 lb and 13 percent, being 2 lb lighter is worth less than being able to eat enough to hike.

What I am injecting and swallowing

When	What	Dose	Evidence
Weekly, split as 1.5 mg twice	Retatrutidemaintenance, for appetite control rather than cutting. The written rule is not to raise it before the race.	3 mg a weekSubQ	Strong. At least two completed trials, and at least 1,000 people given it. 5 completed trials, 1,345 people given it.
AM, training days	KLOWrepair and inflammation. The written rule is not to increase it.	25 units, 5 days a weekSubQ	Preliminary. Studied in people, but no published study has given it to anyone. 1 human study where nobody was given it, 1 registered with no result.
AM	NAD+replaced the daily NAD blend after the B vitamin result	100 mg on four daysIM	Preliminary. Studied in people, but no published study has given it to anyone. 1 human study where nobody was given it.
AM	ARA-290	not recorded on this pageSubQ	Limited. Someone has been given it and a result was published, but fewer than 100 people. 5 completed trials, 95 people given it.
AM, mid week, only if currently cycling	MOTS-c	not recorded on this pageSubQ	Limited. Someone has been given it and a result was published, but fewer than 100 people. 1 completed trial, 30 people given it.
AM only, never at night	Selankdaytime focus	10 mg in a combined 10/10 vial with semax	Moderate. At least one completed trial, and at least 100 people given it. 3 completed trials, 192 people given it.
AM only, never at night	Semax	10 mg in the same vial	Moderate. At least one completed trial, and at least 100 people given it. 4 completed trials, 354 people given it.
Night, 3 or more hours after food, or AM fasted if dinner ran late	Sermorelinthe one compound on the page with its own mixing arithmetic written beside it	200 mcg, 8 units at 2.5 mg/mlSubQ	Moderate. At least one completed trial, and at least 100 people given it. 2 completed trials, 129 people given it.
Evening, short cycle	Epithalon	conservative, not a large nightly milligram figureSubQ	Moderate. At least one completed trial, and at least 100 people given it. 4 completed trials, 266 people given it.
Evening, optional, only on a short cycle	Pinealon	not recorded on this page	Limited. Someone has been given it and a result was published, but fewer than 100 people. 2 completed trials, none stating how many people.
Night, rescue only, not nightly	DSIPsleep rescue	not recorded on this page	Limited. Someone has been given it and a result was published, but fewer than 100 people. 5 completed trials, none stating how many people.
Morning, on scheduled days	Methylene bluetaken in water with salt and lemon	not recorded on this pageoral	Preliminary. Studied in people, but no published study has given it to anyone. 3 human studies where nobody was given it.

Several doses on this sheet are written as a role and a time rather than a milligram figure, and I have transcribed them that way rather than inventing the number. Where the notebook says “AM” and nothing else, that is what it says.

The training, which is the actual protocol

45 minutes on a treadmill or a hill every day wearing a 40 lb weighted vest. The vest goes on for all cardio and all hikes, and the written reason is that the race pack will be lighter than the vest, so the vest is a deliberate overload.
15 minutes of exercise with oxygen therapy daily on the treadmill at a hard running effort, and a second 15 minute session on days he can still walk properly the next morning.
Hyperbaric oxygen three times a week.
A minimum of 5 km of running on coaching days and 10 km on non coaching days.
One long hike a week of 90 to 150 minutes or more, vest on, on terrain that looks like the race.
At least two days in a row on his feet every week, because the race is consecutive days rather than one hard day.
Legs only, 20 to 30 minutes four or five days a week: goblet and suitcase squats, hinges, step ups, split squats, reverse lunges, calf and tibialis work, hip abduction, dead bugs, side planks and single sided carries in the good hand.
Cold plunge, sauna, pulsed electromagnetic field, breath work and red and near infrared light all stay in.
No upper body loading of any kind until after 20 October, which is six months after the repair.
The taper is written as keep the vest, cut volume 20 to 30 percent from 4 October, and do not add any new overload in the last three days.

The eating, now built around hiking rather than shrinking

200 g of protein on eating days, unchanged from every phase since the injury.
Morning after work: a third of a cup of A2 yogurt, ten raspberries, a quarter cup of blueberries, a third of a cup of organic granola.
Lunch: 8 oz of steak and half an avocado. Dinner: 8 oz of filet with asparagus, or another lean protein with a vegetable. A shake if the protein comes up short.
The 48 hour fast is finished and the written rule is not to run another 36 to 48 hour fast unless a day is completely easy, which is a reversal of the February protocol where a 36 hour fast was placed every ten days on a rule.
In the last five days before departure, carbohydrate goes up around the long vest sessions so the legs are full rather than flat.
Salt and lemon daily, and more of it on double oxygen therapy days, because the vest, the sauna and the desert all point the same way.

