GLP-1 blood work: which labs, how often, and what moves

There is no single "GLP-1 panel". The prescribing information for Wegovy and Zepbound names a few situations that call for testing, such as checking blood glucose before starting in people with diabetes and checking kidney function in anyone with severe vomiting or diarrhoea. Everything else on the request form comes from the prescriber, the reason for the drug, and the conditions you already have. What follows is the common shape of it, and the trial numbers behind it.

Before the first dose

A baseline panel does two jobs. It finds anything that changes the plan, such as undiagnosed diabetes or reduced kidney function. And it gives you a starting point to compare against later, which is the part people skip and then regret.

Two tests people ask about are deliberately not on that list. Calcitonin: the Wegovy label says routine monitoring of serum calcitonin or thyroid ultrasound is "of uncertain value" for early detection of medullary thyroid cancer. Lipase and amylase: these rise on the drugs in people without pancreatitis, so a baseline value without symptoms tends to create questions rather than answer them. They are checked when there is abdominal pain.

What the trials saw change

The labels report cardiometabolic results from the pivotal trials. These are intention-to-treat figures with missing data imputed, and the lipid and blood pressure results were secondary measures not controlled for multiple testing, so read them as direction and size rather than proof.

The pattern is consistent. Triglycerides fall the most, LDL moves a little, HDL rises a little, and HbA1c drifts down even in people without diabetes. In people with type 2 diabetes the HbA1c change is much larger: in SURMOUNT-2, the Zepbound label reports a fall of 2.1 points on tirzepatide against 0.5 on placebo from a baseline of 8.0%.

Inflammation. C-reactive protein was a prespecified secondary endpoint in STEP 1 to 3. At week 68, semaglutide 2.4 mg lowered CRP by an estimated 44% relative to placebo in STEP 1 (95% CI, 39 to 49%), 39% in STEP 2 and 48% in STEP 3. The fall tracked weight loss.

Pancreatic enzymes. In the weight trials, the Wegovy label reports mean rises of 15 to 16% in amylase and 39% in lipase, not seen on placebo. The Zepbound label reports 20 to 25% for pancreatic amylase and 28 to 35% for lipase, against 2.1% and 5.8% on placebo. Both labels say the significance of these rises is unknown without other signs of pancreatitis. A raised lipase on a routine panel, with no pain, is a conversation, not an emergency.

Things that can confuse a result

• Creatinine and muscle. eGFR is usually calculated from creatinine, which partly reflects muscle mass. Large weight loss changes body composition, so an eGFR shift can reflect muscle as well as kidneys. Some labs offer cystatin C as a cross-check. The muscle loss page covers the body composition side.
• Dehydration. A week of poor intake after a dose step can push creatinine up temporarily. Note it in the log next to the draw.
• Oral semaglutide and levothyroxine. In a drug interaction study with the semaglutide tablet, levothyroxine exposure rose 33%. The label asks prescribers to consider extra monitoring for drugs like this, which is the main reason TSH comes up.
• Different labs. Reference ranges and methods differ. The lab report guide explains why a change of lab is a break in the series.

How often

For people with diabetes, the ADA's 2026 Standards recommend assessing glycaemic status at least twice a year, and more often, for example every three months, after a treatment change or when not at goal. Starting a GLP-1 is a treatment change. For weight management without diabetes there is no published fixed interval, and your prescriber sets it. HbA1c reflects roughly the previous two to three months of glucose, so a retest much sooner than that mostly measures the time before the drug.

Reading labs against the dose

A lab value without a date on the dose timeline is half a data point. A triglyceride result drawn during week two of a new step, after a week of eating very little, is not the same measurement as one drawn at steady state on a maintenance dose. The useful habit is to record the dose, the week of the step, the weight and anything unusual about the days before each draw, then plot the markers against that line. The trials above are averages. Your own series is the one that answers your question. The peptide blood testing page covers the general rules for timing and reference ranges, and the SELECT page and kidney study page cover the outcome trials behind these markers.

In Peptide AI, the dose log, titration steps, side effects and weight are free to log. Premium adds bloodwork: upload a lab PDF and the markers land on the same timeline as every dose, so a lipid panel at week 30 sits next to the step-up at week 17 rather than in a separate folder. The free GLP-1 tracker page has the detail, and the app is on the download page.