The Deficit Beat the Peptides

Seven days before I injected anything, my insulin-like growth factor 1 was already over the line. IGF-1 is the hormone that growth-hormone drugs exist to raise, and mine measured 251.30 against a reference ceiling of 230. Above range, unmedicated, on a baseline panel drawn specifically to find out what I was starting from.

Then I spent ninety-eight days injecting CJC-1295 and ipamorelin five nights a week. Both are growth-hormone secretagogues. Both exist to push that exact number up.

It came back at 214.70. Inside range. Down fifteen percent.

That is the single most useful result in three panels of blood, and not because it is good news. It is useful because it is the opposite of what the mechanism predicts, which means something bigger was operating on that number than the drugs I was taking to move it.

Three panels, and why there are three

The short version. The first panel is a photograph of a body before anything was done to it. The second is what that body looked like after three months of an eight-compound protocol, drawn with no deliberate pre-draw washout, because the point was to see the physiology as it actually runs rather than an artificial drug-free version of it. TB-500 was the one exception, and only because it had run out. The third exists because the second one left out most of the questions I cared about.

The protocol, so the numbers have a context: testosterone cypionate 100 mg a week, split into two 50 mg doses on Monday and Thursday since late June. Human chorionic gonadotropin at 2,500 IU a week. Retatrutide, a triple agonist hitting the GLP-1, GIP and glucagon receptors, at 0.5 mg weekly for the first month then 1 mg. BPC-157. TB-500, paused since 23 July on a supply gap. CJC-1295 at 200 micrograms with ipamorelin at 200 micrograms, five nights a week. NAD+ at 100 mg, five times a week.

Running underneath all of it: a sustained caloric deficit, a ketogenic diet, and a gym block. Bodyweight went from about 110 kg to 105.5 kg.

That combination is the honest caveat on everything below. This is not an experiment that can assign any single change to any single compound. It is an on-treatment read of a whole system moving at once.

The headline: the deficit beat the peptides

(251.30 − 214.70) / 251.30 × 100 = 14.6% fall

Baseline:  251.30  ██████████████████████░  +21.30 over the ceiling
Day 98:    214.70  ███████████████████░░░░  15.30 under the ceiling
Ceiling:   230.00

The intuition that growth-hormone secretagogues raise IGF-1 is correct. It just lost to something stronger.

IGF-1 is produced mainly by the liver and is dominated by nutritional status. A sustained caloric deficit suppresses hepatic IGF-1 output and shifts the binding proteins that carry it. In this body, over this window, the deficit was the larger lever and it won by enough to drag an above-range value back inside.

This is not evidence that CJC-1295 and ipamorelin lower IGF-1. It is evidence that they were outgunned by something. And to be strict: a deficit, a diet, a training block and eight compounds all moved at once here, so this establishes which direction the number went, not proof of which lever moved it. The mechanism is well described — nutritional status dominates liver IGF-1 output — but a single panel cannot settle attribution on its own. Which has a direct and uncomfortable implication: when the cut ends and the peptides continue, the force currently holding this number down goes away.

The metabolic axis, which is the unambiguous win

If the protocol had one primary job, this was it. Baseline glycated haemoglobin, the three-month average of blood sugar, sat at 5.7 percent. That is not a near miss. It is the exact number where the prediabetic band starts.

Fasting insulin:  (8.7 − 4.5) / 8.7   × 100 = 48.3%
HOMA-IR:          (1.78 − 0.90) / 1.78 × 100 = 49.4%
Triglycerides:    (148 − 87) / 148     × 100 = 41.2%

Glucose was normal at baseline and is still normal, so on its own that line looks like nothing happened. The pairing is what matters: glucose went down while the insulin needed to hold it there nearly halved. That is the difference between a body that is managing and a body that is not having to work at it.

Out of the prediabetic band, insulin resistance halved, triglycerides down 41 percent. That is what a triple incretin agonist plus a sustained deficit is for, and it delivered.

