Beyond a Cold Nodule: Medullary Thyroid Cancer and Calcitonin

Medically reviewed by Dr. Michael Kachur · Frankfurt, Germany··

Thyroid nodules are common, and when one turns out to be cancer, it is usually papillary. So the workup has a default shape: ultrasound, a needle, cytology, and a plan built around a cancer that soaks up iodine and sheds thyroglobulin into the blood. Medullary thyroid cancer breaks every part of that shape. It starts in a different cell, ignores iodine, and is followed by a hormone most nodule workups never order, which is exactly the kind of mismatch Healz's memory is built to notice.

Category
Health Insights
Date
Reading time
11 min
Share

Why a thyroid nodule gets read against the wrong cancer

Beyond a Cold Nodule: Medullary Thyroid Cancer and Calcitonin

Start with the scan that gives this post its title. On a thyroid uptake scan, a cold nodule is one that does not take up the tracer, meaning it is not making thyroid hormone the way the tissue around it is. That sounds specific and is not: it separates the hormonally busy nodules from the quiet ones, and nothing more. Quiet nodules are the ones that get an ultrasound and, at 1 cm or larger, a fine needle aspiration (ATA 2015).

That pathway is sound and built for volume. Most nodules are benign, and most nodule cancers are papillary, so the questions asked along the way are papillary questions. Medullary thyroid cancer accounts for only about 1% to 2% of thyroid cancers in the United States by SEER data, lower than the 3% to 5% figure often quoted (ATA 2015), and a cancer that rare does not get to set the defaults.

Ask Healz.

One chat instead of ten apps. 1M+ rare cases checked. The root cause, not the label. Private.

The problem is that it does not sit at the rare end of the same spectrum. It is a different disease wearing the same clothes, and almost every tool the standard workup reaches for was designed for the other one. Our piece on how a thyroid nodule is worked up covers the ordinary path; this one is about where that path stops answering the question.

The cell type that changes every rule

The thyroid holds two populations of cells with two unrelated jobs. The follicular cells trap iodine, build thyroid hormone and manufacture a protein called thyroglobulin along the way. Scattered between them are the parafollicular cells, also called C cells, which are neuroendocrine and make calcitonin, a hormone involved in keeping blood calcium in a healthy range. Medullary thyroid cancer is a tumour of those C cells (NCI).

Three consequences fall straight out of that one fact.

The first is iodine. C cells do not handle iodine the way follicular cells do, so the tumour does not concentrate it. This is why the National Cancer Institute states plainly that radioactive iodine therapy is not used for medullary thyroid cancer, and why the American Thyroid Association grades it as not indicated after thyroidectomy, citing a series of 293 patients in which disease-specific survival was no different either way (ATA 2015). The exception is a mixed tumour carrying papillary or follicular cancer alongside.

The second is thyroglobulin. It is a follicular product, so for papillary and follicular thyroid cancer it tracks whether disease has come back. For a C cell tumour it reports on the wrong cells, which is why calcitonin and CEA take its place (ATA 2015).

The third is TSH. After surgery for the common thyroid cancers the hormone dose is often set high enough to push TSH down, because TSH drives follicular tissue. The guideline language here is unusually direct: there is no reason to suppress thyrotropin, since this is not a follicular neoplasm. Levothyroxine is dosed to keep TSH in the normal range (ATA 2015).

What calcitonin and CEA actually measure

Calcitonin is useful here for a mundane reason. The tumour is made of the cells that produce it, so the blood level roughly tracks how much tumour there is. The American Thyroid Association recommends measuring basal calcitonin together with carcinoembryonic antigen, CEA, rather than either alone (ATA 2015).

Read together, the pair carries information neither carries alone. Before surgery, the calcitonin level helps estimate how far disease has gone into the neck lymph nodes, which shapes how wide the operation needs to be. A CEA that is markedly high against a lower calcitonin points toward a poorly differentiated tumour, because cells that have drifted from their C cell identity make less of the hormone (ATA 2015).

Two honest limits belong next to that. Neither marker is a screening test, and CEA is not specific to this cancer. Calcitonin can read falsely high or low in a range of other conditions (ATA 2015). A single number out of context is a reason to look again, not a conclusion.

That is also why measuring calcitonin in every thyroid nodule is still an open question: the 2015 guidelines decline to recommend either for or against it, while European guidance has leaned toward including it.

When the needle comes back with the wrong answer

Here is the number that surprises people. In a meta-analysis of six studies that ran both tests on the same histologically proven cases, fine needle aspiration cytology pooled a sensitivity of 54% for medullary thyroid cancer, 95% confidence interval 35% to 73%. Calcitonin measured in the needle washout fluid pooled 98% (Endocrine, 2022). The authors concluded the washout measurement should be used alongside cytology where this cancer is suspected.

Cytology struggles because the cells do not look like what the pathologist is pattern-matching against: this tumour can resemble a follicular lesion, a plasmacytoma or a sarcoma, and amyloid is often present without being diagnostic. Accuracy improves when immunohistochemistry and washout calcitonin are added (ATA 2015), and none of that happens automatically. It happens when someone suspects the diagnosis and asks for it.

So the practical shape of the miss is not a wrong answer on the slide. It is an oddly worded cytology report filed as indeterminate, while the one finding that would have redirected the workup, a calcitonin level, was never ordered. Reading that report closely enough to notice it described something unusual is the kind of job an ai lab report reader is good at, because it never skims.

Why a gene test follows the diagnosis

Medullary thyroid cancer is the one thyroid cancer where the diagnosis routinely triggers a genetic test, and the reason is the RET proto-oncogene. Inherited changes in RET drive the hereditary forms in an autosomal dominant pattern, so one altered copy is enough and an affected parent passes it on half the time (MedlinePlus Genetics).

