The bloodwork worth running before and during a GLP-1 protocol

A mid-protocol lab value means almost nothing without the pre-protocol value it can be compared against. That is not a slogan — it is how every GLP-1 trial was built: blood drawn before the first dose, then on a fixed schedule against that baseline. This article maps what the trial protocols for semaglutide and tirzepatide actually measured, what the FDA labels tell prescribers to watch, and why each marker earned its place. It is a map, not an order sheet: which panels apply to a given person is a decision for the clinician supervising that protocol.

For the efficacy side of the record — STEP, SURMOUNT, and the head-to-head — see semaglutide vs tirzepatide.

The panel, at a glance

Two more things the trials tracked that aren't blood at all — resting heart rate and gallbladder symptoms — get their own section below.

Glycemic markers: the original endpoint

Semaglutide and tirzepatide were diabetes drugs before they were weight drugs, and A1c was the primary endpoint of the SUSTAIN and SURPASS programs. In SURPASS-2 (NEJM 2021), tirzepatide reduced A1c by 2.01 to 2.30 percentage points depending on dose, against 1.86 points for semaglutide 1 mg — effects large enough that the trials monitored glycemia for overshoot, not just progress.

The glycemic story holds in non-diabetic populations too. In SURMOUNT-1 (NEJM 2022), more than 95% of tirzepatide participants who entered with prediabetes reverted to normoglycemia by 72 weeks, against about 62% on placebo. Without a baseline A1c and fasting glucose, that entire axis of the drug's effect is invisible — a reversion can't be documented if the starting point was never measured.

The one glycemic scenario the labels actively warn about is hypoglycemia when a GLP-1 is combined with insulin or an insulin secretagogue. That is a prescriber-managed interaction; the labels put dose-adjustment decisions there.

Lipase and amylase

Both trial programs measured pancreatic enzymes on schedule, and both labels carry pancreatitis warnings: acute pancreatitis has been reported with GLP-1 receptor agonists, and the labels instruct discontinuation if pancreatitis is suspected. Trial protocols routinely excluded people with a history of pancreatitis.

The labels also report a subtler finding: mean increases in serum lipase and amylase in treated patients who never develop symptoms, with the clinical significance of an isolated asymptomatic elevation described as unknown. That cuts both ways. A mildly elevated week-20 lipase is not, by itself, a diagnosis — and a week-20 lipase with no week-0 value next to it is nearly uninterpretable, because some people sit above the reference range before ever taking anything.

Liver enzymes

The expected direction on a GLP-1 is down. Liver fat and ALT tend to improve with substantial weight loss, and both compounds have been formally tested in steatohepatitis: semaglutide in a phase 2 NASH trial (Newsome et al., NEJM 2021) and the phase 3 ESSENCE trial (NEJM 2025), tirzepatide in the phase 2 SYNERGY-NASH trial (NEJM 2024) — with resolution of steatohepatitis exceeding placebo in each.

A baseline ALT and AST does two jobs. It captures the common case of an elevated starting value — fatty liver is prevalent in exactly the population these drugs treat — and it turns any later reading into a trend rather than an isolated flag.

The lipid panel

Both programs reported lipid changes as secondary endpoints: triglycerides fell, and HDL and non-HDL cholesterol improved modestly alongside the weight loss. Tirzepatide's GIP component is thought to contribute additional effects on lipid handling — one of the mechanistic arguments for the dual agonist. A standard lipid panel at baseline is what makes any of that visible in an individual rather than in a trial average.

Kidney function

Both labels warn about acute kidney injury — not from direct renal toxicity, but secondary to dehydration when nausea, vomiting, or diarrhea cut fluid intake, most plausibly during titration. A baseline creatinine gives any later value its context, and the warning is one of the practical reasons the GI-heavy escalation weeks deserve more attention rather than less.

The boxed warning is a history question, not a lab test

Both labels open with the same boxed warning: in rodents, these compounds caused dose-dependent thyroid C-cell tumors, and whether that finding is relevant to humans has not been determined. Both drugs are contraindicated in people with a personal or family history of medullary thyroid carcinoma or Multiple Endocrine Neoplasia syndrome type 2.

Note what the labels do not say: they do not call for routine calcitonin testing. Both state that the value of routine serum calcitonin monitoring or thyroid ultrasound for early MTC detection is uncertain. The screen the labels actually define happens before the first dose, in the family-history conversation — worth naming in a bloodwork article precisely because the most important pre-protocol check isn't a blood draw.

Watched, but not in the blood: gallbladder and heart rate

Gallbladder events appeared in the trial data for both drugs — in STEP 1 (NEJM 2021), cholelithiasis was reported in 2.6% of the semaglutide arm versus 1.2% on placebo — and both labels list acute gallbladder disease among their warnings. Substantial or rapid weight loss raises gallstone risk on its own, which is part of why the signal tracks the drugs that produce the most of it. Monitoring here is symptomatic — right-upper-quadrant pain, fever, jaundice — not a scheduled lab.

Resting heart rate rose by a mean of 1 to 4 beats per minute in the semaglutide trial program, a figure stated on the Wegovy label, which tells prescribers to monitor heart rate at routine visits; small dose-dependent increases were observed with tirzepatide as well. This is the one marker on the list a wearable already tracks continuously — a pre-protocol resting-heart-rate baseline makes a small drift visible instead of deniable.

Why the baseline is the whole game

Run the logic backward from any mid-protocol result. An "elevated" lipase is a different conversation when the baseline already sat at the top of the range. A falling ALT is only a win if you know where it started. An A1c of 5.6 means nothing until it sits next to the 6.1 it came from. The trials measured before dosing because unbaselined data is close to worthless — and that logic doesn't change at n=1.

This is the model PeptideLab is built around: bloodwork stored against the protocol timeline, so a week-20 value renders next to its week-0 baseline and the titration steps between them. The baseline is also the cheap part of the whole endeavor — the lab price comparison tracks what common panels cost across direct-to-consumer labs, and most of the markers above appear on inexpensive standard panels.

What this list is not

It is not an order sheet, and it is deliberately not a recommendation of which tests any individual should run. The labels place monitoring decisions with prescribers, and trial protocols are built for populations, not for one person. What the list is: the documented answer to "what did the people who studied these drugs most carefully decide to watch" — which is the right starting point for a conversation with whoever supervises your protocol.

Sources

• STEP 1 — Wilding et al., New England Journal of Medicine, 2021.
• SURMOUNT-1 — Jastreboff et al., New England Journal of Medicine, 2022.
• SURPASS-2 — Frías et al., New England Journal of Medicine, 2021.
• Semaglutide in NASH, phase 2 — Newsome et al., New England Journal of Medicine, 2021.
• ESSENCE (semaglutide in MASH, phase 3) — New England Journal of Medicine, 2025.
• SYNERGY-NASH (tirzepatide in MASH, phase 2) — Loomba et al., New England Journal of Medicine, 2024.
• Wegovy and Ozempic (semaglutide) prescribing information — U.S. FDA labels, Novo Nordisk.
• Zepbound and Mounjaro (tirzepatide) prescribing information — U.S. FDA labels, Eli Lilly.