Testosterone, oestrogen and the ratio that actually matters

Sex hormone binding globulin, the protein that carries testosterone around and controls how much of it is actually free to act, rose 54 percent. The timing detail on testosterone matters more than the number: 712.60 was drawn roughly three days after a dose, which on a twice-weekly split is a true trough. That is the floor of the week, not the peak.

Oestradiol is the result most likely to be mishandled by someone reading a single line. At 43.84 against a ceiling of 44, it is at 99.6 percent of the male reference limit, and the reflex is to reach for an aromatase inhibitor.

Ceiling check:   43.84 / 44     × 100 = 99.6% of the male reference limit
The ratio:       712.60 / 43.84       = 16.25  →  roughly 16:1 testosterone to oestradiol

The cost side, stated plainly

The pituitary-gonadal axis is suppressed past the point where the assay can see it. Luteinising hormone came back below what the machine can measure at all. That is not the same as zero, but it is as far down as this test can report.

This is the designed consequence of exogenous testosterone, not a malfunction. The body detects plenty of circulating testosterone and stops asking for more. The human chorionic gonadotropin in the protocol substitutes the downstream signal at the testis, which is why it is in there at all, but it does not restore pituitary output and was never going to.

Red cells and iron: the one genuinely open risk

Individually these markers each look like a shrug. Together they are one coherent process, and it is the only thing in the dataset I would call an active risk rather than a resolved question.

Haemoglobin:             17.0 − 15.7                    = +1.3 g/dL
Ferritin:                (129.05 − 78.38) / 129.05 ×100 = −39.3%
Transferrin saturation:  (42.03 − 20.87) / 42.03  ×100 = −50.3%

Read as one story: testosterone is stimulating red cell production, exactly as it is known to. Stores are being drawn down to supply it. Circulating iron has fallen to the floor and transferrin, the iron transport protein, is upregulating to scavenge harder. The new cells are coming out slightly smaller and more variable in size.

That pattern is what iron-restricted erythropoiesis looks like: the factory running faster than the raw material arrives. It is the working interpretation rather than a settled one, and the tests that would settle it are the ones still outstanding. It is emphatically not anaemia, because haemoglobin went up, not down.

One honest complication, and it is the reason I will not call this purely protocol-driven: mean corpuscular volume was already below range at baseline, at 85.3, before anything was injected. A constitutional microcytosis is still on the table, and in that scenario iron supplementation would achieve nothing. The drawdown during treatment is real either way; the starting point just was not clean.

Lipids: six improvements and one regression

Arterial index, checked by hand:
  Baseline:  185 / 40 = 4.63   (reported as 4.6)
  Day 94:    158 / 39 = 4.05   (reported as 4.1)

Five markers improved substantially and one crossed below its floor. High-density lipoprotein going from 40 to 39 is a small absolute move but it is the wrong direction, it put the number below range, and a fall in HDL is a known direction of travel on testosterone. Aggregate improvement does not erase the one line that got worse.

There is also a limit to what this panel can claim. LDL and non-HDL measure cholesterol carried, not the number of atherogenic particles carrying it. Apolipoprotein B counts the particles, and it is one of the six tests I am still missing.

Kidney, liver, inflammation

Creatinine is above range and the estimated filtration rate dropped from 80 to 74. Before anyone panics about kidneys: I supplement creatine, which reliably inflates serum creatinine by 0.1 to 0.3 mg/dL with zero change in actual filtration, and I have trained consistently for three months, which raises creatinine production. My rise was 0.09. That is textbook.

But it is now two consecutive panels drifting the same direction, and it has crossed the limit. The point is not that the kidney is fine. The point is that this number is structurally incapable of telling me either way, because both confounders were documented in writing before the draw. It needs cystatin C, which estimates filtration independently of muscle mass and creatine. That test is pending.

Ninety-eight days of peptides ran through this liver and left no mark on it. AST rose from 18 to 28 while ALT stayed flat at 21, which moves the AST to ALT ratio from 0.95 to 1.33 with a normal ALT. That pattern points at skeletal muscle, not liver. I train. AST is abundant in muscle. Non-finding, correctly interpreted.

