Checkpoint Inhibitor Thyroiditis: When Cancer Immunotherapy Affects the Thyroid

Cancer immunotherapy with PD-1 and PD-L1 inhibitors (pembrolizumab, nivolumab, atezolizumab, durvalumab) causes thyroid dysfunction in 5 to 20% of patients. The typical pattern is a short thyrotoxic phase followed by permanent hypothyroidism. Immunotherapy continues; levothyroxine treats the resulting hypothyroidism.

Why checkpoint inhibitors affect the thyroid

Immune checkpoint inhibitors (ICIs) are monoclonal antibodies that release the brakes on the immune system so it can attack cancer. The two checkpoint molecules most commonly targeted are PD-1 (on T cells) and PD-L1 (on tumor cells), with drugs like pembrolizumab, nivolumab, atezolizumab, and durvalumab. CTLA-4 inhibitors such as ipilimumab work on a different checkpoint and are sometimes combined with PD-1 blockade.

The same mechanism that unleashes anti-tumor immunity also unmasks latent autoimmunity against healthy tissue. The thyroid is one of the most frequently affected organs because it expresses antigens that T cells can recognize, has a rich blood supply, and is patrolled by autoreactive lymphocytes in many healthy adults [C1][C2]. When checkpoint signaling is blocked, these clones expand and trigger a destructive thyroiditis — follicular cells are killed, stored hormone leaks out, and the gland eventually fails [C2][C4].

Patients with a pre-existing autoimmune thyroid condition (Hashimoto's, Graves') or with detectable TPO antibodies at baseline are at substantially higher risk of developing ICI-related thyroiditis on therapy [C2][C4]. Combination regimens (PD-1 plus CTLA-4) carry the highest risk; PD-1 monotherapy is intermediate; CTLA-4 monotherapy is the lowest [C1].

How often it happens and what it looks like

The Barroso-Sousa meta-analysis (38 trials, >7,500 patients) is the canonical epidemiology source [C1]:

• Hypothyroidism occurs in roughly 6 to 7% on PD-1 monotherapy, 4% on PD-L1 monotherapy, and up to 13 to 17% on PD-1 + CTLA-4 combination therapy.
• Thyrotoxicosis (the transient hyperthyroid phase) appears in roughly 3 to 4% on PD-1 monotherapy and up to 8 to 11% on combination therapy.
• Across all regimens, clinically meaningful thyroid dysfunction affects 5 to 20% of patients, depending on the drug class and combination.

The clinical course is biphasic in most patients [C2][C4][C5]:

• Weeks 1–6 (thyrotoxic phase): painless destruction releases pre-formed hormone. TSH suppressed, free T4 and free T3 elevated. Symptoms are often mild — palpitations, heat intolerance, tremor, weight loss, anxiety — and many patients are asymptomatic and picked up only on routine labs.
• Weeks 6–16 (transition): stored hormone is depleted, the gland cannot replace it, free T4 falls.
• Months 2–6 (hypothyroid phase): TSH rises, free T4 drops, and overt hypothyroidism develops. In the majority of cases this is permanent — the destructive process leaves too few follicular cells for recovery [C2][C4].

A minority of patients (mostly thyrotoxic phase only, or subclinical changes) recover spontaneously. Painful thyroid swelling is uncommon — unlike subacute thyroiditis, ICI-thyroiditis is usually painless [C5].

What recovers on adequate levothyroxine

Once the gland fails, levothyroxine is the standard treatment, dosed exactly as for any other cause of primary hypothyroidism [C6][C8]:

• Symptoms improve in weeks. Fatigue, cold intolerance, constipation, and bradycardia respond within 2 to 6 weeks of reaching a normal TSH [C6].
• Cognitive and mood symptoms typically lag — full recovery can take 3 to 6 months [C6].
• Cancer treatment continues uninterrupted. ICI-induced hypothyroidism is not an indication to pause immunotherapy. The thyroid is replaceable; the cancer benefit from the ICI is not [C2][C3].
• TSH targets are the same as for any hypothyroid patient — typically 0.5 to 2.5 mIU/L on stable replacement, with adjustment based on symptoms [C6].

In the thyrotoxic phase, a short course of beta-blocker (propranolol or atenolol) controls tremor and palpitations until the gland transitions to hypofunction. Anti-thyroid drugs (methimazole, propylthiouracil) are not used — this is destructive thyroiditis, not Graves' hyperthyroidism, and the gland is releasing pre-formed hormone, not making new hormone [C2][C4][C5].