The order of the morning

Up at about 4:00 AM, water with salt and lemon.
Morning peptides, then methylene blue with salt and lemon on scheduled days.
Breath work, then pulsed electromagnetic field.
45 minutes on the treadmill in the vest, including or followed by 15 minutes of oxygen therapy at a hard running effort.
A second oxygen session only on a double day. Then cold plunge, coffee, and the yogurt bowl.
Sauna later in the day, only if it will not ruin the next hike.
In bed about 9:30 PM, up about 4:00 AM.

What I write down every day

Bodyweight
Biceps pain, scored 1 to 10
Vest sessions completed
Oxygen therapy sessions, and whether the second one happened
200 g of protein hit, or a written miss
Feet checked every night
Every injection logged in units and time

The sauna is the part I am least sure about

My plan treats sauna after training as heat preparation for the Sahara. The design that actually tested that is weaker than I assumed. A 2025 systematic review and meta-analysis of ten studies in 199 people looked specifically at post exercise heat exposure, sauna or hot water immersion after training, against the same training without it. The effect on performance in hot conditions was trivial, a ratio of means of 1.04 with a confidence interval from 0.94 to 1.15 and a p value of 0.46, graded low to very low certainty. Sweat rate, heart rate and core temperature moved a little; performance in the heat did not. PubMed 39762944.

The distinction that matters is passive against active. The same review notes that exercising in the heat can improve performance; it is doing the training cool and the heat afterwards that showed almost nothing. If I want the adaptation rather than the ritual, the change is to do some of the vest work hot rather than to sit in the sauna after it, and I am keeping the sauna anyway because I sleep better on days I use it.

What this page cannot tell you

It is a plan, not a record. Nothing here has an outcome yet, and the three highest volume items in it, the weighted vest, the oxygen therapy and the hyperbaric sessions, are not compounds and have no profile or evidence band anywhere on this site. I am also still six months out from a tendon repair and doing a 40 lb vest every day on it, which is a loading decision about the rest of my body taken while one part of it is still healing. If any of this goes wrong, it will be on this page too.

Why “25 units” is not a dose

A unit is a volume on a syringe, not an amount of peptide. What it delivers depends entirely on how much water went into the vial. Two people following the same unit figure at different reconstitution volumes are taking different doses, and most protocols posted publicly give the units and leave the concentration out. These are the concentrations I was working from, transcribed from the same notebook page.

Compound	Concentration	Dose	Draw
Retatrutide	6 mg/ml	2 mg	33u
Tesamorelin	5 mg/ml	2 mg	40u
CJC-1295	2 mg/ml	100 mcg	5u
Ipamorelin	2.5 mg/ml	200 mcg	8u
Epitalon	5 mg/ml	1 to 2 mg	20 to 40u
MOTS-c	5 mg/ml	3.5 mg	70u
KLOW	5 mg/ml	1.25 mg	25u
NAD+ Pure	200 mg/ml	1 ml	100u, IM
NAD Blend	not recorded	0.5 to 1 ml	50 to 100u, IM or SubQ
Testosterone cypionate	200 mg/ml	100 mg	0.5 ml on a 3 ml syringe, IM
Sermorelin	2.5 mg/ml, from 5 mg reconstituted with 2 ml	200 mcg	8u, and 10u for 250 mcg, 12u for 300 mcg

The NAD Blend row is blank on purpose. My own notebook does not record its concentration, so the unit figure beside it means nothing without the vial in front of you. The reconstitution calculator does this arithmetic for any vial.

The honest summary

255 lb at 37 percent to 175 lb at 13 percent, over fourteen months. Visceral fat 19 lb to 0.53 lb. Those are the numbers, and they are real.

Five protocols over ten months, escalating until an injury stopped them. Testosterone doubled. The growth hormone secretagogue changed five times and spent two of those phases as something my own notes cannot name. The things that stayed constant longest were the oral pair, SLU-PP-332 and 5-Amino-1MQ, and both of those are the ones with no human study behind them.

What actually changed my body composition is not knowable from this page. I was in a large calorie deficit, I was running, and I started testosterone replacement from a clinically low baseline, any one of which moves body composition on its own. The peptides sat on top of all of that.

The torn tendon is the closest thing here to a natural experiment, and it still is not one. It forced the plan to reverse in writing, which is the most honest part of the whole notebook, and it produced no measurement at all. Six weeks of daily electrical stimulation on an arm I never once tested.

The part I would defend is the arithmetic: the draw volumes are real, the timings are real, and the fasting rule is the one piece here with a stated mechanism I could check against what has actually been measured. The part I would not defend is any suggestion that this is a protocol anyone else should run. In three weeks I take the current one into the Sahara, and whatever it does there will be added to this page whether it flatters it or not.