Do thyroid antibodies fluctuate, and can TPO or thyroglobulin levels change over time?

Qassi Gaba · 1 November 2026 · 5 min read

Yes. Both TPO antibodies and thyroglobulin antibodies can go up, go down, or disappear entirely over months or years, in people who are on no treatment and whose thyroid status has not changed. This is expected, and it is one reason neither test is used to track how you are doing from one appointment to the next.

If that is what you came for, you have it. The rest is why.

What these tests actually measure

TPO antibodies and thyroglobulin antibodies are antibodies your immune system makes against two proteins the thyroid itself produces, thyroid peroxidase and thyroglobulin. Their presence is evidence of autoimmune activity against the thyroid, which is the mechanism behind Hashimoto's thyroiditis. MedlinePlus describes what the test looks for, and the ATA's page on Hashimoto's thyroiditis covers how the diagnosis is made.

What the number is not is a dial for disease severity. A high titer does not mean a worse thyroid than a low one, and the titer does not reliably track with symptoms, TSH, or how much thyroid tissue has been affected. It is mostly used once, to help establish why TSH is behaving the way it is, not watched repeatedly afterward the way TSH or Free T4 are.

Why the number moves on its own

A few things drive genuine fluctuation, separate from anything your doctor is doing or not doing.

The immune response itself is not constant. Antibody-producing activity rises and falls over time in autoimmune conditions generally, and thyroid antibodies are no exception. Someone can have a high TPO antibody titer one year and a much lower one the next, with no change in medication and no change in TSH. Endotext's chapter on thyroid function testing covers the general unreliability of using antibody titer as a quantitative marker.

Pregnancy changes it too. Thyroid antibody levels typically fall during pregnancy, because the immune system is naturally suppressed to tolerate the pregnancy, and then tend to rise again afterward, sometimes substantially, in the postpartum period. Someone testing negative for TPO antibodies while pregnant and positive a year later has not developed a new disease. The underlying autoimmunity was very likely there the whole time.

Assay differences matter as much as biology. Different labs use different antibody assays, calibrated against different reference preparations, with different upper limits of normal. A titer of 60 on one lab's assay is not the same thing as a titer of 60 on another's. If you switch labs, or your doctor's practice switches labs, a number that looks like a big jump can be mostly the assay change. The ATA's page on what each thyroid test measures touches on this, and it is the same issue that affects other antibody and hormone assays in general: the number only means something relative to the reference range it came from.

And some of it is simple assay noise. Any lab test has a margin of run-to-run variability. For a test like TPO antibodies, where the clinically meaningful line is just "positive" versus "negative" rather than a precise number, that noise matters less. But if you are watching the exact figure move from 180 to 210 and back to 160, some of that movement may be the assay, not you.

Why your doctor probably is not retesting it often

Because once the diagnosis is established, the antibody result has usually done its job. NICE's clinical summary on hypothyroidism treats TSH and Free T4 as the tests used to monitor the condition and guide treatment. Antibody status is not part of that ongoing picture, because it does not change what the treatment is or how much of it you need.

This surprises people who are used to thinking of a lab report as something you track like a stock price. Thyroglobulin antibodies in particular have a second, separate use worth being clear about: in people who have had thyroid cancer and had their thyroid removed, thyroglobulin antibody levels are checked because they can interfere with thyroglobulin itself, which is used as a cancer marker in that specific situation. That is a different purpose from diagnosing Hashimoto's, and it is worth not conflating the two uses if you see both mentioned.

What a disappearing or falling titer does and does not mean

A falling TPO antibody titer over the years, especially in someone already on levothyroxine, is common and is not generally read as a sign the thyroid is recovering. The gland damage from chronic autoimmune thyroiditis is not typically reversible, and most people with an established diagnosis stay on replacement hormone regardless of what the antibody number does afterward. The NIDDK's overview of Hashimoto's disease describes the condition as a long-term autoimmune process rather than one whose antibody titer is expected to normalize.

None of this means the number is meaningless, only that it answers a different question than the one people tend to ask of it. It helped answer "is this autoimmune." It is not built to answer "is this getting better or worse this year."

What to do with a changed number

If you have had TPO or thyroglobulin antibodies tested more than once and the number moved, a few questions are worth bringing to whoever ordered it: was the same lab used both times, is there anything about timing (pregnancy, postpartum, a recent illness) that could account for it, and what, if anything, does this change about how you are being monitored. What it means for you specifically, and whether anything should change, is a question for your doctor, not something the number answers by itself.

Where this touches Haloscan

Haloscan logs lab results over time, including antibody titers, and shows them alongside which lab processed each one. The point is not to interpret the trend for you. It is so that when a number has moved, you and your doctor are not trying to reconstruct from memory whether it was the same lab, the same year, or the same you.