High-sensitivity C-reactive protein, the systemic inflammation marker, fell 64 percent from 0.28 to 0.10. That does a second job beyond being good news: ferritin is an acute-phase reactant, meaning inflammation inflates it. With inflammation this low, the ferritin of 78.38 is a real number, which means the iron drawdown above is real too.

Thyroid, vitamin D, homocysteine

Free T3 was the specific worry going in. It is the thyroid hormone that falls first when a sustained caloric deficit starts braking metabolism, and a hard cut running alongside a six-day training week is exactly the setup where you would expect to see it. It went up. Prediction closed, no thyroid problem in this dataset.

Homocysteine moved the wrong way, from 9.21 to 10.25. It is comfortably mid-range and there is no threshold breach, but it is a real change in an unhelpful direction and it belongs on the record rather than in a footnote.

Prolactin deserves a line of its own here. It was 16.07 at baseline, which is 83 percent of its ceiling, and the specific concern was that CJC-1295 and ipamorelin can push it higher. It came back at 9.22. Down 43 percent. Worry closed.

White cells and platelets

Nothing here. Total white count essentially flat, the differential redistributed modestly, platelets rose slightly and stayed well inside range. Mean platelet volume was below range at baseline and fell a little further, which with a normal platelet count supports no conclusion at all. I include it because the claim of this article is completeness, and a section with nothing in it is still a result.

The heart, electrically

One new finding: an isolated T-wave inversion in lead III, with T waves still positive in leads II and aVF and a QRS-T angle of 21 degrees. That combination fits a change in the heart's physical position in the chest, which is the kind of thing that follows a body losing 4.5 kg, rather than a regional problem. The rSr' pattern in the same lead was already there at baseline and is not new.

The blind spot is not electrical, it is biochemical. Potassium and magnesium both move QTc, and neither has been measured on any panel, ever. The trace is normal so this is not urgent. It is just the missing denominator, and it is cheap to add to a draw that has already been paid for.

What the third panel actually settled

The day-94 draw left ten open questions. Here is the scorecard four days later.

Seven of ten closed in four days, by a draw that only happened because someone wrote down the questions the first panel had not asked.

The three-way ledger

The six I am still waiting on

The clinic confirmed today that these are outstanding and is chasing the lab. Five of the six I requested myself; only cystatin C came from the clinic's own order.

A panel is a question set

The prescription for this protocol specifies, in writing, what should be run during treatment: IGF-1, HbA1c, insulin, TSH, free T4, vitamin D, among others. The day-94 draw ran none of those six.

They exist only because a second draw was arranged four days later, and five of the extra tests on it were on a sheet I printed myself and handed to the nurse at the door. Without that, the most interesting finding in this entire dataset — an above-range IGF-1 falling while the drugs that raise it were running — would simply not be known. Neither would the exit from prediabetes.

This is not a complaint about anyone. It is the structural observation that matters most if you are running something like this on yourself: a blood panel is a question set, and if you do not write the questions, you get someone else's defaults run on your blood. The defaults are not wrong. They are just not yours.

The same limit applies to reading the results. Luteinising hormone cannot answer fertility. Creatinine cannot settle filtration in someone taking creatine. LDL cannot count particles. A normal QTc cannot reveal an unmeasured potassium. Ferritin cannot show iron delivery into cells that are being built today. Every one of those is a question the panel looks like it answers and does not.

Six things this dataset taught me

This is the interim read

Ninety-eight days in: the metabolic objective is met and not narrowly. IGF-1 is back inside range. Inflammation is down by nearly two thirds. Testosterone is where it should be, measured at its weekly floor. The liver is untouched, the conduction intervals have not moved, and the thyroid never flinched.

The costs are equally concrete and I am not going to soften them: the pituitary axis is switched off, red cell production is outrunning iron supply with no agreed threshold for when that becomes a problem, HDL has crossed below its floor, and there is a kidney marker sitting above range that nobody can currently interpret.

The trigger for the final version of this article is specific: all six pending results in hand, cystatin C first. Then one more IGF-1, drawn after the deficit ends, because that is the number whose current value belongs to the diet rather than the protocol, and the diet is the part that is going to change.