The guideline position is firm: patients with apparently sporadic medullary thyroid cancer should have germline RET testing, because between 1% and 7% of them carry a germline mutation (ATA 2015). No known family history is not evidence of no inherited cause, partly because small families hide the pattern.

What the test opens up is multiple endocrine neoplasia type 2, which affects roughly 1 in 35,000 people and has medullary thyroid carcinoma as its most common feature. Type 2A, the most common subtype, can also bring pheochromocytoma and hyperparathyroidism, while familial medullary thyroid carcinoma is the form in which the thyroid tumour is the only described feature (MedlinePlus Genetics). Our explainer on multiple endocrine neoplasia covers the syndrome itself.

Two things change once a germline mutation is on the table. Pheochromocytoma and hyperparathyroidism are excluded before the thyroid operation goes ahead, because an undiagnosed adrenal tumour changes the anaesthetic risk. And the mutation stops being one person's result: relatives can be tested, and carriers are managed on a schedule set by which RET change they carry, from thyroidectomy in the first year of life for the highest-risk M918T pattern to surveillance timed largely by calcitonin for moderate-risk changes (ATA 2015). That is cancer prevention running on a gene result.

What treatment and follow-up look like

The treatment is surgical. Total thyroidectomy with dissection of the cervical lymph node compartments, guided by the calcitonin level and the ultrasound, is the standard operation, and the extent of the node dissection is where most of the judgement sits (ATA 2015). The National Cancer Institute describes the same approach, with external radiation for recurrence in the thyroid bed and tyrosine kinase inhibitors such as vandetanib, cabozantinib and selpercatinib for locally advanced or metastatic disease.

Follow-up then runs on the two markers. Calcitonin and CEA are checked about 3 months after surgery, and an undetectable basal calcitonin at that point is considered biochemical cure, although roughly 3% of those patients still develop biochemical recurrence within 7.5 years. A calcitonin under 150 pg/mL prompts neck examination and ultrasound; above 150 pg/mL, broader imaging. Where the markers remain detectable, they are measured at least every 6 months (ATA 2015).

The reason for that cadence is doubling time. In the series the guidelines cite, 5 year and 10 year survival were 25% and 8% when calcitonin doubling time was under 6 months, against 92% and 37% between 6 and 24 months, and everyone over 24 months was alive at the end of the study. Doubling time outperformed every other variable. But it is a slope, not a value, and a reliable estimate needs at least four data points over at least 2 years (ATA 2015).

That is the quiet operational problem here. The single most informative number in the follow-up of this cancer does not exist on any one lab report. It only exists across several of them, in order, with their dates intact.

How Healz reads a calcitonin number against the rest of the case

A calcitonin result means almost nothing on its own and a great deal next to three other things: the nodule it came from, what relatives have had, and what the previous values were. Healz is equipped with memory, so it pulls the medical facts out of every upload, keeps abnormal values up front, dates each finding and connects them across months. The calcitonin from this spring sits beside the one from last year and the CEA beside both, which is how a doubling time becomes something you can see.

The rest of the case lives in the same chat. Frontier AI works the whole file, the ultrasound report, the cytology, the markers and the operative note together, rather than one page at a time. Healz is equipped with root-cause technology, so it does not stop at the first label, asks what a rushed visit skips, and cross-checks the case against 1M+ rare cases, which is how a cancer that is 1% to 2% of thyroid cancers stops being treated as a footnote. One chat, not ten apps and three patient portals. You can bring a board-certified doctor into the same chat for a second opinion.

Frequently asked questions

Does medullary thyroid cancer show up on a thyroid scan?

Not usefully. Because the tumour comes from C cells, it does not take up the tracer, so it reads as a cold nodule, a finding shared by plenty of benign nodules. Ultrasound, a needle sample and calcitonin are what answer the question (ATA 2015).

Why is radioactive iodine not used for medullary thyroid cancer?

Because the cancer cannot concentrate iodine. Radioiodine works on follicular-derived thyroid tissue, and C cell tumours are not that. The National Cancer Institute states it is not used for medullary thyroid cancer, and the American Thyroid Association grades it as not indicated after surgery, citing a 293-patient series with no survival difference (ATA 2015).

Can a needle biopsy miss medullary thyroid cancer?

Often enough to matter. Pooled across six studies, cytology alone had a sensitivity of 54%, while measuring calcitonin in the needle washout fluid reached 98% (Endocrine, 2022). The cells can mimic follicular lesions, plasmacytomas or sarcomas.

Does everyone with medullary thyroid cancer need genetic testing?

Guidelines say yes, including when the case looks sporadic, because 1% to 7% of presumed sporadic patients turn out to carry an inherited RET mutation (ATA 2015). A positive result changes screening for relatives, not just care for the person tested.

What does a rising calcitonin level after surgery mean?

It means the markers are being tracked for a trend rather than a threshold. The rate of rise, expressed as doubling time, carried more prognostic weight than any other variable in the series the guidelines cite (ATA 2015). A reliable doubling time needs at least four values over at least 2 years, so one number in isolation does not supply it.

Most thyroid nodules really are read correctly against the common cancers, and the uncommon one is found when somebody notices the cell type, the marker and the family history do not fit that story, which is a noticing problem before it is a medical one.

Written by Healz Team · Filed under Health Insights

Your health deserves more than guesswork

Get a real investigation, not a symptom list. Root causes found, history remembered, dots connected.

Not a medical provider. Healz.ai is an educational service. Always discuss every decision with your doctor. AI can make mistakes.

© 2026 HealzAI Inc1111B S Governors Ave #92123, Dover, DE 19904, USAMade with for better health