When thyroid dysfunction looks different

A few presentations should prompt a wider workup rather than reflex levothyroxine:

1. Central hypothyroidism with low TSH and low free T4 — this can signal ICI-induced hypophysitis (pituitary inflammation), most common with CTLA-4 inhibitors. Hypophysitis is a medical emergency because of co-existing adrenal insufficiency; cortisol replacement comes before levothyroxine [C2].
2. TSH-receptor antibody-positive thyrotoxicosis — rare but reported, and represents true ICI-triggered Graves' disease rather than destructive thyroiditis [C4]. Anti-thyroid drugs are appropriate in this scenario.
3. Persistent thyrotoxicosis past 6–8 weeks without transition to hypothyroidism — warrants antibody testing and a radioactive iodine uptake scan to distinguish destructive thyroiditis from Graves' [C4][C5].
4. Severe thyrotoxicosis with thyroid storm features — extremely rare with ICIs but requires urgent endocrinology and ICU input.
5. New diabetes, adrenal insufficiency, or hypogonadism alongside thyroid changes — ICIs can hit multiple endocrine glands, and the thyroid finding can be the first signal of a broader endocrinopathy [C2].

What does NOT help

Several approaches lack evidence in ICI-thyroiditis [C2][C3][C6]:

• Stopping immunotherapy. Outside of severe (grade 4) thyrotoxicosis, ICI-related thyroiditis does not require interrupting cancer treatment [C2][C3]. Stopping the drug does not reverse the destructive process already underway and removes a potentially life-saving therapy.
• High-dose steroids. Unlike most other immune-related adverse events, ICI-thyroiditis usually does not respond to glucocorticoids because the damage is already done by the time symptoms appear [C2][C4].
• "Thyroid support" supplements — iodine, kelp, ashwagandha, selenium megadoses — have no role and may complicate management. Excess iodine in particular can worsen autoimmune thyroid damage [C5].
• Anti-thyroid drugs (methimazole/PTU) in the destructive thyrotoxic phase — these block new hormone synthesis, but the problem is leakage of pre-formed hormone, not overproduction [C4][C5].
• Switching to T3 or desiccated thyroid extract without a specific indication. ATA guidelines recommend levothyroxine as first-line [C6].

Practical guidelines

1. Baseline labs before starting immunotherapy. TSH, free T4, and (ideally) TPO antibodies on day 1 — this defines the patient's risk and gives a comparison point [C2][C3].
2. Routine monitoring on therapy. TSH and free T4 every 4 to 6 weeks for the first 6 months, then every 12 weeks for the duration of treatment [C2][C3]. Earlier if symptoms develop.
3. Treat the thyrotoxic phase symptomatically if needed — propranolol 10–40 mg three times daily for palpitations or tremor, tapered as the gland transitions [C2][C4][C5]. No anti-thyroid drugs.
4. Start levothyroxine when TSH rises above 10 mIU/L, or 4–10 with symptoms [C6][C8]. Standard weight-based dosing: about 1.6 mcg/kg/day in otherwise healthy adults, lower starting doses (25–50 mcg) in older or cardiac patients [C6].
5. Your endocrinologist will recheck TSH 6 to 8 weeks after each dose change and titrate to a normal TSH [C6][C7].
6. Continue immunotherapy. Endocrine adverse events from ICIs are managed alongside ongoing cancer treatment; the oncologist and endocrinologist work in parallel [C2][C3].

Frequently asked questions

Will I need levothyroxine forever? In most patients, yes — ICI-induced hypothyroidism is usually permanent because the destructive process eliminates too many follicular cells for recovery [C2][C4]. A minority of patients recover thyroid function and can come off replacement, but this is the exception.

Can immunotherapy be stopped to protect my thyroid? For mild to moderate thyroid dysfunction, no — stopping the drug does not reverse damage already underway and gives up the cancer benefit [C2][C3]. Only severe, life-threatening thyrotoxicosis would prompt a treatment pause, and even that is uncommon.

Does ICI-thyroiditis mean my cancer treatment is working? There is some evidence that patients who develop immune-related adverse events — including thyroiditis — may have better tumor responses on ICIs than those who don't, because both reflect a robust immune activation. Treatment decisions are not made on that basis, but a new endocrinopathy is not a bad prognostic sign [C2].

Will high-dose steroids help my thyroid? No. Unlike colitis or pneumonitis from ICIs, thyroiditis usually does not respond to glucocorticoids — the damage to follicular cells is done by the time labs change [C2][C4]. Levothyroxine is the durable fix.

Can I prevent ICI-thyroiditis if I have Hashimoto's already? Pre-existing Hashimoto's increases the risk but is not a contraindication to immunotherapy [C2][C4]. Closer baseline monitoring (TSH every 3 to 4 weeks early on) is reasonable. Most patients with Hashimoto's tolerate ICIs well and any worsening hypothyroidism is managed by adjusting the levothyroxine dose.

Bottom line

Checkpoint inhibitor thyroiditis is one of the most common endocrine immune-related adverse events of modern cancer immunotherapy, affecting 5 to 20% of patients on PD-1 or PD-L1 blockade [C1]. The typical pattern is a short, mild thyrotoxic phase followed within months by permanent hypothyroidism [C2][C4][C5]. The management is straightforward: monitor TSH every 4 to 6 weeks on therapy, treat the hyperthyroid phase with a beta-blocker if symptomatic, and start levothyroxine once the gland fails [C6][C8]. Immunotherapy continues uninterrupted in nearly all cases — the thyroid is replaceable, the cancer benefit is not [C2][C3].